Congress took the Register's advice. In enacting the Copyright Act of 1976, Congress dropped the reference to "all the writings of an author" and replaced it with the phrase "original works of authorship." 17 U. S. C. § 102(a). In making explicit the originality requirement, Congress announced that it was merely clarifying existing law: "The two fundamental criteria of copyright protection [are] originality and fixation in tangible form. . . . The phrase 'original works of authorship,' which is purposely left undefined, is intended to incorporate without change the standard of originality established by the courts under the present copyright statute." H. R. Rep. No. 94-1476, p. 51 (1976) (emphasis added) (hereinafter H. R. Rep.); S. Rep. No. 94-473, p. 50 (1975) (emphasis added) (hereinafter S. Rep.). This sentiment was echoed by the Copyright Office: "Our intention here is to maintain the established standards of originality. . . ." Supplementary Report of the Register of Copyrights on the General Revision of U.S. Copyright Law, 89th Cong., 1st Sess., pt. 6, p. 3 (H. Judiciary Comm. Print 1965) (emphasis added). To ensure that the mistakes of the "sweat of the brow" courts would not be repeated, Congress took additional measures. For example, § 3 of the 1909 Act had stated that copyright protected only the "copyrightable component parts" of a work, but had not identified originality as the basis for distinguishing those component parts that were copyrightable from those that were not. The 1976 Act deleted this section and replaced it with § 102(b), which identifies specifically those elements of a work for which copyright is not available: " In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work." Section 102(b) is universally understood to prohibit any copyright in facts. Harper & Row, supra, at 547, 556. Accord, Nimmer § 2.03[E] (equating facts with "discoveries"). As with § 102(a), Congress emphasized that § 102(b) did not change the law, but merely clarified it: "Section 102(b) in no way enlarges or contracts the scope of copyright protection under the present law. Its purpose is to restate . . . that the basic dichotomy between expression and idea remains unchanged." H. R. Rep., at 57; S. Rep., at 54. Congress took another step to minimize confusion by deleting the specific mention of "directories . . . and other compilations" in § 5 of the 1909 Act. As mentioned, this section had led some courts to conclude that directories were copyrightable per se and that every element of a directory was protected. In its place, Congress enacted two new provisions. First, to make clear that compilations were not copyrightable per se, Congress provided a definition of the term "compilation." Second, to make clear that the copyright in a compilation did not extend to the facts themselves, Congress enacted § 103.

This case concerns the interaction of two well-established propositions. The first is that facts are not copyrightable; the other, that compilations of facts generally are. Each of these propositions possesses an impeccable pedigree. That there can be no valid copyright in facts is universally understood. The most fundamental axiom of copyright law is that "no author may copyright his ideas or the facts he narrates." Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 556 (1985). Rural wisely concedes this point, noting in its brief that "facts and discoveries, of course, are not themselves subject to copyright protection." Brief for Respondent 24. At the same time, however, it is beyond dispute that compilations of facts are within the subject matter of copyright. Compilations were expressly mentioned in the Copyright Act of 1909, and again in the Copyright Act of 1976. The key to resolving the tension lies in understanding why facts are not copyrightable. The sine qua non of copyright is originality. To qualify for copyright protection, a work must be original to the author. See Harper & Row, supra, at 547-549. Original, as the term is used in copyright, means only that the work was independently created by the author (as opposed to copied from other works), and that it possesses at least some minimal degree of creativity. 1 M. Nimmer & D. Nimmer, Copyright §§ 2.01[A], [B] (1990) (hereinafter Nimmer). To be sure, the requisite level of creativity is extremely low; even a slight amount will suffice. The vast majority of works make the grade quite easily, as they possess some creative spark, "no matter how crude, humble or obvious" it might be. Id., § 1.08[C]. Originality does not signify novelty; a work may be original even though it closely resembles other works so long as the similarity is fortuitous, not the result of copying. To illustrate, assume that two poets, each ignorant of the other, compose identical poems. Neither work is novel, yet both are original and, hence, copyrightable. See Sheldon v. Metro-Goldwyn Pictures Corp., 81 F. 2d 49, 54 (CA2 1936).

Factual compilations, on the other hand, may possess the requisite originality. The compilation author typically chooses which facts to include, in what order to place them, and how to arrange the collected data so that they may be used effectively by readers. These choices as to selection and arrangement, so long as they are made independently by the compiler and entail a minimal degree of creativity, are sufficiently original that Congress may protect such compilations through the copyright laws. Nimmer §§ 2.11[D], 3.03; Denicola 523, n. 38. Thus, even a directory that contains absolutely no protectible written expression, only facts, meets the constitutional minimum for copyright protection if it features an original selection or arrangement. See Harper & Row, 471 U.S., at 547. Accord, Nimmer § 3.03.

This protection is subject to an important limitation. The mere fact that a work is copyrighted does not mean that every element of the work may be protected. Originality remains the sine qua non of copyright; accordingly, copyright protection may extend only to those components of a work that are original to the author. Patterson & Joyce 800-802; Ginsburg, Creation and Commercial Value: Copyright Protection of Works of Information, 90 Colum. L. Rev. 1865, 1868, and n. 12 (1990) (hereinafter Ginsburg). Thus, if the compilation author clothes facts with an original collocation of words, he or she may be able to claim a copyright in this written expression. Others may copy the underlying facts from the publication, but not the precise words used to present them. In Harper & Row, for example, we explained that President Ford could not prevent others from copying bare historical facts from his autobiography, see 471 U.S., at 556-557, but that he could prevent others from copying his "subjective descriptions and portraits of public figures." Id., at 563. Where the compilation author adds no written expression but rather lets the facts speak for themselves, the expressive element is more elusive. The only conceivable expression is the manner in which the compiler has selected and arranged the facts. Thus, if the selection and arrangement are original, these elements of the work are eligible for copyright protection. See Patry, Copyright in Compilations of Facts (or Why the "White Pages" Are Not Copyrightable), 12 Com. & Law 37, 64 (Dec. 1990) (hereinafter Patry). No matter how original the format, however, the facts themselves do not become original through association. See Patterson & Joyce 776. This inevitably means that the copyright in a factual compilation is thin. Notwithstanding a valid copyright, a subsequent compiler remains free to use the facts contained in another's publication to aid in preparing a competing work, so long as the competing work does not feature the same selection and arrangement. As one commentator explains it: "No matter how much original authorship the work displays, the facts and ideas it exposes are free for the taking. . . . The very same facts and ideas may be divorced from the context imposed by the author, and restated or reshuffled by second comers, even if the author was the first to discover the facts or to propose the ideas." Ginsburg 1868.

We think that Whelan's approach to separating idea from expression in computer programs relies too heavily on metaphysical distinctions and does not place enough emphasis on practical considerations. As discussed herein, we think that district courts would be well-advised to undertake a three-step procedure, based on the abstractions test utilized by the district court, in order to determine whether the non-literal elements of two or more computer programs are substantially similar. This approach breaks no new ground; rather, it draws on such familiar copyright doctrines as merger, scenes a faire, and public domain. In taking this approach, however, we are cognizant that computer technology is a dynamic field which can quickly outpace judicial decisionmaking. Thus, in cases where the technology in question does not allow for a literal application of the procedure we outline below, our opinion should not be read to foreclose the district courts of our circuit from utilizing a modified version. I.E. let's do what nimmer tells us. Step One: Abstraction Initially, in a manner that resembles reverse engineering on a theoretical plane, a court should dissect the allegedly copied program's structure and isolate each level of abstraction contained within it. This process begins with the code and ends with an articulation of the program's ultimate function. Along the way, it is necessary essentially to retrace and map each of the designer's steps--in the opposite order in which they were taken during the program's creation. See Background: Computer Program Design, supra. Step Two: Filtration (a) Elements Dictated by Efficiency Efficiency is an industry-wide goal. Since, as we have already noted, there may be only a limited number of efficient implementations for any given program task, it is quite possible that multiple programmers, working independently, will design the identical method employed in the allegedly infringed work. Of course, if this is the case, there is no copyright infringement. See Roth Greeting Cards v. United Card Co., 429 F.2d 1106, 1110 (9th Cir. 1970); Sheldon, 81 F.2d at 54. (b) Elements Dictated By External Factors This process entails examining the structural components at each level of abstraction to determine whether their particular inclusion at that level was "idea" or was dictated by considerations of efficiency, so as to be necessarily incidental to that idea; required by factors external to the program itself; or taken from the public domain and hence is nonprotectable expression. The subject of this litigation originates with one of CA's marketed programs entitled CA-SCHEDULER. CA-SCHEDULER is a job scheduling program designed for IBM mainframe computers. Its primary functions are straightforward: to create a schedule specifying when the computer should run various tasks, and then to control the computer as it executes the schedule. CA-SCHEDULER contains a sub-program entitled ADAPTER, also developed by CA. ADAPTER is not an independently marketed product of CA; it is a wholly integrated component of CA-SCHEDULER and has no capacity for independent use. (c) Elements taken From the Public Domain Closely related to the non-protectability of scenes a faire, is material found in the public domain Step Three: Comparison The third and final step of the test for substantial similarity that we believe appropriate for non-literal program components entails a comparison. Once a court has sifted out all elements of the allegedly infringed program which are "ideas" or are dictated by efficiency or external factors, or taken from the public domain, there may remain a core of protectable expression. In terms of a work's copyright value, this is the golden nugget. Nevertheless, ADAPTER plays an extremely important role. It is an "operating system compatibility component," which means, roughly speaking, it serves as a translator. An "operating system" is itself a program that manages the resources of the computer, allocating those resources to other programs as needed. The IBM System 370 family of computers, for which CA-SCHEDULER was created, is, depending upon the computer's size, designed to contain one of three operating systems: DOS/VSE, MVS, or CMS. As the district court noted, the general rule is that "a program written for one operating system, e.g., DOS/VSE, will not, without modification, run under another operating system such as MVS." Computer Assocs., 775 F. Supp. at 550. ADAPTER's function is to translate the language of a given program into the particular language that the computer's own operating system can understand.

Whelan has fared even more poorly in the academic community, where its standard for distinguishing idea from expression has been widely criticized for being conceptually overbroad. See, e.g., Englund, at 881; Menell, at 1074, 1082; Kretschmer, at 837-39; Spivack, at 747-55; Thomas M. Gage, Note, Whelan Associates v. Jaslow Dental Laboratories: Copyright Protection for Computer Software Structure--What's the Purpose?, 1987 WIS. L. REV. 859, 860-61 (1987). The leading commentator in the field has stated that, "the crucial flaw in [Whelan's] reasoning is that it assumes that only one 'idea,' in copyright law terms, underlies any computer program, and that once a separable idea can be identified, everything else must be expression." The line between idea and expression may be drawn with reference to the end sought to be achieved by the work in question. In other words, the purpose or function of a utilitarian work would be the work's idea, and everything that is not necessary to that purpose or function would be part of the expression of the idea. . . . Where there are various means of achieving the desired purpose, then the particular means chosen is not necessary to the purpose; hence, there is expression, not idea. This criticism focuses not upon the program's ultimate purpose but upon the reality of its structural design. As we have already noted, a computer program's ultimate function or purpose is the composite result of interacting subroutines. Since each subroutine is itself a program, and thus, may be said to have its own "idea," Whelan's general formulation that a program's overall purpose equates with the program's idea is descriptively inadequate.

As discussed earlier, however, the originality requirement is not particularly stringent. A compiler may settle upon a selection or arrangement that others have used; novelty is not required. Originality requires only that the author make the selection or arrangement independently (i. e., without copying that selection or arrangement from another work), and that it display some minimal level of creativity. Presumably, the vast majority of compilations will pass this test, but not all will. There remains a narrow category of works in which the creative spark is utterly lacking or so trivial as to be virtually nonexistent. See generally Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251 (1903) (referring to "the narrowest and most obvious limits"). Such works are incapable of sustaining a valid copyright. Nimmer § 2.01[B].

The Supreme Court has said that fair use is a mixed question of law and fact. Harper & Row, 471 U.S. at 560 (citing Pac. & S. Co. v. Duncan, 744 F.2d 1490, 1495 n.8 (11th Cir. 1984)). Merely characterizing an issue as a mixed question of law and fact does not dictate the applicable standard of review, however. See U.S. Bank Nat’l Ass’n ex rel. CWCapital Asset Mgmt. LLC, No. 15-1509, 2018 WL 1143822, at *5 (U.S. Mar. 5, 2018). That said, the Supreme Court has never clarified whether and to what extent the jury is to play a role in the fair use analysis. Harper & Row involved an appeal from a bench trial where the district court concluded that the use of the copyrighted material was not a fair use. Harper & Row Publishers, Inc. v. Nation Enters., 723 F.2d 195, 199 (2d Cir. 1983). The Court, thus, had no reason to discuss a jury determination of fair use and has not since taken an opportunity to do so.

For these reasons, we conclude that whether the court applied the correct legal standard to the fair use inquiry is a question we review de novo, whether the findings relating to any relevant historical facts were correct are questions which we review with deference, and whether the use at issue is ultimately a fair one is something we also review de novo.

Importantly, the Court noted that “[m]ixed questions are not all alike.” Id. The Court then held that “the standard of review for a mixed question all depends—on whether answering it entails primarily legal or factual work.” Id. Where applying the law to the historical facts “involves developing auxiliary legal principles of use in other cases—appellate courts should typically review a decision de novo.” Id. (citing Salve Regina College v. Russell, 499 U.S. 225, 231-33 (1991)). But where the mixed question requires immersion in case-specific factual issues that are so narrow as to “utterly resist generalization,” the mixed question review is to be deferential. Id. (quoting Pierce v. Underwood, 487 U.S. 552, 561-62 (1988)). Ultimately, the Court found that review of the mixed question at issue in that bankruptcy context should be deferential because de novo review of the question would do little to “clarify legal principles or provide guidance to other courts resolving other disputes.” Id. at *7. While this may be the first time the Supreme Court has so clearly explained how appellate courts are to analyze mixed questions of law and fact, it is not the first time the Supreme Court has told us how to analyze the particular mixed question of law and fact at issue here. In other words, while the Supreme Court has not previously broken the fair use inquiry into its three analytical components as expressly as it did the question in U.S. Bank, it has made clear that both the first and third of those components are subject to de novo review. In Harper & Row, the Court explained that, “[w]here the district court has found facts sufficient to evaluate each of the statutory factors, an appellate court ‘need not remand for further factfinding but may conclude as a matter of law that the challenged use does not qualify as a fair use of the copyrighted work.’” 471 U.S. at 560 (quoting Pac. & S. Co., 744 F.2d at 1495) (internal alterations omitted)). The Ninth Circuit has resolved the question in the same way. Where fair use is resolved on summary judgment, the Ninth Circuit reviews the district court’s ultimate determination de novo. SOFA Entm’t, Inc. v. Dodger Prods., Inc., 709 F.3d 1273, 1277 (9th Cir. 2013) (“Whether Dodger’s use of the clip constitutes fair use is a mixed question of law and fact that we review de novo.”). That court has explained that, “as fair use is a mixed question of fact and law, so long as the record is ‘sufficient to evaluate each of the statutory factors,’ we may reweigh on appeal the inferences to be drawn from that record.’” Mattel Inc. v. Walking Mountain Prods., 353 F.3d 792, 800 (9th Cir. 2003) (quoting L.A. News Serv. v. CBS Broad., Inc., 305 F.3d 924, 942 (9th Cir. 2002)). This treatment of the ultimate question posed when a fair use defense is raised makes sense. The fair use question entails, in the words of U.S. Bank, a primarily legal exercise. It requires a court to assess the inferences to be drawn from the historical facts found in light of the legal standards outlined in the statute and relevant case law and to determine what conclusion those inferences dictate. Because, as noted below, the historical facts in a fair use inquiry are generally few, generally similar from case to case, and rarely debated, resolution of what any set of facts means to the fair use determination definitely does not “resist generalization.” See U.S. Bank, 2018 WL 1143822, at *5. Instead, the exercise of assessing whether a use is fair in one case will help guide resolution of that question in all future cases. For these reasons, we conclude that whether the court applied the correct legal standard to the fair use inquiry is a question we review de novo, whether the findings relating to any relevant historical facts were correct are questions which we review with deference, and whether the use at issue is ultimately a fair one is something we also review de novo.

All jury findings relating to fair use other than its implied findings of historical fact must, under governing Supreme Court and Ninth Circuit case law, be viewed as advisory only. Accordingly, while we might assess the jury's role in the assessment of fair use differently if not bound by Ninth Circuit law, we proceed on the assumption both that: (1) it was not error to send the question to the jury, because the Ninth Circuit has at least implicitly endorsed doing so; and (2) we must assess all inferences to be drawn from the historical facts found by the jury and the ultimate question of fair use de novo, because the Ninth Circuit has explicitly said we must do so.

Before turning to a consideration of the four statutory factors and any relevant underlying factual determinations, we first address the standard of review we are to employ in that consideration. While this section of most appellate opinions presents easily resolvable questions, like much else in the fair use context, that is not completely the case here. There are several components to this inquiry. First, which aspects of the fair use determination are legal in nature and which are factual? Particularly, is the ultimate question of fair use a legal inquiry which is to be reviewed de novo? Second, what factual questions are involved in the fair use determination and under what standard are those determinations to be reviewed? Finally, though neither party addresses the question in detail, we consider what, if any, aspects of the fair use determination are for the jury to decide.

Despite this case law, all aspects of Google’s fair use defense went to the jury with neither party arguing that it should not. Thus, the jury was asked not just what the historical facts were, but what the implications of those facts were for the fair use defense. During the first appeal, Google argued to this court that there were disputed issues of material historical fact relevant to its fair use defense.

As counsel for Oracle noted at oral argument, this is similar to the standard we apply in obviousness cases. Oral Argument at 9:34-10:24. Because obviousness is a mixed question of law and fact, we “first presume that the jury resolved the underlying factual disputes in favor of the verdict and leave those presumed findings undisturbed if they are supported by substantial evidence. Then we examine the [ultimate] legal conclusion [of obviousness] de novo to see whether it is correct in light of the presumed jury fact findings.” Kinetic Concepts, Inc. v. Smith & Nephew, Inc., 688 F.3d 1342, 1356-57 (Fed. Cir. 2012) (quoting Jurgens v. McKasy, 927 F.2d 1552, 1557 (Fed. Cir. 1991)). Likewise, Google cited our decision in Kinetic Concepts for the proposition that we must “presume that the jury made all findings in support of the verdict that are supported by substantial evidence.” Cross-Appellant Br. 35.

We have outlined the legal standard governing fair use above. We consider below whether the court properly applied those standards in the course of its fair use analysis and whether it reached the correct legal conclusion with respect to fair use. Before doing so, we briefly discuss the historical facts relevant to the fair use inquiry and consider the jury’s role in determining those facts. The Supreme Court has described “historical facts” as “a recital of external events.” Thompson v. Keohane, 516 U.S. 99, 110 (1995); see also U.S. Bank, 2018 WL 1143822, at *4 (describing the historical facts at issue there as facts relating to “the attributes of a particular relationship or the circumstances and terms of a prior transaction”). In the fair use context, historical facts include the “origin, history, content, and defendant’s use” of the copyrighted work. Fitzgerald v. CBS Broad., Inc., 491 F. Supp. 2d 177, 184 (D. Mass. 2007); see also Lotus Dev. Corp. v. Borland Int’l, Inc., 788 F. Supp. 78, 95 (D. Mass 1992) (defining historical facts to include “who did what, where, and when”). When asked at oral argument to identify historical facts relevant to the fair use inquiry, counsel for Oracle agreed that they are the “who, what, where, when, how, [and] how much.” Oral Arg. at 3:28- 54, available at http://oralarguments.cafc.uscourts.gov/default.aspx?fl=20 17-1118.mp3. Google did not dispute this characterization. This is, in part, because, in most fair use cases, defendants concede that they have used the copyrighted work, and “there is rarely dispute over the history, content, or origin of the copyrighted work.” See Ned Snow, Judges Playing Jury: Constitutional Conflicts in Deciding Fair Use on Summary Judgment, 44 U.C. Davis L. Rev. 483, 493 (2010). While some courts once treated the entire question of fair use as factual, and, thus, a question to be sent to the jury, that is not the modern view.3 Since Harper & Row, the Ninth Circuit has described fair use as an “equitable defense.” Fisher v. Dees, 794 F.2d 432, 435 (9th Cir. 1986) (“The fair-use doctrine was initially developed by courts as an equitable defense to copyright infringement.”). Indeed, the Supreme Court referred to fair use as “an equitable rule of reason” in Harper & Row. 471 U.S. at 560. Congress did the same when it codified the doctrine of fair use in 1976. See H.R. Rep. No. 94-1476, 94th Cong., 2d Sess. 65-66 (1976), U.S. Code Cong. & Admin. News 1976, 5659, 5679-80 (“[S]ince the doctrine [of fair use] is an equitable rule of reason, no generally applicable definition is possible, and each case raising the question must be decided on its own facts . . . .”). If fair use is equitable in nature, it would seem to be a question for the judge, not the jury, to decide, even when there are factual disputes regarding its application. See Granite State Ins. Co. v. Smart Modular Techs., Inc., 76 F.3d 1023, 1027 (9th Cir. 1996) (“A litigant is not entitled to have a jury resolve a disputed affirmative defense if the defense is equitable in nature.”). In that instance, it would be the judge’s factual determinations that would receive a deferential review—being assessed for clear error on the record before the court. That said, the Supreme Court has never clarified whether and to what extent the jury is to play a role in the fair use analysis. Harper & Row involved an appeal from a bench trial where the district court concluded that the use of the copyrighted material was not a fair use. Harper & Row Publishers, Inc. v. Nation Enters., 723 F.2d 195, 199 (2d Cir. 1983). The Court, thus, had no reason to discuss a jury determination of fair use and has not since taken an opportunity to do so. Perhaps because of this silence, even after Harper & Row, several courts—including the Ninth Circuit—have continued to accept the fact that the question of fair use may go to a jury, albeit without analysis of why it may. Compaq Comput. Corp. v. Ergonome Inc., 387 F.3d 403, 411 (5th Cir. 2004) (“The evidence presented at trial and the reasonable inferences therefrom, when viewed through the lens of the statutory fair use factors, support the jury’s fair use finding.”); Jartech, Inc. v. Clancy, 666 F.2d 403, 407-08 (9th Cir. 1982) (concluding that substan tial evidence supported the jury’s verdict on fair use); Fiset v. Sayles, No. 90-16548, 1992 WL 110263, at *4 (9th Cir. May 22, 1992) (finding that a reasonable jury could have concluded that “the evidence supporting fair use was not substantial”); see also BUC Int’l Corp. v. Int’l Yacht Council, 489 F.3d 1129, 1137 (11th Cir. 2007) (noting that the fair use defense went to the jury); N.Y. Univ. v. Planet Earth Found., 163 F. App’x 13, 14 (2d Cir. 2005) (“As to the copyright infringement claim, the evidence also supports the jury’s finding of fair use, under the four-factored analysis prescribed by statute.”). The Ninth Circuit has clarified, however, that the jury role in this context is limited to determining disputed “historical facts,” not the inferences or conclusions to be drawn from those facts. See Fisher, 794 F.2d at 436. In Fisher, for example, the court explained that “[n]o material historical facts are at issue in this case. The parties dispute only the ultimate conclusions to be drawn from the admitted facts. Because, under Harper & Row, these judgments are legal in nature, we can make them without usurping the function of the jury.” Id.; see also Seltzer v. Green Day, Inc., 725 F.3d 1170, 1175 (9th Cir. 2013) (“As in Fisher, ‘[n]o material historical facts are at issue in this case. The parties dispute only the ultimate conclusion to be drawn from the admitted facts.’” (citing Fisher, 794 F.2d at 436)); Hustler Magazine, Inc. v. Moral Majority, Inc., 606 F. Supp. 1526, 1532 (C.D. Cal. 1985) (noting that “fair use normally is a question of fact for the jury,” but concluding that “the issue of fair use, at least in the context of this case, presents primarily a question of law”). Accordingly, while inferences from the four-factor analysis and the ultimate question of fair use are “legal in nature,” in the Ninth Circuit, disputed historical facts represent questions for the jury. Fisher, 794 F.2d at 436. Where there are no disputed material historical facts, fair use can be decided by the court alone. Id. Despite this case law, all aspects of Google’s fair use defense went to the jury with neither party arguing that it should not. Thus, the jury was asked not just what the historical facts were, but what the implications of those facts were for the fair use defense. During the first appeal, Google argued to this court that there were disputed issues of material historical fact relevant to its fair use defense. As discussed below, the parties stipulated—or at least ceased to dispute—some of those facts, and presented the remaining disputed historical facts to the jury on remand. The jury returned a verdict in favor of Google on its fair use defense. Because the verdict form—though captioned as a “special verdict”—did not ask the jury to articulate its fact findings in any detail, we must assume that the jury resolved all factual issues relating to the historical facts in favor of the verdict.4 Despite the posture of the fair use finding, we must break that finding into its constituent parts. We must then review the subsidiary and controverted findings of historical fact for substantial evidence. See Seltzer, 725 F.3d at 1175; see also Brewer v. Hustler Magazine, Inc., 749 F.2d 527, 528 (9th Cir. 1984) (“We may disturb a jury verdict only if the evidence was insufficient as a matter of law.”).

All jury findings relating to fair use other than its implied findings of historical fact must, under governing Supreme Court and Ninth Circuit case law, be viewed as advisory only.

All jury findings relating to fair use other than its implied findings of historical fact must, under governing Supreme Court and Ninth Circuit case law, be viewed as advisory only. Accordingly, while we might assess the jury’s role in the assessment of fair use differently if not bound by Ninth Circuit law, we proceed on the assumption both that: (1) it was not error to send the question to the jury, because the Ninth Circuit has at least implicitly endorsed doing so; and (2) we must assess all inferences to be drawn from the historical facts found by the jury and the ultimate question of fair use de novo, because the Ninth Circuit has explicitly said we must do so. The parties have identified the following historical facts relating to Google’s use of the copyrighted work: • the history and origin of the copyrighted work, including what declaring code is; • how much of the copyrighted work was copied; • whether there were other ways to write the API packages; • whether the copied material was used for the same purpose as in the original work; • whether the use was commercial in nature; • whether Google acted in bad faith in copying the work; • whether there are functional aspects to the copyrighted work that make it less deserving of protection; and • whether there was harm to the actual or potential markets for the copyrighted work. The parties now agree on the resolution of the first four factual questions: (1) what the declaring code is and what it does in Java SE and Android, and that the code at issue was a work created by Oracle; (2) how many lines of code were copied; (3) that there were other ways for Google to write API packages; and (4) that Google used the API packages in Android for the same purpose they were created for in Java. The parties dispute, however, the remaining historical facts they identified. We address those disputes in the context of our assessment of the statutory factors to which the respective historical fact is relevant.

In Sony, the court found that the defendant’s reverse engineering and intermediate copying of Sony’s copyrighted software system “was a fair use for the purpose of gaining access to the unprotected elements of Sony’s software.” 203 F.3d at 602. The court explained that Sony’s software program contained unprotected functional elements and that the defendant could only access those elements through reverse engineering. Id. at 603. The defendant used that information to create a software program that let consumers play games designed for Sony’s PlayStation console on their computers. The court found that the defendant’s use was only “modestly transformative” where: (1) the defendant created “a wholly new product” with “entirely new . . . code,” and (2) the intermediate copying was performed to “produce a product that would be compatible.” Id. at 606- 07. As Oracle points out, even the “modest” level of transformation at issue in Sony is more transformative than what Google did here: copy code verbatim to attract programmers to Google’s “new and incompatible platform.” Appellant Response Br. 21.

In any event, moving material to a new context is not transformative in and of itself—even if it is a “sharply different context.” TCA Television Corp. v. McCollum, 839 F.3d 168, 181-83 (2d Cir. 2016) (finding that use “at some length, almost verbatim,” of the copyrighted comedy routine “Who’s on First?” in a dramatic play was not transformative where the play neither “imbued the Routine with any new expression, meaning, or message,” nor added “any new dramatic purpose”). As previously explained, a use becomes transformative only if it serves a different purpose or alters the “expression, meaning, or message” of the original work. Kelly, 336 F.3d at 818. As such, “[c]ourts have been reluctant to find fair use when an original work is merely retransmitted in a different medium.” A&M Records, 239 F.3d at 1015. Accordingly, although a change of format may be “useful,” it “is not technically a transformation.” Infinity Broad., 150 F.3d at 108 n.2 (finding that retransmitting copyrighted radio transmissions over telephone lines was not transformative because there was no new expression, meaning, or message).

In evaluating the “purpose and character” factor, the Ninth Circuit applies “the general rule that a party claiming fair use must act in a manner generally compatible with principles of good faith and fair dealing.” Perfect 10, 508 F.3d at 1164 n.8 (citing Harper & Row, 471 U.S. at 562-63). In part, this is based on the fact that, in Harper & Row, the Supreme Court expressly stated that “[f]air use presupposes ‘good faith’ and ‘fair dealing.’” 471 U.S. at 562 (citation omitted). It is also in part true because, as the Ninth Circuit has said, one who acts in bad faith should be barred from invoking the equitable defense of fair use. Fisher, 794 F.2d at 436 (calling the principle of considering the alleged infringer’s “bad conduct” as a “bar [to] his use of the equitable defense of fair use” a sound one).9

In view of the foregoing, this order concludes that our immediate case is controlled by these principles of copyright law: •Under the merger doctrine, when there is only one (or only a few) ways to express something, then no one can claim ownership of such expression by copyright. • Under the names doctrine, names and short phrases are not copyrightable. • Under Section 102(b), copyright protection never extends to any idea, procedure, process, system, method of operation or concept regardless of its form. Functional elements essential for interoperability are not copyrightable. • Under Feist, we should not yield to the temptation to find copyrightability merely to reward an investment made in a body of intellectual property.

But the names are more than just names — they are symbols in a command structure wherein the commands take the form java.package.Class.method()

. . .

Each package was broken into classes and those in turn broken into methods. For example, java.lang (a package) included Math (a class) which in turn included max (a method) to return the greater of two inputs, which was (and remains) callable as java.lang.Math.max with appropriate arguments (inputs) in the precise form required (see the example above).

. . .

This leads to the first holding central to this order and it concerns the method level. The reader will remember that a method is like a subroutine and over six thousand are in play in this proceeding. As long as the specific code written to implement a method is different, anyone is free under the Copyright Act to write his or her own method to carry out exactly the same function or specification of any and all methods used in the Java API. Contrary to Oracle, copyright law does not confer ownership over any and all ways to implement a function or specification, no matter how creative the copyrighted implementation or specification may be. The Act confers ownership only over the specific way in which the author wrote out his version. Others are free to write their own implementation to accomplish the identical function, for, importantly, ideas, concepts and functions cannot be monopolized by copyright. To return to our example, one method in the Java API carries out the function of comparing two numbers and returning the greater. Google — and everyone else in the world —was and remains free to write its own code to carry out the identical function so long as the implementing code in the method body is different from the copyrighted implementation. This is a simple example, but even if a method resembles higher mathematics, everyone is still free to try their hand at writing a different implementation, meaning that they are free to use the same inputs to derive the same outputs (while throwing the same exceptions) so long as the implementation in between is their own. The House Report, quoted above, stated in 1976 that "the actual processes or methods embodied in the program are not within the scope of the copyright law." H.R. REP. NO. 94-1476, at 57 (1976). Much of Oracle's evidence at trial went to show that the design of methods in an API was a creative endeavor. Of course, that is true. Inventing a new method to deliver a new output can be creative, even inventive, including the choices of inputs needed and outputs returned. The same is true for classes. But such inventions — at the concept and functionality level —are protectable only under the Patent Act. The Patent and Trademark Office examines such inventions for validity and if the patent is allowed, it lasts for twenty years. Based on a single implementation, Oracle would bypass this entire patent scheme and claim ownership over any and all ways to carry out methods for 95 years — without any vetting by the Copyright Office of the type required for patents. This order holds that, under the Copyright Act, no matter how creative or imaginative a Java method specification may be, the entire world is entitled to use the same method specification (inputs, outputs, parameters) so long as the line-by-line implementations are different. To repeat the Second Circuit's phrasing, "there might be a myriad of ways in which a programmer may . . . express the idea embodied in a given subroutine." Computer Associates, 982 F.2d at 708. The method specification is the idea. The method implementation is the expression. No one may monopolize the idea. 8 To carry out any given function, way to write something, the merger doctrine bars anyone from claiming exclusive copyright ownership of that expression. Therefore, there can be no copyright violation in using the identical declarations. Nor can there be any copyright violation due to the name given to the method (or to the arguments), for under the law, names and short phrases cannot be copyrighted. In sum, Google and the public were and remain free to write their own implementations to carry out exactly the same functions of all methods in question, using exactly the same method specifications and names. Therefore, at the method level — the level where the heavy lifting is done — Google has violated no copyright, it being undisputed that Google's implementations are different.

As for classes, the rules of the language likewise insist on giving names to classes and the rules insist on strict syntax and punctuation in the lines of code that declare a class. As with methods, for any desired functionality, the declaration line will always read the same (otherwise the functionality would be different) — save only for the name, which cannot be claimed by copyright. Therefore, under the law, the declaration line cannot be protected by copyright. This analysis is parallel to the analysis for methods. This now accounts for virtually all of the three percent of similar code.

To start with a clear-cut rule, names, titles and short phrases are not copyrightable, according to the United States Copyright Office, whose rule thereon states as follows: Copyright law does not protect names, titles, or short phrases or expressions. Even if a name, title, or short phrase is novel or distinctive or lends itself to a play on words, it cannot be protected by copyright. The Copyright Office cannot register claims to exclusive rights in brief combinations of words such as: • Names of products or services. • Names of business organizations, or groups (including the names of performing groups). • Pseudonyms of individuals (including pen or stage names). • Titles of works. • Catchwords, catchphrases, mottoes, slogans, or short advertising expressions. • Listings of ingredients, as in recipes, labels, or formulas. When a recipe or formula is accompanied by an explanation or directions, the text directions may be copyrightable, but the recipe or formula U.S. Copyright Office, Circular 34; see 37 C.F.R. 202.1(a). This rule is followed in the Ninth Circuit. Sega Enters., Ltd. v. Accolade, Inc., 977 F.2d 1510, 1524 n.7 (9th Cir. 1992). This has relevance to Oracle's claim of copyright ownership over names of methods, classes and packages.

Even so, the second major copyright question is whether Google was and remains free to group its methods in the same way as in Java, that is, to organize its Android methods under the same class and package scheme as in Java. For example, the Math classes in both systems have a method that returns a cosine, another method that returns the larger of two numbers, and yet another method that returns logarithmic values, and so on. As Oracle notes, the rules of Java did not insist that these methods be grouped together in any particular class. Google could have placed its trigonometric function (or any other function) under a class other than Math class. Oracle is entirely correct that the rules of the Java language did not require that the same grouping pattern (or even that they be grouped at all, for each method could have been placed in a stand-alone class). Oracle's best argument, therefore, is that while no single name is copyrightable, Java's overall system of organized names — covering 37 packages, with over six hundred classes, with over six thousand methods — is a "taxonomy" and, therefore, copyrightable under American Dental Association v. Delta Dental Plans Association, 126 F.3d 977 (7th Cir. 1997). There was nothing in the rules of the Java language that required that Google replicate the same groupings even if Google was free to replicate the same functionality. The main answer to this argument is that while the overall scheme of file name organization resembles a taxonomy, it is also a command structure for a system or method of operation of the application programming interface.

Turning now to the more difficult question, this trial showcases a distinction between copyright protection and patent protection. It is an important distinction, for copyright exclusivity lasts 95 years whereas patent exclusivity lasts twenty years. . . . In this connection, since the CONTU report was issued in 1980, the number of software patents in force in the United States has dramatically increased from barely a thousand in 1980 to hundreds of thousands today. . . . As software patents gain increasingly broad protection, whatever reasons there once were for broad copyright protection of computer programs disappear. Much of what has been considered the copyrightable "structure, sequence and organization" of a computer program will become a mere incident to the patentable idea of the program or of one of its potentially patentable subroutines. Mark Lemley, Convergence in the Law of Software Copyright?, 10 HIGH TECHNOLOGY LAW JOURNAL 1, 26-27 (1995). . . . But it is nevertheless a command structure, a system or method of operation — a long hierarchy of over six thousand commands to carry out pre-assigned functions. For that reason, it cannot receive copyright protection — patent protection perhaps — but not copyright protection.

Each command calls into action a pre-assigned function. The overall name tree, of course, has creative elements but it is also a precise command structure — a utilitarian and functional set of symbols, each to carry out a pre-assigned function. This command structure is a system or method of operation under Section 102(b) of the Copyright Act and, therefore, cannot be copyrighted. Duplication of the command structure is necessary for interoperability.

Oracle's best argument, therefore, is that while no single name is copyrightable, Java's overall system of organized names — covering 37 packages, with over six hundred classes, with over six thousand methods — is a "taxonomy" and, therefore, copyrightable under American Dental Association v. Delta Dental Plans Association, 126 F.3d 977 (7th Cir. 1997). There was nothing in the rules of the Java language that required that Google replicate the same groupings even if Google was free to replicate the same functionality. . . . This explains why American Dental Association v. Delta Dental Plans Association, 126 F.3d 977 (7th Cir. 1997), is not controlling. Assuming arguendo that a taxonomy is protectable by copyright in our circuit, see Practice Mgmt. Info. Corp. v. Am. Med. Ass'n, 121 F.3d 516 (9th Cir. 1997), the taxonomy in ADA had nothing to do with computer programs. It was not a system of commands, much less a system of commands for a computer language. The taxonomy there subdivided the universe of all dental procedures into an outline of numbered categories with English-language descriptions created by the ADA. This was then to be used by insurance companies and dentists to facilitate billings. By contrast, here the taxonomy is composed entirely of a system of commands to carry out specified computer functions. For a similar reason, Oracle's analogy to stealing the plot and character from a movie is inapt, for movies involve no "system" or "method of operation" — scripts are entirely creative. In ADA, Judge Frank Easterbrook (writing for the panel) suggested that a "system" under Section 102(b) had to come with "instructions for use." 126 F.3d at 980. Because the taxonomy there at issue had no instructions for use, among other reasons, it was held not to be a system. By contrast, the API at issue here does come with instructions for use, namely, the documentation and embedded comments that were much litigated at trial. They describe every package, class and method, what inputs they need, and what outputs they return — the classic form of instructions for use.

Interoperability sheds further light on the character of the command structure as a system or method of operation. Surely, millions of lines of code had been written in Java before Android arrived. These programs necessarily used the java.package.Class.method() command format. These programs called on all or some of the specific 37 packages at issue and necessarily used the command structure of names at issue. Such code was owned by the developers themselves, not by Oracle. In order for at least some of this code to run on Android, Google was required to provide the same java.package.Class.method() command system using the same names with the same "taxonomy" and with the same functional specifications. Google replicated what was necessary to achieve a degree of interoperability —but no more, taking care, as said before, to provide its own implementations. That interoperability is at the heart of the command structure is illustrated by Oracle's preoccupation with what it calls "fragmentation," meaning the problem of having imperfect interoperability among platforms. When this occurs, Java-based applications may not run on the incompatible platforms. For example, Java-based code using the replicated parts of the 37 API packages will run on Android but will not if a 38th package is needed. Such imperfect interoperability leads to a "fragmentation" — a Balkanization — of platforms, a circumstance which Sun and Oracle have tried to curb via their licensing programs. In this litigation, Oracle has made much of this problem, at times almost leaving the impression that if only Google had replicated all 166 Java API packages, Oracle would not have sued. While fragmentation is a legitimate business consideration, it begs the question whether or not a license was required in the first place to replicate some or all of the command structure. (This is especially so inasmuch as Android has not carried the Java trademark, and Google has not held out Android as fully compatible.) The immediate point is this: fragmentation, imperfect interoperability, and Oracle's angst over it illustrate the character of the command structure as a functional system or method of operation. In this regard, the Ninth Circuit decisions in Sega and Sony, although not on all fours, are close analogies. Under these two decisions, interface procedures required for interoperability were deemed "functional requirements for compatibility" and were not copyrightable under Section 102(b). Both decisions held that interface procedures that were necessary to duplicate in order to achieve interoperability were functional aspects not copyrightable under Section 102(b). Here, the command structure for the 37 packages (including inheritances and exception throws), when replicated, at least allows interoperability of code using the replicated commands. To the extent of the 37 packages — which, after all, is the extent of Oracle's copyright claim — Sega and Sony are analogous. Put differently, if someone could duplicate the interfaces of the Sony BIOS in order to run the Playstation games on desktops (taking care to write its own implementations), then Google was free to duplicate the command structure for the 37 packages in Android in order to accommodate third-party source code relying on the 37 packages (taking care to write its own implementations). Contrary to Oracle, "full compatibility" is not relevant to the Section 102(b) analysis. In Sony, the accused product implemented only 137 of the Playstation BIOS's 242 functions because those were the only functions invoked by the games tested. Connectix's Opening Appellate Brief at 18, available at 1999 WL 33623860, (9th Cir. May 27, 1999). Our court of appeals held that the accused product "itself infringe[d] no copyright." Sony, 203 F.3d at 608 n.11. This parallels Google's decision to implement some but not all of the Java API packages in Android. That interoperability is at the heart of the command structure is illustrated by Oracle's preoccupation with what it calls "fragmentation," meaning the problem of having imperfect interoperability among platforms. When this occurs, Java-based applications may not run on the incompatible platforms. For example, Java-based code using the replicated parts of the 37 API packages will run on Android but will not if a 38th package is needed. Such imperfect interoperability leads to a "fragmentation" — a Balkanization — of platforms, a circumstance which Sun and Oracle have tried to curb via their licensing programs. In this litigation, Oracle has made much of this problem, at times almost leaving the impression that if only Google had replicated all 166 Java API packages, Oracle would not have sued. While fragmentation is a legitimate business consideration, it begs the question whether or not a license was required in the first place to replicate some or all of the command structure. (This is especially so inasmuch as Android has not carried the Java trademark, and Google has not held out Android as fully compatible.) The immediate point is this: fragmentation, imperfect In ADA, Judge Frank Easterbrook (writing for the panel) suggested that a "system" under Section 102(b) had to come with "instructions for use." 126 F.3d at 980. Because the taxonomy there at issue had no instructions for use, among other reasons, it was held not to be a system. By contrast, the API at issue here does come with instructions for use, namely, the documentation and embedded comments that were much litigated at trial. They describe every package, class and method, what inputs they need, and what outputs they return — the classic form of instructions for use. In closing, it is important to step back and take in the breadth of Oracle's claim. Of the 166 Java packages, 129 were not violated in any way. Of the 37 accused, 97 percent of the Android lines were new from Google and the remaining three percent were freely replicable under the merger and names doctrines. Oracle must resort, therefore, to claiming that it owns, by copyright, the exclusive right to any and all possible implementations of the taxonomy-like command

Again, the appellate opinion stated that the extent to which the structure, sequence and organization was protectable depended on the facts and circumstances of each case. The circumstances there are not the circumstances here.

. . .

This order holds that, under the Copyright Act, no matter how creative or imaginative a Java method specification may be, the entire world is entitled to use the same method specification (inputs, outputs, parameters) so long as the line-by-line implementations are different. To At odds with the compilation defiinition To repeat the Second Circuit's phrasing, "there might be a myriad of ways in which a programmer may . . . express the idea embodied in a given subroutine." Computer Associates, 982 F.2d at 708. The method specification is the idea. The method implementation is the expression. No one may monopolize the idea. In sum, Google and the public were and remain free to write their own implementations to carry out exactly the same functions of all methods in question, using exactly the same method specifications and names. Therefore, at the method level — the level where the heavy lifting is done — Google has violated no copyright, it being undisputed that Google's implementations are different. As for classes, the rules of the language likewise insist on giving names to classes and the rules insist on strict syntax and punctuation in the lines of code that declare a class. As with methods, for any desired functionality, the declaration line will always read the same (otherwise the functionality would be different) — save only for the name, which cannot be claimed by copyright. Therefore, under the law, the declaration line cannot be protected by copyright. This analysis is parallel to the analysis for methods. This now accounts for virtually all of the three percent of similar code.

This order does not hold that Java API packages are free for all to use without license. It does not hold that the structure, sequence and organization of all computer programs may be stolen. Rather, it holds on the specific facts of this case, the particular elements replicated by Google were free for all to use under the Copyright Act. Therefore, Oracle's claim based on Google's copying of the 37 API packages, including their structure, sequence and organization is DISMISSED. To the extent stated herein, Google's Rule 50 motions regarding copyrightability are GRANTED (Dkt. Nos. 984, 1007). Google's motion for a new trial on copyright infringement is DENIED AS MOOT (Dkt. No. 1105).

In effect, Borland allowed users to choose how they wanted to communicate with Borland's spreadsheet programs: either by using menu commands designed by Borland, or by using the commands and command structure used in Lotus 1-2-3 augmented by Borland-added commands. . . . On July 31, 1992, the district court denied Borland's motion and granted Lotus's motion in part. The district court ruled that the Lotus menu command hierarchy was copyrightable expression because [a] very satisfactory spreadsheet menu tree can be constructed using different commands and a different command structure from those of Lotus 1- 2-3. In fact, Borland has constructed just such an alternate tree for use in Quattro Pro's native mode. Even if one holds the arrangement of menu commands constant, it is possible to generate literally millions of satisfactory menu trees by varying the menu commands employed. . . . If Lotus is granted a monopoly on this pattern, users who have learned the command structure of Lotus 1-2-3 or devised their own macros are locked into Lotus, just as a typist who has learned the QWERTY keyboard would be the captive of anyone who had a monopoly on the production of such a keyboard. . . . But if a better spreadsheet comes along, it is hard to see why customers who have learned the Lotus menu and devised macros for it should remain captives of Lotus because of an investment in learning made by the users and not by Lotus.

The distinction between literal and non-literal aspects of a computer program is separate from the distinction between literal and non-literal copying. See Altai, 982 F.2d at 701-02. “Literal” copying is verbatim copying of original expression. “Non-literal” copying is “paraphrased or loosely paraphrased rather than word for word.” Lotus Dev. Corp. v. Borland Int’l, 49 F.3d 807, 814 (1st Cir. 1995). Here, Google concedes that it copied the declaring code verbatim. Oracle explains that the lines of declaring code “embody the structure of each [API] package, just as the chapter titles and topic sentences represent the structure of a novel.” Appellant Br. 45. As Oracle explains, when Google copied the declaring code in these packages “it also copied the ‘sequence and organization’ of the packages (i.e., the three-dimensional structure with all the chutes and ladders)” employed by Sun/Oracle in the packages. Appellant Br. 27. Oracle also argues that the nonliteral elements of the API packages—the structure, sequence, and organization that led naturally to the implementing code Google created—are entitled to protection. Oracle does not assert “literal” copying of the entire SSO, but, rather, that Google literally copied the declaring code and then paraphrased the remainder of the SSO by writing its own implementing code. It therefore asserts non-literal copying with respect to the entirety of the SSO.

At this stage, it is undisputed that the declaring code and the structure and organization of the Java API packages are original. The testimony at trial revealed that designing the Java API packages was a creative process and that the Sun/Oracle developers had a vast range of options for the structure and organization. In its copyrightability decision, the district court specifically found that the API packages are both creative and original, and Google concedes on appeal that the originality requirements are met. See Copyrightability Decision, 872 F. Supp. 2d at 976 (“The overall name tree, of course, has creative elements . . . .”); Id. at 999 (“Yes, it is creative. Yes, it is original.”); Appellee Br. 5 (“Google does not dispute” the district court’s finding that “the Java API clears the low originality threshold.”). The court found, however, that neither the declaring code nor the SSO was entitled to copyright protection under the Copyright Act.

It is well-established that copyrightability and the scope of protectable activity are to be evaluated at the time of creation, not at the time of infringement. See Apple Computer, Inc. v. Formula Int’l, Inc., 725 F.2d 521, 524 (9th Cir. 1984) (quoting National Commission on New Technological Uses of Copyrighted Works, Final Report at 21 (1979) (“CONTU Report”) (recognizing that the Copyright Act was designed “to protect all works of authorship from the moment of their fixation in any tangible medium of expression”)). The focus is, therefore, on the options that were available to Sun/Oracle at the time it created the API packages. Of course, once Sun/Oracle created “java.lang.Math.max,” programmers who want to use that particular package have to call it by that name. But, as the court acknowledged, nothing prevented Google from writing its own declaring code, along with its own implementing code, to achieve the same result. In such circumstances, the chosen expression simply does not merge with the idea being expressed. It is undisputed, however, that Google wrote its own implementing code, except with respect to: (1) the rangeCheck function, which consisted of nine lines of code; and (2) eight decompiled security files. As to rangeCheck, the court found that the Sun engineer who wrote it later worked for Google and contributed two files he created containing the rangeCheck function— “Timsort.java” and “ComparableTimsort”—to the Android platform. In doing so, the nine-line rangeCheck function was copied directly into Android. As to the eight decompiled files, the district court found that they were copied and used as test files but “never found their way into Android or any handset.” Id. at 983. The district court granted Oracle’s motion for JMOL of infringement as to the eight decompiled files, however. In its order, the court explained that: (1) Google copied the files in their entirety; (2) the trial testimony revealed that the use of those files was “significant”; and (3) no reasonable jury could find the copying de minimis. Order Granting JMOL on Decompiled Files, 2012 U.S. Dist. LEXIS 66417, at *6. The First Circuit, however, held that the Lotus menu command hierarchy was not copyrightable because it was a method of operation under Section 102(b). Put differently, if someone could duplicate the interfaces of the Sony BIOS in order to run the Playstation games on desktops (taking care to write its own implementations), then Google was free to duplicate the command structure for the 37 packages in Android in order to accommodate third-party source code relying on the 37 packages (taking care to write its own implementations). Contrary to Oracle, "full compatibility" is not relevant to the Section 102(b) analysis.

When assessing whether the non-literal elements of a computer program constitute protectable expression, the Ninth Circuit has endorsed an “abstraction-filtration-comparison” test formulated by the Second Circuit and expressly adopted by several other circuits. Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525 (9th Cir. 1992) (“In our view, in light of the essentially utilitarian nature of computer programs, the Second Circuit’s approach is an appropriate one.”). This test rejects the notion that anything that performs a function is necessarily uncopyrightable. See Mitel, 124 F.3d at 1372 (rejecting the Lotus court’s formulation, and concluding that, “although an element of a work may be characterized as a method of operation, that element may nevertheless contain expression that is eligible for copyright protection.”). And it also rejects as flawed the Whelan assumption that, once any separable idea can be identified in a computer program everything else must be protectable expression, on grounds that more than one idea may be embodied in any particular program. Altai, 982 F.2d at 705-06. Thus, this test eschews bright line approaches and requires a more nuanced assessment of the particular program at issue in order to determine what expression is protectable and infringed. As the Second Circuit explains, this test has three steps. In the abstraction step, the court “first break[s] down the allegedly infringed program into its constituent structural parts.” Id. at 706. In the filtration step, the court “sift[s] out all non-protectable material,” including ideas and “expression that is necessarily incidental to those ideas.” Id. In the final step, the court compares the remaining creative expression with the allegedly infringing program.4 In the second step, the court is first to assess whether the expression is original to the programmer or author. Atari, 975 F.2d at 839. The court must then determine whether the particular inclusion of any level of abstraction is dictated by considerations of efficiency, required by factors already external to the program itself, or taken from the public domain—all of which would render the expression unprotectable. Id. These conclusions are to be informed by traditional copyright principles of originality, merger, and scenes a faire. See Mitel, 124 F.3d at 1372 (“Although this core of expression is eligible for copyright protection, it is subject to the rigors of filtration analysis which excludes from protection expression that is in the public domain, otherwise unoriginal, or subject to the doctrines of merger and scenes a faire.”). In all circuits, it is clear that the first step is part of the copyrightability analysis and that the third is an infringement question. It is at the second step of this analysis where the circuits are in less accord. Some treat all aspects of this second step as part of the copyrightability analysis, while others divide questions of originality from the other inquiries, treating the former as a question of copyrightability and the latter as part of the infringement inquiry. Compare Lexmark, 387 F.3d at 537-38 (finding that the district court erred in assessing principles of merger and scenes a faire in the infringement analysis, rather than as a component of copyrightability), with Kregos, 937 F.2d at 705 (noting that the Second Circuit has considered the merger doctrine “in determining whether actionable infringement has occurred, rather than whether a copyright is valid”); see also Lexmark, 387 F.3d at 557 (Feikens, J., dissenting-in-part) (noting the circuit split and concluding that, where a court is assessing merger of an expression with a method of operation, “I would find the merger doctrine can operate only as a defense to infringement in that context, and as such has no bearing on the question of copyrightability.”). We need not assess the wisdom of these respective views because there is no doubt on which side of this circuit split the Ninth Circuit falls.

Importantly, this full analysis only applies where a copyright owner alleges infringement of the non-literal aspects of its work. Where “admitted literal copying of a discrete, easily-conceptualized portion of a work” is at issue—as with Oracle’s declaring code—a court “need not perform a complete abstraction-filtration-comparison analysis” and may focus the protectability analysis on the filtration stage, with attendant reference to standard copyright principles. Mitel, 124 F.3d at 1372-73.

While the trial court mentioned the abstractionfiltration- comparison test when describing the development of relevant law, it did not purport to actually apply that test. Instead, it moved directly to application of familiar principles of copyright law when assessing the copyrightability of the declaring code and interpreted Section 102(b) to preclude copyrightability for any functional element “essential for interoperability” “regardless of its form.” Copyrightability Decision, 872 F. Supp. 2d at 997.

We find that Oracle did not waive arguments based on Google’s literal copying of the declaring code. Prior to trial, both parties informed the court that Oracle’s copyright infringement claims included the declarations of the API elements in the Android class library source code. . . . While Google is correct that the jury instructions and verdict form focused on the structure and organization of the packages, we agree with Oracle that there was no need for the jury to address copying of the declaring code because Google conceded that it copied it verbatim. Indeed, the district court specifically instructed the jury that “Google agrees that it uses the same names and declarations” in Android. Final Charge to the Jury at 10.

The merger doctrine functions as an exception to the idea/expression dichotomy. It provides that, when there are a limited number of ways to express an idea, the idea is said to “merge” with its expression, and the expression becomes unprotected. Altai, 982 F.2d at 707-08. As noted, the Ninth Circuit treats this concept as an affirmative defense to infringement. Ets-Hokin, 225 F.3d at 1082. Accordingly, it appears that the district court’s merger analysis is irrelevant to the question of whether Oracle’s API packages are copyrightable in the first instance. Regardless of when the analysis occurs, we conclude that merger does not apply on the record before us.

On appeal, Oracle argues that the district court: (1) misapplied the merger doctrine; and (2) failed to focus its analysis on the options available to the original author. We agree with Oracle on both points. First, we agree that merger cannot bar copyright protection for any lines of declaring source code unless Sun/Oracle had only one way, or a limited number of ways, to write them. See Satava, 323 F.3d at 812 n.5 (“Under the merger doctrine, courts will not protect a copyrighted work from infringement if the idea underlying the copyrighted work can be expressed in only one way, lest there be a monopoly on the underlying idea.”). The evidence showed that Oracle had “unlimited options as to the selection and arrangement of the 7000 lines Google copied.” Appellant Br. 50. Using the district court’s “java.lang.Math.max” example, Oracle explains that the developers could have called it any number of things, including “Math.maximum” or “Arith.larger.” This was not a situation where Oracle was selecting among preordained names and phrases to create its packages.6 As the district court recognized, moreover “the Android method and class names could have been different from the names of their counterparts in Java and still have worked.” Copyrightability Decision, 872 F. Supp. 2d at 976. Because “alternative expressions [we]re available,” there is no merger. See Atari, 975 F.2d at 840. We further find that the district court erred in focusing its merger analysis on the options available to Google at the time of copying. It is well-established that copyrightability and the scope of protectable activity are to be evaluated at the time of creation, not at the time of infringement. See Apple Computer, Inc. v. Formula Int’l, Inc., 725 F.2d 521, 524 (9th Cir. 1984) (quoting National Commission on New Technological Uses of Copyrighted Works, Final Report at 21 (1979) (“CONTU Report”) (recognizing that the Copyright Act was designed “to protect all works of authorship from the moment of their fixation in any tangible medium of expression”)). The focus is, therefore, on the options that were available to Sun/Oracle at the time it created the API packages. Of course, once Sun/Oracle created “java.lang.Math.max,” programmers who want to use that particular package have to call it by that name. But, as the court acknowledged, nothing prevented Google from writing its own declaring code, along with its own implementing code, to achieve the same result. In such circumstances, the chosen expression simply does not merge with the idea being expressed.7

It seems possible that the merger doctrine, when properly analyzed, would exclude the three packages identified by the district court as core packages from the scope of actionable infringing conduct. This would be so if the Java authors, at the time these packages were created, had only a limited number of ways to express the methods and classes therein if they wanted to write in the Java language. In that instance, the idea may well be merged with the expression in these three packages.8 Interesting footnote 8: At oral argument, counsel for Oracle was asked whether we should view the three core packages “differently vis-à-vis the concept of a method of operation than the other packages.” See Oral Argument at 7:43. He responded: “I think not your Honor. I would view them differently with respect to fair use . . . . It’s not that they are more basic. It’s that there are just several methods, that is, routines, within just those three packages that are necessary to ‘speak the Java language.’ Nothing in the other thirty-four packages is necessary in order to speak in Java, so to speak.” Id. Counsel conceded, however, that this issue “might go to merger. It might go to the question whether someone—since we conceded that it’s okay to use the language—if it’s alright to use the language that there are certain things that the original developers had to say in order to use that language, arguably, although I still think it’s really a fair use analysis.” Id.

[B]y dissecting the individual lines of declaring code at issue into short phrases, the district court further failed to recognize that an original combination of elements can be copyrightable. See Softel, Inc. v. Dragon Med. & Scientific Commc’ns, 118 F.3d 955, 964 (2d Cir. 1997) (noting that, in Feist, “the Court made quite clear that a compilation of nonprotectible elements can enjoy copyright protection even though its constituent elements do not”). By analogy, the opening of Charles Dickens’ A Tale of Two Cities is nothing but a string of short phrases. Yet no one could contend that this portion of Dickens’ work is unworthy of copyright protection because it can be broken into those shorter constituent components. The question is not whether a short phrase or series of short phrases can be extracted from the work, but whether the manner in which they are used or strung together exhibits creativity. Although the district court apparently focused on individual lines of code, Oracle is not seeking copyright protection for a specific short phrase or word. Instead, the portion of declaring code at issue is 7,000 lines, and Google’s own “Java guru” conceded that there can be “creativity and artistry even in a single method declaration.” Joint Appendix (“J.A.”) 20,970. Because Oracle “exercised creativity in the selection and arrangement” of the method declarations when it created the API packages and wrote the relevant declaring code, they contain protectable expression that is entitled to copyright protection. See Atari, 975 F.2d at 840; see also 17 U.S.C. §§ 101, 103 (recognizing copyright protection for “compilations” which are defined as work that is “selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship”). Accordingly, we conclude that the district court erred in applying the short phrases doctrine to find the declaring code not copyrightable.

On appeal, Google refers to scenes a faire concepts briefly, as do some amici, apparently contending that, because programmers have become accustomed to and comfortable using the groupings in the Java API packages, those groupings are so commonplace as to be indispensable to the expression of an acceptable programming platform. As such, the argument goes, they are so associated with the “idea” of what the packages are accomplishing that they should be treated as ideas rather than expression. See Br. of Amici Curiae Rackspace US, Inc., et al. at 19-22. Google cannot rely on the scenes a faire doctrine as an alternative ground upon which we might affirm the copyrightability judgment of the district court. This is so for several reasons. First, as noted, like merger, in the Ninth Circuit, the scenes a faire doctrine is a component of the infringement analysis. “[S]imilarity of expression, whether literal or non-literal, which necessarily results from the fact that the common idea is only capable of expression in more or less stereotyped form, will preclude a finding of actionable similarity.” 4 Nimmer on Copyright § 13.03[B][3]. Thus, the expression is not excluded from copyright protection; it is just that certain copying is forgiven as a necessary incident of any expression of the underlying idea. See Satava, 323 F.3d at 810 n.3 (“The Ninth Circuit treats scenes a faire as a defense to infringement rather than as a barrier to copyrightability.”). Second, Google has not objected to the trial court’s conclusion that Google failed to make a sufficient factual record to support its contention that the groupings and code chosen for the 37 Java API packages were driven by external factors or premised on features that were either commonplace or essential to the idea being expressed. Google provides no record citations indicating that such a showing was made and does not contend that the trial court erred when it expressly found it was not. Indeed, Google does not even make this argument with respect to the core packages. Finally, Google’s reliance on the doctrine below and the amici reference to it here are premised on a fundamental misunderstanding of the doctrine. Like merger, the focus of the scenes a faire doctrine is on the circumstances presented to the creator, not the copier. See Mitel, 124 F.3d at 1375 (finding error to the extent the trial court discussed “whether external factors such as market forces and efficiency considerations justified Iqtel’s copying of the command codes”). The court’s analytical focus must be upon the external factors that dictated Sun’s selection of classes, methods, and code—not upon what Google encountered at the time it chose to copy those groupings and that code. See id. “[T]he scenes a faire doctrine identifies and excludes from protection against infringement expression whose creation ‘flowed naturally from considerations external to the author’s creativity.’” Id. (quoting Nimmer § 13.03[F][3], at 13-131 (1997)). It is this showing the trial court found Google failed to make, and Google cites to nothing in the record which indicates otherwise.

The district court found that the SSO of the Java API packages is creative and original, but nevertheless held that it is a “system or method of operation . . . and, therefore, cannot be copyrighted” under 17 U.S.C. § 102(b). Copyrightability Decision, 872 F. Supp. 2d at 976-77. In reaching this conclusion, the district court seems to have relied upon language contained in a First Circuit decision: Lotus Development Corp. v. Borland International, Inc., 49 F.3d 807 (1st Cir. 1995), aff’d without opinion by equally divided court, 516 U.S. 233 (1996).9

On appeal, Oracle argues that the district court’s reliance on Lotus is misplaced because it is distinguishable on its facts and is inconsistent with Ninth Circuit law. We agree. First, while the defendant in Lotus did not copy any of the underlying code, Google concedes that it copied portions of Oracle’s declaring source code verbatim. Second, the Lotus court found that the commands at issue there (copy, print, etc.) were not creative, but it is undisputed here that the declaring code and the structure and organization of the API packages are both creative and original. Finally, while the court in Lotus found the commands at issue were “essential to operating” the system, it is undisputed that—other than perhaps as to the three core packages—Google did not need to copy the structure, sequence, and organization of the Java API packages to write programs in the Java language. More importantly, however, the Ninth Circuit has not adopted the court’s “method of operation” reasoning in Lotus, and we conclude that it is inconsistent with binding precedent.11 Specifically, we find that Lotus is incon- structure, sequence, and organization of a computer program is eligible for copyright protection where it qualifies as an expression of an idea, rather than the idea itself. See Johnson Controls, 886 F.2d at 1175-76. And, while the court in Lotus held “that expression that is part of a ‘method of operation’ cannot be copyrighted,” 49 F.3d at 818, this court—applying Ninth Circuit law—reached the exact opposite conclusion, finding that copyright protects “the expression of [a] process or method,” Atari, 975 F.2d at 839. We find, moreover, that the hard and fast rule set down in Lotus and employed by the district court here— i.e., that elements which perform a function can never be copyrightable—is at odds with the Ninth Circuit’s endorsement of the abstraction-filtration-comparison analysis discussed earlier. As the Tenth Circuit concluded in expressly rejecting the Lotus “method of operation” analysis, in favor of the Second Circuit’s abstraction-filtrationcomparison test, “although an element of a work may be characterized as a method of operation, that element may nevertheless contain expression that is eligible for copyright protection.” Mitel, 124 F.3d at 1372. Specifically, the court found that Section 102(b) “does not extinguish the protection accorded a particular expression of an idea merely because that expression is embodied in a method of operation at a higher level of abstraction.” Id. Other courts agree that components of a program that can be characterized as a “method of operation” may nevertheless be copyrightable. For example, the Third Circuit rejected a defendant’s argument that operating system programs are “per se” uncopyrightable because an operating system is a “method of operation” for a computer. Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240, 1250-52 (3d Cir. 1983). The court distinguished between the “method which instructs the computer to perform its operating functions” and “the instructions themselves,” and found that the instructions were copyrightable. Id. at 1250-51. In its analysis, the court noted: “[t]hat the words of a program are used ultimately in the implementation of a process should in no way affect their copyrightability.” Id. at 1252 (quoting CONTU Report at 21). The court focused “on whether the idea is capable of various modes of expression” and indicated that, “[i]f other programs can be written or created which perform the same function as [i]n Apple’s operating system program, then that program is an expression of the idea and hence copyrightable.” Id. at 1253. Notably, no other circuit has adopted the First Circuit’s “method of operation” analysis. Courts have likewise found that classifying a work as a “system” does not preclude copyright for the particular expression of that system. See Toro Co. v. R & R Prods. Co., 787 F.2d 1208, 1212 (8th Cir. 1986) (rejecting the district court’s decision that “appellant’s parts numbering system is not copyrightable because it is a ‘system’” and indicating that Section 102(b) does not preclude protection for the “particular expression” of that system); see also Am. Dental Ass’n v. Delta Dental Plans Ass’n, 126 F.3d 977, 980 (7th Cir. 1997) (“A dictionary cannot be called a ‘system’ just because new novels are written using words, all of which appear in the dictionary. Nor is word-processing software a ‘system’ just because it has a command structure for producing paragraphs.”).

Here, the district court recognized that the SSO “resembles a taxonomy,” but found that “it is nevertheless a command structure, a system or method of operation—a long hierarchy of over six thousand commands to carry out pre-assigned functions.” Copyrightability Decision, 872 F. Supp. 2d at 999-1000.12 In other words, the court concluded that, although the SSO is expressive, it is not copyrightable because it is also functional. The problem with the district court’s approach is that computer programs are by definition functional—they are all designed to accomplish some task. Indeed, the statutory definition of “computer program” acknowledges that they function “to bring about a certain result.” See 17 U.S.C. § 101 (defining a “computer program” as “a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result”). If we were to accept the district court’s suggestion that a computer program is uncopyrightable simply because it “carr[ies] out pre-assigned functions,” no computer program is protectable. That result contradicts Congress’s express intent to provide copyright protection to computer programs, as well as binding Ninth Circuit case law finding computer programs copyrightable, despite their utilitarian or functional purpose. Though the trial court did add the caveat that it “does not hold that the structure, sequence and organization of all computer programs may be stolen,” Copyrightability Decision, 872 F. Supp. 2d at 1002, it is hard to see how its method of operation analysis could lead to any other conclusion. While it does not appear that the Ninth Circuit has addressed the precise issue, we conclude that a set of commands to instruct a computer to carry out desired operations may contain expression that is eligible for copyright protection. See Mitel, 124 F.3d at 1372. We agree with Oracle that, under Ninth Circuit law, an original work—even one that serves a function—is entitled to copyright protection as long as the author had multiple ways to express the underlying idea. Section 102(b) does not, as Google seems to suggest, automatically deny copyright protection to elements of a computer program that are functional. Instead, as noted, Section 102(b) codifies the idea/expression dichotomy and the legislative history confirms that, among other things, Section 102(b) was “intended to make clear that the expression adopted by the programmer is the copyrightable element in a computer program.” H.R. Rep. No. 1476, 94th Cong., 2d Sess. 54, reprinted in 1976 U.S.C.C.A.N. 5659, 5670. Therefore, even if an element directs a computer to perform operations, the court must nevertheless determine whether it contains any separable expression entitled to protection.

As the district court acknowledged, Google could have structured Android differently and could have chosen different ways to express and implement the functionality that it copied.14 Specifically, the court found that “the very same functionality could have been offered in Android without duplicating the exact command structure used in Java.” Copyrightability Decision, 872 F. Supp. 2d at 976. The court further explained that Google could have offered the same functions in Android by “rearranging the various methods under different groupings among the various classes and packages.” Id. The evidence showed, moreover, that Google designed many of its own API packages from scratch, and, thus, could have designed its own corresponding 37 API packages if it wanted to do so. Given the court’s findings that the SSO is original and creative, and that the declaring code could have been written and organized in any number of ways and still have achieved the same functions, we conclude that Section 102(b) does not bar the packages from copyright protection just because they also perform functions.

Both Sega and Sony are fair use cases in which copyrightability was addressed only tangentially. In Sega, for example, Sega manufactured a video game console and game cartridges that contained hidden functional program elements necessary to achieve compatibility with the console. Defendant Accolade: (1) reverse-engineered Sega’s video game programs to discover the requirements for compatibility; and (2) created its own games for the Sega console. Sega, 977 F.2d at 1514-15. As part of the reverse-engineering process, Accolade made intermediate copies of object code from Sega’s console. Id. Although the court recognized that the intermediate copying of computer code may infringe Sega’s copyright, it concluded that “disassembly of copyrighted object code is, as a matter of law, a fair use of the copyrighted work if such disassembly provides the only means of access to those elements of the code that are not protected by copyright and the copier has a legitimate reason for seeking such access.” Id. at 1518. The court agreed with Accolade that its copying was necessary to examine the unprotected functional aspects of the program. Id. at 1520. And, because Accolade had a legitimate interest in making its cartridges compatible with Sega’s console, the court found that Accolade’s intermediate copying was fair use. Likewise, in Sony, the Ninth Circuit found that the defendant’s reverse engineering and intermediate copying of Sony’s copyrighted software program “was a fair use for the purpose of gaining access to the unprotected elements of Sony’s software.” Sony, 203 F.3d at 602. The court explained that Sony’s software program contained unprotected functional elements and that the defendant could only access those elements through reverse engineering. Id. at 603. The defendant used that information to create a software program that let consumers play games designed for Sony’s PlayStation console on their computers. Notably, the defendant’s software program did not contain any of Sony’s copyrighted material. Id. at 598. The district court characterized Sony and Sega as “close analogies” to this case. Copyrightability Decision, 872 F. Supp. 2d at 1000. According to the court, both decisions “held that interface procedures that were necessary to duplicate in order to achieve interoperability were functional aspects not copyrightable under Section 102(b).” Id. The district court’s reliance on Sega and Sony in the copyrightability context is misplaced, however. As noted, both cases were focused on fair use, not copyrightability. In Sega, for example, the only question was whether Accolade’s intermediate copying was fair use. The court never addressed the question of whether Sega’s software code, which had functional elements, also contained separable creative expression entitled to protection. Likewise, although the court in Sony determined that Sony’s computer program had functional elements, it never addressed whether it also had expressive elements. Sega and Sony are also factually distinguishable because the defendants in those cases made intermediate copies to understand the functional aspects of the copyrighted works and then created new products. See Sony, 203 F.3d at 606-07; Sega, 977 F.2d at 1522-23. This is not a case where Google reverse-engineered Oracle’s Java packages to gain access to unprotected functional elements contained therein. As the former Register of Copyrights of the United States pointed out in his brief amicus curiae, “[h]ad Google reverse engineered the programming packages to figure out the ideas and functionality of the original, and then created its own structure and its own literal code, Oracle would have no remedy under copyright whatsoever.” Br. for Amicus Curiae Ralph Oman 29. Instead, Google chose to copy both the declaring code and the overall SSO of the 37 Java API packages at issue. We disagree with Google’s suggestion that Sony and Sega created an “interoperability exception” to copyrightability. See Appellee Br. 39 (citing Sony and Sega for the proposition that “compatibility elements are not copyrightable under section 102(b)” (emphasis omitted)). Although both cases recognized that the software programs at issue there contained unprotected functional elements, a determination that some elements are unprotected is not the same as saying that the entire work loses copyright protection. To accept Google’s reading would contradict Ninth Circuit case law recognizing that both the literal and non-literal components of a software program are eligible for copyright protection. See Johnson Controls, 886 F.2d at 1175. And it would ignore the fact that the Ninth Circuit endorsed the abstractionfiltration- comparison inquiry in Sega itself. As previously discussed, a court must examine the software program to determine whether it contains creative expression that can be separated from the underlying function. See Sega, 977 F.2d at 1524-25. In doing so, the court filters out the elements of the program that are “ideas” as well as elements that are “dictated by considerations of efficiency, so as to be necessarily incidental to that idea; required by factors external to the program itself.” Altai, 982 F.2d at 707.

As the district court acknowledged, Google could have structured Android differently and could have chosen different ways to express and implement the functionality that it copied.14 Specifically, the court found that “the very same functionality could have been offered in Android without duplicating the exact command structure used in Java.” Copyrightability Decision, 872 F. Supp. 2d at 976. The court further explained that Google could have offered the same functions in Android by “rearranging the various methods under different groupings among the various classes and packages.” Id. The evidence showed, moreover, that Google designed many of its own API packages from scratch, and, thus, could have designed its own corresponding 37 API packages if it wanted to do so. Given the court’s findings that the SSO is original and creative, and that the declaring code could have been written and organized in any number of ways and still have achieved the same functions, we conclude that Section 102(b) does not bar the packages from copyright protection just because they also perform functions.

To determine “whether certain aspects of an allegedly infringed software are not protected by copyright law, the focus is on external factors that influenced the choice of the creator of the infringed product.” Dun & Bradstreet Software Servs., Inc. v. Grace Consulting, Inc., 307 F.3d 197, 215 (3d Cir. 2002) (citing Altai, 982 F.2d at 714; Mitel, 124 F.3d at 1375). The Second Circuit, for example, has noted that programmers are often constrained in their design choices by “extrinsic considerations” including “the mechanical specifications of the computer on which a particular program is intended to run” and “compatibility requirements of other programs with which a program is designed to operate in conjunction.” Altai, 982 F.2d at 709-10 (citing 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.01 at 13-66-71 (1991)). The Ninth Circuit has likewise recognized that: (1) computer programs “contain many logical, structural, and visual display elements that are dictated by . . . external factors such as compatibility requirements and industry demands”; and (2) “[i]n some circumstances, even the exact set of commands used by the programmer is deemed functional rather than creative for purposes of copyright.” Sega, 977 F.2d at 1524 (internal citation omitted). Because copyrightability is focused on the choices available to the plaintiff at the time the computer program was created, the relevant compatibility inquiry asks whether the plaintiff’s choices were dictated by a need to ensure that its program worked with existing third-party programs. Dun & Bradstreet, 307 F.3d at 215; see also Atari, 975 F.2d at 840 (“External factors did not dictate the design of the 10NES program.”). Whether a defendant later seeks to make its program interoperable with the plaintiff’s program has no bearing on whether the software the plaintiff created had any design limitations dictated by external factors. See Dun & Bradstreet, 307 F.3d at 215 (finding an expert’s testimony on interopera bility “wholly misplaced” because he “looked at externalities from the eyes of the plagiarist, not the eyes of the program’s creator”). Stated differently, the focus is on the compatibility needs and programming choices of the party claiming copyright protection—not the choices the defendant made to achieve compatibility with the plaintiff’s program. Consistent with this approach, courts have recognized that, once the plaintiff creates a copyrightable work, a defendant’s desire “to achieve total compatibility . . . is a commercial and competitive objective which does not enter into the . . . issue of whether particular ideas and expressions have merged.” Apple Computer, 714 F.2d at 1253.

Whether Google’s software is “interoperable” in some sense with any aspect of the Java platform (although as Google concedes, certainly not with the JVM) has no bearing on the threshold question of whether Oracle’s software is copyrightable . . . Google’s interest was in accelerating its development process by “leverag[ing] Java for its existing base of developers.” J.A. 2033, 2092. Although this competitive objective might be relevant to the fair use inquiry, we conclude that it is irrelevant to the copyrightability of Oracle’s declaring code and organization of the API packages.

Finally, to the extent Google suggests that it was entitled to copy the Java API packages because they had become the effective industry standard, we are unpersuaded. Google cites no authority for its suggestion that copyrighted works lose protection when they become popular, and we have found none.16 In fact, the Ninth Circuit has rejected the argument that a work that later becomes the industry standard is uncopyrightable. See Practice Mgmt. Info. Corp. v. Am. Med. Ass’n, 121 F.3d 516, 520 n.8 (9th Cir. 1997) (noting that the district court found plaintiff’s medical coding system entitled to copyright protection, and that, although the system had become the industry standard, plaintiff’s copyright did not prevent competitors “from developing comparative or better coding systems and lobbying the federal government and private actors to adopt them. It simply prevents wholesale copying of an existing system.”). Google was free to develop its own API packages and to “lobby” programmers to adopt them. Instead, it chose to copy Oracle’s declaring code and the SSO to capitalize on the preexisting community of programmers who were accustomed to using the Java API packages. That desire has nothing to do with copyrightability. For these reasons, we find that Google’s industry standard argument has no bearing on the copyrightability of Oracle’s work.

As noted, the jury hung on Google’s fair use defense, and the district court declined to order a new trial given its conclusion that the code and structure Google copied were not entitled to copyright protection. On appeal, Oracle argues that: (1) a remand to decide fair use “is pointless”; and (2) this court should find, as a matter of law, that “Google’s commercial use of Oracle’s work in a market where Oracle already competed was not fair use.” Appellant Br. 68.

. . .

commercial purposes, did so verbatim, and did so to the detriment of Oracle’s market position. These undisputable facts, according to Oracle, should end the fair use inquiry. Oracle’s position is not without force. On many of these points, Google does not debate Oracle’s characterization of its conduct, nor could it on the record evidence. Google contends, however, that, although it admittedly copied portions of the API packages and did so for what were purely commercial purposes, a reasonable juror still could find that: (1) Google’s use was transformative; (2) the Java API packages are entitled only to weak protection; (3) Google’s use was necessary to work within a language that had become an industry standard; and (4) the market impact on Oracle was not substantial. On balance, we find that due respect for the limit of our appellate function requires that we remand the fair use question for a new trial. First, although it is undisputed that Google’s use of the API packages is commercial, the parties disagree on whether its use is “transformative.” Google argues that it is, because it wrote its own implementing code, created its own virtual machine, and incorporated the packages into a smartphone platform. For its part, Oracle maintains that Google’s use is not transformative because: (1) “[t]he same code in Android . . . enables programmers to invoke the same pre-programmed functions in exactly the same way;” and (2) Google’s use of the declaring code and packages does not serve a different function from Java. Appellant Reply Br. 47. While Google overstates what activities can be deemed transformative under a correct application of the law, we cannot say that there are no material facts in dispute on the question of whether Google’s use is “transformative,” even under a correct reading of the law. As such, we are unable to resolve this issue on appeal.

. . .

Ultimately, we conclude that this is not a case in which the record contains sufficient factual findings upon which we could base a de novo assessment of Google’s affirmative defense of fair use. Accordingly, we remand this question to the district court for further proceedings. On remand, the district court should revisit and revise its jury instructions on fair use consistent with this opinion so as to provide the jury with a clear and appropriate picture of the fair use defense.

Next, while we have concluded that it was error for the trial court to focus unduly on the functional aspects of the packages, and on Google’s competitive desire to achieve commercial “interoperability” when deciding whether Oracle’s API packages are entitled to copyright protection, we expressly noted that these factors may be relevant to a fair use analysis. While the trial court erred in concluding that these factors were sufficient to overcome Oracle’s threshold claim of copyrightability, reasonable jurors might find that they are relevant to Google’s fair use defense under the second and third factors of the inquiry. See Sega, 977 F.2d at 1524-25 (discussing the Second Circuit’s approach to “break[ing] down a computer program into its component subroutines and subsubroutines and then identif[ying] the idea or core functional element of each” in the context of the second fair use factor: the nature of the copyrighted work). We find this particularly true with respect to those core packages which it seems may be necessary for anyone to copy if they are to write programs in the Java language. And, it may be that others of the packages were similarly essential components of any Java language-based program. So far, that type of filtration analysis has not occurred.

Importantly for our purposes, the Supreme Court has made clear that “[n]either the Copyright Statute nor any other says that because a thing is patentable it may not be copyrighted.” Mazer v. Stein, 347 U.S. 201, 217 (1954). Indeed, the thrust of the CONTU Report is that copyright is “the most suitable mode of legal protection for computer software.” Peter S. Menell, An Analysis of the Scope of Copyright Protection for Application Programs, 41 Stan. L. Rev. 1045, 1072 (1989); see also CONTU Report at 1 (recommending that copyright law be amended “to make it explicit that computer programs, to the extent that they embody an author’s original creation, are proper subject matter of copyright”). Until either the Supreme Court or Congress tells us otherwise, we are bound to respect the Ninth Circuit’s decision to afford software programs protection under the copyright laws. We thus decline any invitation to declare that protection of software programs should be the domain of patent law, and only patent law.

To repeat, Section 102(b) states that "in no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation . . . regardless of the form . . . "

That a system or method of operation Yes, it is creative. Yes, it is original. Yes, it resembles a taxonomy. But it is nevertheless a command structure, a system or method of operation - a long hierarchy of over six thousand commands to carry out pre-assigned functions. For that reason, it cannot receive copyright protection - patent protection perhaps - but not copyright protection.

All of this would make no difference if Congress had squarely confronted the issue, and given explicit directions as to what should be done. The Copyright Act of 1976 took a different course. While Congress said that computer programs might be subject to copyright protection, it said this in very general terms; and, especially in § 102(b), Congress adopted a string of exclusions that if taken literally might easily seem to exclude most computer programs from protection. The only detailed prescriptions for computers involve narrow issues (like back-up copies) of no relevance here.

Compare Whelan Assocs., Inc. v. Jaslow Dental Lab., Inc., 797 F.2d 1222, 1236 (3d Cir. ORACLE AMERICA, INC. v. GOOGLE INC. 23 1986) (everything not necessary to the purpose or function of a work is expression), with Lotus, 49 F.3d at 815 (methods of operation are means by which a user operates something and any words used to effectuate that operation are unprotected expression). When assessing whether the non-literal elements of a computer program constitute protectable expression, the Ninth Circuit has endorsed an “abstraction-filtration-comparison” test formulated by the Second Circuit and expressly adopted by several other circuits. Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525 (9th Cir. 1992) (“In our view, in light of the essentially utilitarian nature of computer programs, the Second Circuit’s approach is an appropriate one.”). This test rejects the notion that anything that performs a function is necessarily uncopyrightable. See Mitel, 124 F.3d at 1372 (rejecting the Lotus court’s formulation, and concluding that, “although an element of a work may be characterized as a method of operation, that element may nevertheless contain expression that is eligible for copyright protection.”). And it also rejects as flawed the Whelan assumption that, once any separable idea can be identified in a computer program everything else must be protectable expression, on grounds that more than one idea may be embodied in any particular program. Altai, 982 F.2d at 705-06.

We may assume for present purposes that the Jewel-LaSalle decision retains authoritative force in a factual situation like that in which it arose. But, as the Court of Appeals in this case perceived, this Court has in two recent decisions explicitly disavowed the view that the reception of an electronic broadcast can constitute a performance, when the broadcaster himself is licensed to perform the copyrighted material that he broadcasts. Fortnightly Corp. v. United Artists, 392 U.S. 390;Teleprompter Corp. v. CBS, 415 U.S. 394. Respondent did not infringe upon petitioners' exclusive right, under the Copyright Act, "[t]o perform the copyrighted work publicly for profit," since the radio reception did not constitute a "performance" of the copyrighted songs. Fortnightly Corp. v. United Artists, 392 U.S. 390; Teleprompter Corp. v. CBS, 415 U.S. 394

whether the reception of a copyrighted song on a radio broadcast constitutes a copyright violation if the copyright owner has only licensed the broadcaster to "perform the composition publicly for profit<

It is the intent of the conferees that a small commercial establishment of the type involved in Twentieth Century Music Corp. v. Aiken, 422 U.S. 151 [45 L. Ed. 2d 84, 95 S. Ct. 2040] (1975), which merely augmented a home-type receiver and which was not of sufficient size to justify, as a practical matter, a subscription to a commercial background music service, would be exempt. However, where the public communication was by means of something other than a home-type receiver apparatus, or where the establishment actually makes a further transmission to the public, the exemption would not apply.

Such was the state of the law when this Court in 1931 decided Buck v. Jewell-LaSalle Realty Co., 283 U.S. 191. In that case the Court was called upon to answer the following question certified by the Court of Appeals for the Eighth Circuit: "Do the acts of a hotel proprietor, in making available to his guests, through the instrumentality of a radio receiving set and loud speakers installed in his hotel and under his control and for the entertainment of his guests, the hearing of a copyrighted musical composition which has been broadcast from a radio transmitting station, constitute a performance of such composition within the meaning of 17 USC Sec. 1 (e)?" The Court answered the certified question in the affirmative. In stating the facts of the case, however, the Court's opinion made clear that the broadcaster of the musical composition was not licensed to perform it, and at least twice in the course of its opinion the Court indicated that the answer to the [69] certified question might have been different if the broadcast itself had been authorized by the copyright holder.

"Time's cancellation of its projected article and its refusal to pay $ 12,500 were the direct effect of the infringing publication" The court quotes Nimmer: "Fair use, when properly applied, is limited to copying by others which does not materially impair the marketability of the work which is copied.

"Petitioners established a prima facie case of actual damage that respondents failed to rebut. More important, to negate a claim of fair use it need only be shown that if the challenged use should become widespread, it would adversely affect the potential market for the copyrighted work. Here, The Nation's liberal use of verbatim excerpts posed substantial potential for damage to the marketability of first serialization rights in the copyrighted work. Pp. 560-569."

"Although the verbatim quotes in question were an insubstantial portion of the Ford manuscript, they qualitatively embodied Mr. Ford's distinctive expression"); "As the statutory language indicates, a taking may not be excused merely because it is insubstantial with respect to the infringing work. As Judge Learned Hand cogently remarked, "no plagiarist can excuse the wrong by showing how much of his work he did not pirate."

[n]o material historical facts are at issue in this case. The parties dispute only the ultimate conclusions to be drawn from the admitted facts. . . . “the parties dispute only the ultimate conclusions to be drawn from those facts, we may draw those conclusions without usurping the function of the jury.”

Nonetheless, we conclude that the composers have failed to identify any conduct of Dees that is sufficiently blameworthy. For example, Fisher and Segal fault Dees for using the song after Fisher expressly refused him permission to do so. In their view, this shows bad faith on Dees's part. We cannot agree. Parodists will seldom get permission from those whose works are parodied. Self-esteem is seldom strong enough to permit the granting of permission even in exchange for a reasonable fee. See Note, The Parody Defense to Copyright Infringement: Productive Fair Use After Betamax, 97 Harv. L. Rev. 1395, 1397 n.12 (1984) [hereinafter cited as Parody Defense]. The parody defense to copyright infringement exists precisely to make possible a use that generally cannot be bought. See 3 M. Nimmer, supra, § 13.05[C], at 13-89; Gordon, Fair Use as Market Failure: A Structural and Economic Analysis of the Betamax Case and its Predecessors, 82 Colum. L. Rev. 1600, 1633 & n.177. Moreover, to consider Dees blameworthy because he asked permission would penalize him for this modest show of consideration. Even though such gestures are predictably futile, we refuse to discourage them.

The first fair-use factor section 107 directs courts to consider is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. § 107(1) (1982). The parties agree that the parody is a commercial use of the song. This fact "tends to weigh against a finding of fair use," Harper & Row, 105 S. Ct. at 2231, because "every commercial use of copyrighted material is presumptively an unfair exploitation of the monopoly privilege that belongs to the owner of the copyright." Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 451, 78 L. Ed. 2d 574, 104 S. Ct. 774 (1984). We recognize, however, that many parodies distributed commercially may be "more in the nature of an editorial or social commentary than . . . an attempt to capitalize financially on the plaintiff's original work." Milky Way Productions, 215 U.S.P.Q. at 131 (footnote omitted). In such cases, of which this is one, the initial presumption need not be fatal to the defendant's cause. The defendant can rebut the presumption by convincing the court that the parody does not unfairly diminish the economic value of the original. See id. & n.9.

The composers assert that the parody, although it borrows from the original work, was not "directed" at the original. That is, a humorous or satiric work deserves protection under the fair-use doctrine only if the copied work is at least partly the target of the work in question. See Walt Disney Productions v. Air Pirates, 581 F.2d 751, 758 n.15 (9th Cir. 1978), cert. denied, 439 U.S. 1132, 99 S. Ct. 1054, 59 L. Ed. 2d 94 (1979). Otherwise, there is no need to "conjure up" the original in the audience's mind and no justification for borrowing from it. Id; accord MCA, Inc. v. Wilson, 677 F.2d 180, 185 (2d Cir. 1981). We requested counsel to provide us with tapes of both Dees's parody and the original (as sung by Johnny Mathis). Although we have no illusions of musical expertise, it was clear to us that Dees's version was intended to poke fun at the composers' song, and at Mr. Mathis's rather singular vocal range. We reject the notion that the song was used merely as a vehicle to achieve a comedic objective unrelated to the song, its place and time. Cf. id. at 183-85 (purpose of saving the effort of composing original music); infra note 5.

This court has also consistently focused on the third fair-use factor -- the amount and substantiality of the taking, 17 U.S.C. § 107(3). See Air Pirates, 581 F.2d at 756. Thus far, however, we have provided few concrete guidelines; we have merely sketched the outer boundaries of the inquiry. On the one hand, "substantial copying by a defendant, combined with the fact that the portion copied constituted a substantial part of the defendant's work," does not automatically preclude the fair use defense. Id. On the other hand, "copying that is virtually complete or almost verbatim" will not be protected. Id. In Air Pirates, we ultimately based our analysis on the so-called "conjure up" test. See Air Pirates, 581 F.2d at 757 (citing Berlin v. E.C. Publications, Inc., 329 F.2d 541 (2d Cir.), cert. denied, 379 U.S. 822, 85 S. Ct. 46, 13 L. Ed. 2d 33, 144 U.S.P.Q. (BNA) 464 (1964), and Columbia Pictures Corp. v. National Broadcasting Co., 137 F. Supp. 348 (S.D. Cal. 1955)). As the Air Pirates opinion articulated it, the test asks "whether the parodist has appropriated a greater amount of the original work than is necessary to 'recall or conjure up' the object of his satire." Id. The composers interpret this test to limit the amount of permissible copying to that amount necessary to evoke only initial recognition in the listener. We disagree with this rigid view. As the Second Circuit stated in Elsmere Music, Inc. v. National Broadcasting Co., 623 F.2d 252 (2d Cir. 1980) (per curiam): The concept of "conjuring up" an original came into the copyright law not as a limitation on how much of an original may be used, but as a recognition that a parody frequently needs to be more than a fleeting evocation of an original in order to make its humorous point. A parody is entitled at least to "conjure up" the original. Id. at 253 n.1 (citation omitted). Air Pirates does not compel a different view. In that case -- which concerned the near-verbatim copying of Disney characters in the defendants' underground comic book -- we concluded that the defendants "took more than was necessary to place firmly in the reader's mind the parodied work and those specific attributes that [were] to be satirized," 581 F.2d at 758. We did not set a fixed limit on copying, but merely expressed our judgment that that particular parody could easily have been accomplished through more restricted means. We singled out three considerations that we thought important in determining whether a taking is excessive under the circumstances -- the degree of public recognition of the original work, the ease of conjuring up the original work in the chosen medium, and the focus of the parody. See Air Pirates, 581 F.2d at 757-58. Because the Disney characters were familiar and graphics was a relatively easy medium for parody, we concluded that close copying was impermissible. See id. But we expressly noted that media other than the graphic arts might justify greater leeway. We observed: "When the medium involved is a comic book, a recognizable caricature is not difficult to draw, so that an alternative that involves less copying is more likely to be available than if a speech, for instance, is parodied." Id. at 758. The unavailability of viable alternatives is evident in the present case. Like a speech, a song is difficult to parody effectively without exact or near-exact copying. If the would-be parodist varies the music or meter of the original substantially, it simply will not be recognizable to the general audience. This "special need for accuracy," provides some license for "closer" parody. See id. To be sure, that license is not limitless: the parodist's desire to make the best parody must be "balanced against the rights of the copyright owner in his original expressions." Id. We think the balance tips in the parodists' favor here. In view of the parody's medium, its purposes, and its brevity, it takes no more from the original than is necessary to accomplish reasonably its parodic purpose.

Thus, we must turn our attention to the fourth factor in the fair-use analysis -- "the effect of the use upon the potential market for or value of the copyrighted work," 17 U.S.C. § 107(4). This factor, not surprisingly, "is undoubtedly the single most important element of fair use." Harper & Row, 105 S. Ct. at 2234 (footnote omitted). In assessing the economic effect of the parody, the parody's critical impact must be excluded. Through its critical function, a "parody may quite legitimately aim at garroting the original, destroying it commercially as well as artistically." B. Kaplan, An Unhurried View of Copyright 69 (1967). Copyright law is not designed to stifle critics. "'Destructive' parodies play an important role in social and literary criticism and thus merit protection even though they may discourage or discredit an original author." Parody Defense, 97 Harv. L. Rev. at 1411. Accordingly, the economic effect of a parody with which we are concerned is not its potential to destroy or diminish the market for the original -- any bad review can have that effect -- but rather whether it fulfills the demand for the original. Biting criticism suppresses demand; copyright infringement usurps it. Thus, infringement occurs when a parody supplants the original in markets the original is aimed at, or in which the original is, or has reasonable potential to become, commercially valuable. See, e.g., Air Pirates, 581 F.2d at 756; Berlin v. E.C. Publications, Inc., 329 F.2d 541, 545 (2d Cir.), cert. denied, 379 U.S. 822, 85 S. Ct. 46, 13 L. Ed. 2d 33 (1964); Parody Defense, supra, at 1409-11. This is not a case in which commercial substitution is likely. "When Sunny Gets Blue" is "a lyrical song concerning or relating to a woman's feelings about lost love and her chance for . . . happiness again." Appellants' Opening Brief at 3. By contrast, the parody is a 29-second recording concerning a woman who sniffs glue, which "ends with noise and laughter mixed into the song." Id. at 7. We do not believe that consumers desirous of hearing a romantic and nostalgic ballad such as the composers' song would be satisfied to purchase the parody instead. Nor are those fond of parody likely to consider "When Sunny Gets Blue" a source of satisfaction. The two works do not fulfill the same demand. Consequently, the parody has no cognizable economic effect on the original.

1. Purpose of the use. Hamilton's book fits comfortably within several of the statutory categories of uses illustrative of uses that can be fair. The book may be considered "criticism," "scholarship," and "research." See 17 U.S.C. § 107. The proposed use is not an attempt to rush to the market just ahead of the copyright holder's imminent publication, as occurred in Harper & Row. Whether Random House plans to "exploit the headline value of its infringement," Harper & Row, supra, 471 U.S. at 561, as The Nation did, is not clear on the record thus far developed. Though no evidence has yet been presented on the advertising materials Random House plans to use, it is hard to believe that some emphasis will not be placed upon the fact that the book draws generously upon Salinger's unpublished letters. We agree with Judge Leval that Hamilton's purpose in using the Salinger letters to enrich his scholarly biography weighs the first fair use factor in Hamilton's favor, notwithstanding that he and his publisher anticipate profits. However, we do not agree that a biographer faces a dilemma that entitles him to a generous application of the fair use doctrine. Judge Leval perceived the dilemma in these terms: To the extent [the biographer] quotes (or closely paraphrases), he risks a finding of infringement and an injunction effectively destroying his biographical work. To the extent he departs from the words of the letters, he distorts, sacrificing both accuracy and vividness of description. 650 F. Supp. at 424. This dilemma is not faced by the biographer who elects to copy only the factual content of letters. The biographer who copies only facts incurs no risk of an injunction; he has not taken copyrighted material. And it is unlikely that the biographer will distort those facts by rendering them in words of his own choosing. On the other hand, the biographer who copies the letter writer's expression of facts properly faces an unpleasant choice. If he copies more than minimal amounts of (unpublished) expressive content, he deserves to be enjoined; if he "distorts" the expressive content, he deserves to be criticized for "sacrificing accuracy and vividness." But the biographer has no inherent right to copy the "accuracy" or the "vividness" of the letter writer's expression. Indeed, "vividness of description" is precisely an attribute of the author's expression that he is entitled to protect. The point is sharply, though unwittingly, made by defendant Hamilton in the course of his deposition in this case. On cross-examination, he is pressed as to why he copied a stylistic device that Salinger had employed in one of the letters. He responds: "I wanted to convey the fact that [Salinger] was adopting an ironic term. . . ." When the cross-examiner asks, "Couldn't you have stated that he had an ironic tone," Hamilton replies, "That would make a pedestrian sentence I didn't wish to put my name to." (A-925c). But when dealing with copyrighted expression, a biographer (or any other copier) may frequently have to content himself with reporting only the fact of what his subject did, even if he thereby pens a "pedestrian" sentence. The copier is not at liberty to avoid "pedestrian" reportage by appropriating his subject's literary devices. In sum, we agree with the District Court that the first fair use factor weighs in Hamilton's favor, but not that the purpose of his use entitles him to any special consideration.

2. Nature of the Copyrighted Work. "The fact that a work is unpublished is a critical element of its 'nature '". Harper & Row, supra, 471 U.S. at 564. Salinger's letters are unpublished, and they have not lost that attribute by their placement in libraries where access has been explicitly made subject to observance of at least the protections of copyright law. In considering this second factor, we encounter some ambiguity arising from the Supreme Court's observation that "the scope of fair use is narrower with respect to unpublished works." Id. (emphasis added). This could mean either that the circumstances in which copying will be found to be fair use are fewer in number for unpublished works than for published works or that the amount of copyrighted material that may be copied as fair use is a lesser quantity for unpublished works than for published works. Some support for the latter view can be derived from the statement in Harper & Row that, though "substantial" quotations might be used in a review of a published work, the author's right to control first publication weighs against "such use" prior to publication. Id. However, we think that the tenor of the Court's entire discussion of unpublished works conveys the idea that such works normally enjoy complete protection against copying any protected expression. Narrower "scope" seems to refer to the diminished likelihood that copying will be fair use when the copyrighted material is unpublished. The District Judge considered the nature of the copyrighted work, especially its unpublished nature, primarily in rejecting the plaintiff's argument that fair use was inapplicable to unpublished works. However, in analyzing and weighing the fair use factors, Judge Leval gave no explicit consideration to this second factor. Since the copyrighted letters are unpublished, the second factor weighs heavily in favor of Salinger.

3. Amount and Substantiality of the Portion Used. It is with regard to this third factor that we have the most serious disagreement with the District Judge's legal analysis, both as to the pertinent standard and its application. As to the standard, we start, as did Judge Leval, by recognizing that what is relevant is the amount and substantiality of the copyrighted expression that has been used, not the factual content of the material in the copyrighted works. However, that protected expression has been "used" whether it has been quoted verbatim or only paraphrased. See Nichols v. Universal Pictures Corp., 45 F.2d 119, 121 (2d Cir. 1930), cert. denied, 282 U.S. 902, 51 S. Ct. 216, 75 L. Ed. 795 (1931); 3 Nimmer, supra, § 13.03[A] at 13-20.1. We cannot be certain that Judge Leval included close paraphrases of the Salinger letters in his determination of the quantity of copyrighted material that has been used in the Hamilton biography. At one point in his decision, Judge Leval states that he has identified "approximately 30 letters from which Hamilton has taken, either by brief quotation or paraphrase, a few words of copyright protected material." 650 F. Supp. at 420 (emphasis added). At other points, however, his opinion suggests that only direct quotations have been counted. Noting that Hamilton had revised the May galleys "by eliminating and writing around most of the quoted matter" previously appearing in that version of the book, Judge Leval concluded that what Hamilton accomplished "was to reduce from a large number to fewer than 30 the instances of use of copyrighted expression." Id. at 423. This strongly suggests that passages quoted in the May galleys but paraphrased in the October galleys were not counted as instances of use of copyrighted expression. Moreover, of the 59 passages of the Hamilton biography complained of by Salinger, 24 contain direct quotations from the letters, perhaps constituting the "fewer than 30" instances the District Judge had in mind. Furthermore, Judge Leval states, "In the rarest case, a complete sentence is taken," id. at 423, citing to a passage in the October galleys that contains a full sentence quoted from one of Salinger's letters. If the District Judge had included paraphrases, he would have noticed numerous instances where the quantity of material taken from a single letter greatly exceeded a single sentence. Even if Judge Leval included paraphrases, we conclude that he misapplied the governing standard in determining the number of instances in which Hamilton has used Salinger's copyrighted expression. The District Judge rejected Salinger's claim of infringement as to several passages of the letters because they "employ[ed] a cliche or a word-combination that is so ordinary that it does not qualify for the copyright law's protection." Id. at 419. Though a cliche or an "ordinary" word-combination by itself will frequently fail to demonstrate even the minimum level of creativity necessary for copyright protection, see, e.g., Alberto-Culver Co. v. Andrea Dumon, Inc., 466 F.2d 705, 711 (7th Cir. 1972); W. Patry, Latman's The Copyright Law, supra, at 23-24, such protection is available for the "association, presentation, and combination of the ideas and thought which go to make up the [author's] literary composition." Nutt v. National Institute Inc. for the Improvement of Memory, 31 F.2d 236, 237 (2d Cir. 1929). Or as we have more recently stated, "What is protected is the manner of expression, the author's analysis or interpretation of events, the way he structures his material and marshals facts, his choice of words and the emphasis he gives to particular developments." Wainwright Securities, Inc. v. Wall Street Transcript Corp., 558 F.2d 91, 95-96 (2d Cir. 1977), cert. denied, 434 U.S. 1014, 98 S. Ct. 730, 54 L. Ed. 2d 759, 196 U.S.P.Q. (BNA) 864 (1978). The "ordinary" phrase may enjoy no protection as such, but its use in a sequence of expressive words does not cause the entire passage to lose protection. And though the "ordinary" phrase may be quoted without fear of infringement, a copier may not quote or paraphrase the sequence of creative expression that includes such a phrase. In almost all of those instances where the quoted or paraphrased passages from Salinger's letters contain an "ordinary" phrase, the passage as a whole displays a sufficient degree of creativity as to sequence of thoughts, choice of words, emphasis, and arrangement to satisfy the minimal threshold of required creativity. And in all of the instances where that minimum threshold is met, the Hamilton paraphrasing tracks the original so closely as to constitute infringement. We have carefully analyzed all 59 of the passages from Hamilton's book cited by Salinger as instances of infringing copying from 44 of his letters. Of these 44 letters, the Hamilton biography copies (with some use of quotation or close paraphrase) protected sequences constituting at least one-third of 17 letters and at least 10 percent of 42 letters. These sequences are protected, notwithstanding that they include some reporting of facts and an occasional use of a commonplace word or expression. Hamilton's use of these sequences "exceeds that necessary to disseminate the facts," Harper & Row, supra, 471 U.S. at 564. Judge Leval found that "in the rarest case a complete sentence was taken." 650 F. Supp. at 423. That is true only with respect to material directly quoted. The material closely paraphrased frequently exceeds ten lines from a single letter. Even if in one or two instances the portions of the letters copied could be said to lack sufficient creativity to warrant copyright protection, there remains sufficient copying of protected material to constitute a very substantial appropriation. The taking is significant not only from a quantitative standpoint but from a qualitative one as well. The copied passages, if not the "'heart of the book,'" Harper & Row, supra, 471 U.S. at 565 (quoting the District Court's opinion, 557 F. Supp. at 1072 (1983)), are at least an important ingredient of the book as it now stands. To a large extent, they make the book worth reading. The letters are quoted or paraphrased on at least 40 percent of the book's 192 pages. In sum, the third fair use factor weighs heavily in Salinger's favor.

4. Effect on the Market. The Supreme Court has called the fourth factor -- effect on the market for the copyrighted work -- "the single most important element of fair use." Harper & Row, supra, 471 U.S. at 566 (footnote omitted). As Judge Leval recognized, the need to assess the effect on the market for Salinger's letters is not lessened by the fact that their author has disavowed any intention to publish them during his lifetime. First, the proper inquiry concerns the "potential market" for the copyrighted work, 17 U.S.C. § 107(4); see Pacific and Southern Co. v. Duncan, 744 F.2d 1490, 1496-97 (11th Cir. 1984), cert. denied, 471 U.S. 1004, 85 L. Ed. 2d 161, 105 S. Ct. 1867 (1985). Second, Salinger has the right to change his mind. He is entitled to protect his opportunity to sell his letters, an opportunity estimated by his literary agent to have a current value in excess of $500,000. Proceeding from his conclusion that only a few fragments of the letters have been used in Hamilton's book, Judge Leval expressed the view that such use would have "no effect" on the marketability of the letters. 650 F. Supp. at 425. Concluding as we do that substantial portions of the letters have been copied, we do not share the District Judge's view that marketability of the letters will be totally unimpaired. To be sure, the book would not displace the market for the letters. Indeed, we think it likely that most of the potential purchasers of a collection of the letters would not be dissuaded by publication of the biography. Yet some impairment of the market seems likely. The biography copies virtually all of the most interesting passages of the letters, including several highly expressive insights about writing and literary criticism. Perhaps few readers of the biography would refrain from purchasing a published collection of the letters if they appreciated how inadequately Hamilton's paraphrasing has rendered Salinger's chosen form of expression. The difficulty, however, is that some readers of the book will gain the impression that they are learning from Hamilton what Salinger has written. Hamilton frequently laces his paraphrasing with phrases such as "he wrote," "said Salinger," "he speaks of," "Salinger declares," "he says," and "he said." For at least some appreciable number of persons, these phrases will convey the impression that they have read Salinger's words, perhaps not quoted verbatim, but paraphrased so closely as to diminish interest in purchasing the originals. The fourth fair use factor weighs slightly in Salinger's favor.

"Though a cliche or an "ordinary" word-combination by itself will frequently fail to demonstrate even the minimum level of creativity necessary for copyright protection, see, e.g., Alberto-Culver Co. v. Andrea Dumon, Inc., 466 F.2d 705, 711 (7th Cir. 1972); W. Patry, Latman's The Copyright Law, supra, at 23-24, such protection is available for the "association, presentation, and combination of the ideas and thought which go to make up the [author's] literary composition." Nutt v. National Institute Inc. for the Improvement of Memory, 31 F.2d 236, 237 (2d Cir. 1929). Or as we have more recently stated, "What is protected is the manner of expression, the author's analysis or interpretation of events, the way he structures his material and marshals facts, his choice of words and the emphasis he gives to particular developments." Wainwright Securities, Inc. v. Wall Street Transcript Corp., 558 F.2d 91, 95-96 (2d Cir. 1977), cert. denied, 434 U.S. 1014, 98 S. Ct. 730, 54 L. Ed. 2d 759 (1978). The "ordinary" phrase may enjoy no protection as such, but its use in a sequence of expressive words does not cause the entire passage to lose protection. And though the "ordinary" phrase may be quoted without fear of infringement, a copier may not quote or paraphrase the sequence of creative expression that includes such a phrase."

The Court of Appeals determined that the use of Woolrich's story in petitioners' motion picture was not fair use. We agree. The motion picture neither falls into any of the categories enumerated in § 107 nor meets the four criteria set forth in § 107. "[E]very [unauthorized] commercial use of copyrighted material is presumptively an unfair exploitation of the monopoly privilege that belongs to the owner of the copyright." Sony Corp. of America v. Universal Studios, Inc., supra, at 451. Petitioners received $ 12 million from the re-release of the motion picture during the renewal term. 863 F. 2d, at 1468. Petitioners asserted before the Court of Appeals that their use was educational rather than commercial. The Court of Appeals [1628] found nothing in the record to support this assertion, nor do we.

Applying the second factor, the Court of Appeals pointed out that "[a] use is less likely to be deemed fair when the copyrighted work is a creative product." 863 F. 2d, at 1481 (citing Brewer v. Hustler Magazine, Inc., 749 F. 2d 527, 529 (CA9 1984)). In general, fair use is more likely to be found in factual works than in fictional works. See 3 Nimmer § 13.05[A], pp. 13-77 to 13-78 ("[A]pplication of the fair use defense [is] greater . . . in the case of factual works than in the case of works of fiction or fantasy"); cf. Harper & Row, supra, at 563 ("The law generally recognizes a greater need to diseminate factual works than works of fiction or fantasy"). A motion picture based on a fictional short story obviously falls into the latter category.

Examining the third factor, the Court of Appeals determined that the story was a substantial portion of the motion picture. See id., at 564-565 (finding unfair use where quotation from book "'took what was essentially the heart of the book'"). The motion picture expressly uses the story's unique setting, characters, plot, and sequence of events. Petitioners argue that the story constituted only 20% of the motion picture's story line, Brief for Petitioners 40, n. 69, but that does not mean that a substantial portion of the story was not used in the motion picture. "[A] taking may not be excused merely because it is insubstantial with respect to the infringing work." Harper & Row, supra, at 565.

The fourth factor is the "most important, and indeed, central fair use factor." 3 Nimmer § 13.05[A], p. 13-81. The record supports the Court of Appeals' conclusion that re-release of the film impinged on the ability to market new versions of the story. Common sense would yield the same conclusion. Thus, all four factors point to unfair use. "This case presents a classic example of an unfair use: a commercial use of a fictional story that adversely affects the story owner's adaptation rights." 863 F. 2d, at 1482.

As a general matter, and to varying degrees, copyright protection extends beyond a literary work's strictly textual form to its non-literal components. As we have said, "it is of course essential to any protection of literary property . . . that the right cannot be limited literally to the text, else a plagiarist would escape by immaterial variations." Nichols v. Universal Pictures Co., 45 F.2d 119, 121 (2d Cir. 1930) (L. Hand, J.), cert. denied, 282 U.S. 902, 75 L. Ed. 795, 51 S. Ct. 216 (1931). Thus, where "the fundamental essence or structure of one work is duplicated in another," 3 Nimmer, § 13.03[A][1], at 13-24, courts have found copyright infringement. See, e.g., Horgan v. Macmillan, 789 F.2d 157, 162 (2d Cir. 1986) (recognizing that a book of photographs might infringe ballet choreography); Twentieth Century-Fox Film Corp. v. MCA, Inc., 715 F.2d 1327, 1329 (9th Cir. 1983) (motion picture and television series); Sid & Marty Krofft Television Prods., Inc. v. McDonald's Corp., 562 F.2d 1157, 1167 (9th Cir. 1977) (television commercial and television series); Sheldon v. Metro-Goldwyn Pictures Corp., 81 F.2d 49, 55 (2d Cir.), cert. denied, 298 U.S. 669, 80 L. Ed. 1392, 56 S. Ct. 835 (1936) (play and motion picture); accord Stewart v. Abend, 495 U.S. 207, 238, 109 L. Ed. 2d 184, 110 S. Ct. 1750 (1990) (recognizing that motion picture may infringe copyright in book by using its "unique setting, characters, plot, and sequence of events"). This black letter proposition is the springboard for our discussion.

Moving to the district court's evaluation of OSCAR 3.5's structural components, we agree with Judge Pratt's systematic exclusion of non-protectable expression. With respect to code, the district court observed that after the rewrite of OSCAR 3.4 to OSCAR 3.5, "there remained virtually no lines of code that were identical to ADAPTER." Id. at 561. Accordingly, the court found that the code "presented no similarity at all." Id. at 562. Next, Judge Pratt addressed the issue of similarity between the two programs' parameter lists and macros. He concluded that, viewing the conflicting evidence most favorably to CA, it demonstrated that "only a few of the lists and macros were similar to protected elements in ADAPTER; the others were either in the public domain or dictated by the functional demands of the program." Id. As discussed above, functional elements and elements taken from the public domain do not qualify for copyright protection. With respect to the few remaining parameter lists and macros, the district court could reasonably conclude that they did not warrant a finding of infringement given their relative contribution to the overall program. See Warner Bros., Inc. v. American Broadcasting Cos., Inc., 720 F.2d 231, 242 (2d Cir. 1983) (discussing de minimis exception which allows for literal copying of a small and usually insignificant portion of the plaintiff's work); 3 Nimmer § 13.03[F][5], at 13-74. In any event, the district court reasonably found that, for lack of persuasive evidence, CA failed to meet its burden of proof on whether the macros and parameter lists at issue were substantially similar. See Computer Assocs., 775 F. Supp. at 562. The district court also found that the overlap exhibited between the list of services required for both ADAPTER and OSCAR 3.5 was "determined by the demands of the operating system and of the applications program to which it [was] to be linked through ADAPTER or OSCAR. . . ." Id. In other words, this aspect of the program's structure was dictated by the nature of other programs with which it was designed to interact and, thus, is not protected by copyright. Finally, in his infringement analysis, Judge Pratt accorded no weight to the similarities between the two programs' organizational charts, "because [the charts were] so simple and obvious to anyone exposed to the operation of the programs." Id. CA argues that the district court's action in this regard "is not consistent with copyright law"--that "obvious" expression is protected, and that the district court erroneously failed to realize this. However, to say that elements of a work are "obvious," in the manner in which the district court used the word, is to say that they "follow naturally from the work's theme rather than from the author's creativity." 3 Nimmer § 13.03[F][3], at 13-65. This is but one formulation of the scenes a faire doctrine, which we have already endorsed as a means of weeding out unprotectable expression. CA argues, at some length, that many of the district court's factual conclusions regarding the creative nature of its program's components are simply wrong. Of course, we are limited in our review of factual findings to setting aside only those that we determine are clearly erroneous. See Fed. R. Civ. P. 52. Upon a thorough review of the voluminous record in this case, which is comprised of conflicting testimony and other highly technical evidence, we discern no error on the part of Judge Pratt, let alone clear error. Since we accept Judge Pratt's factual conclusions and the results of his legal analysis, we affirm his dismissal of CA's copyright infringement claim based upon OSCAR 3.5. We emphasize that, like all copyright infringement cases, those that involve computer programs are highly fact specific. The amount of protection due structural elements, in any given case, will vary according to the protectable expression found to exist within the program at issue.

D. Legal Analysis of Substantial Similarity In the Krofft case, we established a two-part test for substantial similarity. 562 F.2d at 1164. Under the original formulation of Krofft's "extrinsic" component, a plaintiff needed to prove only the "similarity of ideas" in the two programs. Id. Today, however, the extrinsic test looks at more than just the similarity of ideas. As the district court understood, the extrinsic analysis is " 'an objective test which rests upon specific criteria that can be listed and analyzed.' " 12 U.S.P.Q.2D (BNA) at 1994 (quoting Data East USA, Inc. v. Epyx, Inc., 862 F.2d 204, 208 (9th Cir. 1988)). The extrinsic test thus has become an "objective . . . analysis of expression." Shaw, 919 F.2d at 1357 (original emphasis); see also 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.03[E]3, at 13-78.19 to -78.20 (1991) (hereinafter, "Nimmer") (discussing reformulation of Krofft analysis achieved in Shaw). To establish infringement of a copyright, the two works in question must also meet the second component of the test set forth in Krofft, the "intrinsic test." 562 F.2d at 1164. The intrinsic test, according to Krofft, should measure "substantial similarity in expressions . . . depending on the response of the ordinary reasonable person….It does not depend on the type of external criteria and analysis which marks the extrinsic test." Id. (citations omitted). In applying the intrinsic test, therefore, "analytic dissection and expert testimony are not appropriate." Id. Brown Bag contends that the district court misapplied the intrinsic test because the court improperly performed "analytic dissection" in comparing the expressions of ideas embodied in the two programs. Indeed, the district court explicitly performed "analytic dissection," comparing specific screens, menus, and keystrokes, in part to detect substantial similarity. 12 U.S.P.Q.2D (BNA) at 1995-96. The court performed the dissection at Brown Bag's behest and in response to a list of allegedly substantially similar features Brown Bag proffered in opposition to summary judgment. Brown Bag's argument ignores the evolution of the "extrinsic" component of the Krofft analysis. Properly understood, the district court's analysis was not a misapplication of the "intrinsic test," but a proper application of the revised "extrinsic test." Under the reformulated extrinsic test, we mean to perpetuate "analytic dissection" as a tool for comparing not only ideas but also expression. Thus, it was not error for the district court to use "analytic dissection" in comparing the expressions of the two programs. Specifically, the district court properly compared the expressions embodied in the two screens presenting the programs' respective opening menus. See 12 U.S.P.Q.2D (BNA) at 1996. The district court's analytic dissection of the five groups of features involving unprotected or unprotectable work was also proper, but for an entirely different reason. This analysis had nothing to do with assessing substantial similarity, and therefore was not subject to the old limitations Krofft imposed on the use of analytic dissection. Instead, as to these groups, the district court applied analytic dissection for the purpose of determining "whether similarities [between the programs] result from unprotectable [or unprotected] expression." Data East, 862 F.2d at 208. In this application, analytic dissection is used not for the purposes of comparing similarities and identifying infringement, but for the purpose of defining the scope of plaintiff's copyright. This point bears emphasis. Analytic dissection is relevant not only to the copying element of a copyright infringement claim, but also to the claim's ownership element. One aspect of the ownership element is the copyrightability of the subject matter and, more particularly, the scope of whatever copyright lies therein. See Nimmer § 13.01[A]. To the extent a plaintiff's work is unprotected or unprotectable under copyright, the scope of the copyright must be limited. See, e.g., Data East, 862 F.2d at 209 (holding that, because "the similarities result from unprotectable expression," "the lower court erred by not limiting the scope of [plaintiff's] copyright protection"). Thus, where two works are found to be similar without regard to the scope of the copyright in the plaintiff's work, Data East teaches that the source of the similarity must be identified and a determination made as to whether this source is covered by plaintiff's copyright. It is this process in which the district court engaged with regard to the five groups of features found to be unprotected or unprotectable by Brown Bag's copyright in PC-Outline. Again, this use of analytic dissection was proper. Brown Bag further argues that the district court became so engrossed in its analytic dissection that it neglected to evaluate the "overall look and feel" of the two programs for substantial similarity. Stated differently, Brown Bag contends that the district court failed to consider the "intrinsic" component of the Krofft test, which involves a subjective analysis of similarities in expression. See Shaw, 919 F.2d at 1357. Even assuming this is true, we find no reversible error because the record fails to include any evidence indicating that Brown Bag requested the district court to make this analysis. Although the district court "entertained an in-court demonstration of the two programs and . . . carefully reviewed submitted copies of relevant outlining programs," 12 U.S.P.Q.2D (BNA) at 1992, no such evidence was ever included as part of Brown Bag's opposition to Symantec's summary judgment motion. Therefore, there is no reason we should expect the district court to have made the analysis Brown Bag now requests. Finally, Brown Bag argues that, because PC-Outline and Grandview shared a common idea - specifically the idea of an outlining program - summary judgment is precluded by our decision in Shaw. We held in Shaw that "satisfaction of the extrinsic test creates a triable issue of fact in a copyright action involving a literary work." 919 F.2d at 1359. In response, Symantec argues we should distinguish Shaw from the instant case on the basis that the Shaw rule applies by its own terms only to literary works. We decline to distinguish Shaw for that reason, however, because at least some computer programs bear significant similarities to literary works. On the scant record available to us in this case, we cannot conclude as a matter of law that the programs at issue here, or computer programs in general, are so nonliterary or so limited in their variety of expression as to avoid application of the rule announced in Shaw. We nevertheless reject Brown Bag's argument on the basis that it misapprehends the nature of the "extrinsic" test following Shaw. If the rule in Shaw was applied without recognizing that the extrinsic test now includes some analysis of expression, Shaw would ring "the death-knell in the Ninth Circuit for defense summary judgments based on purported absence of substantial similarity." Nimmer § 13.03[E]3, at 13-78.20. But, as we have discussed, after Shaw the extrinsic test examines not only similarity of ideas, but also objective similarity of expressions. In particular, the extrinsic test for literary works requires analytic dissection of several "objective components of expression," Shaw, 919 F.2d at 1361, within a literary work such as plot, theme, characters, and dialogue. Id. at 1356-57. Similarity of these objective components in two literary works logically gives rise to a triable issue of similarity. Id. at 1357. Likewise, computer programs are subject to a Shawtype analytic dissection of various standard components, e.g., screens, menus, and keystrokes. Because the district court found that Brown Bag made no showing of similarity along these lines with regard to copyrighted components of PC-Outline, summary judgment was not precluded by the rule announced in Shaw. See also Nimmer § 13.03[E]3, at 13-78.20 ("given Shaw's reformulation of the extrinsic test, the decision is better viewed as standing for the unremarkable proposition that when objectively speaking a triable case of expressive similarity exists, the case must be sent to the trier of fact; but when no such objective argument is possible, summary judgment will lie.") (footnotes omitted).

At this stage in the proceedings of this case, Nintendo has made a sufficient showing that its 10NES program contains protectable expression. After filtering unprotectable elements out of the 10NES program, this court finds no error in the district court's conclusion that 10NES contains protectable expression. Nintendo independently created the 10NES program and exercised creativity in the selection and arrangement of its instruction lines. See Bellsouth Advertising & Publishing v. Donnelly Info. Publishing, 933 F.2d 952, 957 (11th Cir. 1991) (selection, coordination, or arrangement of information constitutes originality). The security function of the program necessitated an original signal combination to act as a lock and key for the NES console. To generate an original signal, Nintendo had to design an original program. In sum, the district court properly discerned that the 10NES program contains protectable expression. At a minimum, Nintendo may protect under copyright the unique and creative arrangement of instructions in the 10NES program.

Fair use to discern a work's ideas, however, does not justify extensive efforts to profit from replicating protected expression. Subparagraphs 1 and 4 of section 107 clarify that the fair use in intermediate copying does not extend to commercial exploitation of protected expression. Sony Corp., 464 U.S. at 451. The fair use reproductions of a computer program must not exceed what is necessary to understand the unprotected elements of the work. This limited exception is not an invitation to misappropriate protectable expression. Any reproduction of protectable expression must be strictly necessary to ascertain the bounds of protected information within the work. In this case, the source code obtained from the Copyright Office facilitated Atari's intermediate copying of the 10NES program. To invoke the fair use exception, an individual must possess an authorized copy of a literary work. See Harper & Row, 471 U.S. at 562-63 (Knowing exploitation of purloined manuscript not compatible with "good faith" and "fair dealings" underpinnings of fair use doctrine.). Because Atari was not in authorized possession of the Copyright Office copy of 10NES, any copying or derivative copying of 10NES source code from the Copyright Office does not qualify as a fair use. Reverse engineering, untainted by the purloined copy of the 10NES program and necessary to understand 10NES, is a fair use. An individual cannot even observe, let alone understand, the object code on Nintendo's chip without reverse engineering. Atari retrieved this object code from NES security chips in its efforts to reverse engineer the 10NES program. Atari chemically removed layers from Nintendo's chips to reveal the 10NES object code. Through microscopic examination of the "peeled" chip, Atari engineers transcribed the 10NES object code into a handwritten list of ones and Zeros. While these ones and zeros represent the configuration of machine readable software, the ones and zeros convey little, if any, information to the normal unaided observer. Atari then keyed this handwritten copy into a computer. The computer then "disassembled" the object code or otherwise aided the observer in understanding the program's method or functioning. This "reverse engineering" process, to the extent untainted by the 10NES copy purloined from the Copyright Office, qualified as a fair use. The district court assumed that reverse engineering (intermediate copying) was copyright infringement. Atari Games v. Nintendo of Am., Nos. 88-4805, 89-0027, 89-0824, slip op. at 11-13 (N.D. Cal. Apr. 11, 1991). This court disagrees. Atari did not violate Nintendo's copyright by deprocessing computer chips in Atari's rightful possession. Atari could lawfully deprocess Nintendo's 10NES chips to learn their unprotected ideas and processes. This fair use did not give Atari more than the right to understand the 10NES program and to distinguish the protected from the unprotected elements of the 10NES program. Any copying beyond that necessary to understand the 10NES program was infringement. Atari could not use reverse engineering as an excuse to exploit commercially or otherwise misappropriate protected expression.

IV. Trademark Issues Ordinarily in a trademark case, a trademark holder contends that another party is misusing the holder's mark or is attempting to pass off goods or services as those of the trademark holder. The other party usually protests that the mark is not being misused, that there is no actual confusion, or that for some other reason no violation has occurred. This case is different. Here, both parties agree that there is a misuse of a trademark, both agree that there is unlawful mislabeling, and both agree that confusion may result. The issue, here, is - which party is primarily responsible? Which is the wrongdoer - the violator? Is it Sega, which has adopted a security system governing access to its Genesis III console that displays its trademark and message whenever the initialization code for the security system is utilized, even when the video game program was manufactured by a Sega competitor? Or is it Accolade, which, having discovered how to gain access to the Genesis III through the initialization code, uses that code even though doing so triggers the display of Sega's trademark and message in a manner that leads observers to believe that Sega manufactured the Accolade game cartridge? In other words, is Sega the injured party because its mark is wrongfully attached to an Accolade video game by Accolade? Or is Accolade wronged because its game is mislabeled as a Sega product by Sega? The facts are relatively straightforward and we have little difficulty answering the question. Sega's trademark security system (TMSS) initialization code not only enables video game programs to operate on the Genesis III console, but also prompts a screen display of the SEGA trademark and message. As a result, Accolade's inclusion of the TMSS initialization code in its video game programs has an effect ultimately beneficial neither to Sega nor to Accolade. A Genesis III owner who purchases a video game made by Accolade sees Sega's trademark associated with Accolade's product each time he inserts the game cartridge into the console. Sega claims that Accolade's inclusion of the TMSS initialization code in its games constitutes trademark infringement and false designation of origin in violation of sections 32(1)(a) and 43(a) of the Lanham Trademark Act, 15 U.S.C. §§ 1114(1)(a), 1125(a), respectively. Accolade counterclaims that Sega's use of the TMSS to prompt a screen display of its trademark constitutes false designation of origin under Lanham Act section 43(a), 15 U.S.C. § 1125(a). Because the TMSS has the effect of regulating access to the Genesis III console, and because there is no indication in the record of any public or industry awareness of any feasible alternate method of gaining access to the Genesis III, we hold that Sega is primarily responsible for any resultant confusion. Thus, it has not demonstrated a likelihood of success on the merits of its Lanham Act claims. Accordingly, the preliminary injunction it obtained must be dissolved with respect to the trademark claim also. However, we decline to instruct the district court to grant Accolade's request for preliminary injunctive relief at this time. The decision whether to grant such relief requires the making of factual and equitable determinations in light of the legal conclusions we express here. Such determinations are best left in the first instance to the district court. A. False Labeling Section 32(1)(a) of the Lanham Act creates a cause of action for trademark infringement against any person who, without the consent of the trademark owner, "uses in commerce any reproduction . . . of a registered mark in connection with the sale, offering for sale, distribution, or advertising of any goods or services on or in connection with which such use is likely to cause confusion, or to cause mistake, or to deceive. . . ." 15 U.S.C. § 1114(1)(a). Section 43(a) proscribes the use in commerce of a false designation of origin in connection with goods or services where such use is "likely to cause confusion, or . . . mistake." Id. § 1125(a). Both Sega and Accolade agree that the screen display of the Sega trademark and message creates a likelihood of consumer confusion regarding the origin of Accolade's games. The question is: which party is legally responsible for that confusion? We disagree with the answer given by the district court. The district court found that Accolade bore primary responsibility for any consumer confusion that resulted from the display of the false Sega Message. However, Accolade had no desire to cause the Sega Message to appear or otherwise to create any appearance of association between itself and Sega; in fact, it had precisely the opposite wish. It used the TMSS initialization code only because it wanted to gain access for its products to the Genesis III, and was aware of no other method for doing so. On the other hand, while it may not have been Sega's ultimate goal to mislabel Accolade's products, the record is clear that the false labeling was the result of a deliberate decision on the part of Sega to include in the Genesis III a device which would both limit general access and cause false labeling. The decision to use the SEGA trademark as an essential element of a functional device that regulates access and to cause the SEGA trademark and message to be displayed whenever that functional device was triggered compels us to place primary responsibility for consumer confusion squarely on Sega. With respect to Accolade, we emphasize that the record clearly establishes that it had only one objective in this matter: to make its video game programs compatible with the Genesis III console. That objective was a legitimate and a lawful one. There is no evidence whatsoever that Accolade wished Sega's trademark to be displayed when Accolade's games were played on Sega's consoles. To the contrary, Accolade included disclaimers on its packaging materials which stated that "Accolade, Inc. is not associated with Sega Enterprises, Ltd." When questioned regarding the Sega Message and its potential effect on consumers, Alan Miller testified that Accolade does not welcome the association between its product and Sega and would gladly avoid that association if there were a way to do so. Miller testified that Accolade's engineers had not been able to discover any way to modify their game cartridges so that the games would operate on the Genesis III without prompting the screen display of the Sega Message. In contrast, Sega officials testified that Sega incorporated the TMSS into the Genesis console, known in Asia as the Mega-Drive, in order to lay the groundwork for the trademark prosecution of software pirates who sell counterfeit cartridges in Taiwan and South Korea, as well as in the United States. Sega then marketed the redesigned console worldwide. Sega intended that when Sega game programs manufactured by a counterfeiter were played on its consoles, the Sega Message would be displayed, thereby establishing the legal basis for a claim of trademark infringement. However, as Sega certainly knew, the TMSS also had the potential to affect legitimate competitors adversely. First, Sega should have foreseen that a competitor might discover how to utilize the TMSS, and that when it did and included the initialization code in its cartridges, its video game programs would also end up being falsely labeled. Sega should also have known that the TMSS might discourage some competitors from manufacturing independently developed games for use with the Genesis III console, because they would not want to become the victims of such a labeling practice. Thus, in addition to laying the groundwork for lawsuits against pirates, Sega knowingly risked two significant consequences: the false labeling of some competitors' products and the discouraging of other competitors from manufacturing Genesis-compatible games. Under the Lanham Act, the former conduct, at least, is clearly unlawful. "Trademark policies are designed '(1) to protect consumers from being misled . . . (2) to prevent an impairment of the value of the enterprise which owns the trademark; and (3) to achieve these ends in a manner consistent with the objectives of free competition.'" Anti-Monopoly, Inc. v. General Mills Fun Group, 611 F.2d 296, 300-01 (9th Cir. 1979) (quoting HMH Publishing Co. v. Brincat, 504 F.2d 713, 716 (9th Cir. 1974)). Sega violated the first and the third of these principles. "The trademark is misused if it serves to limit competition in the manufacture and sales of a product. That is the special province of the limited monopolies provided pursuant to the patent laws." Id. at 301 (citation omitted). Sega makes much of the fact that it did not adopt the TMSS in order to wage war on Accolade in particular, but rather as a defensive measure against software counterfeiters. It is regrettable that Sega is troubled by software pirates who manufacture counterfeit products in other areas of the world where adequate copyright remedies are not available. However, under the Lanham Act, which governs the use of trademarks and other designations of origin in this country, it is the effect of the message display that matters. Whatever Sega's intent with respect to the TMSS, the device serves to limit competition in the market for Genesis-compatible games and to mislabel the products of competitors. Moreover, by seeking injunctive relief based on the mislabeling it has itself induced, Sega seeks once again to take advantage of its trademark to exclude its competitors from the market. The use of a mark for such purpose is inconsistent with the Lanham Act. B. Functionality Sega argues that even if the legal analysis we have enunciated is correct, the facts do not support its application to this case. Specifically, Sega contends that the TMSS does not prevent legitimate unlicensed competitors from developing and marketing Genesis III-compatible cartridges that do not trigger a display of the Sega trademark and message. In other words, Sega claims that Accolade could have "engineered around" the TMSS. Accolade strongly disagrees with Sega's factual assertions. It contends that the TMSS initialization sequence is a functional feature that must be included in a video game program by a manufacturer in order for the game to operate on the Genesis III. Sega's factual argument stands or falls on the Nagashima declaration and the accompanying modified game cartridges that Sega introduced at the hearing. Having carefully reviewed the declaration, we conclude that Sega has not met its burden of establishing nonfunctionality. Based on the Nagashima declaration and on the modified cartridges, the district court concluded that the TMSS initialization sequence was not a necessary component of a Genesis-compatible game. The court found that Accolade could have created a game cartridge that lacked the TMSS initialization code but would still operate on the Genesis III, or could have programmed its games in such a way that the false Sega Message would not be displayed on the screen. The court further found that either modification could have been accomplished at minimal additional expense to Accolade. Accordingly, the court ruled that Accolade could not assert a functionality defense. The question whether a product feature is functional is a question of fact. Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 855, 72 L. Ed. 2d 606, 102 S. Ct. 2182 (1982). Determination of the correct legal standard to apply in evaluating functionality, however, is a question of law which we review de novo. Id. at 855 n.15. The burden of proving nonfunctionality is on Sega. See Rachel v. Banana Republic, Inc., 831 F.2d 1503, 1506 (9th Cir. 1987). In the case before us, we conclude that the district court's finding of nonfunctionality was based on its use of an incorrect legal standard. Viewed in the correct light, the record before us supports only one conclusion: The TMSS initialization code is a functional feature of a Genesis-compatible game and Accolade may not be barred from using it. "Functional features of a product are features 'which constitute the actual benefit that the consumer wishes to purchase, as distinguished from an assurance that a particular entity made, sponsored, or endorsed a product.'" Vuitton et Fils S.A. v. J. Young Enterprises, Inc., 644 F.2d 769, 774 (9th Cir. 1981) (quoting International Order of Job's Daughters v. Lindeburg & Co., 633 F.2d 912, 917 (9th Cir. 1980), cert. denied, 452 U.S. 941, 69 L. Ed. 2d 956, 101 S. Ct. 3086 (1981)). A product feature thus is functional "if it is essential to the use or purpose of the article or if it affects the cost or quality of the article." Inwood Laboratories, 456 U.S. at 850 n.10. The Lanham Act does not protect essentially functional or utilitarian product features because such protection would constitute a grant of a perpetual monopoly over features that could not be patented. Keene Corp. v. Paraflex Industries, Inc., 653 F.2d 822, 824 (3d Cir. 1981). Even when the allegedly functional product feature is a trademark, the trademark owner may not enjoy a monopoly over the functional use of the mark. Job's Daughters, 633 F.2d at 918-19. In determining whether a product feature is functional, a court may consider a number of factors, including - but not limited to - "the availability of alternative designs; and whether a particular design results from a comparatively simple or cheap method of manufacture." Clamp Mfg. Co. v. Enco Mfg. Co., Inc., 870 F.2d 512, 516 (9th Cir.), cert. denied, 493 U.S. 872107 L. Ed. 2d 155, 110 S. Ct. 202 (1989). The availability of alternative methods of manufacture must be more than merely theoretical or speculative, however. The court must find "that commercially feasible alternative configurations exist." Id. (emphasis added). Moreover, some cases have even suggested that in order to establish nonfunctionality the party with the burden must demonstrate that the product feature "'serves no purpose other than identification.'" Keene Corp., 653 F.2d at 826 (quoting SK&F Co. v. Premo Pharmaceutical Laboratories, Inc., 625 F.2d 1055, 1063 (3d Cir. 1980)). With these principles in mind, we turn to the question whether the TMSS initialization code is a functional feature of a Genesis-compatible game. It is indisputable that, in the case before us, part of "the actual benefit that the consumer wishes to purchase" is compatibility with the Genesis III console. The TMSS initialization code provides that compatibility. Sega argues that the modified cartridges that were introduced in the district court establish the actual existence of technically and commercially feasible alternative methods of gaining access to the Genesis III. The cartridges were prepared by Nagashima, an employee in Sega's Hardware Research and Development Department who was "familiar with the TMSS system". At most, the Nagashima affidavit establishes that an individual familiar with the operation of the TMSS can discover a way to engineer around it. It does not establish that a competitor with no knowledge of the workings of the TMSS could do so. Nor is there any evidence that there was any public or industry awareness of any alternate method for gaining access to the Genesis III. Evidence that an individual, even an independent expert, produced one or more cartridges is not sufficient proof that an alternate method exists. What is needed for proof of that fact is proof of the method itself. Here, such proof is totally lacking. What is also needed is proof that knowledge of the alternate method exists or is readily available to knowledgeable persons in the industry. That proof also is totally lacking here. Accordingly, the district court erred as a matter of law in concluding that the Nagashima declaration and the modified cartridges were sufficient to establish nonfunctionality. Because the TMSS serves the function of regulating access to the Genesis III, and because a means of access to the Genesis III console without using the TMSS initialization code is not known to manufacturers of competing video game cartridges, there is an insufficient basis for a finding of nonfunctionality. Moreover, we note that the only evidence in the record (other than the Nagashima declaration) relating to Accolade's ability to gain access to the Genesis III through the use of any process other than the TMSS is the affidavit of Alan Miller. Miller stated that Accolade's software engineers - who, absent any evidence to the contrary, we presume to be reasonably competent representatives of their profession - have not been able to discover such a method. This evidence supports our conclusion that Sega has not met its burden of establishing nonfunctionality. Sega argues that it is not required to share with Accolade or with any other competitor the secrets of how the TMSS works, and how to engineer around it. Sega is correct - the law does not require that it disclose its trade secrets to Accolade in connection with its effort to prevail on its Lanham Act claim, nor in connection with its effort to defend itself against Accolade's counterclaim. Nevertheless, a Lanham Act plaintiff is not entitled to prevail in litigation solely on the basis of unsupported assertions. Rather, it has a choice. It can take its chances and proceed to trial without the sensitive evidence. Alternatively, if it believes the evidence important to the resolution of the dispute, it may seek a protective order from the court pursuant to Federal Rule of Civil Procedure 26(c)(7) governing discovery. "The protective order is not a substitute for [evidence relevant to the merits]. Its purpose . . . is to prevent harm by limiting disclosure of relevant and necessary information." Micro Motion, Inc. v. Kane Steel Co., Inc., 894 F.2d 1318, 1325 (Fed. Cir. 1990) (emphasis omitted). Upon a showing that a protective order is warranted, see American Standard, Inc. v. Pfizer Inc., 828 F.2d 734, 739-44 (Fed. Cir. 1987), the court may restrict access to the disputed material to the opposing party's counsel, or may allow the parties to retain independent experts to evaluate material that is subject to the protective order. See, e.g., Safe Flight Instrument Corp. v. Sundstrand Data Control, Inc., 682 F. Supp. 20, 22 (D. Del. 1988) (listing cases). The latter solution is particularly helpful to the court in a case such as this one, in which the dispute is highly technical in nature. However, neither the district court nor Sega took advantage of this procedure. Thus there is no independent evidence to support the conclusion offered by Nagashima. In summary, because Sega did not produce sufficient evidence regarding the existence of a feasible alternative to the use of the TMSS initialization code, it did not carry its burden and its claim of nonfunctionality fails. Possibly, Sega will be able to meet its burden of proof at trial. We cannot say. However, we conclude that in light of the record before the district court, Sega was not entitled to preliminary injunctive relief under the Lanham Act. 10

With respect to the first statutory factor, we observe initially that the fact that copying is for a commercial purpose weighs against a finding of fair use. Harper & Row, 471 U.S. at 562. However, the presumption of unfairness that arises in such cases can be rebutted by the characteristics of a particular commercial use. Hustler Magazine, Inc. v. Moral Majority, Inc., 796 F.2d 1148, 1152 (9th Cir. 1986); see also Maxtone-Graham v. Burtchaell, 803 F.2d 1253, 1262 (2d Cir. 1986), cert. denied, 481 U.S. 1059, 95 L. Ed. 2d 856, 107 S. Ct. 2201 (1987). Further "the commercial nature of a use is a matter of degree, not an absolute. . . ." Maxtone-Graham, 803 F.2d at 1262. Sega argues that because Accolade copied its object code in order to produce a competing product, the Harper & Row presumption applies and precludes a finding of fair use. That analysis is far too simple and ignores a number of important considerations. We must consider other aspects of "the purpose and character of the use" as well. As we have noted, the use at issue was an intermediate one only and thus any commercial "exploitation" was indirect or derivative. The declarations of Accolade's employees indicate, and the district court found, that Accolade copied Sega's software solely in order to discover the functional requirements for compatibility with the Genesis console - aspects of Sega's programs that are not protected by copyright. 17 U.S.C. § 102(b). With respect to the video game programs contained in Accolade's game cartridges, there is no evidence in the record that Accolade sought to avoid performing its own creative work. Indeed, most of the games that Accolade released for use with the Genesis console were originally developed for other hardware systems. Moreover, with respect to the interface procedures for the Genesis console, Accolade did not seek to avoid paying a customarily charged fee for use of those procedures, nor did it simply copy Sega's code; rather, it wrote its own procedures based on what it had learned through disassembly. Taken together, these facts indicate that although Accolade's ultimate purpose was the release of Genesis-compatible games for sale, its direct purpose in copying Sega's code, and thus its direct use of the copyrighted material, was simply to study the functional requirements for Genesis compatibility so that it could modify existing games and make them usable with the Genesis console. Moreover, as we discuss below, no other method of studying those requirements was available to Accolade. On these facts, we conclude that Accolade copied Sega's code for a legitimate, essentially non-exploitative purpose, and that the commercial aspect of its use can best be described as of minimal significance. We further note that we are free to consider the public benefit resulting from a particular use notwithstanding the fact that the alleged infringer may gain commercially. See Hustler, 796 F.2d at 1153 (quoting MCA, Inc. v. Wilson, 677 F.2d 180, 182 (2d Cir. 1981)). Public benefit need not be direct or tangible, but may arise because the challenged use serves a public interest. Id. In the case before us, Accolade's identification of the functional requirements for Genesis compatibility has led to an increase in the number of independently designed video game programs offered for use with the Genesis console. It is precisely this growth in creative expression, based on the dissemination of other creative works and the unprotected ideas contained in those works, that the Copyright Act was intended to promote. See Feist Publications, Inc. v. Rural Tel. Serv. Co., ___ U.S. ___, 111 S. Ct. 1282, 1290, 113 L. Ed. 2d 358 (1991) (citing Harper & Row, 471 U.S. at 556-57). The fact that Genesis-compatible video games are not scholarly works, but works offered for sale on the market, does not alter our judgment in this regard. We conclude that given the purpose and character of Accolade's use of Sega's video game programs, the presumption of unfairness has been overcome and the first statutory factor weighs in favor of Accolade.

The second statutory factor, the nature of the copyrighted work, reflects the fact that not all copyrighted works are entitled to the same level of protection. The protection established by the Copyright Act for original works of authorship does not extend to the ideas underlying a work or to the functional or factual aspects of the work. 17 U.S.C. § 102(b). To the extent that a work is functional or factual, it may be copied, Baker v. Selden, 101 U.S. 99, 102-04, 25 L. Ed. 841 (1879), as may those expressive elements of the work that "must necessarily be used as incident to" expression of the underlying ideas, functional concepts, or facts, id. at 104. Works of fiction receive greater protection than works that have strong factual elements, such as historical or biographical works, Maxtone-Graham, 803 F.2d at 1263 (citing Rosemont Enterprises, Inc. v. Random House, Inc., 366 F.2d 303, 307 (2d Cir. 1966), cert. denied, 385 U.S. 1009, 17 L. Ed. 2d 546, 87 S. Ct. 714 (1967)), or works that have strong functional elements, such as accounting textbooks, Baker, 101 U.S. at 104. Works that are merely compilations of fact are copyrightable, but the copyright in such a work is "thin." Feist Publications, 111 S. Ct. at 1289. Computer programs pose unique problems for the application of the "idea/expression distinction" that determines the extent of copyright protection. To the extent that there are many possible ways of accomplishing a given task or fulfilling a particular market demand, the programmer's choice of program structure and design may be highly creative and idiosyncratic. However, computer programs are, in essence, utilitarian articles - articles that accomplish tasks. As such, they contain many logical, structural, and visual display elements that are dictated by the function to be performed, by considerations of efficiency, or by external factors such as compatibility requirements and industry demands. Computer Assoc. Int'l, Inc. v. Altai, Inc., 1992 U.S. App. LEXIS 14305, 23 U.S.P.Q.2D (BNA) 1241, 1253-56 (2d Cir. 1992) ("CAI"). In some circumstances, even the exact set of commands used by the programmer is deemed functional rather than creative for purposes of copyright. "When specific instructions, even though previously copyrighted, are the only and essential means of accomplishing a given task, their later use by another will not amount to infringement." CONTU Report at 20; see CAI, 23 U.S.P.Q. 2d at 1254. Because of the hybrid nature of computer programs, there is no settled standard for identifying what is protected expression and what is unprotected idea in a case involving the alleged infringement of a copyright in computer software. We are in wholehearted agreement with the Second Circuit's recent observation that "thus far, many of the decisions in this area reflect the courts' attempt to fit the proverbial square peg in a round hole." CAI, 23 U.S.P.Q. 2d at 1257. In 1986, the Third Circuit attempted to resolve the dilemma by suggesting that the idea or function of a computer program is the idea of the program as a whole, and "everything that is not necessary to that purpose or function [is] part of the expression of that idea." Whelan Assoc., Inc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222, 1236 (3d Cir. 1986) (emphasis omitted). The Whelan rule, however, has been widely - and soundly - criticized as simplistic and overbroad. See CAI, 23 U.S.P.Q. 2d at 1252 (citing cases, treatises, and articles). In reality, "a computer program's ultimate function or purpose is the composite result of interacting subroutines. Since each subroutine is itself a program, and thus, may be said to have its own 'idea,' Whelan's general formulation . . . is descriptively inadequate." Id.. For example, the computer program at issue in the case before us, a video game program, contains at least two such subroutines - the subroutine that allows the user to interact with the video game and the subroutine that allows the game cartridge to interact with the console. Under a test that breaks down a computer program into its component subroutines and sub-subroutines and then identifies the idea or core functional element of each, such as the test recently adopted by the Second Circuit in CAI, 23 U.S.P.Q.2D (BNA) at 1252-53, many aspects of the program are not protected by copyright. In our view, in light of the essentially utilitarian nature of computer programs, the Second Circuit's approach is an appropriate one. Sega argues that even if many elements of its video game programs are properly characterized as functional and therefore not protected by copyright, Accolade copied protected expression. Sega is correct. The record makes clear that disassembly is wholesale copying. Because computer programs are also unique among copyrighted works in the form in which they are distributed for public use, however, Sega's observation does not bring us much closer to a resolution of the dispute. The unprotected aspects of most functional works are readily accessible to the human eye. The systems described in accounting textbooks or the basic structural concepts embodied in architectural plans, to give two examples, can be easily copied without also copying any of the protected, expressive aspects of the original works. Computer programs, however, are typically distributed for public use in object code form, embedded in a silicon chip or on a floppy disk. For that reason, humans often cannot gain access to the unprotected ideas and functional concepts contained in object code without disassembling that code - i.e., making copies. Atari Games Corp. v. Nintendo of America, 975 F.2d 832 (Fed. Cir. 1992). Sega argues that the record does not establish that disassembly of its object code is the only available method for gaining access to the interface specifications for the Genesis console, and the district court agreed. An independent examination of the record reveals that Sega misstates its contents, and demonstrates that the district court committed clear error in this respect. First, the record clearly establishes that humans cannot read object code. Sega makes much of Mike Lorenzen's statement that a reverse engineer can work directly from the zeros and ones of object code but "it's not as fun." In full, Lorenzen's statements establish only that the use of an electronic decompiler is not absolutely necessary. Trained programmers can disassemble object code by hand. Because even a trained programmer cannot possibly remember the millions of zeros and ones that make up a program, however, he must make a written or computerized copy of the disassembled code in order to keep track of his work. See generally Johnson-Laird, Technical Demonstration of "Decompilation", reprinted in Reverse Engineering: Legal and Business Strategies for Competitive Design in the 1990's 102 (Prentice Hall Law & Business ed. 1992). The relevant fact for purposes of Sega's copyright infringement claim and Accolade's fair use defense is that translation of a program from object code into source code cannot be accomplished without making copies of the code. Second, the record provides no support for a conclusion that a viable alternative to disassembly exists. The district court found that Accolade could have avoided a copyright infringement claim by "peeling" the chips contained in Sega's games or in the Genesis console, as authorized by section 906 of the SCPA, 17 U.S.C. § 906. Even Sega's amici agree that this finding was clear error. The declaration of Dr. Harry Tredennick, an expert witness for Accolade, establishes that chip peeling yields only a physical diagram of the object code embedded in a ROM chip. It does not obviate the need to translate object code into source code. Atari Games Corp., slip op. at 22. The district court also suggested that Accolade could have avoided a copyright infringement suit by programming in a "clean room". That finding too is clearly erroneous. A "clean room" is a procedure used in the computer industry in order to prevent direct copying of a competitor's code during the development of a competing product. Programmers in clean rooms are provided only with the functional specifications for the desired program. As Dr. Tredennick explained, the use of a clean room would not have avoided the need for disassembly because disassembly was necessary in order to discover the functional specifications for a Genesis-compatible game. In summary, the record clearly establishes that disassembly of the object code in Sega's video game cartridges was necessary in order to understand the functional requirements for Genesis compatibility. The interface procedures for the Genesis console are distributed for public use only in object code form, and are not visible to the user during operation of the video game program. Because object code cannot be read by humans, it must be disassembled, either by hand or by machine. Disassembly of object code necessarily entails copying. Those facts dictate our analysis of the second statutory fair use factor. If disassembly of copyrighted object code is per se an unfair use, the owner of the copyright gains a de facto monopoly over the functional aspects of his work - aspects that were expressly denied copyright protection by Congress. 17 U.S.C. § 102(b). In order to enjoy a lawful monopoly over the idea or functional principle underlying a work, the creator of the work must satisfy the more stringent standards imposed by the patent laws. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 159-64, 103 L. Ed. 2d 118, 109 S. Ct. 971 (1989). Sega does not hold a patent on the Genesis console. Because Sega's video game programs contain unprotected aspects that cannot be examined without copying, we afford them a lower degree of protection than more traditional literary works. See CAI, 23 U.S.P.Q. 2d at 1257. In light of all the considerations discussed above, we conclude that the second statutory factor also weighs in favor of Accolade. 8

As to the third statutory factor, Accolade disassembled entire programs written by Sega. Accordingly, the third factor weighs against Accolade. The fact that an entire work was copied does not, however, preclude a finding a fair use. Sony Corp., 464 U.S. at 449-50; Hustler, 795 F.2d at 1155 ("Sony Corp. teaches us that the copying of an entire work does not preclude fair use per se."). In fact, where the ultimate (as opposed to direct) use is as limited as it was here, the factor is of very little weight. Cf. Wright v. Warner Books, Inc., 953 F.2d 731, 738 (2d Cir. 1991). (e)

As applied, the fourth statutory factor, effect on the potential market for the copyrighted work, bears a close relationship to the "purpose and character" inquiry in that it, too, accommodates the distinction between the copying of works in order to make independent creative expression possible and the simple exploitation of another's creative efforts. We must, of course, inquire whether, "if [the challenged use] should become widespread, it would adversely affect the potential market for the copyrighted work," Sony Corp. v. Universal City Studios, 464 U.S. 417, 451, 78 L. Ed. 2d 574, 104 S. Ct. 774 (1984), by diminishing potential sales, interfering with marketability, or usurping the market, Hustler, 796 F.2d at 1155-56. If the copying resulted in the latter effect, all other considerations might be irrelevant. The Harper & Row Court found a use that effectively usurped the market for the copyrighted work by supplanting that work to be dispositive. 471 U.S. at 567-69. However, the same consequences do not and could not attach to a use which simply enables the copier to enter the market for works of the same type as the copied work. Unlike the defendant in Harper & Row, which printed excerpts from President Ford's memoirs verbatim with the stated purpose of "scooping" a Time magazine review of the book, 471 U.S. at 562, Accolade did not attempt to "scoop" Sega's release of any particular game or games, but sought only to become a legitimate competitor in the field of Genesis-compatible video games. Within that market, it is the characteristics of the game program as experienced by the user that determine the program's commercial success. As we have noted, there is nothing in the record that suggests that Accolade copied any of those elements. By facilitating the entry of a new competitor, the first lawful one that is not a Sega licensee, Accolade's disassembly of Sega's software undoubtedly "affected" the market for Genesis-compatible games in an indirect fashion. We note, however, that while no consumer except the most avid devotee of President Ford's regime might be expected to buy more than one version of the President's memoirs, video game users typically purchase more than one game. There is no basis for assuming that Accolade's "Ishido" has significantly affected the market for Sega's "Altered Beast", since a consumer might easily purchase both; nor does it seem unlikely that a consumer particularly interested in sports might purchase both Accolade's "Mike Ditka Power Football" and Sega's "Joe Montana Football", particularly if the games are, as Accolade contends, not substantially similar. In any event, an attempt to monopolize the market by making it impossible for others to compete runs counter to the statutory purpose of promoting creative expression and cannot constitute a strong equitable basis for resisting the invocation of the fair use doctrine. Thus, we conclude that the fourth statutory factor weighs in Accolade's, not Sega's, favor, notwithstanding the minor economic loss Sega may suffer.

1. Abstraction The first step in the analysis involves dissecting the allegedly infringed program according to the abstractions test. As Judge Learned Hand wrote in Nichols v. Universal Pictures Corp., 45 F.2d 119 (1930), cert. denied, 282 U.S. 902, 75 L. Ed. 795, 51 S. Ct. 216 (1931), a case dealing with the alleged copyright infringement of a theatrical script: Upon any work, and especially upon a play, a great number of patterns of increasing generality will fit equally well, as more and more of the incident is left out. The last may be no more than the most general statement of what the play is about, and at times might only consist of its title; but there is a point in this series of abstractions where they are no longer protected, since otherwise the playwright could prevent the use of his "ideas," to which, apart from their expression, his property is never extended. Id. at 121. The abstractions test is especially well suited to the dissection of computer programs because the test breaks down a program in a way that parallels the typical development of a program. See Autoskill, 994 F.2d at 1491-92; Whelan, 797 F.2d at 1229-30; 3 Nimmer 13.03[F], at 13-102.9 to 13-102.19. We agree with Professor Nimmer that "applying [the abstractions] test conscientiously and systematically can help a court separate ideas from expression and eliminate from the substantial similarity analysis those portions of a work that are not eligible for copyright protection." 3Nimmer 13.03[F] 13-102.17. However, in and of itself, the abstraction test does not identify the protectable elements of a program. Rather, it is merely one tool that can be utilized to accomplish this task. Abstraction is particularly useful in enabling a court to filter out ideas and processes from protectable expression. Application of the abstractions test will necessarily vary from case-to-case and program-to-program. Given the complexity and ever-changing nature of computer technology, we decline to set forth any strict methodology for the abstraction of computer programs. See generally AndrewH. Rosen, Virtual Reality: Copyrightable Subject Matter and the Scope of Judicial Protection, 33 Jurimetrics J. 35 (1992). Indeed, in most cases we foresee that the use of experts will provide substantial guidance to the court in applying an abstractions test. However, a computer program can often be parsed into at least six levels of generally declining abstraction: (i) the main purpose, (ii) the program structure or architecture, (iii) modules, (iv) algorithms and data structures, (v) source code, and (vi) object code. See JohnW.L.Ogilvie, Defining Computer Program Parts Under Learned Hand's Abstractions Test in Software Copyright Infringement Cases, 91 Mich. L. Rev. 526 (1992) (hereinafter "Ogilvie"). The main purpose of a program is a description of the program's function or what it is intended to do. See Ogilvie at 534; Altai, 982 F.2d at 697 ("ultimate function or purpose"); Whelan, 797 F.2d at 1238 ("purpose"). When defining a program's main purpose, the court must take care to describe the program's function as specifically as possible without reference to the technical aspects of the program. The program's architecture or structure is a description of how the program operates in terms of its various functions, which are performed by discrete modules, and how each of these modules interact with each other. The architecture or structure of a program is often reduced to a flowchart, which a programmer uses visually to depict the inner workings of a program. Paperback, 740 F. Supp. at 44-45. Structure exists at nearly every level of a program and can be conceived of as including control flow, data flow, and substructure or nesting. Control flow is the sequence in which the modules perform their respective tasks. See Ogilvie at 535; Whelan, 797 F.2d at 1230 (arrangement of modules); Gates Rubber, 798 F. Supp. at 1514 (sequence of events). Data flow describes the movement of information through the program and the sequence with which it is operated on by the modules. See Ogilvie at 535; Gates Rubber, 798 F.2d at 1514 ("described as being analogous to 'receipt'"). Substructure or nesting describes the inner structure of a module whereby one module is subsumed within another and performs part of the second module's task. Ogilvie at 535. The next level of abstraction consists of the modules. See StevenR. Englund, Note, Idea, Process, or Protected Expression?: Determining the Scope of Copyright Protection of the Structure of Computer Programs, 88 Mich. L. Rev. 866, 871, 899 (1990) (hereinafter "Englund"). A module typically consists of two components: operations and data types. An operation identifies a particular result or set of actions that may be performed. For example, operations in a calculator program might include adding or printing data. A data type defines the type of item that an operator acts upon such as a student record or a daily balance. See Ogilvie at 534-36. Algorithms and data structures are more specific manifestations of operations and data types, respectively. An algorithm is a specific series of steps that accomplish a particular operation. Ogilvie at 536; see Whelan, 797 F.2d at 1229-30. Data structure is a precise representation or specification of a data type that consists of (i)basic data type groupings such as integers or characters, (ii)values, (iii)variables, (iv)arrays or groupings of the same data type, (v)records or groupings of different date types, and (vi)pointers or connections between records that set aside space to hold the record's values. Ogilvie at 536-39; see Whelan, 797 F.2d at 1230. The computer program is written first in a programming language, such as Pascal or Fortran, and then in a binary language consisting of zeros and ones. Source code is the literal text of a programs' instructions written in a particular programming language. Trandes Corp. v. Guy F. Atkinson Co., 996 F.2d 655, 663 (4th Cir. 1993); Whelan, 797 F.2d at 1230. Object code is the literal text of a computer program written in a binary language through which the computer directly receives its instructions. Autoskill, 994 F.2d at 1492 n.18; Altai, 982 F.2d at 698. These generalized levels of abstraction will not, of course, fit all computer codes. Ordinarily, expert testimony will be helpful to organize a particular program into various levels of abstraction. In any event, as pointed out earlier, the organization of a program into abstraction levels is not an end in itself, but it is only a tool that facilitates the critical next step of filtering out unprotectable elements of the program.

2. Filtration Once a court has succeeded in identifying the various levels of abstraction of a computer program, it must filter out those elements of the program that are not protected by copyright. The Copyright Act provides that "ideas, procedures, processes, systems, methods of operation, concepts, principles, or discoveries" are not protectable. 17 U.S.C. 102(b) (Supp. 1993). Accordingly, the court must filter out these unprotectable elements. In order to effectuate the purposes behind the copyright laws, it is also appropriate to apply the doctrines of merger and scenes a faire to filter out unprotectable elements. (a) The Idea-Expression Dichotomy One of the fundamental tenets of copyright law is that protection extends only to the author's original expression and not to the ideas embodied in that expression. 17 U.S.C. 102(b) codifies this tenet by denying copyright to ideas, concepts and principles. Numerous cases have expounded on the idea-expression dichotomy. See e.g., Feist, 111 S. Ct. at 1287-90 (dicta); Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 556-57, 85 L. Ed. 2d 588, 105 S. Ct. 2218 (1985); Baker v. Selden, 101 U.S. 99, 104, 107, 25 L. Ed. 841 (1879); Autoskill, 994 F.2d at 1491. Distinguishing between ideas and the expression of those ideas is not an easy endeavor, and given the varying nature of computer programs it must necessarily be ad hoc. With the framework provided by the abstractions test, however, we believe that a court can adequately filter out the ideas of a program. For example, the main purpose or function of a program will always be an unprotectable idea. See Whelan, 797 F.2d at 1236, 1238; Paperback, 740 F. Supp. at 65. Likewise, each module may typically be described by its individual purpose or function, and the basic function or purpose of a module will nearly always be an unprotectable idea or process. At the other end of the abstractions spectrum, source and object code, which are the literal elements of a program, will almost always be found to be protectable expression unless the doctrines of merger and scenes a faire come into play. See Altai, 982 F.2d at 702; Johnson Controls, Inc. v. Phoenix Control Systems, Inc., 886 F.2d 1173, 1175 (9th Cir. 1989); Whelan, 797 F.2d at 1233. The intermediate levels of abstraction, such as structure, sequence, organization, and the like, are less prone to generalizations. While the structure of one program may be unprotectable because it constitutes an idea, the organization and arrangement of another program may be expressive and thereby protectable. And, of course, within a program certain structural elements may be protectable expression while other elements are unprotectable ideas. (b) The Process-Expression Dichotomy Although it is the idea-expression distinction that has received the primary attention in the cases construing copyright protection, the Copyright Act denies protection to other equally important program elements. When considering utilitarian works such as computer programs one of the most important of these elements is process. See Englund, 88 Mich. L. Rev. 866. 17U.S.C. 102(b) denies protection to procedures, processes, systems and methods of operation. The legislative history of the Copyright Act clarifies any ambiguity about the status of processes. Some concern has been expressed lest copyright in computer programs should extend protection to the methodology or processes adopted by the programmer, rather than merely to the "writing" expressing his ideas. Section 102(b) is intended, among other things, to make clear that the expression adopted by the programmer is the copyrightable element in a computer program, and that the actual processes or methods embodied in the program are not within the scope of the copyright law. H.R. Rep. No. 1476, 94th Cong., 2d Sess. 57 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5670. The Supreme Court addressed the copyrightability of a utilitarian process in Baker v. Selden, 101 U.S. 99, 25 L. Ed. 841 (1879). There, the Court considered whether the author of a book describing an accounting system could obtain protection over the system itself through copyright of the book. The Court distinguished the "art" or process from the author's explanation thereof and found the former unprotectable. Baker, 101 U.S. at 104. Other courts have similarly found processes unprotectable. See e.g., Altai, 982 F.2d at 704; Atari Games Corp. v. Nintendo of America, Inc., 975 F.2d 832, 838-39 (Fed. Cir. 1992). Certain processes may be the subject of patent law protection under Title 35 of the United States Code. See Atari Games, 975 F.2d at 839-40; Englund, 88 Mich. L. Rev. at 893-94. Although processes themselves are not copyrightable, an author's description of that process, so long as it incorporates some originality, may be protectable. See Applied Innovations, Inc. v. Regents of the Univ. of Minnesota, 876 F.2d 626, 636 (8th Cir. 1989); Apple Computer, Inc. v. Franklin Computer Corp., 714 F.2d 1240, 1250-51 (3d Cir. 1983), cert. dismissed, 464 U.S. 1033 (1984). Returning then to our levels of abstraction framework, we note that processes can be found at any level, except perhaps the main purpose level of abstraction. Most commonly, processes will be found as part of the system architecture, as operations within modules, or as algorithms. (c) Facts The Copyright Act also denies protection to discoveries. 17U.S.C. 102(b). The Supreme Court squarely addressed the issue of the protectability of facts in its recent opinion in Feist Publications, Inc. v. Rural Telephone Services Co., 499 U.S. 340, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (1991). In Feist, the Court considered the copyrightability of a telephone directory comprised merely of names, addresses, and phone numbers organized in alphabetical order. The Court rejected the notion that copyright law was meant to reward authors for the "sweat of the brow," and instead concluded that protection only extends to the original components of an author's work. As to facts, the Court found that No one may claim originality as to facts. This is because facts do not owe their origin to an act of authorship. The distinction is one between creation and discovery: the first person to find and report a particular fact has not created the fact; he or she has merely discovered its existence. . . . One who discovers a fact is not its maker or originator. The discoverer merely finds and records. 111 S. Ct. at 1288 (internal quotations and citations omitted). Like ideas and processes, facts themselves are not protectable; however, an author's original compilation, arrangement or selection of facts can be protected by copyright. Feist, 111 S. Ct. at 1289; Applied Innovations, 876 F.2d at 636. However, "the copyright is limited to the particular selection or arrangement. In no event may copyright extend to the facts themselves." Feist, 111 S. Ct. at 1290. In computer programs facts may be found at a number of levels of abstraction, but, will most often be found as part of data structures or literally expressed in the source or object codes. (d) Public Domain "The two fundamental criteria of copyright protection [are] originality and fixation in tangible form . . .." H.R. Rep. No. 1476, 94th Cong., 2d Sess. 51 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5664. 17 U.S.C. 102(a) establishes that copyright can only subsist in "original works of authorship." (emphasis added). Originality in the field of copyright requires that the work be independently created by the author and that it possesses a minimal degree of creativity. Feist, 111 S. Ct. at 1296. Accordingly, in determining copyright infringement, a court must filter out all unoriginal elements of a program, including those elements that are found in the public domain. See Comprehensive Technologies International, Inc. v. Software Artisans, Inc., 3 F.3d 7301993 U.S. App. LEXIS 21706, * 8 (4th Cir. 1993); Altai, 982 F.2d at 710; Atari Games, 975 F.2d at 839; Brown Bag Software v. Symantec Corp., 960 F.2d 1465, 1474-75 (9th Cir. 1992), cert. denied, 121 L. Ed. 2d 140, 113 S. Ct. 198 (1992); E.F. Johnson Co. v. Uniden Corp. of America, 623 F. Supp. 1485, 1499 (D.Minn. 1985). Unoriginal elements of a program may be found at any level of abstraction. Once a court has filtered out a program's ideas, processes, facts, and unoriginal elements, it has eliminated most of those elements that are unprotected based on the first principles of copyright law. However, in order to give effect to these principles, the courts have devised two additional filtration doctrines. These are the doctrines of merger and scenes a faire. (e) The Merger Doctrine Under the merger doctrine, copyright protection is denied to expression that is inseparable from or merged with the ideas, processes, or discoveries underlying the expression. Concrete Machinery Co. v. Classic Lawn Ornaments, Inc., 843 F.2d 600, 606-607 (1st Cir. 1988); Apple Computer, Inc. v. Microsoft Corp., 799 F. Supp. 1006, 1021 (N.D. Cal. 1992), order clarified, 821 F. Supp. 616 (1993); Englund, 88 Mich. L. Rev. at 877. The merger doctrine is applied as a prophylactic device to ensure that courts do not unwittingly grant protection to an idea by granting exclusive rights to the only, or one of only a few, means of expressing that idea. See Herbert Rosenthal Jewelry Corp. v. Kalpakian, 446 F.2d 738, 742 (9th Cir. 1971). If protection were granted to these expressions, it would so increase the cost of creation for others who seek to build on the work that it would impede progress in the arts. Such a result is contrary to the goals of copyright as embedded in the Constitution. See U.S. Const., art. I, 8, cl. 8. (f) Scenes A Faire Under the scenes a faire doctrine, we deny protection to those expressions that are standard, stock, or common to a particular topic or that necessarily follow from a common theme or setting. Autoskill, 994 F.2d at 1494 (citing 3 Nimmer 13.03 [B][4], at 13-70); Atari, 672 F.2d at 616. Granting copyright protection to the necessary incidents of an idea would effectively afford a monopoly to the first programmer to express those ideas. Whelan, 797 F.2d at 1236-37; Landsberg v. Scrabble Crossword Game Players, Inc., 736 F.2d 485, 489 (9th Cir.), cert. denied, 469 U.S. 1037, 83 L. Ed. 2d 403, 105 S. Ct. 513 (1984). Furthermore, where a particular expression is common to the treatment of a particular idea, process, or discovery, it is lacking in the originality that is the sine qua non for copyright protection. Feist, 111 S. Ct. at 1289. The scenes a faire doctrine also excludes from protection those elements of a program that have been dictated by external factors. See Plains Cotton Co-op Ass'n. v. Goodpasture Computer Serv., Inc., 807 F.2d 1256, 1262 (5th Cir.), cert. denied, 484 U.S. 821, 98 L. Ed. 2d 42, 108 S. Ct. 80 (1987); Apple Computer, 799 F. Supp. at 1022-26. In the area of computer programs these external factors may include: hardware standards and mechanical specifications, see Manufacturers Technologies, Inc. v. Cams, Inc., 706 F. Supp. 984, 995 (D.Conn. 1989), software standards and compatibility requirements, Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525-27 (9th Cir. 1993), computer manufacturer design standards, target industry practices and demands, see Plains Cotton, 807 F.2d at 1262, and computer industry programming practices, see Apple Computer, 799 F. Supp. at 1033. 3 Nimmer 13.03[F][3][a]-[e], at 13-102.21 to 13-102.28.

3. Comparison After the court has filtered out those elements of the original program that it has found to be unprotectable, it is left with a core of protected elements that can be compared to the alleged infringing program. Ultimately the court must decide whether those protectable portions of the original work that have been copied constitute a substantial part of the original work -- i.e. matter that is significant in the plaintiff's program. See Altai, 982 F.2d at 710-11; Atari, Inc., 672 F.2d at 619; Paperback, 740 F. Supp. at 61; 3 Nimmer 13.03[A][2], at 13-46 to 13-48. This is primarily a qualitative rather than a purely quantitative analysis, see Whelan, 797 F.2d at 1245, and must be performed on a case-by-case basis. B. Review of Constitutional and Statutory Principles and Key Judicial Precedent. The test that we adopt today is firmly rooted in traditional principles of copyright law. The Constitution establishes an affirmative duty for Congress "to promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." U.S. Const. art I, 8, cl. 8; see Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 428-29, 78 L. Ed. 2d 574, 104 S. Ct. 774 (1984); Mazer v. Stein, 347 U.S. 201, 219, 98 L. Ed. 630, 74 S. Ct. 460 (1954). Copyright policy is meant to balance protection, which seeks to ensure a fair return to authors and inventors and thereby to establish incentives for development, with dissemination, which seeks to foster learning, progress and development. In order to secure these goals Congress enacted The Copyright Act, 17 U.S.C. 101 et seq., which establishes that (a) Copyright protection subsists, in accordance with this title, in original works of authorship fixed in any tangible medium of expression, now known or later developed, from which they can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device. Works of authorship include the following categories: (1) literary works; . . . . (6) motion pictures and other audiovisual works; . . . . 17 U.S.C. 102(a) (Supp. 1993). Computer programs are considered literary works for purposes of copyright analysis. 17 U.S.C. 101 states: "Literary works" are works… expressed in words, numbers, or other verbal or numerical symbols or indicia, regardless of the nature of the material objects, such as… film, tapes, disks, or cards, in which they are embodied. . . . . A "computer program" is a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result. 17 U.S.C. 101 (1977 and Supp. 1993). Atari Games Corp. v. Nintendo of America, Inc., 975 F.2d 832, 838 (Fed. Cir. 1992); H.R. Rep. No. 1476, 94th Cong., 2d Sess. 54, reprinted in 1976 U.S.C.C.A.N. 5659, 5667 (The term "literary works" "also includes computer data bases, and computer programs to the extent that they incorporate authorship in the programmer's expression of original ideas, as distinguished from the ideas themselves."). A computer program is protectable under copyright if two requirements are met: (1) it is fixed in a tangible medium of expression, and (2) the program is original. 17 U.S.C. 102(a)(Supp. 1993); M. Kramer Mfg. Co. v. Andrews, 783 F.2d 421, 433 (4th Cir. 1986); H.R. Rep. No. 1476, 94th Cong., 2d Sess. 51- 2, reprinted in 1976 U.S.C.C.A.N. 5659, 5664-65. Much of the modern doctrine concerning the copyright of utilitarian works and the process-expression dichotomy owes its origins to the Supreme Court's opinion in Baker v. Selden, 101 U.S. 99, 25 L. Ed. 841 (1879), in which the Court considered the copyright of a book setting out a system of accounting. The Court held that neither the system itself nor the forms necessary to operate the system were protectable, although the author's particular expression of the system was protectable. The copyright of a work on mathematical science cannot give to the author an exclusive right to the methods of operation which he propounds, or to the diagrams which he employs to explain them, so as to prevent an engineer from using them whenever occasion requires. The very object of publishing a book on science or the useful arts is to communicate to the world the useful knowledge which it contains. But this object would be frustrated if the knowledge could not be used without incurring the guilt of piracy of the book. And where the art it teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are to be considered as necessary incidents to the art, and given therewith to the public; not given for the purpose of publication in other works explanatory of the art, but for the purpose of practical application.Baker, 101 U.S. at 103. One of the early attempts to apply the doctrine of copyright to computer programs came in Whelan Assoc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222 (3d Cir. 1986), cert. denied, 479 U.S. 1031, 93 L. Ed. 2d 831, 107 S. Ct. 877 (1987). The Whelan court extended copyright protection to the structure, sequence and organization of a computer program even though the literal elements were not copied. 797 F.2d at 1248. Moreover, the Third Circuit attempted to establish a principle to distinguish between idea and expression. It concluded that: The line between idea and expression may be drawn with reference to the end sought to be achieved by the work in question. In other words, the purpose or function of a utilitarian work would be the work's idea, and everything that is not necessary to that purpose or function would be part of the expression of the idea. 797 F.2d at 1236 (emphasis omitted). The Whelan court's articulation of this general rule for distinguishing idea from expression has been criticized extensively. However, Whelan did use the limiting doctrines of merger and scenes a faire to restrict the scope of protection. In any event, when a program is understood to encompass more than one idea, the general principle of Whelan provides a useful means to distinguish idea from expression. At its base, Whelan is premised upon traditional principles of copyright law, and its conclusion that the structure of a program may be protectable is sound. In Lotus Development Corp. v. Paperback Software International, 740 F. Supp. 37 (D. Mass. 1990), Judge Keaton, who has written extensively on the subject of software protection, considered the copyright of the Lotus 1-2-3 spreadsheet program. Like Whelan, the court concluded that the "structure, sequence and organization" of the software may be protectable. The Paperback court developed a three-part test that is a forerunner of the standard that we adopt in this case. 740 F. Supp. at 59-62. First, the court formulated a conception of the program's idea based on the various abstractions that it found to describe the program. 740 F. Supp. at 60. Second, the court separated out those expressions that it found to be essential to the ideas. Finally, the court determined whether the remaining expressions were a substantial part of the copyrighted work. 740 F. Supp. at 61, 70. The Second Circuit in Computer Associates International, Inc. v. Altai, Inc., 982 F.2d 693 (2d Cir. 1992), built on the case law that proceeded it. The court agreed with Whelan and Paperback that copyright can protect the non-literal structures of computer programs. 982 F.2d at 702-703. In analyzing the similarity of the computer programs before it, the Altai court utilized a three-part test that bore some similarity to the three-part test in Paperback. The Altai court articulated its test as an "Abstration-Filtration-Comparison" test. Under the first prong the court dissected the program and identified the various levels of abstraction in the program. Altai, 982 F.2d at 707. Under the second prong, the court examined the separate elements within the program and eliminated those elements that it found to be either (1) ideas, (2) in the public domain, (3) required by factors external to the program, or (4) dictated by efficiency concerns such that the idea and expression were inseparable. Altai, 982 F.2d at 707-710. Those unprotectable elements were then filtered out before the two programs were compared. Finally, in the third prong, the court examined the remaining "core of protectable expression" to determine whether the defendant's program misappropriated substantial portions of the plaintiff's program. Altai, 982 F.2d at 710-711. We accept this basic three-part analysis and seek only to elaborate upon the various steps and to clarify the role of the abstraction test. We suggest that a court will often be assisted in determining the factual issue of copying if both programs are first compared in their entirety without filtering out the unprotected elements. Such a preliminary step does not obviate the ultimate need to compare just the protected elements of the copyrighted program with the alleged infringing program. However, an initial holistic comparison may reveal a pattern of copying that is not obvious when only certain components are examined. The Supreme Court's most recent pronouncements in the area of copyright law came in Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (1991). Although Feist concerned the protectability of telephone directories rather than computer programs, the Court clarified several principles of copyright law applicable to the area of computers. The Court held that facts are not copyrightable regardless of the effort the author undertook to discover or compile them. In doing so the court rejected the "sweat of the brow" doctrine, which maintained that copyright was meant to protect and reward the efforts of an author. The court established that it is originality, not effort, that is the basis of copyright protection. However, the Court concluded that the original organization, selection, or arrangement in a work that is comprised exclusively of unprotectable facts could nevertheless be protected. A panel of this circuit recently considered computer program copyright infringement in Autoskill, Inc. v. National Educational Support Systems, Inc., 994 F.2d 1476 (10th Cir. 1993), cert. denied, U.S. LEXIS 6372 (U.S. Oct.12, 1993). We upheld the grant of a preliminary injunction against the developers of a program designed to test and train individuals with reading deficiencies. Because the case involved a preliminary injunction, we only reviewed the district court's ruling to determine if the plaintiff had established a prima facie case with reasonable probability of success, and we expressly refrained from dictating a precise test for determining copyright infringement. Albeit, we affirmed the district court's application of the "Abstraction-Filtration-Comparison" test and application of the doctrines of merger, scenes a faire, and public domain. We find in these and other cases that have considered the copyrightability of computer programs that there has begun to be developed a coherent approach to the protectability analysis. The approach that we outline today is consistent with this evolving approach to the copyright protection of computer programs.

C. Applying the Abstraction-Filtration-Comparison Test to the Computer Programs at Issue in This Case. The district court undertook an analysis similar to that which we set forth today. First, it determined whether, as a factual matter, the Chauffeur program had been copied from the Design Flex program. The court initially found that the defendants had access to the Design Flex program. It then considered the testimony of the expert witnesses and identified a number of similarities between the two programs. The court concluded that "the only explanation regarding the constants, the install files, and the overall data flow and presentation, is that the Chauffeur program was copied from the Design Flex program." Gates Rubber Co. v. Bando American, Inc., 798 F. Supp. 1499, 1516 (D.Colo. 1992). Next, the court applied the abstractions test to identify "unprotectable ideas." The court dissected the program and considered whether certain of the elements more closely approximated idea or expression. The court concluded that the merger doctrine and the scenes a faire doctrine were inapplicable. Finally, the court evaluated what it regarded as the protectable elements and determined that they were sufficiently significant to the Design Flex program to conclude that the defendants had infringed on Gates' copyright in the Design Flex program when it copied such elements. Gates Rubber, 798 F. Supp. at 1519. The copyright issues on appeal all revolve around whether the district court erroneously extended copyright protection to unprotectable elements of the Design Flex program. The appellants argue that the court failed to consider whether three of the elements of the Design Flex program were protectable, and that it applied the wrong legal standard in protecting certain elements of the Design Flex program. The appellants argue that if a proper filtration analysis were conducted, the Chauffeur program would not be found to have infringed any protectable elements of the Design Flex program. The district court ultimately concluded that the defendants had misappropriated ten elements of the Design Flex program that were protected by Gates' copyright. They included: menus, constants, sorting criteria, control flow, data flow, the engineering calculation module, the design module, common errors, fundamental tasks, and install files. We conclude that the district court failed to undertake a proper filtration analysis with respect to several elements and that it erroneously found other elements to be protectable. Accordingly, we remand for further consideration by the district court. Constants Constants are the invariable integers that comprise part of the formulas used to perform the calculations in the programs. The district court failed to consider whether the constants were unprotectable processes or facts or whether they were subject to the merger doctrine. In failing to undergo the filtration process, the court skipped an essential step in the copyright infringement analysis and as a result extended copyright protection to an unprotectable element of Design Flex. The record reveals that these constants are facts that are unprotectable under copyright law. The constants in the Design Flex program represent scientific observations of physical relationships concerning the load that a particular belt can carry around certain sized gears at certain speeds given a number of other variables. These relationships are not invented or created; they already exist and are merely observed, discovered and recorded. Such a discovery does not give rise to copyright protection. See Feist, 111 S. Ct. at 1288-19 ("Facts do not owe their origin to an act of authorship. The distinction is one between creation and discovery: the first person to find and report a particular fact has not created the fact; he or she has merely discovered its existence."). Gates claims that the constants should be protected because it spent thousands of hours testing the relationships and because engineers ultimately had to determine the best figure to represent the test results. However, this argument amounts to an assertion of the "sweat of the brow" doctrine which has been rejected by the Supreme Court. Feist, 111 S. Ct. at 1291-92. Accordingly, while the constants were probative of the factual question whether the defendants copied from the Design Flex program, they were unprotectable and therefore should have been filtered out before the district court addressed the ultimate question of whether the defendants infringed on Gates' copyright. Instead, the court relied heavily on the similarity of the constants in the two programs when it concluded that Chauffeur infringed on Gates' copyright. It noted that the constants "lie at the heart of the dispute," and gave "special regard to the mathematical constants" in finding that the defendants had violated Gates' copyright. Gates Rubber, 798 F. Supp. at 1518-19. Because of the court's emphasis on the constants in its reasoning, and our finding that the constants are unprotectable facts, we remand for the district court to consider whether the remaining protectable elements found to have been copied are such a substantial part of the Design Flex program that their misappropriation constitutes copyright infringement. Menus and Sorting Criteria These terms are not defined in the district court opinion. However, menus may mean the visual screen displays that present a computer operator with a limited number of commands available at a given stage in the computer program's operation. Lotus Development Corp. v. Borland International, Inc., 799 F. Supp. 203, 206 (D.Mass. 1992). Sorting criteria would ordinarily mean the factors that determine how the data in the program is organized. The district court failed to undertake a proper filtration analysis with respect to the menus and sorting criteria. Because the record before us is ambiguous and incomplete with regard to these terms, we are unable to determine on appeal whether these elements are protectable. The defendants claim that the menus and the sorting criteria are visual components of the programs as to which Gates has waived all claims of infringement. At trial, counsel for Gates stated that the "plaintiff is willing to stipulate that we are making no claim as to trade secret or copyright as to the screens of these computer programs…" Aplt. Apx. at 266. When asked about its claims concerning the menus and sorting criteria at oral argument in this appeal, however, Gates asserted that it had only waived its claims concerning the visual screen displays, and that it maintained a copyright claim on the written elements of the program that created those displays. The district court failed to clarify whether it was referring to the visual screen displays or some other aspect of the program when it discussed the menus and sorting criteria, and it failed to address whether the visual screen display can be separated from the program generating the visual screen display for purposes of the stipulation. Accordingly, we remand for a clarification of the terms "menu" and "sorting criteria"; for a determination of whether Gates has waived its infringement claims with respect to these elements; and for a determination of the protectability of such elements if Gates is determined not to have waived its infringement claim with regard to them. Control and Data Flow Control flow refers to the overall sequence of actions and events in a program. Data flow is the sequence of actions taken on each piece of information, that is, how the data travels through the program. The district court found that "the flow of these two types of information are closer to the expression of how the task is performed than the idea of undertaking calculation of belts and drives by use of a computer program." Gates Rubber, 798 F. Supp. at 1518. The defendants object that this conclusion is erroneously predicated on the dicta in Whelan that each program has only one idea. See Whelan, 797 F.2d at 1236. They assert that the control and data flow elements occur at such high levels of abstraction that they are inappropriate for copyright protection. A fair reading of the district court opinion suggests that the court recognized that a computer program may contain more than one idea, and that unprotected ideas may be found throughout the program. However, we are more concerned by the district court's failure to examine the control and data flow in light of the process-expression dichotomy. Again, the district court failed to define exactly what it meant by control flow and data flow and the record is unclear. We remand for clarification of the terms "control flow" and "data flow" and for reconsideration of these elements in light of the idea-expression and process-expression dichotomies as they have been set forth in this opinion. Engineering Calculation and Design Modules The program performs its calculations and selects the proper belt through a series of operating instructions contained in modules. The district court concluded that: The modules of the program were also found to be substantially similar. These modules, like chapters of a novel or scenes from a play, performed particular functions - here the engineering calculations and the design aspect of the program (containing the V-belt design algorithms). Concerning the former module, it was noted that behavior was similar, and with the latter the similarity concerned overall structure and organization. With regard to the former module, this falls closer to the expression range because, although the pure engineering aspect in a broad sense may be more likely to be not protected, the relevant engineering modules in the two programs contain particular elements which perform in similar manners. The latter module is somewhat more difficult as it involves algorithms, or procedures for solving given types of mathematical problems. The Court here rejects the argument that while the remaining portion [of the program] is properly protected under copyright law, the algorithms, as a "process" can only be covered by patent law. Such a holding would tend to fragment further the rather tenuous continuity found in copyright law concerning computer programs. This conclusion is supported by the Whelan decision. 797 F.2d at 1229. Gates Rubber, 798 F. Supp. at 1518. (emphasis added). The district court identified two modules - the engineering module and the design module - and it found them to be protected for different reasons. The appellant contests this finding, asserting that calculation module, design module, and engineering module are just different names for the same module. Gates has failed to brief this issue. The expert testimony casts doubt on the district court's interpretation that the engineering module and the design module are two separate modules; however, the record is not sufficiently clear to resolve the dispute. The district court found that the engineering module was expression rather than idea because of "particular elements which perform in similar manners." Gates Rubber, 798 F. Supp. at 1518. However, the district court failed to identify those "particular elements" upon which it based its decision, and it failed to analyze with any particularity whether those elements were ideas, processes, or facts. If, as the district court seems to suggest, the algorithms are "process," then they would not be protectable under copyright law. There remains the possibility that they are expressions, but, the court will then need to address the merger doctrine. Further, defendants have drawn our attention to evidence that was before the district court that tends to show that certain module functions were standard in the industry. There is no indication that the district court analyzed those components of the module under the scenes a faire doctrine. Accordingly, we remand for clarification of the district court's analysis and consideration of the modules in light of the process-expression dichotomy, merger and scenes a faire doctrines. Unless the district court provides us with sufficient detail to understand its ruling, we are unable to provide meaningful review of this finding. Common Errors Common errors or misbehaviors are instances in which the two programs share similar errors when not performing correctly or as intended. See Atari Games Corp. v. Nintendo of America, 975 F.2d 832, 845 (Fed. Cir. 1992); M. Kramer Manufacturing Co. v. Andrews, 783 F.2d 421, 446 (4th Cir. 1986). The experts in the instant case identified two such common errors. First, both programs identify that it is impossible to compute a maximum center distance that is less than a minimum center distance, but both programs attempt to do so anyway. Second, both programs will jump to the wrong menu if the cursor happens to be over a certain character during one of the input sequences. While common errors may often be strong evidence of copying as a factual matter, they do not assist in determining what material is protectable under copyright law. BellSouth Advertising & Publishing Corp. v. Donnelley Information Publishing, Inc., 999 F.2d 1436 (11th Cir. 1993). Relying on several cases that have held that "common errors are the strongest evidence of copying," Gates argues that the common errors should be considered protectable." Aplee. Br. at 33. See Atari Games, 975 F.2d at 845; M. Kramer, 783 at 446; E.F. Johnson Co. v. Uniden Corp. of America, 623 F. Supp. 1485, 1496 (D.Minn. 1985). However, none of the cases cited by Gates held that the actual misbehavior or faulty operation of the computer was protected. Rather, the cases stand for the limited proposition that evidence that two programs contain the same unnecessary or defective textual instructions is probative on the issue of copying. None of the cases cited considered the protectability of those instructions. Errors per se are not protectable, although the expression containing the error may be protectable if it otherwise meets the test for protectability set forth in this opinion. That analysis is lacking here. Thus, it is necessary for us to remand on this issue. Fundamental Tasks We normally would associate the term "fundamental tasks" with the highest level of abstraction -- that is, the ideas or purposes underlying a program. In the instant case, the district court found the fundamental tasks of the Design Flex program to be protectable expression, but it failed to explain what it thought the fundamental tasks of the Design Flex program were. The district court stated that although "fundamental tasks" may in a broader sense merely describe what are necessary means to effect a particular end, the term here is more specific due to the types of tasks which were available to achieve the particular end of designing belts and drives. Gates Rubber, 798 F. Supp. at 1519. This discussion suggests the district court held a different understanding of the term "fundamental tasks" than what we would normally understand it to mean. However, we do not understand what the district court meant, nor do we have the benefit of the district court analysis beyond its mere conclusion that the merger doctrine is not applicable. Accordingly, we must remand for further analysis of this element as well. Install Files The install files here are separate utility programs that are used to load the program from a floppy disk onto a hard disk. They are part of the Disk Operating System and apparently are not part of the Design Flex program. Accordingly, it is not clear that Gates has a copyright claim on the install files. The district court failed to make adequate findings on this element to enable us to determine if the install files were covered by Gate's copyright and it failed to engage in a filtration analysis as to this element. Accordingly, we remand for a determination of whether Gates held a copyright on the install files and for a reconsideration of the install files in light of the test we have set forth herein. 24

The first factor in a fair use enquiry is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." § 107(1). This factor draws on Justice Story's formulation, "the nature and objects of the selections made." Folsom v. Marsh, supra, at 348. The enquiry here may be guided by the examples given in the preamble to § 107, looking to whether the use is for criticism, or comment, or news reporting, and the like, see § 107. The central purpose of this investigation is to see, in Justice Story's words, whether the new work merely "supersede[s] the objects" of the original creation, Folsom v. Marsh, supra, at 348; accord, Harper & Row, supra, at 562 ("supplanting" the original), or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is "transformative." Leval 1111. Although such transformative use is not absolutely necessary for a finding of fair use, Sony, supra, at 455, n. 40, the goal of copyright, to promote science and the arts, is generally furthered by the creation of transformative works. Such works thus lie at the heart of the fair use doctrine's guarantee of breathing space within the confines of copyright, see, e. g., Sony, supra, at 478-480 (BLACKMUN, J., dissenting), and the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use. This Court has only once before even considered whether parody may be fair use, and that time issued no opinion because of the Court's equal division. Benny v. Loew's Inc., 239 F.2d 532 (CA9 1956), aff'd sub nom. Columbia Broadcasting System, Inc. v. Loew's Inc., 356 U.S. 43, 2 L. Ed. 2d 583, 78 S. Ct. 667 (1958). Suffice it to say now that parody has an obvious claim to transformative value, as Acuff-Rose itself does not deny. Like less ostensibly humorous forms of criticism, it can provide social benefit, by shedding light on an earlier work, and, in the process, creating a new one. We thus line up with the courts that have held that parody, like other comment or criticism, may claim fair use under § 107. See, e. g., Fisher v. Dees, 794 F.2d 432 (CA9 1986) ("When Sonny Sniffs Glue," a parody of "When Sunny Gets Blue," is fair use); Elsmere Music, Inc. v. National Broadcasting Co., 482 F. Supp. 741 (SDNY), aff'd, 623 F.2d 252 (CA2 1980) ("I Love Sodom," a "Saturday Night Live" television parody of "I Love New York," is fair use); see also House Report, p. 65; Senate Report, [1966] p. 61 ("Use in a parody of some of the content of the work parodied" may be fair use). The germ of parody lies in the definition of the Greek parodeia, quoted in Judge Nelson's Court of Appeals dissent, as "a song sung alongside another." 972 F.2d at 1440, quoting 7 Encyclopedia Britannica 768 (15th ed. 1975). Modern dictionaries accordingly describe a parody as a "literary or artistic work that imitates the characteristic style of an author or a work for comic effect or ridicule," or as a "composition in prose or verse in which the characteristic turns of thought and phrase in an author or class of authors are imitated in such a way as to make them appear ridiculous." For the purposes of copyright law, the nub of the definitions, and the heart of any parodist's claim to quote from existing material, is the use of some elements of a prior author's composition to create a new one that, at least in part, comments on that author's works. See, e. g., Fisher v. Dees, supra, at 437; MCA, Inc. v. Wilson, 677 F.2d 180, 185 (CA2 1981). If, on the contrary, the commentary has no critical bearing on the substance or style of the original composition, which the alleged infringer merely uses to get attention or to avoid the drudgery in working up something fresh, the claim to fairness in borrowing from another's work diminishes accordingly (if it does not vanish), and other factors, like the extent of its commerciality, loom larger. Parody needs to mimic an original to make its point, and so has some claim to use the creation of its victim's (or collective victims') imagination, whereas satire can stand on its own two feet and so requires justification for the very act of borrowing. See ibid.; Bisceglia, Parody and Copyright Protection: Turning the Balancing Act Into a Juggling Act, in ASCAP, Copyright Law Symposium, No. 34, p. 25 (1987). The fact that parody can claim legitimacy for some appropriation does not, of course, tell either parodist or judge much about where to draw the line. Like a book review quoting the copyrighted material criticized, parody may or may not be fair use, and petitioners' suggestion that any parodic use is presumptively fair has no more justification in law or fact than the equally hopeful claim that any use for news reporting should be presumed fair, see Harper & Row, 471 U.S. at 561. The Act has no hint of an evidentiary preference for parodists over their victims, and no workable presumption for parody could take account of the fact that parody often shades into satire when society is lampooned through its creative artifacts, or that a work may contain both parodic and nonparodic elements. Accordingly, parody, like any other use, has to work its way through the relevant factors, and be judged case by case, in light of the ends of the copyright law. Here, the District Court held, and the Court of Appeals assumed, that 2 Live Crew's "Pretty Woman" contains parody, commenting on and criticizing the original work, whatever it may have to say about society at large. As the District Court remarked, the words of 2 Live Crew's song copy the original's first line, but then "quickly degenerate into a play on words, substituting predictable lyrics with shocking ones . . . [that] derisively demonstrate how bland and banal the Orbison song seems to them." 754 F. Supp. at 1155 (footnote omitted). Judge Nelson, dissenting below, came to the same conclusion, that the 2 Live Crew song "was clearly intended to ridicule the white-bread original" and "reminds us that sexual congress with nameless streetwalkers is not necessarily the stuff of romance and is not necessarily without its consequences. The singers (there are several) have the same thing on their minds as did the lonely man with the nasal voice, but here there is no hint of wine and roses." 972 F.2d at 1442. Although the majority below had difficulty discerning any criticism of the original in 2 [1967] Live Crew's song, it assumed for purposes of its opinion that there was some. Id., at 1435-1436, and n.8. We have less difficulty in finding that critical element in 2 Live Crew's song than the Court of Appeals did, although having found it we will not take the further step of evaluating its quality. The threshold question when fair use is raised in defense of parody is whether a parodic character may reasonably be perceived. Whether, going beyond that, parody is in good taste or bad does not and should not matter to fair use. As Justice Holmes explained, "it would be a dangerous undertaking for persons trained only to the law to constitute themselves final judges of the worth of [a work], outside of the narrowest and most obvious limits. At the one extreme some works of genius would be sure to miss appreciation. Their very novelty would make them repulsive until the public had learned the new language in which their author spoke." Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251, 47 L. Ed. 460, 23 S. Ct. 298 (1903) (circus posters have copyright protection); cf. Yankee Publishing Inc. v. News America Publishing, Inc., 809 F. Supp. 267, 280 (SDNY 1992) (Leval, J.) ("First Amendment protections do not apply only to those who speak clearly, whose jokes are funny, and whose parodies succeed") (trademark case). While we might not assign a high rank to the parodic element here, we think it fair to say that 2 Live Crew's song reasonably could be perceived as commenting on the original or criticizing it, to some degree. 2 Live Crew juxtaposes the romantic musings of a man whose fantasy comes true, with degrading taunts, a bawdy demand for sex, and a sigh of relief from paternal responsibility. The later words can be taken as a comment on the naivete of the original of an earlier day, as a rejection of its sentiment that ignores the ugliness of street life and the debasement that it signifies. It is this joinder of reference and ridicule that marks off the author's choice of parody from the other types of comment and criticism that traditionally have had a claim to fair use protection as transformative works. The Court of Appeals, however, immediately cut short the enquiry into 2 Live Crew's fair use claim by confining its treatment of the first factor essentially to one relevant fact, the commercial nature of the use. The court then inflated the significance of this fact by applying a presumption ostensibly culled from Sony, that "every commercial use of copyrighted material is presumptively . . . unfair . . . ." Sony, 464 U.S. at 451. In giving virtually dispositive weight to the commercial nature of the parody, the Court of Appeals erred. The language of the statute makes clear that the commercial or nonprofit educational purpose of a work is only one element of the first factor enquiry into its purpose and character. Section 107(1) uses the term "including" to begin the dependent clause referring to commercial use, and the main clause speaks of a broader investigation into "purpose and character." As we explained in Harper & Row, Congress resisted attempts to narrow the ambit of this traditional enquiry by adopting categories of presumptively fair use, and it urged courts to preserve the breadth of their traditionally ample view of the universe of relevant evidence. 471 U.S. at 561; House Report, p. 66. Accordingly, the mere fact that a use is educational and not for profit does not insulate it from a finding of infringement, any more than the commercial character of a use bars a finding of fairness. If, indeed, commerciality carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107, including news reporting, comment, criticism, teaching, scholarship, and research, since these activities "are generally conducted for profit in this country." Harper & Row, supra, at 592 (Brennan, J., dissenting). Congress could not have intended such a rule, which certainly is not inferable from the common-law cases, arising as they did from the world of letters in which Samuel Johnson could pronounce that "no man but a blockhead ever wrote, except for money." 3 Boswell's Life of Johnson 19 (G. Hill ed. 1934). Sony itself called for no hard evidentiary presumption. There, we emphasized the need for a "sensitive balancing of interests," 464 [1968] U.S. at 455, n.40, noted that Congress had "eschewed a rigid, bright-line approach to fair use," id., at 449, n.31, and stated that the commercial or nonprofit educational character of a work is "not conclusive," id., at 448-449, but rather a fact to be "weighed along with other[s] in fair use decisions," id., at 449, n.32 (quoting House Report, p. 66). The Court of Appeals's elevation of one sentence from Sony to a per se rule thus runs as much counter to Sony itself as to the long common-law tradition of fair use adjudication. Rather, as we explained in Harper & Row, Sony stands for the proposition that the "fact that a publication was commercial as opposed to nonprofit is a separate factor that tends to weigh against a finding of fair use." 471 U.S. at 562. But that is all, and the fact that even the force of that tendency will vary with the context is a further reason against elevating commerciality to hard presumptive significance. The use, for example, of a copyrighted work to advertise a product, even in a parody, will be entitled to less indulgence under the first factor of the fair use enquiry than the sale of a parody for its own sake, let alone one performed a single time by students in school. See generally Patry & Perlmutter 679-680; Fisher v. Dees, 794 F.2d at 437; Maxtone-Graham v. Burtchaell, 803 F.2d 1253, 1262 (CA2 1986); Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510, 1522 (CA9 1992). 18

The second statutory factor, "the nature of the copyrighted work," § 107(2), draws on Justice Story's expression, the "value of the materials used." Folsom v. Marsh, 9 F. Cas. at 348. This factor calls for recognition that some works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied. See, e.g., Stewart v. Abend, 495 U.S. at 237-238 (contrasting fictional short story with factual works); Harper & Row, 471 U.S. at 563-564 (contrasting soon-to-be-published memoir with published speech); Sony, 464 U.S. at 455, n.40 (contrasting motion pictures with news broadcasts); Feist, 499 U.S. at 348-351 (contrasting creative works with bare factual compilations); 3 M. Nimmer & D. Nimmer, Nimmer on Copyright § 13.05[A][2] (1993) (hereinafter Nimmer); Leval 1116. We agree with both the District Court and the Court of Appeals that the Orbison original's creative expression for public dissemination falls within the core of the copyright's protective purposes. 754 F. Supp. at 1155-1156; 972 F.2d at 1437. This fact, however, is not much help in this case, or ever likely to help much in separating the fair use sheep from the infringing goats in a parody case, since parodies almost invariably copy publicly known, expressive works.

The third factor asks whether "the amount and substantiality of the portion used in relation to the copyrighted work as a whole," § 107(3) (or, in Justice Story's words, "the quantity and value of the materials used," Folsom v. Marsh, supra, at 348) are reasonable in relation to the purpose of the copying. Here, attention turns to the persuasiveness of a parodist's justification for the particular copying done, and the enquiry will harken back to the first of the statutory factors, for, as in prior cases, we recognize that the extent of permissible copying varies with the purpose and character of the use. See Sony, supra, at 449-450 (reproduction of entire work "does not have its ordinary effect of militating against a finding of fair use " as to home videotaping of television programs); Harper & Row, supra, at 564 ("Even substantial quotations might qualify as fair use in a review of a published work or a news account of a speech" but not in a scoop of a soon-to-be-published memoir). The facts bearing on this factor will also tend to address the fourth, by revealing the degree to which the parody may serve as a market substitute for the original or potentially licensed derivatives. See Leval 1123. The District Court considered the song's parodic purpose in finding that 2 Live Crew had not helped themselves overmuch. 754 F. [1969] Supp. at 1156-1157. The Court of Appeals disagreed, stating that "while it may not be inappropriate to find that no more was taken than necessary, the copying was qualitatively substantial. . . . We conclude that taking the heart of the original and making it the heart of a new work was to purloin a substantial portion of the essence of the original." 972 F.2d at 1438. The Court of Appeals is of course correct that this factor calls for thought not only about the quantity of the materials used, but about their quality and importance, too. In Harper & Row, for example, the Nation had taken only some 300 words out of President Ford's memoirs, but we signaled the significance of the quotations in finding them to amount to "the heart of the book," the part most likely to be news-worthy and important in licensing serialization. 471 U.S. at 564-566, 568 (internal quotation marks omitted). We also agree with the Court of Appeals that whether "a substantial portion of the infringing work was copied verbatim" from the copyrighted work is a relevant question, see id., at 565, for it may reveal a dearth of transformative character or purpose under the first factor, or a greater likelihood of market harm under the fourth; a work composed primarily of an original, particularly its heart, with little added or changed, is more likely to be a merely superseding use, fulfilling demand for the original. Where we part company with the court below is in applying these guides to parody, and in particular to parody in the song before us. Parody presents a difficult case. Parody's humor, or in any event its comment, necessarily springs from recognizable allusion to its object through distorted imitation. Its art lies in the tension between a known original and its parodic twin. When parody takes aim at a particular original work, the parody must be able to "conjure up" at least enough of that original to make the object of its critical wit recognizable. See, e.g., Elsmere Music, 623 F.2d at 253, n.1; Fisher v. Dees, 794 F.2d at 438-439. What makes for this recognition is quotation of the original's most distinctive or memorable features, which the parodist can be sure the audience will know. Once enough has been taken to assure identification, how much more is reasonable will depend, say, on the extent to which the song's overriding purpose and character is to parody the original or, in contrast, the likelihood that the parody may serve as a market substitute for the original. But using some characteristic features cannot be avoided. We think the Court of Appeals was insufficiently appreciative of parody's need for the recognizable sight or sound when it ruled 2 Live Crew's use unreasonable as a matter of law. It is true, of course, that 2 Live Crew copied the characteristic opening bass riff (or musical phrase) of the original, and true that the words of the first line copy the Orbison lyrics. But if quotation of the opening riff and the first line may be said to go to the "heart" of the original, the heart is also what most readily conjures up the song for parody, and it is the heart at which parody takes aim. Copying does not become excessive in relation to parodic purpose merely because the portion taken was the original's heart. If 2 Live Crew had copied a significantly less memorable part of the original, it is difficult to see how its parodic character would have come through. See Fisher v. Dees, supra, at 439. This is not, of course, to say that anyone who calls himself a parodist can skim the cream and get away scot free. In parody, as in news reporting, see Harper & Row, supra, context is everything, and the question of fairness asks what else the parodist did besides go to the heart of the original. It is significant that 2 Live Crew not only copied the first line of the original, but thereafter departed markedly from the Orbison lyrics for its own ends. 2 Live Crew not only copied the bass riff and repeated it, but also produced otherwise distinctive sounds, interposing "scraper" noise, overlaying the music with solos in different keys, and altering the drum beat. See 754 F. Supp. at 1155. This is not a case, then, where "a substantial portion" of the parody itself is composed of a "verbatim" copying of the original. It is not, that is, a case where the parody is so insubstantial, as compared to the copying, that the third factor must be resolved as a matter of law against the parodists. Suffice it to say here that, as to the lyrics, we think the Court of Appeals correctly suggested that "no more was taken than necessary," 972 F.2d at 1438, but just for that reason, we fail to see how the copying can be excessive in relation to its parodic purpose, even if the portion taken is the original's "heart." As to the music, we express no opinion whether repetition of the bass riff is excessive copying, and we remand to permit evaluation of the amount taken, in [1970] light of the song's parodic purpose and character, its transformative elements, and considerations of the potential for market substitution sketched more fully below.

The fourth fair use factor is "the effect of the use upon the potential market for or value of the copyrighted work." § 107(4). It requires courts to consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also "whether unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market" for the original. Nimmer § 13.05[A][4], p. 13-102.61 (footnote omitted); accord, Harper & Row, 471 U.S. at 569; Senate Report, p. 65; Folsom v. Marsh, 9 F. Cas. at 349. The enquiry "must take account not only of harm to the original but also of harm to the market for derivative works." Harper & Row, supra, at 568. Since fair use is an affirmative defense, its proponent would have difficulty carrying the burden of demonstrating fair use without favorable evidence about relevant markets. In moving for summary judgment, 2 Live Crew left themselves at just such a disadvantage when they failed to address the effect on the market for rap derivatives, and confined themselves to uncontroverted submissions that there was no likely effect on the market for the original. They did not, however, thereby subject themselves to the evidentiary presumption applied by the Court of Appeals. In assessing the likelihood of significant market harm, the Court of Appeals quoted from language in Sony that "'if the intended use is for commercial gain, that likelihood may be presumed. But if it is for a noncommercial purpose, the likelihood must be demonstrated.'" 972 F.2d at 1438, quoting Sony, 464 U.S. at 451. The court reasoned that because "the use of the copyrighted work is wholly commercial, . . . we presume that a likelihood of future harm to Acuff-Rose exists." 972 F.2d at 1438. In so doing, the court resolved the fourth factor against 2 Live Crew, just as it had the first, by applying a presumption about the effect of commercial use, a presumption which as applied here we hold to be error. No "presumption" or inference of market harm that might find support in Sony is applicable to a case involving something beyond mere duplication for commercial purposes. Sony's discussion of a presumption contrasts a context of verbatim copying of the original in its entirety for commercial purposes, with the noncommercial context of Sony itself (home copying of television programming). In the former circumstances, what Sony said simply makes common sense: when a commercial use amounts to mere duplication of the entirety of an original, it clearly "supersede[s] the objects," Folsom v. Marsh, supra, at 348, of the original and serves as a market replacement for it, making it likely that cognizable market harm to the original will occur. Sony, supra, at 451. But when, on the contrary, the second use is transformative, market substitution is at least less certain, and market harm may not be so readily inferred. Indeed, as to parody pure and simple, it is more likely that the new work will not affect the market for the original in a way cognizable under this factor, that is, by acting as a substitute for it ("superseding [its] objects"). See Leval 1125; Patry & Perlmutter 692, 697-698. This is so because the parody and the original usually serve different market functions. Bisceglia, ASCAP, Copyright Law Symposium, No. 34, at 23. We do not, of course, suggest that a parody may not harm the market at all, but when a lethal parody, like a scathing theater review, kills demand for the original, it does not produce a harm cognizable under the Copyright Act. Because "parody may quite legitimately aim at garroting the original, destroying it commercially as well as artistically," B. Kaplan, An Unhurried View of Copyright 69 (1967), the role of the courts is to distinguish between "biting criticism [that merely] suppresses demand [and] copyright infringement[, which] usurps it. " Fisher v. Dees, 794 F.2d at 438. This distinction between potentially remediable displacement and unremediable disparagement is reflected in the rule that there is no protectible derivative market for criticism. The market for potential derivative uses includes only those that creators of [1971] original works would in general develop or license others to develop. Yet the unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions removes such uses from the very notion of a potential licensing market. "People ask . . . for criticism, but they only want praise." S. Maugham, Of Human Bondage 241 (Penguin ed. 1992). Thus, to the extent that the opinion below may be read to have considered harm to the market for parodies of "Oh, Pretty Woman," see 972 F.2d at 1439, the court erred. Accord, Fisher v. Dees, supra, at 437; Leval 1125; Patry & Perlmutter 688-691. In explaining why the law recognizes no derivative market for critical works, including parody, we have, of course, been speaking of the later work as if it had nothing but a critical aspect (i.e., "parody pure and simple," supra, at 591). But the later work may have a more complex character, with effects not only in the arena of criticism but also in protectible markets for derivative works, too. In that sort of case, the law looks beyond the criticism to the other elements of the work, as it does here. 2 Live Crew's song comprises not only parody but also rap music, and the derivative market for rap music is a proper focus of enquiry, see Harper & Row, supra, at 568; Nimmer § 13.05B. Evidence of substantial harm to it would weigh against a finding of fair use, because the licensing of derivatives is an important economic incentive to the creation of originals. See 17 U.S.C. § 106(2) (copyright owner has rights to derivative works). Of course, the only harm to derivatives that need concern us, as discussed above, is the harm of market substitution. The fact that a parody may impair the market for derivative uses by the very effectiveness of its critical commentary is no more relevant under copyright than the like threat to the original market. Although 2 Live Crew submitted uncontroverted affidavits on the question of market harm to the original, neither they, nor Acuff-Rose, introduced evidence or affidavits addressing the likely effect of 2 Live Crew's parodic rap song on the market for a nonparody, rap version of "Oh, Pretty Woman." And while Acuff-Rose would have us find evidence of a rap market in the very facts that 2 Live Crew recorded a rap parody of "Oh, Pretty Woman" and another rap group sought a license to record a rap derivative, there was no evidence that a potential rap market was harmed in any way by 2 Live Crew's parody, rap version. The fact that 2 Live Crew's parody sold as part of a collection of rap songs says very little about the parody's effect on a market for a rap version of the original, either of the music alone or of the music with its lyrics. The District Court essentially passed on this issue, observing that Acuff-Rose is free to record "whatever version of the original it desires," 754 F. Supp. at 1158; the Court of Appeals went the other way by erroneous presumption. Contrary to each treatment, it is impossible to deal with the fourth factor except by recognizing that a silent record on an important factor bearing on fair use disentitled the proponent of the defense, 2 Live Crew, to summary judgment. The evidentiary hole will doubtless be plugged on remand.

We think that abstracting menu command hierarchies down to their individual word and menu levels and then filtering idea from expression at that stage, as both the Altai and the district court tests require, obscures the more fundamental question of whether a menu command hierarchy can be copyrighted at all. The initial inquiry should not be whether individual components of a menu command hierarchy are expressive, but rather whether the menu command hierarchy as a whole can be copyrighted. But see Gates Rubber Co. v. Bando Chem. Indus., Ltd., 9 F.3d 823 (10th Cir. 1993) (endorsing Altai's abstraction-filtration-comparison test as a way of determining whether "menus and sorting criteria" are copyrightable).
. . .
The district court held that the Lotus menu command hierarchy, with its specific choice and arrangement of command terms, constituted an "expression" of the "idea" of operating a computer program with commands arranged hierarchically into menus and submenus. Borland II, 799 F. Supp. at 216. Under the district court's reasoning, Lotus's decision to employ hierarchically arranged command terms to operate its program could not foreclose its competitors from also employing hierarchically arranged command terms to operate their programs, but it did foreclose them from employing the specific command terms and arrangement that Lotus had used. In effect, the district court limited Lotus 1-2-3's "method of operation" to an abstraction.

Accepting the district court's finding that the Lotus developers made some expressive choices in choosing and arranging the Lotus command terms, we nonetheless hold that that expression is not copyrightable because it is part of Lotus 1-2-3's "method of operation." We do not think that "methods of operation" are limited to abstractions; rather, they are the means by which a user operates something. If specific words are essential to operating something, then they are part of a "method of operation" and, as such, are unprotectable. This is so whether they must be highlighted, typed in, or even spoken, as computer programs no doubt will soon be controlled by spoken words.

The fact that Lotus developers could have designed the Lotus menu command hierarchy differently is immaterial to the question of whether it is a "method of operation." In other words, our initial inquiry is not whether the Lotus menu command hierarchy incorporates any expression. Rather, our initial inquiry is whether the Lotus menu command hierarchy is a "method of operation." Concluding, as we do, that users operate Lotus 1-2-3 by using the Lotus menu command hierarchy, and that the entire Lotus menu command hierarchy is essential to operating Lotus 1-2-3, we do not inquire further whether that method of operation could have been designed differently. The "expressive" choices of what to name the command terms and how to arrange them do not magically change the uncopyrightable menu command hierarchy into copyrightable subject matter.

7]The fourth fair use factor is "the effect of the use upon the potential market for or value of the copyrighted work." § 107(4). It requires courts to consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also "whether unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market" for the original. Nimmer § 13.05[A][4], p. 13-102.61 (footnote omitted); accord, Harper & Row, 471 U.S. at 569; Senate Report, p. 65; Folsom v. Marsh, 9 F. Cas. at 349. The enquiry "must take account not only of harm to the original but also of harm to the market for derivative works." Harper & Row, supra, at 568. Since fair use is an affirmative defense, its proponent would have difficulty carrying the burden of demonstrating fair use without favorable evidence about relevant markets. In moving for summary judgment, 2 Live Crew left themselves at just such a disadvantage when they failed to address the effect on the market for rap derivatives, and confined themselves to uncontroverted submissions that there was no likely effect on the market for the original. They did not, however, thereby subject themselves to the evidentiary presumption applied by the Court of Appeals. In assessing the likelihood of significant market harm, the Court of Appeals quoted from language in Sony that "'if the intended use is for commercial gain, that likelihood may be presumed. But if it is for a noncommercial purpose, the likelihood must be demonstrated.'" 972 F.2d at 1438, quoting Sony, 464 U.S. at 451. The court reasoned that because "the use of the copyrighted work is wholly commercial, . . . we presume that a likelihood of future harm to Acuff-Rose exists." 972 F.2d at 1438. In so doing, the court resolved the fourth factor against 2 Live Crew, just as it had the first, by applying a presumption about the effect of commercial use, a presumption which as applied here we hold to be error.
. . .
[19]We do not, of course, suggest that a parody may not harm the market at all, but when a lethal parody, like a scathing theater review, kills demand for the original, it does not produce a harm cognizable under the Copyright Act. Because "parody may quite legitimately aim at garroting the original, destroying it commercially as well as artistically," B. Kaplan, An Unhurried View of Copyright 69 (1967), the role of the courts is to distinguish between "biting criticism [that merely] suppresses demand [and] copyright infringement[, which] usurps it. " Fisher v. Dees, 794 F.2d at 438. This distinction between potentially remediable displacement and unremediable disparagement is reflected in the rule that there is no protectible derivative market for criticism. The market for potential derivative uses includes only those that creators of [1971] original works would in general develop or license others to develop. Yet the unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions removes such uses from the very notion of a potential licensing market. "People ask . . . for criticism, but they only want praise." S. Maugham, Of Human Bondage 241 (Penguin ed. 1992). Thus, to the extent that the opinion below may be read to have considered harm to the market for parodies of "Oh, Pretty Woman," see 972 F.2d at 1439, the court erred. Accord, Fisher v. Dees, supra, at 437; Leval 1125; Patry & Perlmutter 688-691. In explaining why the law recognizes no derivative market for critical works, including parody, we have, of course, been speaking of the later work as if it had nothing but a critical aspect (i.e., "parody pure and simple," supra, at 591). But the later work may have a more complex character, with effects not only in the arena of criticism but also in protectible markets for derivative works, too. In that sort of case, the law looks beyond the criticism to the other elements of the work, as it does here. 2 Live Crew's song comprises not only parody but also rap music, and the derivative market for rap music is a proper focus of enquiry, see Harper & Row, supra, at 568; Nimmer § 13.05B. Evidence of substantial harm to it would weigh against a finding of fair use, because the licensing of derivatives is an important economic incentive to the creation of originals. See 17 U.S.C. § 106(2) (copyright owner has rights to derivative works). Of course, the only harm to derivatives that need concern us, as discussed above, is the harm of market substitution. The fact that a parody may impair the market for derivative uses by the very effectiveness of its critical commentary is no more relevant under copyright than the like threat to the original market. Although 2 Live Crew submitted uncontroverted affidavits on the question of market harm to the original, neither they, nor Acuff-Rose, introduced evidence or affidavits addressing the likely effect of 2 Live Crew's parodic rap song on the market for a nonparody, rap version of "Oh, Pretty Woman." And while Acuff-Rose would have us find evidence of a rap market in the very facts that 2 Live Crew recorded a rap parody of "Oh, Pretty Woman" and another rap group sought a license to record a rap derivative, there was no evidence that a potential rap market was harmed in any way by 2 Live Crew's parody, rap version. The fact that 2 Live Crew's parody sold as part of a collection of rap songs says very little about the parody's effect on a market for a rap version of the original, either of the music alone or of the music with its lyrics. The District Court essentially passed on this issue, observing that Acuff-Rose is free to record "whatever version of the original it desires," 754 F. Supp. at 1158; the Court of Appeals went the other way by erroneous presumption. Contrary to each treatment, it is impossible to deal with the fourth factor except by recognizing that a silent record on an important factor bearing on fair use disentitled the proponent of the defense, 2 Live Crew, to summary judgment. The evidentiary hole will doubtless be plugged on remand.

We also note that in most contexts, there is no need to "build" upon other people's expression, for the ideas conveyed by that expression can be conveyed by someone else without copying the first author's expression. In the context of methods of operation, however, "building" requires the use of the precise method of operation already employed; otherwise, "building" would require dismantling, too. Original developers are not the only people entitled to build on the methods of operation they create; anyone can. Thus, Borland may build on the method of operation that Lotus designed and may use the Lotus menu command hierarchy in doing so.

Concurrence follows. BOUDIN, Circuit Judge, concurring. The importance of this case, and a slightly different emphasis in my view of the underlying problem, prompt me to add a few words to the majority's tightly focused discussion. I. Most of the law of copyright and the "tools" of analysis have developed in the context of literary works such as novels, plays, and films. In this milieu, the principal problem--simply stated, if difficult to resolve--is to stimulate creative expression without unduly limiting access by others to the broader themes and concepts deployed by the author. The middle of the spectrum presents close cases; but a "mistake" in providing too much protection involves a small cost: subsequent authors treating the same themes must take a few more steps away from the original expression. The problem presented by computer programs is fundamentally different in one respect. The computer program is a means for causing something to happen; it has a mechanical utility, an instrumental role, in accomplishing the world's work. Granting protection, in other words, can have some of the consequences of patent protection in limiting other people's ability to perform a task in the most efficient manner. Utility does not bar copyright (dictionaries may be copyrighted), but it alters the calculus. Of course, the argument for protection is undiminished, perhaps even enhanced, by utility: if we want more of an intellectual product, a temporary monopoly for the creator provides incentives for others to create other, different items in this class. But the "cost" side of the equation may be different where one places a very high value on public access to a useful innovation that may be the most efficient means of performing a given task. Thus, the argument for extending protection may be the same; but the stakes on the other side are much higher. It is no accident that patent protection has preconditions that copyright protection does not--notably, the requirements of novelty and non-obviousness--and that patents are granted for a shorter period than copyrights. This problem of utility has sometimes manifested itself in copyright cases, such as Baker v. Selden, 101 U.S. 99, 25 L. Ed. 841 (1879), and been dealt with through various formulations that limit copyright or create limited rights to copy. But the case law and doctrine addressed to utility in copyright have been brief detours in the general march of copyright law. Requests for the protection of computer menus present the concern with fencing off access to the commons in an acute form. A new menu may be a creative work, but over time its importance may come to reside more in the investment that has been made by users in learning the menu and in building their own mini-programs--macros--in reliance upon the menu. Better typewriter keyboard layouts may exist, but the familiar QWERTY keyboard dominates the market because that is what everyone has learned to use. See P. David, CLIO and the Economics of QWERTY, 75 Am. Econ. Rev. 332 (1985). The QWERTY keyboard is nothing other than a menu of letters. Thus, to assume that computer programs are just one more new means of expression, like a filmed play, may be quite wrong. The "form"--the written source code or the menu structure depicted on the screen--look hauntingly like the familiar stuff of copyright; but the "substance" probably has more to do with problems presented in patent law or, as already noted, in those rare cases where copyright law has confronted industrially useful expressions. Applying copyright law to computer programs is like assembling a jigsaw puzzle whose pieces do not quite fit. All of this would make no difference if Congress had squarely confronted the issue, and given explicit directions as to what should be done. The Copyright Act of 1976 took a different course. While Congress said that computer programs might be subject to copyright protection, it said this in very general terms; and, especially in § 102(b), Congress adopted a string of exclusions that if taken literally might easily seem to exclude most computer programs from protection. The only detailed prescriptions for computers involve narrow issues (like back-up copies) of no relevance here. Of course, one could still read the statute as a congressional command that the familiar doctrines of copyright law be taken and applied to computer programs, in cookie cutter fashion, as if the programs were novels or play scripts. Some of the cases involving computer programs embody this approach. It seems to me mistaken on two different grounds: the tradition of copyright law, and the likely intent of Congress. The broad-brush conception of copyright protection, the time limits, and the formalities have long been prescribed by statute. But the heart of copyright doctrine--what may be protected and with what limitations and exceptions--has been developed by the courts through experience with individual cases. B. Kaplan, An Unhurried View of Copyright 40 (1967). Occasionally Congress addresses a problem in detail. For the most part the interstitial development of copyright through the courts is our tradition. Nothing in the language or legislative history of the 1976 Act, or at least nothing brought to our attention, suggests that Congress meant the courts to abandon this case-by-case approach. Indeed, by setting up § 102(b) as a counterpoint theme, Congress has arguably recognized the tension and left it for the courts to resolve through the development of case law. And case law development is adaptive: it allows new problems to be solved with help of earlier doctrine, but it does not preclude new doctrines to meet new situations. II. In this case, the raw facts are mostly, if not entirely, undisputed. Although the inferences to be drawn may be more debatable, it is very hard to see that Borland has shown any interest in the Lotus menu except as a fall-back option for those users already committed to it by prior experience or in order to run their own macros using 1-2-3 commands. At least for the amateur, accessing the Lotus menu in the Borland Quattro or Quattro Pro program takes some effort. Put differently, it is unlikely that users who value the Lotus menu for its own sake--independent of any investment they have made themselves in learning Lotus' commands or creating macros dependent upon them--would choose the Borland program in order to secure access to the Lotus menu. Borland's success is due primarily to other features. Its rationale for deploying the Lotus menu bears the ring of truth. Now, any use of the Lotus menu by Borland is a commercial use and deprives Lotus of a portion of its "reward," in the sense that an infringement claim if allowed would increase Lotus' profits. But this is circular reasoning: broadly speaking, every limitation on copyright or privileged use diminishes the reward of the original creator. Yet not every writing is copyrightable or every use an infringement. The provision of reward is one concern of copyright law, but it is not the only one. If it were, copyrights would be perpetual and there would be no exceptions. The present case is an unattractive one for copyright protection of the menu. The menu commands (e.g., "print," "quit") are largely for standard procedures that Lotus did not invent and are common words that Lotus cannot monopolize. What is left is the particular combination and sub-grouping of commands in a pattern devised by Lotus. This arrangement may have a more appealing logic and ease of use than some other configurations; but there is a certain arbitrariness to many of the choices. If Lotus is granted a monopoly on this pattern, users who have learned the command structure of Lotus 1-2-3 or devised their own macros are locked into Lotus, just as a typist who has learned the QWERTY keyboard would be the captive of anyone who had a monopoly on the production of such a keyboard. Apparently, for a period Lotus 1-2-3 has had such sway in the market that it has represented the de facto standard for electronic spreadsheet commands. So long as Lotus is the superior spreadsheet--either in quality or in price--there may be nothing wrong with this advantage. But if a better spreadsheet comes along, it is hard to see why customers who have learned the Lotus menu and devised macros for it should remain captives of Lotus because of an investment in learning made by the users and not by Lotus. Lotus has already reaped a substantial reward for being first; assuming that the Borland program is now better, good reasons exist for freeing it to attract old Lotus customers: to enable the old customers to take advantage of a new advance, and to reward Borland in turn for making a better product. If Borland has not made a better product, then customers will remain with Lotus anyway. Thus, for me the question is not whether Borland should prevail but on what basis. Various avenues might be traveled, but the main choices are between holding that the menu is not protectable by copyright and devising a new doctrine that Borland's use is privileged. No solution is perfect and no intermediate appellate court can make the final choice. To call the menu a "method of operation" is, in the common use of those words, a defensible position. After all, the purpose of the menu is not to be admired as a work of literary or pictorial art. It is to transmit directions from the user to the computer, i.e., to operate the computer. The menu is also a "method" in the dictionary sense because it is a "planned way of doing something," an "order or system," and (aptly here) an "orderly or systematic arrangement, sequence or the like." Random House Webster's College Dictionary 853 (1991). A different approach would be to say that Borland's use is privileged because, in the context already described, it is not seeking to appropriate the advances made by Lotus' menu; rather, having provided an arguably more attractive menu of its own, Borland is merely trying to give former Lotus users an option to exploit their own prior investment in learning or in macros. The difference is that such a privileged use approach would not automatically protect Borland if it had simply copied the Lotus menu (using different codes), contributed nothing of its own, and resold Lotus under the Borland label. The closest analogue in conventional copyright is the fair use doctrine. E.g., Harper & Row, Publishers, Inc. v. Nation Enters., 471 U.S. 539, 85 L. Ed. 2d 588, 105 S. Ct. 2218 (1985). Although invoked by Borland, it has largely been brushed aside in this case because the Supreme Court has said that it is "presumptively" unavailable where the use is a "commercial" one. See id. at 562. But see Campbell v. Acuff-Rose Music, Inc., 127 L. Ed. 2d 500, 114 S. Ct. 1164, 1174 (1994). In my view, this is something less than a definitive answer; "presumptively" does not mean "always" and, in any event, the doctrine of fair use was created by the courts and can be adapted to new purposes. But a privileged use doctrine would certainly involve problems of its own. It might more closely tailor the limits on copyright protection to the reasons for limiting that protection; but it would entail a host of administrative problems that would cause cost and delay, and would also reduce the ability of the industry to predict outcomes. Indeed, to the extent that Lotus' menu is an important standard in the industry, it might be argued that any use ought to be deemed privileged. In sum, the majority's result persuades me and its formulation is as good, if not better, than any other that occurs to me now as within the reach of courts. Some solutions (e.g., a very short copyright period for menus) are not options at all for courts but might be for Congress. In all events, the choices are important ones of policy, not linguistics, and they should be made with the underlying considerations in view.

The first factor listed in section 107 is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. § 107(1). Especially pertinent to an assessment of the first fair use factor are the precise circumstances under which copies of the eight Catalysis articles were made. After noticing six of these articles when the original copy of the journal issue containing each of them was circulated to him, Chickering had them photocopied, at least initially, for the same basic purpose that one would normally seek to obtain the original -- to have it available on his shelf for ready reference if and when he needed to look at it. The library circulated one copy and invited all the researchers to make their own photocopies. It is a reasonable inference that the library staff wanted each journal issue moved around the building quickly and returned to the library so that it would be available for others to look at. Making copies enabled all researchers who might one day be interested in examining the contents of an article in the issue to have the article readily available in their own offices. In Chickering's own words, the copies of the articles were made for "my personal convenience," since it is "far more convenient to have access in my office to a photocopy of an article than to have to go to the library each time I wanted to refer to it." Affidavit of Donald Chickering at 11 (submitted as direct trial testimony) [hereinafter Chickering testimony]. Significantly, Chickering did not even have occasion to use five of the photocopied articles at all, further revealing that the photocopies of the eight Catalysis articles were primarily made just for "future retrieval and reference." Id. It is true that photocopying these articles also served other purposes. The most favorable for Texaco is the purpose of enabling Chickering, if the need should arise, to go into the lab with pieces of paper that (a) were not as bulky as the entire issue or a bound volume of a year's issues, and (b) presented no risk of damaging the original by exposure to chemicals. And these purposes might suffice to tilt the first fair use factor in favor of Texaco if these purposes were dominant. For example, if Chickering had asked the library to buy him a copy of the pertinent issue of Catalysis and had placed it on his shelf, and one day while reading it had noticed a chart, formula, or other material that he wanted to take right into the lab, it might be a fair use for him to make a photocopy, and use that copy in the lab (especially if he did not retain it and build up a mini-library of photocopied articles). This is the sort of "spontaneous" copying that is part of the test for permissible nonprofit classroom copying. See Agreement on Guidelines for Classroom Copying in Not-For-Profit Educational Institutions, quoted in Patry, The Fair Use Privilege, at 308. But that is not what happened here as to the six items copied from the circulated issues. As to the other two articles, the circumstances are not quite as clear, but they too appear more to serve the purpose of being additions to Chickering's office "library" than to be spontaneous copying of a critical page that he was reading on his way to the lab. One was copied apparently when he saw a reference to it in another article, which was in an issue circulated to him. The most likely inference is that he decided that he ought to have copies of both items -- again for placement on his shelf for later use if the need arose. The last article was copied, according to his affidavit, when he saw a reference to it "elsewhere." Chickering testimony at 22. What is clear is that this item too was simply placed "on the shelf." As he testified, "I kept a copy to refer to in case I became more involved in support effects research." Id. The photocopying of these eight Catalysis articles may be characterized as "archival" -- i.e., done for the primary purpose of providing numerous Texaco scientists (for whom Chickering served as an example) each with his or her own personal copy of each article without Texaco's having to purchase another original journal. The photocopying "merely 'supersedes the objects' of the original creation," Campbell, 114 S. Ct. at 1171 (quoting Folsom v. Marsh, 9 F. Cas. 342, 348(C.C.D. Mass. 1841) (No. 4,901) ), and tilts the first fair use factor against Texaco. We do not mean to suggest that no instance of archival copying would be fair use, but the first factor tilts against Texaco in this case because the making of copies to be placed on the shelf in Chickering's office is part of a systematic process of encouraging employee researchers to copy articles so as to multiply available copies while avoiding payment. Texaco criticizes three aspects of the District Court's analysis of the first factor. Relying largely on the Supreme Court's discussion of fair use in Sony, the District Court suggested that a secondary user will "win" this first factor by showing a "transformative (or productive) nonsuperseding use of the original, or [a] noncommercial use, generally for a socially beneficial or widely accepted purpose." 802 F. Supp. at 12. The District Court then concluded that Texaco's copying is "neither transformative nor noncommercial," id. at 13: not transformative because Texaco "simply makes mechanical photocopies of the entirety of relevant articles" and the "primary aspect" of Texaco's photocopying is to multiply copies, see id. at 13-15; and not noncommercial because, though it facilitates research, this research is conducted solely for commercial gain, see id. at 15-16. Texaco asserts that the District Court mischaracterized the inquiry under the first factor and overlooked several relevant considerations. First, Texaco contends that the District Court inappropriately focussed on the character of the user rather than the nature of the use in labeling Texaco's copying as commercial. Texaco claims that its status as a for-profit corporation has no bearing on the fair use analysis, and that its use should be considered noncommercial since it photocopied articles in order to aid Chickering's research. Texaco emphasizes that "research" is explicitly listed in the preamble of section 107, a circumstance that Texaco contends should make its copying favored under the first factor and throughout the entire fair use analysis. Second, Texaco contends that the District Court put undue emphasis on whether its use was "transformative," especially since the Supreme Court appears to have rejected the view that a use must be transformative or productive to be a fair use. See Sony, 464 U.S. at 455 n.40 ("The distinction between 'productive' and 'unproductive' uses may be helpful in calibrating the balance [of interests], but it cannot be wholly determinative."). Texaco asserts that the "transformative use" concept is valuable only to the extent that it focusses attention upon whether a second work unfairly competes with the original. Texaco states that in this case, where the photocopies it made were not sold or distributed in competition with the original, the nontransformative nature of its copying should not prevent a finding of fair use. Texaco also suggests that its use should be considered transformative: photocopying the article separated it from a bulky journal, made it more amenable to markings, and provided a document that could be readily replaced if damaged in a laboratory, all of which "transformed" the original article into a form that better served Chickering's research needs. Finally, Texaco claims that it should prevail on the first factor because, as the District Court acknowledged, the type of photocopying it conducted is widespread and has long been considered reasonable and customary. Texaco stresses that some courts and commentators regard custom and common usage as integral to the fair use analysis. See, e.g., Williams & Wilkins Co. v. United States, 487 F.2d 1345, 1353-56, 203 Ct. Cl. 74 (Ct. Cl. 1973), aff'd by equally divided Court, 420 U.S. 376, 43 L. Ed. 2d 264, 95 S. Ct. 1344 (1975); Lloyd L. Weinreb, Fair's Fair: A Comment on the Fair Use Doctrine, 103 Harv. L. Rev. 1137, 1140 (1990) [hereinafter Weinreb, Fair's Fair]. We consider these three lines of attack separately. 1. Commercial use. We generally agree with Texaco's contention that the District Court placed undue emphasis on the fact that Texaco is a for-profit corporation conducting research primarily for commercial gain. Since many, if not most, secondary users seek at least some measure of commercial gain from their use, unduly emphasizing the commercial motivation of a copier will lead to an overly restrictive view of fair use. See Campbell, 114 S. Ct. at 1174; see also Maxtone-Graham v. Burtchaell, 803 F.2d 1253, 1262 (2d Cir. 1986) (noting that if "commercial" nature of a secondary use is over-emphasized in the analysis, "fair use would be virtually obliterated"), cert. denied, 481 U.S. 1059, 95 L. Ed. 2d 856, 107 S. Ct. 2201 (1987). See generally 3 Nimmer on Copyright § 13.05[A][1][c], at 13-162 to 13-163 (categorical rule against commercial uses unwarranted since this "would cause the fair use analysis to collapse in all but the exceptional case of nonprofit exploitation"). Though the Supreme Court had stated in Sony that every commercial use was "presumptively" unfair, see 464 U.S. at 451, that Court and lower courts have come to explain that the commercial nature of a secondary use simply "'tends to weigh against a finding of fair use.'" Campbell, 114 S. Ct. at 1174 (quoting Harper & Row, 471 U.S. at 562); accord Rogers v. Koons, 960 F.2d 301, 309 (2d Cir.), cert. denied, 121 L. Ed. 2d 278, 113 S. Ct. 365 (1992); Sega Enterprises Limited v. Accolade, Inc., 977 F.2d 1510, 1522 (9th Cir. 1992); Maxtone-Graham, 803 F.2d at 1262. Indeed, Campbell warns against 'elevating . . . to a per se rule' Sony's language about a presumption against fair use arising from commercial use. 114 S. Ct. at 1174. Campbell discards that language in favor of a more subtle, sophisticated approach, which recognizes that 'the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.' Id. at 1171. The Court states that 'the commercial or nonprofit educational purpose of a work is only one element of the first factor enquiry' id. at 1174, and points out that 'if, indeed, commerciality carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107 . . . .' Id. We do not mean to suggest that the District Court overlooked these principles; in fact, the Court discussed them insightfully, see 802 F. Supp. at 12-13. Rather, our concern here is that the Court let the for-profit nature of Texaco's activity weigh against Texaco without differentiating between a direct commercial use and the more indirect relation to commercial activity that occurred here. Texaco was not gaining direct or immediate commercial advantage from the photocopying at issue in this case -- i.e., Texaco's profits, revenues, and overall commercial performance were not tied to its making copies of eight Catalysis articles for Chickering. Cf. Basic Books, Inc. v. Kinko's Graphics Corp., 758 F. Supp. 1522 (S.D.N.Y. 1991) (revenues of reprographic business stemmed directly from selling unauthorized photocopies of copyrighted books). Rather, Texaco's photocopying served, at most, to facilitate Chickering's research, which in turn might have led to the development of new products and technology that could have improved Texaco's commercial performance. Texaco's photocopying is more appropriately labeled an "intermediate use." See Sega Enterprises, 977 F.2d at 1522-23 (labeling secondary use "intermediate" and finding first factor in favor of for-profit company, even though ultimate purpose of copying was to develop competing commercial product, because immediate purpose of copying computer code was to study idea contained within computer program). We do not consider Texaco's status as a for-profit company irrelevant to the fair use analysis. Though Texaco properly contends that a court's focus should be on the use of the copyrighted material and not simply on the user, it is overly simplistic to suggest that the "purpose and character of the use" can be fully discerned without considering the nature and objectives of the user. Ultimately, the somewhat cryptic suggestion in section 107(1) to consider whether the secondary use "is of a commercial nature or is for nonprofit educational purposes" connotes that a court should examine, among other factors, the value obtained by the secondary user from the use of the copyrighted material. See Rogers, 960 F.2d at 309 ("The first factor . . . asks whether the original was copied in good faith to benefit the public or primarily for the commercial interests of the infringer."); MCA, Inc. v. Wilson, 677 F.2d 180, 182 (2d Cir. 1981) (court is to consider "whether the alleged infringing use was primarily for public benefit or for private commercial gain"). The commercial/nonprofit dichotomy concerns the unfairness that arises when a secondary user makes unauthorized use of copyrighted material to capture significant revenues as a direct consequence of copying the original work. See Harper & Row, 471 U.S. at 562 ("The crux of the profit/nonprofit distinction is . . . whether the user stands to profit from exploitation of the copyrighted material without paying the customary price."). Consistent with these principles, courts will not sustain a claimed defense of fair use when the secondary use can fairly be characterized as a form of "commercial exploitation," i.e., when the copier directly and exclusively acquires conspicuous financial rewards from its use of the copyrighted material. See Harper & Row, 471 U.S. at 562-63; Twin Peaks, 996 F.2d at 1375; Rogers, 960 F.2d at 309; Iowa State University Research Foundation, Inc. v. American Broadcasting Companies, Inc., 621 F.2d 57, 61 (2d Cir. 1980); Meeropol v. Nizer, 560 F.2d 1061, 1069 (2d Cir. 1977) (examining whether use was "predominantly for commercial exploitation"), cert. denied, 434 U.S. 1013, 54 L. Ed. 2d 756, 98 S. Ct. 727 (1978). Conversely, courts are more willing to find a secondary use fair when it produces a value that benefits the broader public interest. See Twin Peaks, 996 F.2d at 1375; Sega Enterprises, 977 F.2d at 1523; Rosemont Enterprises, Inc. v. Random House, Inc., 366 F.2d 303, 307-09 (2d Cir. 1966), cert. denied, 385 U.S. 1009, 17 L. Ed. 2d 546, 87 S. Ct. 714 (1967). The greater the private economic rewards reaped by the secondary user (to the exclusion of broader public benefits), the more likely the first factor will favor the copyright holder and the less likely the use will be considered fair. As noted before, in this particular case the link between Texaco's commercial gain and its copying is somewhat attenuated: the copying, at most, merely facilitated Chickering's research that might have led to the production of commercially valuable products. Thus, it would not be accurate to conclude that Texaco's copying of eight particular Catalysis articles amounted to "commercial exploitation," especially since the immediate goal of Texaco's copying was to facilitate Chickering's research in the sciences, an objective that might well serve a broader public purpose. See Twin Peaks, 996 F.2d at 1375; Sega Enterprises, 977 F.2d at 1522. Still, we need not ignore the for-profit nature of Texaco's enterprise, especially since we can confidently conclude that Texaco reaps at least some indirect economic advantage from its photocopying. As the publishers emphasize, Texaco's photocopying for Chickering could be regarded simply as another "factor of production" utilized in Texaco's efforts to develop profitable products. Conceptualized in this way, it is not obvious why it is fair for Texaco to avoid having to pay at least some price to copyright holders for the right to photocopy the original articles. 2. Transformative Use. The District Court properly emphasized that Texaco's photocopying was not "transformative." After the District Court issued its opinion, the Supreme Court explicitly ruled that the concept of a "transformative use" is central to a proper analysis under the first factor, see Campbell, 114 S. Ct. at 1171-73. The Court explained that though a "transformative use is not absolutely necessary for a finding of fair use, . . . the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Id. at 1171. The "transformative use" concept is pertinent to a court's investigation under the first factor because it assesses the value generated by the secondary use and the means by which such value is generated. To the extent that the secondary use involves merely an untransformed duplication, the value generated by the secondary use is little or nothing more than the value that inheres in the original. Rather than making some contribution of new intellectual value and thereby fostering the advancement of the arts and sciences, an untransformed copy is likely to be used simply for the same intrinsic purpose as the original, thereby providing limited justification for a finding of fair use. See Weissmann v. Freeman, 868 F.2d 1313, 1324 (2d Cir.) (explaining that a use merely for the same "intrinsic purpose" as original "moves the balance of the calibration on the first factor against" secondary user and "seriously weakens a claimed fair use"), cert. denied, 493 U.S. 883, 107 L. Ed. 2d 172, 110 S. Ct. 219 (1989). In contrast, to the extent that the secondary use "adds something new, with a further purpose or different character," the value generated goes beyond the value that inheres in the original and "the goal of copyright, to promote science and the arts, is generally furthered." Campbell, 114 S. Ct. at 1171,; see also Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1111 (1990) [hereinafter Leval, Toward a Fair Use Standard]. It is therefore not surprising that the "preferred" uses illustrated in the preamble to section 107, such as criticism and comment, generally involve some transformative use of the original work. See 3 Nimmer on Copyright § 13.05[A][1][b], at 13-160. Texaco suggests that its conversion of the individual Catalysis articles through photocopying into a form more easily used in a laboratory might constitute a transformative use. However, Texaco's photocopying merely transforms the material object embodying the intangible article that is the copyrighted original work. See 17 U.S.C. §§ 101, 102 (explaining that copyright protection in literary works subsists in the original work of authorship "regardless of the nature of the material objects . . . in which they are embodied"). Texaco's making of copies cannot properly be regarded as a transformative use of the copyrighted material. See Steven D. Smit, "Make a Copy for the File . . .": Copyright Infringement by Attorneys, 46 Baylor L. Rev. 1, 15 & n.58 (1994); see also Basic Books, 758 F. Supp. at 1530-31 (repackaging in anthology form of excerpts from copyrighted books not a transformative use). Even though Texaco's photocopying is not technically a transformative use of the copyrighted material, we should not overlook the significant independent value that can stem from conversion of original journal articles into a format different from their normal appearance. See generally Sony, 464 U.S. at 454, 455 n.40 (acknowledging possible benefits from copying that might otherwise seem to serve "no productive purpose"); Weinreb, Fair's Fair, at 1143 & n.29 (discussing potential value from nontransformative copying). As previously explained, Texaco's photocopying converts the individual Catalysis articles into a useful format. Before modern photocopying, Chickering probably would have converted the original article into a more serviceable form by taking notes, whether cursory or extended; today he can do so with a photocopying machine. Nevertheless, whatever independent value derives from the more usable format of the photocopy does not mean that every instance of photocopying wins on the first factor. In this case, the predominant archival purpose of the copying tips the first factor against the copier, despite the benefit of a more usable format. 3. Reasonable and Customary Practice. Texaco contends that Chickering's photocopying constitutes a use that has historically been considered "reasonable and customary." We agree with the District Court that whatever validity this argument might have had before the advent of the photocopying licensing arrangements discussed below in our consideration of the fourth fair use factor, the argument today is insubstantial. As the District Court observed, "To the extent the copying practice was 'reasonable' in 1973 [when Williams & Wilkins was decided], it has ceased to be 'reasonable' as the reasons that justified it before [photocopying licensing] have ceased to exist." 802 F. Supp. at 25. In amplification of Texaco's arguments, our dissenting colleague makes two further points about the first factor analysis that merit a response. First, the dissent disputes our characterization of Chickering's use as "archival" on the ground that such a use would occur in an institutional setting, whereas Chickering copied for his personal use. Second, the dissent contends that Chickering's use is transformative because it is an important step in the process of doing research. We think the proper response to these observations emerges from considering how they would fare if the Texaco library had sent around entire books, rather than issues of a journal. Clearly, Chickering (and all the other researchers at the Beacon facility) would be making archival use of the circulating books if they made photocopies of the books for their individual offices and thereby spared Texaco the expense of buying them all individual volumes. An individual copies for archival purposes even if the resulting archive remains in a private office. When a corporation invites such archival copying by circulating items likely to be worth copying (whether articles or entire books), any distinction between individual and institutional archiving loses all significance. Moreover, the concept of a "transformative" use would be extended beyond recognition if it was applied to Chickering's copying simply because he acted in the course of doing research. The purposes illustrated by the categories listed in section 107 refer primarily to the work of authorship alleged to be a fair use, not to the activity in which the alleged infringer is engaged. Texaco cannot gain fair use insulation for Chickering's archival photocopying of articles (or books) simply because such copying is done by a company doing research. It would be equally extravagant for a newspaper to contend that because its business is "news reporting" it may line the shelves of its reporters with photocopies of books on journalism or that schools engaged in "teaching" may supply its faculty members with personal photocopies of books on educational techniques or substantive fields. Whatever benefit copying and reading such books might contribute to the process of "teaching" would not for that reason satisfy the test of a "teaching" purpose. On balance, we agree with the District Court that the first factor favors the publishers, primarily because the dominant purpose of the use is a systematic institutional policy of multiplying the available number of copies of pertinent copyrighted articles by circulating the journals among employed scientists for them to make copies, thereby serving the same purpose for which additional subscriptions are normally sold, or, as will be discussed, for which photocopying licenses may be obtained.

"Though a significant measure of creativity was undoubtedly used in the creation of the eight articles copied from Catalysis, even a glance at their content immediately reveals the predominantly factual nature of these works.11 Moreover, though we have previously recognized the importance of strong copyright protection to provide sufficient incentives for the creation of scientific works, see Weissmann, 868 F.2d at 1325, nearly every category of copyrightable works could plausibly assert that broad copyright protection was essential to the continued vitality of that category of works. Ultimately, then, the manifestly factual character of the eight articles precludes us from considering the articles as "within the core of the copyright's protective purposes," Campbell, --- U.S. at ----, 114 S. Ct. at 1175; see also Harper & Row, 471 U.S. at 563, 105 S. Ct. at 2232 ("The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy."). Thus, in agreement with the District Court, we conclude that the second factor favors Texaco."

As we emphasized at the outset, each article enjoys independent copyright protection, which the authors transferred to Academic Press, and what the publishers claim has been infringed is the copyright that subsists in each individual article--not the distinct copyright that may subsist in each journal issue or volume by virtue of the publishers' original compilation of these articles. The only other appellate court to consider the propriety of photocopying articles from journals also recognized that each article constituted an entire work in the fair use analysis. See Williams & Wilkins, 487 F.2d at 1353. Despite Texaco's claims that we consider its amount of copying "minuscule" in relation to the entirety of Catalysis, we conclude, as did the District Court, that Texaco has copied entire works. Though this conclusion does not preclude a finding of fair use, it militates against such a finding, "though we are sensitive to Texaco's claim that the third factor serves merely as a proxy for determining whether a secondary use significantly interferes with demand for the original--a concern echoed by some commentators, see William W. Fisher III, Reconstructing the Fair Use Doctrine, 101 Harv. L. Rev. 1661, 1678 (1988) [hereinafter Fisher, Reconstructing Fair Use ]--we think this factor serves a further end that advances the fair use analysis. Specifically, by focussing on the amount and substantiality of the original work used by the secondary user, we gain insight into the purpose and character of the use as we consider whether the quantity of the material used was "reasonable in relation to the purpose of the copying."" "In this case, the fact that Texaco photocopied the eight Catalysis articles in their entirety weakens its assertion that the overriding purpose and character of its use was to enable the immediate use of the article in the laboratory and strengthens our view that the predominant purpose and character of the use was to establish a personal library of pertinent articles for Chickering. Cf. id. at ----, 114 S. Ct. at 1176 (intimating that extent of copying can provide insight into primary purpose of copying)."

The fourth statutory fair use factor is "the effect of the use upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). Assessing this factor, the District Court detailed the range of procedures Texaco could use to obtain authorized copies of the articles that it photocopied and found that "whatever combination of procedures Texaco used, the publishers' revenues would grow significantly." 802 F. Supp. at 19. The Court concluded that the publishers "powerfully demonstrated entitlement to prevail as to the fourth factor," since they had shown "a substantial harm to the value of their copyrights" as the consequence of Texaco's copying. See id. at 18-21. Prior to Campbell, the Supreme Court had characterized the fourth factor as "the single most important element of fair use," Harper & Row, 471 U.S. at 566; accord 3 Nimmer on Copyright § 13.05[A][4], at 13-183. However, Campbell's discussion of the fourth factor conspicuously omits this phrasing. Apparently abandoning the idea that any factor enjoys primacy, Campbell instructs that 'all [four factors] are to be explored, and the results weighed together, in light of the purposes of copyright.' 114 S. Ct. at 1171. In analyzing the fourth factor, it is important (1) to bear in mind the precise copyrighted works, namely the eight journal articles, and (2) to recognize the distinctive nature and history of "the potential market for or value of" these particular works. Specifically, though there is a traditional market for, and hence a clearly defined value of, journal issues and volumes , in the form of per-issue purchases and journal subscriptions, there is neither a traditional market for, nor a clearly defined value of, individual journal articles. As a result, analysis of the fourth factor cannot proceed as simply as would have been the case if Texaco had copied a work that carries a stated or negotiated selling price in the market. Like most authors, writers of journal articles do not directly seek to capture the potential financial rewards that stem from their copyrights by personally marketing copies of their writings. Rather, like other creators of literary works, the author of a journal article "commonly sells his rights to publishers who offer royalties in exchange for their services in producing and marketing the author's work." Harper & Row, 471 U.S. at 547. In the distinctive realm of academic and scientific articles, however, the only form of royalty paid by a publisher is often just the reward of being published, publication being a key to professional advancement and prestige for the author, see Weissmann, 868 F.2d at 1324 (noting that "in an academic setting, profit is ill-measured in dollars. Instead, what is valuable is recognition because it so often influences professional advancement and academic tenure."). The publishers in turn incur the costs and labor of producing and marketing authors' articles, driven by the prospect of capturing the economic value stemming from the copyrights in the original works, which the authors have transferred to them. Ultimately, the monopoly privileges conferred by copyright protection and the potential financial rewards therefrom are not directly serving to motivate authors to write individual articles; rather, they serve to motivate publishers to produce journals, which provide the conventional and often exclusive means for disseminating these individual articles. It is the prospect of such dissemination that contributes to the motivation of these authors. Significantly, publishers have traditionally produced and marketed authors' individual articles only in a journal format, i.e., in periodical compilations of numerous articles. In other words, publishers have conventionally sought to capture the economic value from the "exclusive rights" to "reproduce" and "distribute copies" of the individual articles, see 17 U.S.C. § 106(1) & (3), solely by compiling many such articles together in a periodical journal and then charging a fee to subscribe. Publishers have not traditionally provided a simple or efficient means to obtain single copies of individual articles; reprints are usually available from publishers only in bulk quantities and with some delay. This marketing pattern has various consequences for our analysis of the fourth factor. First, evidence concerning the effect that photocopying individual journal articles has on the traditional market for journal subscriptions is of somewhat less significance than if a market existed for the sale of individual copies of articles. Second, this distinctive arrangement raises novel questions concerning the significance of the publishers' establishment of an innovative licensing scheme for the photocopying of individual journal articles. 1. Sales of Additional Journal Subscriptions, Back Issues, and Back Volumes. Since we are concerned with the claim of fair use in copying the eight individual articles from Catalysis, the analysis under the fourth factor must focus on the effect of Texaco's photocopying upon the potential market for or value of these individual articles. Yet, in their respective discussions of the fourth statutory factor, the parties initially focus on the impact of Texaco's photocopying of individual journal articles upon the market for Catalysis journals through sales of Catalysis subscriptions, back issues, or back volumes. As a general matter, examining the effect on the marketability of the composite work containing a particular individual copyrighted work serves as a useful means to gauge the impact of a secondary use "upon the potential market for or value of" that individual work, since the effect on the marketability of the composite work will frequently be directly relevant to the effect on the market for or value of that individual work. Quite significantly, though, in the unique world of academic and scientific articles, the effect on the marketability of the composite work in which individual articles appear is not obviously related to the effect on the market for or value of the individual articles. Since (1) articles are submitted unsolicited to journals, (2) publishers do not make any payment to authors for the right to publish their articles or to acquire their copyrights, and (3) there is no evidence in the record suggesting that publishers seek to reprint particular articles in new composite works, we cannot readily conclude that evidence concerning the effect of Texaco's use on the marketability of journals provides an effective means to appraise the effect of Texaco's use on the market for or value of individual journal articles. These considerations persuade us that evidence concerning the effect of Texaco's photocopying of individual articles within Catalysis on the traditional market for Catalysis subscriptions is of somewhat limited significance in determining and evaluating the effect of Texaco's photocopying "upon the potential market for or value of" the individual articles. We do not mean to suggest that we believe the effect on the marketability of journal subscriptions is completely irrelevant to gauging the effect on the market for and value of individual articles. Were the publishers able to demonstrate that Texaco's type of photocopying, if widespread, would impair the marketability of journals, then they might have a strong claim under the fourth factor. Likewise, were Texaco able to demonstrate that its type of photocopying, even if widespread, would have virtually no effect on the marketability of journals, then it might have a strong claim under this fourth factor. On this record, however, the evidence is not resounding for either side. The District Court specifically found that, in the absence of photocopying, (1) "Texaco would not ordinarily fill the need now being supplied by photocopies through the purchase of back issues or back volumes . . . [or] by enormously enlarging the number of its subscriptions," but (2) Texaco still "would increase the number of subscriptions somewhat." 802 F. Supp. at 19. This moderate conclusion concerning the actual effect on the marketability of journals, combined with the uncertain relationship between the market for journals and the market for and value of individual articles, leads us to conclude that the evidence concerning sales of additional journal subscriptions, back issues, and back volumes does not strongly support either side with regard to the fourth factor. Cf. Sony, 464 U.S. at 451-55 (rejecting various predictions of harm to value of copyrighted work based on speculation about possible consequences of secondary use). At best, the loss of a few journal subscriptions tips the fourth factor only slightly toward the publishers because evidence of such loss is weak evidence that the copied articles themselves have lost any value. 2. Licensing Revenues and Fees. The District Court, however, went beyond discussing the sales of additional journal subscriptions in holding that Texaco's photocopying affected the value of the publishers' copyrights. Specifically, the Court pointed out that, if Texaco's unauthorized photocopying was not permitted as fair use, the publishers' revenues would increase significantly since Texaco would (1) obtain articles from document delivery services (which pay royalties to publishers for the right to photocopy articles), (2) negotiate photocopying licenses directly with individual publishers, and/or (3) acquire some form of photocopying license from the Copyright Clearance Center Inc. ("CCC"). See 802 F. Supp. at 19. Texaco claims that the District Court's reasoning is faulty because, in determining that the value of the publishers' copyrights was affected, the Court assumed that the publishers were entitled to demand and receive licensing royalties and fees for photocopying. Yet, continues Texaco, whether the publishers can demand a fee for permission to make photocopies is the very question that the fair use trial is supposed to answer. It is indisputable that, as a general matter, a copyright holder is entitled to demand a royalty for licensing others to use its copyrighted work, see 17 U.S.C. § 106 (copyright owner has exclusive right "to authorize" certain uses), and that the impact on potential licensing revenues is a proper subject for consideration in assessing the fourth factor, see, e.g., Campbell, 114 S. Ct. at 1178; Harper & Row, 471 U.S. at 568-69; Twin Peaks, 996 F.2d at 1377; DC Comics Inc. v. Reel Fantasy, Inc., 696 F.2d 24, 28 (2d Cir. 1982); United Telephone Co. of Missouri v. Johnson Publishing Co., Inc., 855 F.2d 604, 610 (8th Cir. 1988). However, not every effect on potential licensing revenues enters the analysis under the fourth factor. Specifically, courts have recognized limits on the concept of "potential licensing revenues" by considering only traditional, reasonable, or likely to be developed markets when examining and assessing a secondary use's "effect upon the potential market for or value of the copyrighted work." See Campbell, 114 S. Ct. at 1178 ("The market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop."); Harper & Row, 471 U.S. at 568 (fourth factor concerned with "use that supplants any part of the normal market for a copyrighted work") (emphasis added) (quoting S. Rep. No. 473, 94th Cong., 1st Sess. 65 (1975)); see also Mathieson v. Associated Press, 1992 U.S. Dist. LEXIS 9269, 23 U.S.P.Q.2d 1685, 1690-91 (S.D.N.Y. 1992) (refusing to find fourth factor in favor of copyright holder because secondary use did not affect any aspect of the normal market for copyrighted work). For example, the Supreme Court recently explained that because of the "unlikelihood that creators of imaginative works will license critical reviews or lampoons" of their works, "the law recognizes no derivative market for critical works," Campbell, 114 S. Ct. at 1178. Similarly, other courts have found that the fourth factor will favor the secondary user when the only possible adverse effect occasioned by the secondary use would be to a potential market or value that the copyright holder has not typically sought to, or reasonably been able to, obtain or capture. See Twin Peaks, 996 F.2d at 1377 (noting that fourth factor will favor secondary user when use "filled a market niche that the [copyright owner] simply had no interest in occupying"); Pacific and Southern Co. v. Duncan, 744 F.2d 1490, 1496 (11th Cir. 1984) cert. denied, 471 U.S. 1004, 85 L. Ed. 2d 161, 105 S. Ct. 1867 (1985) (noting that the fourth factor may not favor copyright owner when the secondary user "profits from an activity that the owner could not possibly take advantage of"). Thus, Texaco is correct, at least as a general matter, when it contends that it is not always appropriate for a court to be swayed on the fourth factor by the effects on potential licensing revenues. Only an impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets should be legally cognizable when evaluating a secondary use's "effect upon the potential market for or value of the copyrighted work." Though the publishers still have not established a conventional market for the direct sale and distribution of individual articles, they have created, primarily through the CCC, a workable market for institutional users to obtain licenses for the right to produce their own copies of individual articles via photocopying. The District Court found that many major corporations now subscribe to the CCC systems for photocopying licenses. 802 F. Supp. at 25. Indeed, it appears from the pleadings, especially Texaco's counterclaim, that Texaco itself has been paying royalties to the CCC. See Complaint P 38; First Counterclaim P 71. Since the Copyright Act explicitly provides that copyright holders have the "exclusive rights" to "reproduce" and "distribute copies" of their works, see 17 U.S.C. § 106(1) & (3), and since there currently exists a viable market for licensing these rights for individual journal articles, it is appropriate that potential licensing revenues for photocopying be considered in a fair use analysis. Despite Texaco's claims to the contrary, it is not unsound to conclude that the right to seek payment for a particular use tends to become legally cognizable under the fourth fair use factor when the means for paying for such a use is made easier. This notion is not inherently troubling: it is sensible that a particular unauthorized use should be considered "more fair" when there is no ready market or means to pay for the use, while such an unauthorized use should be considered "less fair" when there is a ready market or means to pay for the use. The vice of circular reasoning arises only if the availability of payment is conclusive against fair use. Whatever the situation may have been previously, before the development of a market for institutional users to obtain licenses to photocopy articles, see Williams & Wilkins, 487 F.2d at 1357-59, it is now appropriate to consider the loss of licensing revenues in evaluating "the effect of the use upon the potential market for or value of" journal articles. It is especially appropriate to do so with respect to copying of articles from Catalysis, a publication as to which a photocopying license is now available. We do not decide how the fair use balance would be resolved if a photocopying license for Catalysis articles were not currently available. In two ways, Congress has impliedly suggested that the law should recognize licensing fees for photocopying as part of the "potential market for or value of" journal articles. First, section 108 of the Copyright Act narrowly circumscribes the conditions under which libraries are permitted to make copies of copyrighted works. See 17 U.S.C. § 108. Though this section states that it does not in any way affect the right of fair use, see id. § 108(f)(4), the very fact that Congress restricted the rights of libraries to make copies implicitly suggests that Congress views journal publishers as possessing the right to restrict photocopying, or at least the right to demand a licensing royalty from nonpublic institutions that engage in photocopying. Second, Congress apparently prompted the development of CCC by suggesting that an efficient mechanism be established to license photocopying, see S. Rep. No. 983, 93d Cong., 2d Sess. 122 (1974); S. Rep. No. 473, 94th Cong., 1st Sess. 70-71 (1975); H.R. Rep. No. 83, 90th Cong., 1st Sess. 33 (1968). It is difficult to understand why Congress would recommend establishing such a mechanism if it did not believe that fees for photocopying should be legally recognized as part of the potential market for journal articles. Primarily because of lost licensing revenue, and to a minor extent because of lost subscription revenue, we agree with the District Court that "the publishers have demonstrated a substantial harm to the value of their copyrights through [Texaco's] copying," 802 F. Supp. at 21, and thus conclude that the fourth statutory factor favors the publishers.

Note that we do not conclude that the Code is a compilation covered by 17 U.S.C. § 103. It could be a compilation only if its elements existed independently and the ADA merely put them in order. A taxonomy is a way of describing items in a body of knowledge or practice; it is not a collection or compilation of bits and pieces of "reality". The 1991 and 1994 versions of the Code may be recompilations of earlier editions, but the original Code is covered by § 102(a) as an "original work of authorship", and its amendments by § 106(2) as derivative works.

Any original literary work may be copyrighted. The necessary degree of "originality" is low, and the work need not be aesthetically pleasing to be "literary." Feist Publications, Inc v. Rural Telephone Service Co., 499 U.S. 340, 345-46, 113 L. Ed. 2d 358, 111 S. Ct. 1282 (1991). Term papers by college sophomores are as much within the domain of copyright as Saul Bellow's latest novel. See Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 47 L. Ed. 460, 23 S. Ct. 298 (1903). Scholarship that explicates important facts about the universe likewise is well within this domain. Einstein's articles laying out the special and general theories of relativity were original works even though many of the core equations, such as the famous E = mc2, express "facts" and therefore are not copyrightable. Einstein could have explained relativity in any of a hundred different ways; another physicist could expound the same principles differently. So too with a taxonomy--of butterflies, legal citations, or dental procedures. Facts do not supply their own principles of organization. Classification is a creative endeavor. Butterflies may be grouped by their color, or the shape of their wings, or their feeding or breeding habits, or their habitats, or the attributes of their caterpillars, or the sequence of their DNA; each scheme of classification could be expressed in multiple ways. Dental procedures could be classified by complexity, or by the tools necessary to perform them, or by the parts of the mouth involved, or by the anesthesia employed, or in any of a dozen different ways. The Code's descriptions don't "merge with the facts" any more than a scientific description of butterfly attributes is part of a butterfly. Cf. Nash v. CBS, Inc., 899 F.2d 1537 (7th Cir. 1990) (discussing the fact expression dichotomy). There can be multiple, and equally original, biographies of the same person's life, and multiple original taxonomies of a field of knowledge. Creativity marks the expression even after the fundamental scheme has been devised. This is clear enough for the long description of each procedure in the ADA's Code. The long description is part of the copyrighted work, and original long descriptions make the work as a whole copyrightable. But we think that even the short description and the number are original works of authorship.

The first factor in a fair use inquiry is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." § 107(1). While this inquiry does not specify which purpose might render a given use "fair," the preamble to § 107 provides an illustrative, though not limitative, listing which includes "criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research." § 107. Under this factor, the inquiry is whether The Cat NOT in the Hat! merely supersedes the Dr. Seuss creations, or whether and to what extent the new work is "transformative," i.e., altering The Cat in the Hat with new expression, meaning or message. Parody is regarded as a form of social and literary criticism, having a socially significant value as free speech under the First Amendment. This court has adopted the "conjure up" test where the parodist is permitted a fair use of a copyrighted work if it takes no more than is necessary to "recall" or "conjure up" the object of his parody. See MCA, Inc. v. Wilson, 677 F.2d 180, 184 (2d Cir. 1981); Warner Bros., Inc. v. American Broadcasting Cos., 523 F. Supp. 611, 617 (S.D.N.Y.), aff'd, 654 F.2d 204 (2d Cir. 1981); Walt Disney Prods. v. Air Pirates, 581 F.2d 751, 757 (9th Cir. 1978), cert. denied, 439 U.S. 1132, 59 L. Ed. 2d 94, 99 S. Ct. 1054 (1979). Accordingly, the critical issue under this factor is whether The Cat NOT in the Hat! is a parody. See Acuff-Rose, 114 S. Ct. at 1173 ("The threshold question when fair use is raised in defense of parody is whether a parodic character may reasonably be perceived."). We first examine the definition of parody. The parties disagree over the appropriate interpretation of Acuff-Rose's holding with respect to the definition of parody under the fair use exception. The Supreme Court of the United States in the Acuff-Rose case held that a rap group's version of Ray Orbison's song "Oh, Pretty Woman" was a candidate for a parody fair use defense. Justice Souter, the opinion's author, defined parody: For the purposes of copyright law, the nub of the definitions, and the heart of any parodist's claim to quote from existing material, is the use of some elements of a prior author's composition to create a new one that, at least in part, comments on that author's works. . . . If, on the contrary, the commentary has no critical bearing on the substance or style of the original composition, which the alleged infringer merely uses to get attention or to avoid the drudgery in working up something fresh, the claim to fairness in borrowing from another's work diminishes accordingly (if it does not vanish), and other factors, like the extent of its commerciality, loom larger. Id. at 1172 (citations omitted). The Court pointed out the difference between parody (in which the copyrighted work is the target) and satire (in which the copyrighted work is merely a vehicle to poke fun at another target): "Parody needs to mimic an original to make its point, and so has some claim to use the creation of its victim's (or collective victims') imagination, whereas satire can stand on its own two feet and so requires justification for the very act of borrowing." Id. As Justice Kennedy put it in his concurrence: "The parody must target the original, and not just its general style, the genre of art to which it belongs, or society as a whole (although if it targets the original, it may target those features as well)." Id. at 1180. The Second Circuit in Rogers v. Koons, 960 F.2d 301, 310 (2d Cir. 1992), also emphasized that unless the plaintiff's copyrighted work is at least in part the target of the defendant's satire, then the defendant's work is not a "parody" in the legal sense: It is the rule in this Circuit that though the satire need not be only of the copied work and may . . . also be a parody of modern society, the copied work must be, at least in part, an object of the parody, otherwise there would be no need to conjure up the original work. . . . By requiring that the copied work be an object of the parody, we merely insist that the audience be aware that underlying the parody there is an original and separate expression, attributable to a different artist. Similarly, the American Heritage Dictionary defines "parody" as a "literary or artistic work that broadly mimics an author's characteristic style and holds it up to ridicule." We now turn our attention to The Cat NOT in the Hat! itself. The first two pages present a view of Los Angeles, with particular emphasis on the connection with Brentwood, given the depiction of the news camera lights. The story begins as follows: A happy town Inside L.A. Where rich folks play The day away. But under the moon The 12th of June. Two victims flail Assault! Assail! Somebody will go to jail! Who will it be? Oh my! Oh me! The third page reads: "One Knife? / Two Knife? / Red Knife / Dead Wife." This stanza no doubt mimics the first poem in Dr. Seuss' One Fish Two Fish Red Fish Blue Fish: "One fish / two fish / red fish / blue fish. Black fish / blue fish / old fish / new fish." For the next eighteen pages, Katz writes about Simpson's trip to Chicago, the noise outside Kato Kaelin's room, the bloody glove found by Mark Fuhrman, the Bronco chase, the booking, the hiring of lawyers, the assignment of Judge Ito, the talk show interest, the comment on DNA, and the selection of a jury. On the hiring of lawyers for Simpson, Katz writes: A plea went out to Rob Shapiro Can you save the fallen hero? And Marcia Clark, hooray, hooray Was called in with a justice play. A man this famous Never hires Lawyers like Jacoby-Meyers. When you're accused of a killing scheme You need to build a real Dream Team. Cochran! Cochran! Doodle-doo Johnnie, won't you join the crew? Cochran! Cochran! Deedle-dee The Dream Team needs a victory. These stanzas and the illustrations simply retell the Simpson tale. Although The Cat NOT in the Hat! does broadly mimic Dr. Seuss' characteristic style, it does not hold his style up to ridicule. The stanzas have "no critical bearing on the substance or style of" The Cat in the Hat. Katz and Wrinn merely use the Cat's stove-pipe hat, the narrator ("Dr. Juice), and the title (The Cat NOT in the Hat!) "to get attention" or maybe even "to avoid the drudgery in working up something fresh." Acuff-Rose, 114 S. Ct. at 1172. While Simpson is depicted 13 times in the Cat's distinctively scrunched and somewhat shabby red and white stove-pipe hat, the substance and content of The Cat in the Hat is not conjured up by the focus on the Brown-Goldman murders or the O.J. Simpson trial. Because there is no effort to create a transformative work with "new expression, meaning, or message," the infringing work's commercial use further cuts against the fair use defense. Id. at 1171.

he second statutory factor, "the nature of the copyrighted work," § 107(2), recognizes that creative works are "closer to the core of intended copyright protection" than informational and functional works, "with the consequence that fair use is more difficult to establish when the former works are copied." Acuff-Rose, 114 S. Ct. at 1175. While this factor typically has not been terribly significant in the overall fair use balancing, the creativity, imagination and originality embodied in The Cat in the Hat and its central character tilts the scale against fair use.

The third factor asks whether "the amount and substantiality of the portion used in relation to the copyrighted work as a whole," § 107(3), are reasonable in relation to the purpose of the copying. This factor really raises the question of substantial similarity discussed in the preceding section, rather than whether the use is "fair." The district court concluded that "The Cat in the Hat" is the central character, appearing in nearly every image of The Cat in the Hat. Penguin and Dove appropriated the Cat's image, copying the Cat's Hat and using the image on the front and back covers and in the text (13 times). We have no doubt that the Cat's image is the highly expressive core of Dr. Seuss' work. Under this factor, we also turn our attention "to the persuasiveness of a parodist's justification for the particular copying done, and the enquiry will harken back to the first of the statutory factors, for, as in prior cases, we recognize that the extent of permissible copying varies with the purpose and character of the use." Acuff-Rose, 114 S. Ct. at 1175. Katz and Wrinn insist that they selected The Cat in the Hat as the vehicle for their parody because of the similarities between the two stories: Nicole Brown and Ronald Goldman were surprised by a "Cat" (O.J. Simpson) who committed acts contrary to moral and legal authority. The prosecution of Simpson created a horrible mess, in which the defense team seemed to impose "tricks" on an unwilling public, resulting in a verdict that a substantial segment of the public regarded as astonishing. Just as The Cat in the Hat ends with the moral dilemma of whether the children should tell their mother about their visitor that afternoon, Katz and Wrinn maintain that The Cat NOT in the Hat! ends with a similar moral dilemma: JUICE +ST JUSTICE Hmm. . . take the word JUICE. Then add ST. Between the U and I, you see. And then you have JUSTICE. Or maybe you don't. Maybe we will. And maybe we won't. 'Cause if the Cat didn't do it? Then who? Then who? Was it him? Was it her? Was it me? Was it you? Oh me! Oh my! Oh my! Oh me! The murderer is running free. In their Opening Brief, Penguin and Dove characterize The Cat NOT in the Hat! ("Parody") as follows: The Parody is a commentary about the events surrounding the Brown/Goldman murders and the O.J. Simpson trial, in the form of a Dr. Seuss parody that transposes the childish style and moral content of the classic works of Dr. Seuss to the world of adult concerns. The Parody's author felt that, by evoking the world of The Cat in the Hat, he could: (1) comment on the mix of frivolousness and moral gravity that characterized the culture's reaction to the events surrounding the Brown/Goldman murders, (2) parody the mix of whimsy and moral dilemma created by Seuss works such as The Cat in the Hat in a way that implied that the work was too limited to conceive the possibility of a real trickster "cat" who creates mayhem along with his friends Thing 1 and Thing 2, and then magically cleans it up at the end, leaving a moral dilemma in his wake. We completely agree with the district court that Penguin and Dove's fair use defense is "pure shtick" and that their post-hoc characterization of the work is "completely unconvincing."

The fourth fair use factor is "the effect of the use upon the potential market for or value of the copyrighted work." § 107(4). Under this factor, we consider both the extent of market harm caused by the publication and distribution of The Cat NOT in the Hat! and whether unrestricted and widespread dissemination would hurt the potential market for the original and derivatives of The Cat in the Hat. The Second Circuit has characterized this factor as calling for the striking of a balance "between the benefit the public will derive if the use is permitted and the personal gain the copyright owner will receive if the use is denied. The less adverse effect that an alleged infringing use has on the copyright owner's expectation of gain, the less public benefit need be shown to justify the use." MCA, Inc. v. Wilson, 677 F.2d 180, 183 (2d Cir. 1981) (citations omitted). The good will and reputation associated with Dr. Seuss' work is substantial. Because, on the facts presented, Penguin and Dove's use of The Cat in the Hat original was nontransformative, and admittedly commercial, we conclude that market substitution is at least more certain, and market harm may be more readily inferred. Since fair use is an affirmative defense, Penguin and Dove must bring forward favorable evidence about relevant markets. Given their failure to submit evidence on this point, instead confining "themselves to uncontroverted submissions that there was no likely effect on the market for the original," we conclude that "it is impossible to deal with the fourth factor except by recognizing that a silent record on an important factor bearing on fair use disentitles the proponent of the defense," Penguin and Dove, to relief from the preliminary injunction. Acuff-Rose, 114 S. Ct. at 1177-1179. In light of the fair use analysis, we conclude that the district court's finding that Seuss showed a likelihood of success on the merits of the copyright claim was not clearly erroneous. 10

D. Legal Analysis of Substantial Similarity In the Krofft case, we established a two-part test for substantial similarity. 562 F.2d at 1164. Under the original formulation of Krofft's "extrinsic" component, a plaintiff needed to prove only the "similarity of ideas" in the two programs. Id. Today, however, the extrinsic test looks at more than just the similarity of ideas. As the district court understood, the extrinsic analysis is " 'an objective test which rests upon specific criteria that can be listed and analyzed.' " 12 U.S.P.Q.2D (BNA) at 1994 (quoting Data East USA, Inc. v. Epyx, Inc., 862 F.2d 204, 208 (9th Cir. 1988)). The extrinsic test thus has become an "objective . . . analysis of expression." Shaw, 919 F.2d at 1357 (original emphasis); see also 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.03[E]3, at 13-78.19 to -78.20 (1991) (hereinafter, "Nimmer") (discussing reformulation of Krofft analysis achieved in Shaw). To establish infringement of a copyright, the two works in question must also meet the second component of the test set forth in Krofft, the "intrinsic test." 562 F.2d at 1164. The intrinsic test, according to Krofft, should measure "substantial similarity in expressions . . . depending on the response of the ordinary reasonable person….It does not depend on the type of external criteria and analysis which marks the extrinsic test." Id. (citations omitted). In applying the intrinsic test, therefore, "analytic dissection and expert testimony are not appropriate." Id. Brown Bag contends that the district court misapplied the intrinsic test because the court improperly performed "analytic dissection" in comparing the expressions of ideas embodied in the two programs. Indeed, the district court explicitly performed "analytic dissection," comparing specific screens, menus, and keystrokes, in part to detect substantial similarity. 12 U.S.P.Q.2D (BNA) at 1995-96. The court performed the dissection at Brown Bag's behest and in response to a list of allegedly substantially similar features Brown Bag proffered in opposition to summary judgment. Brown Bag's argument ignores the evolution of the "extrinsic" component of the Krofft analysis. Properly understood, the district court's analysis was not a misapplication of the "intrinsic test," but a proper application of the revised "extrinsic test." Under the reformulated extrinsic test, we mean to perpetuate "analytic dissection" as a tool for comparing not only ideas but also expression. Thus, it was not error for the district court to use "analytic dissection" in comparing the expressions of the two programs. Specifically, the district court properly compared the expressions embodied in the two screens presenting the programs' respective opening menus. See 12 U.S.P.Q.2D (BNA) at 1996. The district court's analytic dissection of the five groups of features involving unprotected or unprotectable work was also proper, but for an entirely different reason. This analysis had nothing to do with assessing substantial similarity, and therefore was not subject to the old limitations Krofft imposed on the use of analytic dissection. Instead, as to these groups, the district court applied analytic dissection for the purpose of determining "whether similarities [between the programs] result from unprotectable [or unprotected] expression." Data East, 862 F.2d at 208. In this application, analytic dissection is used not for the purposes of comparing similarities and identifying infringement, but for the purpose of defining the scope of plaintiff's copyright. This point bears emphasis. Analytic dissection is relevant not only to the copying element of a copyright infringement claim, but also to the claim's ownership element. One aspect of the ownership element is the copyrightability of the subject matter and, more particularly, the scope of whatever copyright lies therein. See Nimmer § 13.01[A]. To the extent a plaintiff's work is unprotected or unprotectable under copyright, the scope of the copyright must be limited. See, e.g., Data East, 862 F.2d at 209 (holding that, because "the similarities result from unprotectable expression," "the lower court erred by not limiting the scope of [plaintiff's] copyright protection"). Thus, where two works are found to be similar without regard to the scope of the copyright in the plaintiff's work, Data East teaches that the source of the similarity must be identified and a determination made as to whether this source is covered by plaintiff's copyright. It is this process in which the district court engaged with regard to the five groups of features found to be unprotected or unprotectable by Brown Bag's copyright in PC-Outline. Again, this use of analytic dissection was proper. Brown Bag further argues that the district court became so engrossed in its analytic dissection that it neglected to evaluate the "overall look and feel" of the two programs for substantial similarity. Stated differently, Brown Bag contends that the district court failed to consider the "intrinsic" component of the Krofft test, which involves a subjective analysis of similarities in expression. See Shaw, 919 F.2d at 1357. Even assuming this is true, we find no reversible error because the record fails to include any evidence indicating that Brown Bag requested the district court to make this analysis. Although the district court "entertained an in-court demonstration of the two programs and . . . carefully reviewed submitted copies of relevant outlining programs," 12 U.S.P.Q.2D (BNA) at 1992, no such evidence was ever included as part of Brown Bag's opposition to Symantec's summary judgment motion. Therefore, there is no reason we should expect the district court to have made the analysis Brown Bag now requests. Finally, Brown Bag argues that, because PC-Outline and Grandview shared a common idea - specifically the idea of an outlining program - summary judgment is precluded by our decision in Shaw. We held in Shaw that "satisfaction of the extrinsic test creates a triable issue of fact in a copyright action involving a literary work." 919 F.2d at 1359. In response, Symantec argues we should distinguish Shaw from the instant case on the basis that the Shaw rule applies by its own terms only to literary works. We decline to distinguish Shaw for that reason, however, because at least some computer programs bear significant similarities to literary works. On the scant record available to us in this case, we cannot conclude as a matter of law that the programs at issue here, or computer programs in general, are so nonliterary or so limited in their variety of expression as to avoid application of the rule announced in Shaw. We nevertheless reject Brown Bag's argument on the basis that it misapprehends the nature of the "extrinsic" test following Shaw. If the rule in Shaw was applied without recognizing that the extrinsic test now includes some analysis of expression, Shaw would ring "the death-knell in the Ninth Circuit for defense summary judgments based on purported absence of substantial similarity." Nimmer § 13.03[E]3, at 13-78.20. But, as we have discussed, after Shaw the extrinsic test examines not only similarity of ideas, but also objective similarity of expressions. In particular, the extrinsic test for literary works requires analytic dissection of several "objective components of expression," Shaw, 919 F.2d at 1361, within a literary work such as plot, theme, characters, and dialogue. Id. at 1356-57. Similarity of these objective components in two literary works logically gives rise to a triable issue of similarity. Id. at 1357. Likewise, computer programs are subject to a Shawtype analytic dissection of various standard components, e.g., screens, menus, and keystrokes. Because the district court found that Brown Bag made no showing of similarity along these lines with regard to copyrighted components of PC-Outline, summary judgment was not precluded by the rule announced in Shaw. See also Nimmer § 13.03[E]3, at 13-78.20 ("given Shaw's reformulation of the extrinsic test, the decision is better viewed as standing for the unremarkable proposition that when objectively speaking a triable case of expressive similarity exists, the case must be sent to the trier of fact; but when no such objective argument is possible, summary judgment will lie.") (footnotes omitted).

Surely a text testing one's knowledge of Joyce's Ulysses, or Shakespeare's Hamlet, would qualify as "criticism, comment, scholarship, or research," or such. The same must be said, then, of a text testing one's knowledge of Castlerock's Seinfeld. Id. ("A comment is as eligible for fair use protection when it concerns `Masterpiece Theater' and appears in the New York Review of Books as when it concerns `As the World Turns' and appears in Soap Opera Digest."). Thus, the Court is satisfied that SAT "serves one or more of the non-exclusive purposes that section 107 identifies as examples of purposes for which a protected fair use may be made." Id. As the Supreme Court recently explained, the "central purpose" of the Court's inquiry into the character and purpose of an allegedly infringing work must be to determine whether that work is "transformative." Campbell, 510 U.S. at 578-80, 114 S. Ct. at 1171; see also Twin Peaks, 996 F.2d at 1375. Though plaintiff insists that SAT is not at all creative, the Court concludes otherwise. Given the absence of any case law addressing the copyright status of a work completely devoted to posing trivia questions about the fictional elements of another work, it is clear that SAT is itself an "original creation."

Commerciality - Defendants' initial advantage must be tempered, if only slightly, by the fact that their creation and publication of SAT was a commercial endeavor. The Copyright Act "plainly assigns a higher value to a use that serves `nonprofit educational purposes' than to one of a `commercial nature.'" Twin Peaks, 996 F.2d at 1374. This factor must not be unduly emphasized, however. As the Supreme Court reasoned in Campbell, "[i]f ... commerciality carried presumptive force against a finding of fairness, the presumption would swallow nearly all of the illustrative uses listed in the preamble paragraph of § 107 ... since these activities `are generally conducted for profit in this country.'" Campbell, 510 U.S. at 584, 114 S. Ct. at 1174 (citations omitted); see also Robinson v. Random House, Inc., 877 F. Supp. 830, 840 (S.D.N.Y. 1995) ("because nearly all authors hope to make a profit with their work, courts should be wary of placing too much emphasis on the commercial nature in a fair use determination."). Thus, the commercial nature of SAT reduces but does not nearly eliminate the significance properly ascribed to the transformative quality of defendants' work.

Seinfeld is a highly successful fictional and creative work. As defendants concede in their opposition papers, plaintiff thereby has a decisive advantage with respect to the second factor of the fair use analysis. (Defendant's Memorandum of Law in Opposition to Plaintiffs Motion for Partial Summary Judgment at 14-15).

Though the Court recognizes that a quantitative approach to addressing the substantiality question has a place in the analysis, it is clear that even small amounts of material extracted from an original work can suffice to counter a claim of fair use. Twin Peaks, 996 F.2d at 1372 (finding infringement where defendant excerpted a total of 89 lines of dialogue from several episodes of a protected television program); Harper & Row, 471 U.S. at 564-65, 105 S. Ct. at 2232-33 (finding infringement where defendant excerpted approximately 400 words of a full length book); In other words, the substantiality factor "has both a quantitative and a qualitative element to it." Wright v. Warner Books, 953 F.2d 731, 738 (2d Cir.1991). If a challenged work appropriates what amounts to "the heart" of an original work, even if only in a few words, then such an appropriation is substantial for purposes of the fair use inquiry. See Harper & Row, 471 U.S. at 565, 105 S. Ct. at 2233. Beginning with the significance that the appropriated material has in relation to the Seinfeld show, a brief review of SAT confirms that the book invokes all of the show's main characters, and many of the show's most humorous plot elements. Perhaps more to the point, SAT seizes upon the notion which lies at the very heart of Seinfeld that there is humor in the mundane, seemingly trivial, aspects of every day life. Indeed, by inviting its readers to recall literally 643 bits of information from various Seinfeld episodes, SAT "follow[s] the basic premise of the Seinfeld show by focusing on minutiae in the day-to-day lives of the show's characters." (Shostak Dep. Ex. 2 at 000604). As defendants boasted before the onset of this litigation, SAT succeeds at "capturing [Seinfeld's] flavor in quiz book fashion." (Golub Dep. Ex. 5 at 00606). Simply put, without Seinfeld, there can be no SAT. See Salinger v. Random House, 811 F.2d 90, 99 (2d Cir.) (finding that quantitatively modest excerpting from plaintiffs personal letters was substantial where those excerpts, "[t]o a large extent, [made defendant's] book worth reading."), cert. denied, 484 U.S. 890, 108 S. Ct. 213, 98 L. Ed. 2d 177 (1987); see also Addison-Wesley Publishing Co. v. Brown, 223 F. Supp. 219, 223-24 (E.D.N.Y.1963) (stressing that defendant's book, a manual consisting of the answers to a set of physics problems included in plaintiffs college course book, had "no independent viability."); Midway Mfg. Co. v. Artic Int'l, Inc., 1981 WL 1390, * 9 (N.D.Ill. 1981) ("[If] defendant's device would only have value because of plaintiffs particular copyrighted audio visual image, then plainly defendant's device would only have value because of plaintiffs particular copyrighted audio visual work. Defendant, thus, by selling its device reaps the benefits of plaintiffs artistic endeavor."). In sum, defendants have identified and appropriated the most important elements of Seinfeld, and have made them the most important elements of SAT. Previously, the Court emphasized that its finding that SAT is transformative of Seinfeld cannot be dispositive for defendants, because such a holding would discredit the proposition that the unauthorized production of a derivative work can be infringing. On similar logic, the Court's finding that SAT incorporates a substantial amount from Seinfeld cannot be dispositive in plaintiffs favor. Because a finding of substantial similarity is a prerequisite to a prima facie claim of infringement, such a finding cannot negate the possibility of fair use. Otherwise, the fair use provision of the Copyright Act would amount to little more than a false promise of a viable defense; there would be no real chance that a prima facie case of infringement could ever be negated by a showing of fair use. The first three factors of the fair use analysis, then, suggest a somewhat unsatisfying result; plaintiff has an advantage, but one that is hardly compelling or dispositive.

This does not end the analysis, however; "the proper inquiry concerns the `potential market' for the copyrighted work." Salinger, 811 F.2d at 99. More broadly, the inquiry must extend even to the potential market for as yet nonexistent derivative works. n other words, the Court must consider not only whether SAT detracts from interest in Seinfeld, or even whether SAT occupies markets that plaintiff intends to enter; the analysis is whether SAT occupies derivative markets that plaintiff may potentially enter. Id.; see also Rogers, 751 F. Supp. at 480 ("I do not think the case turns upon Rogers' past conduct or present intention as much as it does upon the existence of a recognized market for new versions or new uses of the photograph, which unauthorized use clearly undermines."). The term "potential markets" does not properly encompass all conceivable markets for an original and its derivatives. "The market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop." Campbell, 510 U.S. at 592, 114 S. Ct. at 1178. The examples of parody and criticism should serve to clarify and illustrate this proposition. By the very nature of such endeavors, persons other than the copyright holder are undoubtedly better equipped, and more likely, to fill these particular market and intellectual niches. See Campbell, 510 U.S. at 592, 114 S. Ct. at 1178 ("there is no protectable derivative market for criticism."); New Era Publications, Int'l v. Carol Publishing Group, 904 F.2d 152, 160 (2d Cir.) ("a critical biography serves a different function than does an authorized, favorable biography, and thus injury to the potential market for the favorable biography by the publication of the unfavorable biography does not affect application of factor four."), cert. denied, 498 U.S. 921, 111 S. Ct. 297, 112 L. Ed. 2d 251 (1990); Leibovitz, 948 F. Supp. at 1216 ("although derivative markets are an appropriate consideration in a fair use analysis, there is no protectable derivative market for criticism."). Here, the Court sees no reason that the market for derivative game versions of Seinfeld is a market that should be reserved for persons other than plaintiff. A Seinfeld trivia game is not critical of the *272 program, nor does it parody the program; if anything, SAT pays homage to Seinfeld. The market for such works is one that should properly be left to plaintiffs exclusive control. The Court's resolution of the "potential markets" inquiry is not effected by the prospect that plaintiff will choose to leave this particular derivative market unsatisfied. See Salinger, 811 F.2d at 99 ("the need to assess the effect on the market for Salinger's letters is not lessened by the fact that their author has disavowed any intention to publish them during his lifetime."). The Court is persuaded that there is a meaningful difference, for purposes of the Copyright Act, between a copyright holder's failure to occupy a particular market as a matter of choice, and a failure to occupy such a market as a matter of neglect. Id. In a manner of speaking, plaintiff has exercised its control over derivative markets for Seinfeld products, if only by its decision to refrain from inundating those markets. Indeed, artists express themselves not merely by deciding what to create from their original work, but by deciding what not to create as well. Cf. Harper & Row, 471 U.S. at 559, 105 S. Ct. at 2230 ("freedom of thought and expression `includes both the right to speak freely and the right to refrain from speaking at all.'") (citations omitted). It would therefor not serve the ends of the Copyright Act i.e., to advance the arts if artists were denied their monopoly over derivative versions of their creative works merely because they made the artistic decision not to saturate those markets with variations of their original. Where nothing in the nature of criticism or parody is at issue, this creative choice must be respected.

The Copyright Act defines a derivative work as a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a "derivative work." Id. § 101. The statutory language is hopelessly overbroad, however, for " [e]very book in literature, science and art, borrows and must necessarily borrow, and use much which was well known and used before." Emerson v. Davies, 8 F. Cas. 615, 619 (C.C.D. Mass. 1845) (No. 4436), quoted in 1 Nimmer on Copyright, § 3.01, at 3-2 (1997). To narrow the statute to a manageable level, we have developed certain criteria a work must satisfy in order to qualify as a derivative work. One of these is that a derivative work must exist in a "concrete or permanent form," Galoob, 964 F.2d at 967 (internal quotation marks omitted), and must substantially incorporate protected material from the preexisting work, see Litchfield v. Spielberg, 736 F.2d 1352, 1357 (9th Cir. 1984). Micro Star argues that N/I is not a derivative work because the audiovisual displays generated when D/N-3D is run with N/I's MAP files are not incorporated in any concrete or permanent form, and the MAP files do not copy any of D/N-3D's protected expression. It is mistaken on both counts. The requirement that a derivative work must assume a concrete or permanent form was recognized without much discussion in Galoob. There, we noted that all the Copyright Act's examples of derivative works took some definite, physical form and concluded that this was a requirement of the Act. See Galoob, 964 F.2d at 967-68; see also Edward G. Black & Michael H. Page, Add-On Infringements, 15 Hastings Comm/Ent. L.J. 615, 625 (1993) (noting that in Galoob the Ninth Circuit "re-examined the statutory definition of derivative works offered in section 101 and found an independent fixation requirement of sorts built into the statutory definition of derivative works"). Obviously, N/I's MAP files themselves exist in a concrete or permanent form; they are burned onto a CD-ROM. See ProCD, Inc. v. Zeidenberg, 86 F.3d 1447, 1453 (7th Cir. 1996) (computer files on a CD are fixed in a tangible medium of expression). But what about the audiovisual displays generated when D/N-3D runs the N/I MAP files--i.e., the actual game level as displayed on the screen? Micro Star argues that, because the audiovisual displays in Galoob didn't meet the "concrete or permanent form" requirement, neither do N/I's. In Galoob, we considered audiovisual displays created using a device called the Game Genie, which was sold for use with the Nintendo Entertainment System. The Game Genie allowed players to alter individual features of a game, such as a character's strength or speed, by selectively "blocking the value for a single data byte sent by the game cartridge to the [Nintendo console] and replacing it with a new value." Galoob, 964 F.2d at 967. Players chose which data value to replace by entering a code; over a billion different codes were possible. The Game Genie was dumb; it functioned only as a window into the computer program, allowing players to temporarily modify individual aspects of the game. See Lewis Galoob Toys, Inc. v. Nintendo of Am., Inc., 780 F. Supp. 1283, 1289 (N.D. Cal. 1991). Nintendo sued, claiming that when the Game Genie modified the game system's audiovisual display, it created an infringing derivative work. We rejected this claim because " [a] derivative work must incorporate a protected work in some concrete or permanent form." Galoob, 964 F.2d at 967 (internal quotation marks omitted). The audiovisual displays generated by combining the Nintendo System with the Game Genie were not incorporated in any permanent form; when the game was over, they were gone. Of course, they could be reconstructed, but only if the next player chose to reenter the same codes.4 Micro Star argues that the MAP files on N/I are a more advanced version of the Game Genie, replacing old values (the MAP files in the original game) with new values (N/I's MAP files). But, whereas the audiovisual displays created by Game Genie were never recorded in any permanent form, the audiovisual displays generated by D/N-3D from the N/I MAP files are in the MAP files themselves. In Galoob, the audiovisual display was defined by the original game cartridge, not by the Game Genie; no one could possibly say that the data values inserted by the Game Genie described the audiovisual display. In the present case the audiovisual display that appears on the computer monitor when a N/I level is played is described--in exact detail--by a N/I MAP file. This raises the interesting question whether an exact, down to the last detail, description of an audiovisual display (and--by definition--we know that MAP files do describe audiovisual displays down to the last detail) counts as a permanent or concrete form for purposes of Galoob. We see no reason it shouldn't. What, after all, does sheet music do but describe in precise detail the way a copyrighted melody sounds? See 1 William F. Patry, Copyright Law and Practice 168 (1994) (" [A] musical composition may be embodied in sheet music...."). To be copyrighted, pantomimes and dances may be "described in sufficient detail to enable the work to be performed from that description." Id. at 243 (citing Compendium II of Copyright Office Practices § 463); see also Horgan v. Macmillan, Inc., 789 F.2d 157, 160 (2d Cir. 1986). Similarly, the N/I MAP files describe the audiovisual display that is to be generated when the player chooses to play D/N-3D using the N/I levels. Because the audiovisual displays assume a concrete or permanent form in the MAP files, Galoob stands as no bar to finding that they are derivative works. In addition, " [a] work will be considered a derivative work only if it would be considered an infringing work if the material which it has derived from a preexisting work had been taken without the consent of a copyright proprietor of such preexisting work." Mirage Editions v. Albuquerque A.R.T. Co., 856 F.2d 1341, 1343 (quoting 1 Nimmer on Copyright § 3.01 (1986)) (internal quotation marks omitted). "To prove infringement, [FormGen] must show that [D/N-3D's and N/I's audiovisual displays] are substantially similar in both ideas and expression." Litchfield v. Spielberg, 736 F.2d 1352, 1356 (9th Cir. 1984) (emphasis omitted). Similarity of ideas may be shown by comparing the objective details of the works: plot, theme, dialogue, mood, setting, characters, etc. See id. Similarity of expression focuses on the response of the ordinary reasonable person, and considers the total concept and feel of the works. See id. at 1356-57. FormGen will doubtless succeed in making these showings since the audiovisual displays generated when the player chooses the N/I levels come entirely out of D/N-3D's source art library. Cf. Atari, Inc. v. North Am. Philips Consumer Elec. Corp., 672 F.2d 607, 620 (7th Cir. 1982) (finding two video games substantially similar because they shared the same "total concept and feel"). Micro Star further argues that the MAP files are not derivative works because they do not, in fact, incorporate any of D/N-3D's protected expression. In particular, Micro Star makes much of the fact that the N/I MAP files reference the source art library, but do not actually contain any art files themselves. Therefore, it claims, nothing of D/N-3D's is reproduced in the MAP files. In making this argument, Micro Star misconstrues the protected work. The work that Micro Star infringes is the D/N-3D story itself--a beefy commando type named Duke who wanders around post-Apocalypse Los Angeles, shooting Pig Cops with a gun, lobbing hand grenades, searching for medkits and steroids, using a jetpack to leap over obstacles, blowing up gas tanks, avoiding radioactive slime. A copyright owner holds the right to create sequels, see Trust Co. Bank v. MGM/UA Entertainment Co., 772 F.2d 740 (11th Cir. 1985), and the stories told in the N/I MAP files are surely sequels, telling new (though somewhat repetitive) tales of Duke's fabulous adventures. A book about Duke Nukem would infringe for the same reason, even if it contained no pictures.5

Micro Star nonetheless claims that its use of D/N-3D's protected expression falls within the doctrine of fair use, which permits unauthorized use of copyrighted works "for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research." 17 U.S.C. § 107; see Narell v. Freeman, 872 F.2d 907, 913 (9th Cir. 1989). Section 107 instructs courts "determining whether the use made of a work in any particular case is a fair use" to consider four factors: (1) the purpose and character of the use, including whether it is commercial in nature; (2) the nature of the copyrighted work; (3) the amount and substantiality of the copied material in relation to the copyrighted work as a whole; and (4) the effect of the use on the potential market for the copyrighted work. 17 U.S.C. § 107. As a preliminary matter, Micro Star asks us to focus on the player's use of the N/I CD in evaluating the fair use claim, because-according to Micro Star-the player actually creates the derivative work. In Galoob, after we assumed for purposes of argument that the Game Genie did create derivative works, we went on to consider the fair use defense from the player's point of view. See Galoob, 964 F.2d at 970. But the fair use analysis in Galoob was not necessary and therefore is clearly dicta. More significantly, Nintendo alleged only contributory infringement--that Galoob was helping consumers create derivative works; FormGen here alleges direct infringement by Micro Star, because the MAP files encompass new Duke stories, which are themselves derivative works. Our examination of the section 107 factors yields straightforward results. Micro Star's use of FormGen's protected expression was made purely for financial gain. While that does not end our inquiry, see Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 584, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (1994), "every commercial use of copyrighted material is presumptively an unfair exploitation of the monopoly privilege that belongs to the owner of the copyright." Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 451, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984).6 The Supreme Court has explained that the second factor, the nature of the copyrighted work, is particularly significant because "some works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied." Campbell, 510 U.S. at 586, 114 S. Ct. 1164. The fair use defense will be much less likely to succeed when it is applied to fiction or fantasy creations, as opposed to factual works such as telephone listings. See United Tel. Co. v. Johnson Publ'g Co., 855 F.2d 604, 609 (8th Cir. 1988); see also Stewart v. Abend, 495 U.S. 207, 237, 110 S. Ct. 1750, 109 L. Ed. 2d 184 (1990). Duke Nukem's world is made up of aliens, radioactive slime and freezer weapons-clearly fantasies, even by Los Angeles standards. N/I MAP files "expressly use the [D/N-3D] story's unique setting, characters, [and] plot," Stewart, 495 U.S. at 238, 110 S. Ct. 1750; both the quantity and importance of the material Micro Star used are substantial. Finally, by selling N/I, Micro Star "impinged on [FormGen's] ability to market new versions of the [D/N-3D] story." Stewart, 495 U.S. at 238, 110 S. Ct. 1750; see also Twin Peaks Productions, Inc. v. Publications Int'l, Ltd., 996 F.2d 1366, 1377 (2d Cir. 1993). Only FormGen has the right to enter that market; whether it chooses to do so is entirely its business. " [N/I] neither falls into any of the categories enumerated in section 107 nor meets the four criteria set forth in section 107." Stewart, 495 U.S. at 237, 110 S. Ct. 1750. It is not protected by fair use.

The first statutory factor is the purpose and character of the alleged infringing use. The focus of this factor is whether the new work merely supersedes the objects of the original creation [,] or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is "transformative." Campbell, 510 U.S. at 579, 114 S. Ct. 1164 (internal citations omitted). The district court found that this factor "cuts to some extent in Kirkwood's direction," primarily because Kirkwood "is using [Infinity's] broadcasts for a quite different purpose," namely for information rather than entertainment. 965 F. Supp. at 557. Kirkwood, too, argues at length about the differences between Infinity's purposes for its broadcasts and the reasons his customers use Dial-Up. We agree that the difference in purposes tends to support Kirkwood's fair use claim. However, difference in purpose is not quite the same thing as transformation, and Campbell instructs that transformativeness is the critical inquiry under this factor. Although it "is not absolutely necessary" for a fair use to be transformative, "the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Campbell, 510 U.S. at 579, 114 S. Ct. 1164. Here, as the district judge observed, "Kirkwood's retransmissions leave the character of the original broadcasts unchanged. There is neither new expression, new meaning nor new message." 965 F. Supp. at 557. In short, there is no transformation. As then-District Judge Leval noted in his frequently-cited article on fair use, a use of copyrighted material that "merely repackages or republishes the original" is unlikely to be deemed a fair use. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1111 (1990). Kirkwood argues that Dial-Up's users transform the broadcasts by using them for their factual, not entertainment, content. However, it is Kirkwood's own retransmission of the broadcasts, not the acts of his end-users, that is at issue here and all Kirkwood does is sell access to unaltered radio broadcasts.2 Also, it is not clear that all of Kirkwood's target audience "transforms" the broadcasts as he suggests. Talent scouts, who admittedly would not be listening in order to be entertained themselves, would nevertheless be listening for the entertainment value of the broadcasts rather than the factual content. Kirkwood also argues that his service is "of substantial benefit to society" because, among other things, it enables advertisers to confirm that their commercials were properly aired. According to Infinity, however, 75 to 80 percent of the advertising on its stations is from companies within the station's broadcast range who may therefore monitor Infinity merely by turning on a radio. Similarly, the public benefit from other uses of Dial-Up can either be accomplished by other methods or is simply not enough to justify Kirkwood's non-transformative retransmissions. See Texaco, 60 F.3d at 922 (finding no fair use even though photocopying to facilitate "research in the sciences ... might well serve a broader public purpose."). We also note that the district court placed little or no emphasis on the commercial nature of Dial-Up. We agree that, notwithstanding its mention in the text of the statute, commerciality has only limited usefulness to a fair use inquiry; most secondary uses of copyrighted material, including nearly all of the uses listed in the statutory preamble, are commercial. As the Supreme Court observed in Campbell, to give commerciality a "presumptive force against a finding of fairness," would render the preamble a nullity. 510 U.S. at 584, 114 S. Ct. 1164. See also, Castle Rock Entertainment, Inc. v. Carol Publishing Group, Inc., 150 F.3d 132, 141-42 (2d Cir. 1998); Texaco, 60 F.3d at 921.3 In sum, we think the different, and possibly beneficial, purposes of Kirkwood's customers are outweighed by the total absence of transformativeness in Kirkwood's acts of retransmission. We find that the first statutory factor leans in Infinity's favor, rather than in Kirkwood's.

The second factor is the nature of the copyrighted work, which recognizes that creative works are "closer to the core of intended copyright protection" than more factual works. Campbell, 510 U.S. at 586, 114 S. Ct. 1164. The district court found that this factor favored Infinity. 965 F. Supp. at 557. Kirkwood admits that some portions of the broadcasts are original and therefore deserving of copyright protection, but argues that Infinity also broadcasts material such as music and advertisements to which it does not own the copyright. The district court properly recognized, however, that "the compilation of these and other elements to make the allegedly infringed works is both unique and creative." Id. The second factor favors Infinity.

3. Amount and Substantiality of the Portion Used The third factor, amount and substantiality of the portion used, recognizes that the more of a copyrighted work that is taken, the less likely the use is to be fair, and that even a less substantial taking may be unfair if it captures the essence of the copyrighted work. See e.g., Harper & Row, 471 U.S. at 564-65, 105 S. Ct. 2218 (although "words actually quoted were an insubstantial portion" of copyrighted book, they were "essentially the heart of the book"). Though not an absolute rule, "generally, it may not constitute a fair use if the entire work is reproduced." Nimmer on Copyright, § 13.05 [A] at 13-178 (1997). See also, Weissmann v. Freeman, 868 F.2d 1313, 1325 (2d Cir. 1989). It is somewhat difficult to apply this factor to Dial-Up. Depending on the interest of Kirkwood's customers, mere snippets of Infinity's broadcasts may be retransmitted or a single station may be listened to for hours at a time. According to the district court " [t]here is a virtually complete absence of proof concerning the extent to which the three copyrighted programs at issue were retransmitted by Kirkwood's service." 965 F. Supp. at 558. Since Kirkwood has the burden of proof on fair use, Texaco, 60 F.3d at 918, this seems to cut against Kirkwood. Moreover, like the district court, we are convinced that in this case the potential scope of retransmission is more relevant than evidence of actual retransmission by Dial-Up users thus far. Dial-Up permits essentially unlimited access to radio broadcasts in the cities in which it has receivers and there is thus the potential for retransmission of entire copyrighted programs. As the district court observed, [t]he more successful Kirkwood becomes in selling his service to interested parties, the more likely it is that any given broadcast will be retransmitted ... [and] that a subscriber, or the collective action of a plurality of subscribers, will cause Kirkwood to retransmit most or all of a given program. 965 F. Supp. at 558. Again, Kirkwood argues that because his users' motives are beneficial to the purpose of copyright it makes no difference if entire broadcasts are heard (see discussion in section B. 1, above). As there discussed, however, societal benefit does not guarantee a finding of fair use. Nor does it, by itself, answer the question most relevant to this factor: whether "no more was taken than necessary." Campbell, 510 U.S. at 589, 114 S. Ct. 1164. Even if Kirkwood is correct that society benefits from his provision of access to Infinity's broadcasts, he still must justify potentially providing his subscribers with access to every radio station in the cities Kirkwood serves, 24 hours a day, seven days a week. We agree with the district court that the third factor favors Infinity.

4. Effect of the Use upon the Potential Market The fourth factor, effect of the use on the potential market for the copyrighted work, requires courts to consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also "whether unrestricted and wide-spread conduct of the sort engaged in by the defendant ... would result in a substantially adverse impact on the potential market" for the original. Campbell, 510 U.S. at 590, 114 S. Ct. 1164 (quoting Nimmer, § 13.05 [A], at 13-102.61). Not all harms are cognizable under this factor. Specifically, fair use does not condemn the suppression or even destruction of the market by, for example, a "scathing theater review" or a parody. See Campbell, 510 U.S. at 591-92, 114 S. Ct. 1164. Rather, it is concerned with secondary uses that, by offering a substitute for the original, usurp a market that properly belongs to the copyright-holder. As always, Kirkwood bears the burden of showing that his use does not cause this type of harm to Infinity's interests, including Infinity's own listen lines. But cf., Leibovitz v. Paramount Pictures Corp., 137 F.3d 109, 116 n. 6 (2d Cir. 1998) (in case where plaintiff conceded lack of market for derivative works, "defendant had no obligation to present evidence showing lack of harm"). Historically, the fourth factor has been seen as central to fair use analysis, Harper & Row, 471 U.S. at 566, 105 S. Ct. 2218 ("This last factor is undoubtedly the single most important element of fair use."), although the Supreme Court appears to have backed away from this position. See Texaco, 60 F.3d at 926 (observing that the Supreme Court in Campbell omitted mention of the fourth factor's primacy and instructed that all factors are to be explored); Castle Rock, 150 F.3d at 145-46 ("Supreme Court has recently retreated ..." from suggestion that fourth factor is most important); Leibovitz, 137 F.3d at 113. But see Princeton University Press v. Michigan Document Svcs., 99 F.3d 1381, 1385 (6th Cir. 1996) (in banc) (acknowledging Second Circuit's observation but calling fourth factor first among equals). See generally, Leval, 103 Harv. L. Rev. at 1124 (Supreme Court has "somewhat overstated" importance of fourth factor; fact that secondary use does not harm the market "gives no assurance" that secondary use is justified). The district court stated that it would have reached the same result regardless of the comparative importance of the fourth factor, 965 F. Supp. at 559 n. 21, and so would we. The district court, in applying this factor, started from the assumption that the value of Infinity's copyrighted programs "lies in the ability to attract listeners" by providing the programs to the general public either on its own stations or, by syndication, on stations owned by others. Id. at 559. The court then concluded that Kirkwood's service, which is "directed to a narrow and specialized audience," was likely to have no material effect on the value of the copyrighted material. Id. at 560. The district court regarded as particularly relevant that Infinity is neither in the business of operating commercial listen lines nor has it "attempted to license the retransmission of its copyrighted programming for purposes such as Kirkwood's," and that the potential revenues to be obtained from such licensing, based on Kirkwood's experience, are "extremely modest." Id. We are not as impressed by these facts as was the district court. Although Infinity does not operate commercial listen lines such as Kirkwood's, certain of its stations have their own listen lines and Infinity described these lines as being included in a "package" offered to certain advertisers. It is true that Infinity does not charge separately for these lines, and that Kirkwood has had limited success in doing so himself. But this does not necessarily mean that Infinity derives no economic benefit from its listen lines or that Kirkwood's use has only a negligible effect on Infinity's potential to exploit this market. Infinity, in the exercise of its business judgment, has decided that its best current use of listen lines is to offer them at no additional cost to certain "valued customers." Dial-Up disrupts this practice by removing Infinity's control over who should have access to such lines. Kirkwood is selling Infinity's copyrighted material in a market that Infinity, as the copyright owner, is exclusively entitled to exploit. Kirkwood does not suppress demand for Infinity's broadcasts in the manner of a reviewer, but instead replaces Infinity as the supplier of those broadcasts to meet the demand of his customers. This is precisely the kind of harm the fourth factor aims to prevent. As Infinity points out in its brief on appeal, "the revenues that Kirkwood generates for himself come not from a market that is only 'likely to be developed,' Texaco, 60 F.3d at 930, but from a market that Infinity currently occupies," albeit in different form. Infinity's presence in Kirkwood's market weighs in Infinity's favor on this factor. Kirkwood also implies that Infinity lacks an incentive to exploit the listen line market because enabling advertisers to check up on Infinity's advertising billing practices is against its interest. This is not obvious, as evidenced by the fact that Infinity does offer listen lines to some advertisers. If we assume that advertisers will prefer to use broadcasters that can be relied upon to be honest, and consider the provision of monitoring services like listen lines to be an indicator of such reliability, then broadcasters who do not offer listen lines may be at a competitive disadvantage. Also, the possible unavailability of a substitute for Kirkwood's service is something that may be considered in determining the remedy, if any, to which Infinity may be entitled. The Supreme Court suggested in Campbell that injunctive relief may be withheld where there is "a strong public interest in the publication of the secondary work [and] the copyright owner may be adequately protected by an award of damages." 510 U.S. at 578 n. 10, 114 S. Ct. 1164 (quoting Leval, 103 Harv. L. Rev. at 1132). We express no opinion as to the appropriate remedy in this case if Kirkwood is finally adjudged an infringer.

Purpose and character of the use. The first factor, 17 U.S.C. § 107(1), requires a court to consider "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." The central purpose of this inquiry is to determine whether and to what extent the new work is "transformative." Campbell, 510 U.S. at 579. A work is "transformative" when the new work does not "merely supersede the objects of the original creation" but rather "adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Id. (internal quotation and alteration omitted). Conversely, if the new work "supersede[s] the use of the original," the use is likely not a fair use. Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 550-51, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985) (internal quotation omitted) (publishing the "heart" of an unpublished work and thus supplanting the copyright holder's first publication right was not a fair use); see also Wall Data Inc. v. L.A. County Sheriff's Dep't, 447 F.3d 769, 778-82 (9th Cir. 2006) (using a copy to save the cost of buying additional copies of a computer program was not a fair use). As noted in Campbell, a "transformative work" is one that alters the original work "with new expression, meaning, or message." Campbell, 510 U.S. at 579. "A use is considered transformative only where a defendant changes a plaintiff's copyrighted work or uses the plaintiff's copyrighted work in a different context such that the plaintiff's work is transformed into a new creation." Wall Data, 447 F.3d at 778. Google's use of thumbnails is highly transformative. In Kelly, we concluded that Arriba's use of thumbnails was transformative because "Arriba's use of the images serve[d] a different function than Kelly's use--improving access to information on the [I]nternet versus artistic expression." Kelly, 336 F.3d at 819. Although an image may have been created originally to serve an entertainment, aesthetic, or informative function, a search engine transforms the image into a pointer directing a user to a source of information. Just as a "parody has an obvious claim to transformative value" because "it can provide social benefit, by shedding light on an earlier work, and, in the process, creating a new one,"Campbell, 510 U.S. at 579, a search engine provides social benefit by incorporating an original work into a new work, namely, an electronic reference tool. Indeed, a search engine may be more transformative than a parody because a search engine provides an entirely new use for the original work, while a parody typically has the same entertainment purpose as the original work. See, e.g., id. at 594-96 (holding that 2 Live Crew's parody of "Oh, Pretty Woman" using the words "hairy woman" or "bald headed woman" was a transformative work, and thus constituted a fair use); Mattel, Inc. v. Walking Mountain Prods., 353 F.3d 792, 796-98, 800-06 (9th Cir 2003) (concluding that photos parodying Barbie by depicting "nude Barbie dolls juxtaposed with vintage kitchen appliances" was a fair use). In other words, a search engine puts images "in a different context" so that they are "transformed into a new creation." Wall Data, 447 F.3d at 778. The fact that Google incorporates the entire Perfect 10 image into the search engine results does not diminish the transformative nature of Google's use. As the district court correctly noted, Perfect 10, 416 F. Supp. 2d at 848-49, we determined in Kelly that even making an exact copy of a work may be transformative so long as the copy serves a different function than the original work, Kelly, 336 F.3d at 818-19. For example, the First Circuit has held that the republication of photos taken for a modeling portfolio in a newspaper was transformative because the photos served to inform, as well as entertain. See Nez v. Caribbean Int'l News Corp., 235 F.3d 18, 22-23 (1st Cir. 2000). In contrast, duplicating a church's religious book for use by a different church was not transformative. See Worldwide Church of God v. Phila. Church of God, Inc., 227 F.3d 1110, 1117 (9th Cir. 2000). Nor was a broadcaster's simple retransmission of a radio broadcast over telephone lines transformative, where the original radio shows were given no "new expression, meaning, or message." Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir. 1998). Here, Google uses Perfect 10's images in a new context to serve a different purpose. The district court nevertheless determined that Google's use of thumbnail images was less transformative than Arriba's use of thumbnails in Kelly because Google's use of thumbnails superseded Perfect 10's right to sell its reduced-size images for use on cell phones. See Perfect 10, 416 F. Supp. 2d at 849. The district court stated that "mobile users can download and save the thumbnails displayed by Google Image Search onto their phones," and concluded "to the extent that users may choose to download free images to their phone rather than purchase [Perfect 10's] reduced-size images, Google's use supersedes [Perfect 10's]." Id. Additionally, the district court determined that the commercial nature of Google's use weighed against its transformative nature. Id. Although Kelly held that the commercial use of the photographer's images by Arriba's search engine was less exploitative than typical commercial use, and thus weighed only slightly against a finding of fair use, Kelly, 336 F.3d at 818-20, the district court here distinguished Kelly on the ground that some website owners in the AdSense program had infringing Perfect 10 images on their websites, Perfect 10, 416 F. Supp. 2d at 846-47. The district court held that because Google's thumbnails "lead users to sites that directly benefit Google's bottom line," the AdSense program increased the commercial nature of Google's use of Perfect 10's images. Id. at 847. In conducting our case-specific analysis of fair use in light of the purposes of copyright, Campbell, 510 U.S. at 581, we must weigh Google's superseding and commercial uses of thumbnail images against Google's significant transformative use, as well as the extent to which Google's search engine promotes the purposes of copyright and serves the interests of the public. Although the district court acknowledged the "truism that search engines such as Google Image Search provide great value to the public," Perfect 10, 416 F. Supp. 2d at 848-49, the district court did not expressly consider whether this value outweighed the significance of Google's superseding use or the commercial nature of Google's use. Id. at 849. The Supreme Court, however, has directed us to be mindful of the extent to which a use promotes the purposes of copyright and serves the interests of the public. See Campbell, 510 U.S. at 579; Harper & Row, 471 U.S. at 556-57; Sony, 464 U.S. at 431-32. We note that the superseding use in this case is not significant at present: the district court did not find that any downloads for mobile phone use had taken place. See Perfect 10, 416 F. Supp. 2d at 849. Moreover, while Google's use of thumbnails to direct users to AdSense partners containing infringing content adds a commercial dimension that did not exist in Kelly, the district court did not determine that this commercial element was significant. See id. at 848-49. The district court stated that Google's AdSense programs as a whole contributed "$ 630 million, or 46% of total revenues" to Google's bottom line, but noted that this figure did not "break down the much smaller amount attributable to websites that contain infringing content." Id. at 847 & n.12 (internal quotation omitted). We conclude that the significantly transformative nature of Google's search engine, particularly in light of its public benefit, outweighs Google's superseding and commercial uses of the thumbnails in this case. In reaching this conclusion, we note the importance of analyzing fair use flexibly in light of new circumstances. Sony, 464 U.S. at 431-32; id. at 448 n.31 ("'[Section 107] endorses the purpose and general scope of the judicial doctrine of fair use, but there is no disposition to freeze the doctrine in the statute, especially during a period of rapid technological change.'" (quoting H.R. Rep. No. 94-1476, p. 65-66 (1976), U.S. Code Cong. & Admin. News 1976, p. 5680)). We are also mindful of the Supreme Court's direction that "the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Campbell, 510 U.S. at 579. Accordingly, we disagree with the district court's conclusion that because Google's use of the thumbnails could supersede Perfect 10's cell phone download use and because the use was more commercial than Arriba's, this fair use factor weighed "slightly" in favor of Perfect 10. Perfect 10, 416 F. Supp. 2d at 849. Instead, we conclude that the transformative nature of Google's use is more significant than any incidental superseding use or the minor commercial aspects of Google's search engine and website. Therefore, this factor weighs heavily in favor of Google.

The nature of the copyrighted work. With respect to the second factor, "the nature of the copyrighted work," 17 U.S.C. § 107(2), our decision in Kelly is directly on point. There we held that the photographer's images were "creative in nature" and thus "closer to the core of intended copyright protection than are more fact-based works." Kelly, 336 F.3d at 820 (internal quotation omitted). However, because the photos appeared on the Internet before Arriba used thumbnail versions in its search engine results, this factor weighed only slightly in favor of the photographer. Id. Here, the district court found that Perfect 10's images were creative but also previously published. Perfect 10, 416 F. Supp. 2d at 850. The right of first publication is "the author's right to control the first public appearance of his expression." Harper & Row, 471 U.S. at 564. Because this right encompasses "the choices of when, where, and in what form first to publish a work," id., an author exercises and exhausts this one-time right by publishing the work in any medium. See, e.g., Batjac Prods. Inc. v. GoodTimes Home Video Corp., 160 F.3d 1223, 1235 (9th Cir. 1998) (noting, in the context of the common law right of first publication, that such a right "does not entail multiple first publication rights in every available medium"). Once Perfect 10 has exploited this commercially valuable right of first publication by putting its images on the Internet for paid subscribers, Perfect 10 is no longer entitled to the enhanced protection available for an unpublished work. Accordingly the district court did not err in holding that this factor weighed only slightly in favor of Perfect 10. See Perfect 10, 416 F. Supp. 2d at 849-50.

The amount and substantiality of the portion used. "The third factor asks whether the amount and substantiality of the portion used in relation to the copyrighted work as a whole . . . are reasonable in relation to the purpose of the copying." Campbell, 510 U.S. at 586 (internal quotation omitted); see also 17 U.S.C. § 107(3). In Kelly, we held Arriba's use of the entire photographic image was reasonable in light of the purpose of a search engine. Kelly, 336 F.3d at 821. Specifically, we noted, "[i]t was necessary for Arriba to copy the entire image to allow users to recognize the image and decide whether to pursue more information about the image or the originating [website]. If Arriba only copied part of the image, it would be more difficult to identify it, thereby reducing the usefulness of the visual search engine." Id. Accordingly, we concluded that this factor did not weigh in favor of either party. Id. Because the same analysis applies to Google's use of Perfect 10's image, the district court did not err in finding that this factor favored neither party.

Effect of use on the market. The fourth factor is "the effect of the use upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). In Kelly, we concluded that Arriba's use of the thumbnail images did not harm the market for the photographer's full-size images. See Kelly, 336 F.3d at 821-22. We reasoned that because thumbnails were not a substitute for the full-sized images, they did not harm the photographer's ability to sell or license his full-sized images. Id. The district court here followed Kelly's reasoning, holding that Google's use of thumbnails did not hurt Perfect 10's market for full-size images. See Perfect 10, 416 F. Supp. 2d at 850-51. We agree. Perfect 10 argues that the district court erred because the likelihood of market harm may be presumed if the intended use of an image is for commercial gain. However, this presumption does not arise when a work is transformative because "market substitution is at least less certain, and market harm may not be so readily inferred." Campbell, 510 U.S. at 591. As previously discussed, Google's use of thumbnails for search engine purposes is highly transformative, and so market harm cannot be presumed. Perfect 10 also has a market for reduced-size images, an issue not considered in Kelly. The district court held that "Google's use of thumbnails likely does harm the potential market for the downloading of [Perfect 10's] reduced-size images onto cell phones." Perfect 10, 416 F. Supp. 2d at 851 (emphasis omitted). The district court reasoned that persons who can obtain Perfect 10 images free of charge from Google are less likely to pay for a download, and the availability of Google's thumbnail images would harm Perfect 10's market for cell phone downloads. Id. As we discussed above, the district court did not make a finding that Google users have downloaded thumbnail images for cell phone use. This potential harm to Perfect 10's market remains hypothetical. We conclude that this factor favors neither party.

3. Purpose and character of the use Under the first factor, purpose and character of the use, we inquire into whether Connectix's Virtual Game Station merely supersedes the objects of the original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is "transformative." Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579, 127 L. Ed. 2d 500, 114 S. Ct. 1164 (1994) (internal quotation marks and citations omitted). As an initial matter, we conclude that the district court applied an erroneous legal standard; the district court held that Connectix's commercial purpose in copying the Sony BIOS gave rise to a "presumption of unfairness that . . . can be rebutted by the characteristics of a particular commercial use." Order at 14-15 (citing Sega, 977 F.2d at 1522). Since Sega, however, the Supreme Court has rejected this presumption as applied to the first and fourth factor of the fair use analysis. Acuff-Rose, 510 U.S. at 584, 594 (clarifying Sony, 464 U.S. at 451). Instead, the fact that Connectix's copying of the Sony BIOS was for a commercial purpose is only a "separate factor that tends to weigh against a finding of fair use." Id. at 585 (internal quotation marks omitted). We find that Connectix's Virtual Game Station is modestly transformative. The product creates a new platform, the personal computer, on which consumers can play games designed for the Sony PlayStation. This innovation affords opportunities for game play in new environments, specifically anywhere a Sony PlayStation console and television are not available, but a computer with a CD-ROM drive is. More important, the Virtual Game Station itself is a wholly new product, notwithstanding the similarity of uses and functions between the Sony PlayStation and the Virtual Game Station. The expressive element of software lies as much in the organization and structure of the object code that runs the computer as it does in the visual expression of that code that appears on a computer screen. See 17 U.S.C. § 102(a) (extending copyright protection to original works of authorship that "can be perceived, reproduced, or otherwise communicated, either directly or with the aid of a machine or device"). Sony does not claim that the Virtual Game Station itself contains object code that infringes Sony's copyright. We are therefore at a loss to see how Connectix's drafting of entirely new object code for its VGS program could not be transformative, despite the similarities in function and screen output. Finally, we must weigh the extent of any transformation in Connectix's Virtual Game Station against the significance of other factors, including commercialism, that militate against fair use. See Acuff-Rose, 510 U.S. at 579. Connectix's commercial use of the copyrighted material was an intermediate one, and thus was only "indirect or derivative." Sega, 977 F.2d at 1522. Moreover, Connectix reverse-engineered the Sony BIOS to produce a product that would be compatible with games designed for the Sony PlayStation. We have recognized this purpose as a legitimate one under the first factor of the fair use analysis. See id. Upon weighing these factors, we find that the first factor favors Connectix. The district court ruled, however, that the Virtual Game Station was not transformative on the rationale that a computer screen and a television screen are interchangeable, and the Connectix product therefore merely "supplants" the Sony PlayStation console. Order at 15. The district court clearly erred. For the reasons stated above, the Virtual Game Station is transformative and does not merely supplant the PlayStation console. In reaching its decision, the district court apparently failed to consider the expressive nature of the Virtual Game Station software itself. Sony's reliance on Infinity Broadcast Corp. v. Kirkwood, 150 F.3d 104 (2d Cir. 1998), suffers from the same defect. The Infinity court reasoned that a "change of format, though useful, is not technically a transformation." Id. at 108 n.2. But the infringing party in that case was merely taking copyrighted radio transmissions and retransmitting them over telephone lines; there was no new expression. Id. at 108. Infinity does not change our conclusion; the purpose and character of Connectix's copying points toward fair use.

1. Nature of the copyrighted work Under our analysis of the second statutory factor, nature of the copyrighted work, we recognize that "some works are closer to the core of intended copyright protection than others." Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 586, 127 L. Ed. 2d 500, 114 S. Ct. 1164 (1994). Sony's BIOS lies at a distance from the core because it contains unprotected aspects that cannot be examined without copying. See Sega, 977 F.2d at 1526. We consequently accord it a "lower degree of protection than more traditional literary works." Id. As we have applied this standard, Connectix's copying of the Sony BIOS must have been "necessary" to have been fair use. See id. at 1524-26. We conclude that it was. There is no question that the Sony BIOS contains unprotected functional elements. Nor is it disputed that Connectix could not gain access to these unprotected functional elements without copying the Sony BIOS. Sony admits that little technical information about the functionality of the Sony BIOS is publicly available. The Sony BIOS is an internal operating system that does not produce a screen display to reflect its functioning. Consequently, if Connectix was to gain access to the functional elements of the Sony BIOS it had to be through a form of reverse engineering that required copying the Sony BIOS onto a computer. Sony does not dispute this proposition. The question then becomes whether the methods by which Connectix reverse-engineered the Sony BIOS were necessary to gain access to the unprotected functional elements within the program. We conclude that they were. Connectix employed several methods of reverse engineering (observation and observation with partial disassembly) each of which required Connectix to make intermediate copies of copyrighted material. Neither of these methods renders fair use protection inapplicable. Sega expressly sanctioned disassembly. See id. at 1527-28. We see no reason to distinguish observation of copyrighted software in an emulated computer environment. Both methods require the reverse engineer to copy protected as well as unprotected elements of the computer program. Because this intermediate copying is the gravamen of the intermediate infringement claim, see 17 U.S.C. § 106(1); Sega, 977 F.2d at 1518-19, and both methods of reverse engineering require it, we find no reason inherent in these methods to prefer one to another as a matter of copyright law. Connectix presented evidence that it observed the Sony BIOS in an emulated environment to observe the functional aspects of the Sony BIOS. When this method of reverse engineering was unsuccessful, Connectix engineers disassembled discrete portions of the Sony BIOS to view directly the ideas contained therein. We conclude that intermediate copying in this manner was "necessary" within the meaning of Sega. We decline to follow the approach taken by the district court. The district court did not focus on whether Connectix's copying of the Sony BIOS was necessary for access to functional elements. Instead, it found that Connectix's copying and use of the Sony BIOS to develop its own software exceeded the scope of Sega. See Order at 17 ("They disassembled Sony's code not just to study the concepts. They actually used that code in the development of [their] product."). This rationale is unpersuasive. It is true that Sega referred to "studying or examining the unprotected aspects of a copyrighted computer program." 977 F.2d at 1520 (emphasis added). But in Sega, Accolade's copying, observation and disassembly of Sega's game cartridges was held to be fair use, even though Accolade "loaded the disassembled code back into a computer, and experimented to discover the interface specifications for the Genesis console by modifying the programs and studying the results." Id. at 1515. Thus, the distinction between "studying" and "use" is unsupported in Sega. Moreover, reverse engineering is a technically complex, frequently iterative process. Johnson-Laird, 19 U. Dayton L. Rev. at 843-44. Within the limited context of a claim of intermediate infringement, we find the semantic distinction between "studying" and "use" to be artificial, and decline to adopt it for purposes of determining fair use. We also reject the argument, urged by Sony, that Connectix infringed the Sony copyright by repeatedly observing the Sony BIOS in an emulated environment, thereby making repeated copies of the Sony BIOS. These intermediate copies could not have been "necessary" under Sega, contends Sony, because Connectix engineers could have disassembled the entire Sony BIOS first, then written their own Connectix BIOS, and used the Connectix BIOS to develop the Virtual Game Station hardware emulation software. We accept Sony's factual predicate for the limited purpose of this appeal. Our doing so, however, does not aid Sony. Sony contends that Connectix's reverse engineering of the Sony BIOS should be considered unnecessary on the rationale that Connectix's decision to observe the Sony BIOS in an emulated environment required Connectix to make more intermediate copies of the Sony BIOS than if Connectix had performed a complete disassembly of the program. Under this logic, at least some of the intermediate copies were not necessary within the meaning of Sega. This construction stretches Sega too far. The "necessity" we addressed in Sega was the necessity of the method, i.e., disassembly, not the necessity of the number of times that method was applied. See 977 F.2d at 1524-26. In any event, the interpretation advanced by Sony would be a poor criterion for fair use. Most of the intermediate copies of the Sony BIOS were made by Connectix engineers when they booted up their computers and the Sony BIOS was copied into RAM. But if Connectix engineers had left their computers turned on throughout the period during which they were observing the Sony BIOS in an emulated environment, they would have made far fewer intermediate copies of the Sony BIOS (perhaps as few as one per computer). Even if we were inclined to supervise the engineering solutions of software companies in minute detail, and we are not, our application of the copyright law would not turn on such a distinction. Such a rule could be easily manipulated. More important, the rule urged by Sony would require that a software engineer, faced with two engineering solutions that each require intermediate copying of protected and unprotected material, often follow the least efficient solution. (In cases in which the solution that required the fewest number of intermediate copies was also the most efficient, an engineer would pursue it, presumably, without our urging.) This is precisely the kind of "wasted effort that the proscription against the copyright of ideas and facts . . . [is] designed to prevent." Feist Publications, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 354, 113 L. Ed. 2d 358, 111 S. Ct. 1282 (1991) (internal quotation marks omitted). Such an approach would erect an artificial hurdle in the way of the public's access to the ideas contained within copyrighted software programs. These are "aspects that were expressly denied copyright protection by Congress." Sega, 977 F.2d at 1526 (citing 17 U.S.C. § 102(b)). We decline to erect such a barrier in this case. If Sony wishes to obtain a lawful monopoly on the functional concepts in its software, it must satisfy the more stringent standards of the patent laws. See Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 160-61, 103 L. Ed. 2d 118, 109 S. Ct. 971 (1989); Sega, 977 F.2d at 1526. This Sony has not done. The second statutory factor strongly favors Connectix.

2. Amount and substantiality of the portion used With respect to the third statutory factor, amount and substantiality of the portion used in relation to the copyrighted work as a whole, Connectix disassembled parts of the Sony BIOS and copied the entire Sony BIOS multiple times. This factor therefore weighs against Connectix. But as we concluded in Sega, in a case of intermediate infringement when the final product does not itself contain infringing material, this factor is of "very little weight." Sega, 977 F.2d at 1526-27; see also Sony Corp. of Am. v. Universal City Studios, Inc., 464 U.S. 417, 449-50, 78 L. Ed. 2d 574, 104 S. Ct. 774 (1984) (copying of entire work does not preclude fair use).

4. Effect of the use upon the potential market We also find that the fourth factor, effect of the use upon the potential market, favors Connectix. Under this factor, we consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also "whether unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market" for the original. Acuff-Rose, 510 U.S. at 590 (quoting 3 M. Nimmer & D. Nimmer, Nimmer on Copyright, § 13.05[A][4], at 13-102.61 (1993)). Whereas a work that merely supplants or supersedes another is likely to cause a substantially adverse impact on the potential market of the original, a transformative work is less likely to do so. See id. at 591; Harper & Row, Publishers, Inc. v. Nation Enters, Inc., 471 U.S. 539, 567-69, 85 L. Ed. 2d 588, 105 S. Ct. 2218 (1985). The district court found that "to the extent that such a substitution [of Connectix's Virtual Game Station for Sony PlayStation console] occurs, Sony will lose console sales and profits." Order at 19. We recognize that this may be so. But because the Virtual Game Station is transformative, and does not merely supplant the PlayStation console, the Virtual Game Station is a legitimate competitor in the market for platforms on which Sony and Sony-licensed games can be played. See Sega, 977 F.2d at 1522-23. For this reason, some economic loss by Sony as a result of this competition does not compel a finding of no fair use. Sony understandably seeks control over the market for devices that play games Sony produces or licenses. The copyright law, however, does not confer such a monopoly. See id. at 1523-24 ("An attempt to monopolize the market by making it impossible for others to compete runs counter to the statutory purpose of promoting creative expression and cannot constitute a strong equitable basis for resisting the invocation of the fair use doctrine."). This factor favors Connectix. B. Tarnishment The district court found that Connectix's sale of the Virtual Game Station tarnished Sony's "PlayStation" mark under 15 U.S.C. § 1125(c)(1). The district court based its preliminary injunction, however, exclusively on Sony's copyright claim, and did not cite its tarnishment finding as a ground for the injunction. Although we can "affirm the district court on any ground supported by the record," Charley's Taxi Radio Dispatch Corp. v. SIDA of Haw., Inc., 810 F.2d 869, 874 (9th Cir. 1987), we decline to affirm on this alternative ground. Sony has not shown a likelihood of success on each element of the tarnishment claim. To prevail on its tarnishment claim, Sony must show that (1) the PlayStation "mark is famous;" (2) Connectix is "making a commercial use of the mark;" (3) Connectix's "use began after the mark became famous;" and (4) Connectix's "use of the mark dilutes the quality of the mark by diminishing the capacity of the mark to identify and distinguish goods and services." Films of Distinction, Inc. v. Allegro Film Prods., Inc., 12 F. Supp. 2d 1068, 1078 (C.D. Cal. 1998); 15 U.S.C. §§ 1125 (c)(1), 1127 (definition of "dilution"). Connectix does not dispute the first and third of these elements. We address only the fourth element. Because Sony proceeds under a tarnishment theory of dilution, it must show under this fourth element that its PlayStation mark will "suffer negative associations" through Connectix's use. Hormel Foods Corp. v. Jim Henson Prods., Inc., 73 F.3d 497, 507 (2d Cir. 1996); see also 4 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 24.95 (4th ed. 1996 & Supp. 1999). The district court found the Virtual Game Station does not play PlayStation games as well as the PlayStation console, and that although the Virtual Game Station's packaging contains a disclaimer to this effect, "game players do not comprehend this distinction." Order at 24-25. The Sony PlayStation mark therefore suffers negative associations because of this confusion on the part of consumers who play Sony games on the Virtual Game Station software. Id. at 25. The evidence on the record does not support such a finding of misattribution. The district court relied primarily on a series of semi-anonymous reviews posted on the Internet and submitted by Connectix. As the district court acknowledged, these reviews were neither authenticated nor identified. More important, the print-out of the comments does not reveal the context in which the comments were made; this omission makes the extent of any confusion by game players difficult to assess reliably. The district court also referred to two focus group studies conducted by market research firms at Sony's bequest. These studies address the difference of quality between the Virtual Game Station and PlayStation, but shed no light on the question of misattribution. Thus, we reject as clearly erroneous the district court's finding that the Virtual Game Station tarnishes the Sony PlayStation mark on a misattribution theory of tarnishment. Nor are we persuaded by Sony's argument that the difference in quality between the two platforms is itself sufficient to find tarnishment. See Deere & Co. v. MTD Prods, Inc., 41 F.3d 39, 43 (2d Cir. 1994) ("'Tarnishment' generally arises when the plaintiff's trademark is linked to products of shoddy quality," diminishing the value of the mark "because the public will associate the lack of quality . . . with the plaintiff's unrelated goods."). Even if we assume, without deciding, that the concept of tarnishment is applicable to the present factual scenario, there is insufficient evidence to support a finding of tarnishment. "The sine qua non of tarnishment is a finding that plaintiff's mark will suffer negative associations through defendant's use." Hormel Foods, 73 F.3d at 507. The evidence here fails to show or suggest that Sony's mark or product was regarded or was likely to be regarded negatively because of its performance on Connectix's Virtual Game Station. The evidence is not even substantial on the quality of that performance. The Sony studies, each of included eight participants, presented a range of conclusions. One study concluded that "on balance, the results of this focus group study show that the testers preferred the PlayStation gaming experience over the Virtual Game Station gaming experience." The other concluded that consumers found the Virtual Game Station was "generally acceptable" for one game, but "nearly unplayable" on another. The internet reviews submitted by Connectix also presented a range of opinion; while some anonymous reviewers loved the Virtual Game Station, some were ambivalent, and a relative few hated the Virtual Game Station emulation. In the only review for attribution, Newsweek said the software played "surprisingly well," and that some games on the Virtual Game Station "rocked." Steven Levy, "Play it Your Way," Newsweek, Mar. 15, 1999, at 84. This evidence is insufficient to support a conclusion that the shoddiness of the Virtual Game Station alone tarnishes the Sony mark. Sony's tarnishment claim cannot support the injunction.

1. The first factor calls for consideration of "the purposes and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. 107(1). "The central purpose of this investigation is to see, in Justice Story's words, whether the new work merely `supersede [s] the objects' of the original creation [citations omitted] or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is`transformative.' " Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994). As Justice Story put it: "There must be real, substantial condensation of the materials, and intellectual labor and judgment bestowed thereon; and not merely the facile use of the scissors; or extracts of the essential parts, constituting the chief value of the original work." Folsom, 9 F. Cas. at 345. PCG's copying of WCG's MOA in its entirety bespeaks no "intellectual labor and judgment." It merely "supersedes the object" of the original MOA, to serve religious practice and education. Although "transformative use is not absolutely necessary for a finding of fair use," Campbell, 510 U.S. at 579, where the "use is for the same intrinsic purpose as [the copyright holder's] . . . such use seriously weakens a claimed fair use." Weissmann v. Freeman, 868 F.2d 1313, 1324 (2d Cir. 1989). Nevertheless, PCG argues that this factor favors fair use because its use is not commercial or for profit. The Supreme Court has cautioned that "the commercial or nonprofit educational purpose of a work is only one element of the first factor inquiry into its purpose and character." Campbell, 510 U.S. at 584. While the fact that a publication is commercial tends to weigh against fair use, the absence of a commercial use merely eliminates the presumption of unfairness. " [T]he mere fact that a use is educational and not for profit does not insulate it from a finding of infringement . . . ." Id.; see also Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 450 (1984) ("Even copying for noncommercial purposes may impair the copyright holder's ability to obtain the rewards that Congress intended him to have."); Marcus v. Rowley, 695 F.2d 1171, 1175 (9th Cir. 1983)."The crux of the profit/nonprofit distinction is not whether the sole motive of the use is monetary gain but whether the user stands to profit from exploitation of the copyrighted material without paying the customary price." Harper & Row, 471 U.S. at 562. We agree with the Second Circuit that in weighing whether the purpose was for "profit," " [m]onetary gain is not the sole criterion . . . [p]articularly in [a] . . . setting [where] profit is ill-measured in dollars." Weissmann, 868 F.2d at 1324 (holding that a professor's verbatim copying of an academic work was not fair use, in part because "the profit/nonprofit distinction is context specific, not dollar dominated" and a professor can "profit" by gaining recognition among his peers and authorship credit). See also WEBSTER'S THIRD NEW INTERNATIONAL DICTIONARY (1971) 1811 (defining "profit" as "an advantage, [a] benefit"). Putting aside the disputed question whether PCG uses MOA to generate income, and having in mind that like academia, religion is generally regarded as "not dollar dominated," MOA's use unquestionably profits PCG by providing it at no cost with the core text essential to its members' religious observance, by attracting through distribution of MOA new members who tithe ten percent of their income to PCG, and by enabling the ministry's growth. During the time of PCG's production and distribution of copies of MOA its membership grew to some seven thousand members. It is beyond dispute that PCG "profited" from copying MOA--it gained an "advantage" or "benefit" from its distribution and use of MOA without having to account to the copyright holder. The first factor weighs against fair use.

2. The second statutory factor,"the nature of the copyrighted work," turns on whether the work is informational or creative. See Harper & Row, 471 U.S. at 563 ("The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy."); see also Sony, 464 U.S. at 455 n.40 ("Copying a news broadcast may have a stronger claim to fair use than copying a motion picture."); Hustler, 796 F.2d at 1153-54 ("The scope of fair use is greater when `informational' as opposed to more `creative' works are involved."). PCG's brief describes MOA as "primarily a textual, historical account of [Armstrong's] views of the `the truth' of the Bible." While it may be viewed as "factual" by readers who share Armstrong's religious beliefs, the creativity, imagination and originality embodied in MOA tilt the scale against fair use. See Dr. Seuss Enter., L.P. v. Penguin Books USA, Inc., 109 F.3d 1394, 1402 (9th Cir. 1997).

3. The third factor directs us to consider "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." 17 U.S.C. S107(3). PCG copied the entire MOA verbatim, deleting only the "Suggested Readings" and the reference to "Worldwide Church of God " from the copyright page. While "wholesale copying does not preclude fair use per se," copying an entire work "militates against a finding of fair use." Hustler, 796 F.2d at 1155. Moreover, "the fact that a substantial portion of the infringing work was copied verbatim is evidence of the qualitative value of the copied material, both to the originator and to the plagiarist who seeks to profit from marketing someone else's copyrighted expression." Harper & Row, 471 U.S. at 565. PCG argues its verbatim copying of the whole work is reasonable because its use of MOA is religious in nature. " [T]he extent of permissible copying varies with the purpose and character of the use." Campbell, 510 U.S. at 586-87. In Campbell, the Court held that " [c]opying does not become excessive in relation to parodic purpose merely because the portion taken was the original's heart." Id. at 588. PCG's copying stands on a different footing for the purpose for which it uses the MOA is the same as WCG's. This court has held "that a finding that the alleged infringers copied the material to use it for the same intrinsic purpose for which the copyright owner intended it to be used is strong indicia of no fair use." Marcus, 695 F.2d at 1175. Reliance on Sony would be misplaced. There, the Supreme Court held that reproduction of the entire work " [did] not have its ordinary effect of militating against a finding of fair use" under the unique circumstances of that case, to wit: copying of videotapes for time-shifting for personal use to "enable a viewer to see such a work which he had been invited to witness in its entirety free of charge." Sony, 464 U.S. at 449-50. No such circumstances exist here to justify PCG's reproduction of the entire work. PCG uses the MOA as a central element of its members' religious observance; a reasonable person would expect PCG to pay WCG for the right to copy and distribute MOA created by WCG with its resources. The third factor, therefore, weighs against fair use.

4. The fourth factor considers "the effect of the use upon the potential market for or value of the copyrighted work." 17 U.S.C. 107(4). It has been said that " [f]air use, when properly applied, is limited to copying by others which does not materially impair the marketability of the work which is copied." Harper & Row, 471 U.S. at 566-67 (quoting Nimmer, Copyright 1.10 [D], at 1-87). This case presents a novel application of the fair use doctrine where the copyright owner is a not-for-profit organization. As might be expected, published case law deals with works marketed for profit. However, it cannot be inferred from that fact that the absence of a conventional market for a work, the copyright to which is held by a nonprofit, effectively deprives the holder of copyright protection. If evidence of actual or potential monetary loss were required, copyrights held by nonprofits would be essentially worthless. Religious, educational and other public interest institutions would suffer if their publications invested with an institution's reputation and goodwill could be freely appropriated by anyone. The statute by its terms is not limited to market effect but includes also "the effect of the use on the value of the copyrighted work." 17 U.S.C. 107(4) (emphasis added). As Sony states, " [e]ven copying for noncommercial purposes may impair the copyright holder's ability to obtain the rewards that Congress intended him to have." Sony, 464 U.S. at 450. Those rewards need not be limited to monetary rewards; compensation may take a variety of forms. Id. at 447 n.28 ("The copyright law does not require a copyright owner to charge a fee for the use of his works. . . . It is not the role of the courts to tell copyright holders the best way for them to exploit their copyrights"). WCG points out that those who respond to PCG's ads are the same people who would be interested in WCG's planned annotated version or any future republication of the original version. With an annotated MOA, WCG hopes to reach out to those familiar with Armstrong's teachings and those in the broader Christian community. PCG's distribution of its unauthorized version of MOA thus harms WCG's goodwill by diverting potential members and contributions from WCG. While the district court found that PCG's MOA and WCG's proposed annotated MOA "would not in any sense `compete' in the same market," undisputed evidence shows that individuals who received copies of MOA from PCG are present or could be potential adherents of WCG. MOA's value is as a marketing device; that is how PCG uses it and both PCG and WCG are engaged in evangelizing in the Christian community. PCG argues that WCG's failure to exploit MOA for ten years and its lack of a concrete plan to publish a new version show that "MOA has no economic value to the WCG that the PCG's dissemination of the work would adversely affect." We disagree. Even an author who had disavowed any intention to publish his work during his lifetime was entitled to protection of his copyright, first, because the relevant consideration was the "potential market" and, second, because he has the right to change his mind. See Salinger, 811 F.2d at 99. WCG explained that it ceased distribution because the Church's position on various doctrines had changed, continued distribution would offend cultural standards of social sensitivity, and dissemination would perpetuate what the Church considered ecclesiastical error. For those reasons, WCG planned an annotated edition of MOA.2 Finally, PCG argues that if WCG published an annotated version it would be so different as not to be competitive with PCG's MOA. The argument, aside from being speculative, misses the point. The fact remains that PCG has unfairly appropriated MOA in its entirety for the very purposes for which WCG created MOA. We have found no published case holding that fair use protected the verbatim copying, without criticism, of a written work in its entirety. As the 1967 House Report notes, the market factor "must almost always be judged in conjunction with the other three criteria. " H.R. REP. NO. 83, at 35 (1967). Judge Pierre N. Leval has written: When the secondary use does substantially interfere with the market for the copyrighted work, as was the case in [Harper & Row], this factor powerfully opposes a finding of fair use. But the inverse does not follow. The fact that the secondary use does not harm the market for the original gives no assurance that the secondary use is justified. Thus, notwithstanding the importance of the market factor, especially when the market is impaired by the secondary use, it should not overshadow the requirement of justification under the first factor, without which there can be no fair use.

The district court first concluded that downloading MP3 files does not transform the copyrighted work. Napster, 114 F. Supp. 2d at 912. This conclusion is supportable. Courts have been reluctant to find fair use when an original work is merely retransmitted in a different medium. See, e.g., Infinity Broadcast Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir. 1994) (concluding that retransmission of radio broadcast over telephone lines is not transformative); UMG Recordings, Inc. v. MP3. com, Inc., 92 F. Supp. 2d 349, 351 (S.D.N.Y.) (finding that reproduction of audio CD into MP3 format does not "transform" the work), certification denied, 2000 U.S. Dist. LEXIS 7439, 2000 WL 710056 (S.D.N.Y. June 1, 2000) ("Defendant's copyright infringement was clear, and the mere fact that it was clothed in the exotic webbing of the Internet does not disguise its illegality."). This "purpose and character "element also requires the district court to determine whether the allegedly infringing use is commercial or noncommercial. See Campbell, 510 U.S. at 584-85. A commercial use weighs against a finding of fair use but is not conclusive on the issue. Id. The district court determined that Napster users engage in commercial use of the copyrighted materials largely because (1) "a host user sending a file cannot be said to engage in a personal use when distributing that file to an anonymous requester "and (2) "Napster users get for free something they would ordinarily have to buy." Napster, 114 F. Supp. 2d at 912. The district court's findings are not clearly erroneous. Direct economic benefit is not required to demonstrate a commercial use. Rather, repeated and exploitative copying of copyrighted works, even if the copies are not offered for sale, may constitute a commercial use. See Worldwide Church of God v. Philadelphia Church of God, 227 F.3d 1110, 1118 (9th Cir. 2000) (stating that church that copied religious text for its members "unquestionably profited" from the unauthorized "distribution and use of [the text] without having to account to the copyright holder"); American Geophysical Union v. Texaco, Inc., 60 F.3d 913, 922 (2d Cir. 1994) (finding that researchers at for-profit laboratory gained indirect economic advantage by photocopying copyrighted scholarly articles). In the record before us, commercial use is demonstrated by a showing that repeated and exploitative unauthorized copies of copyrighted works were made to save the expense of purchasing authorized copies. See Worldwide Church, 227 F.3d at 1117-18; Sega Enters. Ltd. v. MAPHIA, 857 F. Supp. 679, 687 (N.D. Cal. 1994) (finding commercial use when individuals downloaded copies of video games "to avoid having to buy video game cartridges"); see also American Geophysical, 60 F.3d at 922. Plaintiffs made such a showing before the district court. We also note that the definition of a financially motivated transaction for the purposes of criminal copyright actions includes trading infringing copies of a work for other items, "including the receipt of other copyrighted works." See No Electronic Theft Act ("NET Act"), Pub. L. No. 105-147, 18 U.S.C. § 101 (defining "Financial Gain").

Works that are creative in nature are "closer to the core of intended copyright protection" than are more fact-based works. See Campbell, 510 U.S. at 586. The district court determined that plaintiffs' "copyrighted musical compositions and sound recordings are creative in nature . . .which cuts against a finding of fair use under the second factor." Napster, 114 F. Supp. 2d at 913. We find no error in the district court's conclusion.

While 'wholesale copying does not preclude fair use per se, 'copying an entire work 'militates against a finding of fair use.'" Worldwide Church, 227 F.3d at 1118 (quoting Hustler Magazine, Inc. v. Moral Majority, Inc., 796 F.2d 1148, 1155 (9th Cir. 1986)). The district court determined that Napster users engage in "wholesale copying" of copyrighted work because file transfer necessarily "involves copying the entirety of the copyrighted work." Napster, 114 F. Supp. 2d at 913. We agree. We note, however, that under certain circumstances, a court will conclude that a use is fair even when the protected work is copied in its entirety. See, e.g., Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 449-50, 78 L. Ed. 2d 574, 104 S. Ct. 774 (1984) (acknowledging that fair use of time-shifting necessarily involved making a full copy of a protected work).

Fair use, when properly applied, is limited to copying by others which does not materially impair the marketability of the work which is copied." Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 566-67, 85 L. Ed. 2d 588, 105 S. Ct. 2218 (1985)."The importance of this [fourth] factor will vary, not only with the amount of harm, but also with the relative strength of the showing on the other factors." Campbell, 510 U.S. at 591 n.21. The proof required to demonstrate present or future market harm varies with the purpose and character of the use: A challenge to a noncommercial use of a copy-righted work requires proof either that the particular use is harmful, or that if it should become wide-spread, it would adversely affect the potential market for the copyrighted work. … If the intended use is for commercial gain, that likelihood [of market harm] may be presumed. But if it is for a noncommercial purpose, the likelihood must be demonstrated. Sony, 464 U.S. at 451 (emphases added). Addressing this factor, the district court concluded that Napster harms the market in "at least" two ways: it reduces audio CD sales among college students and it "raises barriers to plaintiffs' entry into the market for the digital downloading of music." Napster, 114 F. Supp. 2d at 913. The district court relied on evidence plaintiffs submitted to show that Napster use harms the market for their copyrighted musical compositions and sound recordings. In a separate memorandum and order regarding the parties' objections to the expert reports, the district court examined each report, finding some more appropriate and probative than others. A&M Records, Inc. v. Napster, Inc., 114 F. Supp. 2d 896, 2000 WL 1170106 (N.D. Cal. 2000). Notably, plaintiffs' expert, Dr. E. Deborah Jay, conducted a survey (the "Jay Report") using a random sample of college and university students to track their reasons for using Napster and the impact Napster had on their music purchases. Id. at *2. The court recognized that the Jay Report focused on just one segment of the Napster user population and found "evidence of lost sales attributable to college use to be probative of irreparable harm for purposes of the preliminary injunction motion." 114 F. Supp. 2d at 923, Id. at *3. Plaintiffs also offered a study conducted by Michael Fine, Chief Executive Officer of Soundscan, (the "Fine Report") to determine the effect of online sharing of MP3 files in order to show irreparable harm. Fine found that online file sharing had resulted in a loss of "album" sales within college markets. After reviewing defendant's objections to the Fine Report and expressing some concerns regarding the methodology and findings, the district court refused to exclude the Fine Report insofar as plaintiffs offered it to show irreparable harm. Id. at *6. Plaintiffs' expert Dr. David J. Teece studied several issues ("Teece Report"), including whether plaintiffs had suffered or were likely to suffer harm in their existing and planned businesses due to Napster use. Id. Napster objected that the report had not undergone peer review. The district court noted that such reports generally are not subject to such scrutiny and overruled defendant's objections. Id. As for defendant's experts, plaintiffs objected to the report of Dr. Peter S. Fader, in which the expert concluded that Napster is beneficial to the music industry because MP3 music file-sharing stimulates more audio CD sales than it displaces. Id. at *7. The district court found problems in Dr. Fader's minimal role in overseeing the administration of the survey and the lack of objective data in his report. The court decided the generality of the report rendered it "of dubious reliability and value." The court did not exclude the report, however, but chose "not to rely on Fader's findings in determining the issues of fair use and irreparable harm." 114 F. Supp. 2d at 912, Id. at *8. The district court cited both the Jay and Fine Reports in support of its finding that Napster use harms the market for plaintiffs' copyrighted musical compositions and sound recordings by reducing CD sales among college students. The district court cited the Teece Report to show the harm Napster use caused in raising barriers to plaintiffs' entry into the market for digital downloading of music. Napster, 114 F. Supp. 2d at 910. The district court's careful consideration of defendant's objections to these reports and decision to rely on the reports for specific issues demonstrates a proper exercise of discretion in addition to a correct application of the fair use doctrine. Defendant has failed to show any basis for disturbing the district court's findings.
. . .
We, therefore, conclude that the district court made sound findings related to Napster's deleterious effect on the present and future digital download market. Moreover, lack of harm to an established market cannot deprive the copyright holder of the right to develop alternative markets for the works. See L.A. Times v. Free Republic, 2000 U.S. Dist. LEXIS 5669, 54 U.S.P.Q.2D (BNA) 1453, 1469-71 (C.D. Cal. 2000) (stating that online market for plaintiff newspapers' articles was harmed because plaintiffs demonstrated that "[defendants] are attempting to exploit the market for viewing their articles online"); see also UMG Recordings, 92 F. Supp. 2d at 352 (" Any allegedly positive impact of defendant's activities on plaintiffs' prior market in no way frees defendant to usurp a further market that directly derives from reproduction of the plaintiffs' copyrighted works."). Here, similar to L.A. Times and UMG Recordings, the record supports the district court's finding that the "record company plaintiffs have already expended considerable funds and effort to commence Internet sales and licensing for digital downloads." 114 F. Supp. 2d at 915. Having digital downloads available for free on the Napster system necessarily harms the copyright holders' attempts to charge for the same downloads.

Napster maintains that its identified uses of sampling and space-shifting were wrongly excluded as fair uses by the district court. a. Sampling Napster contends that its users download MP3 files to "sample" the music in order to decide whether to purchase the recording. Napster argues that the district court: (1) erred in concluding that sampling is a commercial use because it conflated a noncommercial use with a personal use; (2) erred in determining that sampling adversely affects the market for plaintiffs' copyrighted music, a requirement if the use is noncommercial; and (3) erroneously concluded that sampling is not a fair use because it determined that samplers may also engage in other infringing activity. The district court determined that sampling remains a commercial use even if some users eventually purchase the music. We find no error in the district court's determination. Plaintiffs have established that they are likely to succeed in proving that even authorized temporary downloading of individual songs for sampling purposes is commercial in nature. See Napster, 114 F. Supp. 2d at 913. The record supports a finding that free promotional downloads are highly regulated by the record company plaintiffs and that the companies collect royalties for song samples available on retail Internet sites. Id. Evidence relied on by the district court demonstrates that the free downloads provided by the record companies consist of thirty-to-sixty second samples or are full songs programmed to "time out," that is, exist only for a short time on the downloader's computer. Id. at 913-14. In comparison, Napster users download a full, free and permanent copy of the recording. Id. at 914-15. The determination by the district court as to the commercial purpose and character of sampling is not clearly erroneous. The district court further found that both the market for audio CDs and market for online distribution are adversely affected by Napster's service. As stated in our discussion of the district court's general fair use analysis: the court did not abuse its discretion when it found that, overall, Napster has an adverse impact on the audio CD and digital download markets. Contrary to Napster's assertion that the district court failed to specifically address the market impact of sampling, the district court determined that " [e]ven if the type of sampling supposedly done on Napster were a non-commercial use, plaintiffs have demonstrated a substantial likelihood that it would adversely affect the potential market for their copyrighted works if it became widespread." Napster, 114 F. Supp. 2d at 914. The record supports the district court's preliminary determinations that: (1) the more music that sampling users download, the less likely they are to eventually purchase the recordings on audio CD; and (2) even if the audio CD market is not harmed, Napster has adverse effects on the developing digital download market. Napster further argues that the district court erred in rejecting its evidence that the users' downloading of "samples" increases or tends to increase audio CD sales. The district court, however, correctly noted that "any potential enhancement of plaintiffs' sales . . . would not tip the fair use analysis conclusively in favor of defendant." Id. at 914. We agree that increased sales of copyrighted material attributable to unauthorized use should not deprive the copyright holder of the right to license the material. See Campbell, 510 U.S. at 591 n.21 ("Even favorable evidence, without more, is no guarantee of fairness. Judge Leval gives the example of the film producer's appropriation of a composer's previously unknown song that turns the song into a commercial success; the boon to the song does not make the film's simple copying fair."); see also L.A. Times, 54 U.S.P.Q.2d at 1471-72. Nor does positive impact in one market, here the audio CD market, deprive the copyright holder of the right to develop identified alternative markets, here the digital download market. See id. at 1469-71. We find no error in the district court's factual findings or abuse of discretion in the court's conclusion that plaintiffs will likely prevail in establishing that sampling does not constitute a fair use. b. Space-Shifting Napster also maintains that space-shifting is a fair use. Space-shifting occurs when a Napster user downloads MP3 music files in order to listen to music he already owns on audio CD. See id. at 915-16. Napster asserts that we have already held that space-shifting of musical compositions and sound recordings is a fair use. See Recording Indus. Ass'n of Am. v. Diamond Multimedia Sys., Inc., 180 F.3d 1072, 1079 (9th Cir. 1999) ("Rio [a portable MP3 player] merely makes copies in order to render portable, or 'space-shift,' those files that already reside on a user's hard drive. . . . Such copying is a paradigmatic noncommercial personal use."). See also generally Sony, 464 U.S. at 423 (holding that "time-shifting," where a video tape recorder owner records a television show for later viewing, is a fair use). We conclude that the district court did not err when it refused to apply the "shifting" analyses of Sony and Diamond. Both Diamond and Sony are inapposite because the methods of shifting in these cases did not also simultaneously involve distribution of the copyrighted material to the general public; the time or space-shifting of copyrighted material exposed the material only to the original user. In Diamond, for example, the copyrighted music was transferred from the user's computer hard drive to the user's portable MP3 player. So too Sony, where "the majority of VCR purchasers . . . did not distribute taped television broadcasts, but merely enjoyed them at home." Napster, 114 F. Supp. 2d at 913. Conversely, it is obvious that once a user lists a copy of music he already owns on the Napster system in order to access the music from another location, the song becomes "available to millions of other individuals," not just the original CD owner. See UMG Recordings, 92 F. Supp. 2d at 351-52 (finding space-shifting of MP3 files not a fair use even when previous ownership is demonstrated before a download is allowed); cf. Religious Tech. Ctr. v. Lerma, No. 95-1107A, 1996 WL 633131, at *6 (E.D. Va. Oct. 4, 1996) (suggesting that storing copyrighted material on computer disk for later review is not a fair use).

As the district court recognized, the most important component of the inquiry into the "purpose and character of the use" is the question whether the allegedly fair use was "transformative," i.e., whether the second use "adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Id. at 579. "The more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Id. We have previously held that despite the newsworthiness of LANS's riot videos, their mere rebroadcast was not in itself transformative. See, e.g., Reuters, 149 F.3d at 993 ("Although [Reuters's] service does have a news reporting purpose, its use of the works was not very transformative. Reuters copies footage and transmits it to news reporting organizations; Reuters does not explain the footage, edit the content of the footage, or include editorial comment."); L.A. News Serv. v. KCAL-TV Channel 9, 108 F.3d 1119, 1122 (9th Cir. 1997) ("Although KCAL apparently ran its own voice-over, it does not appear to have added anything new or transformative to what made the LANS work valuable -- a clear, visual recording of the beating itself."). [4] For the most part, it does not appear that Court TV's use of the video clip was transformative. Merely plucking the most visually arresting excerpt from LANS's nine minutes of footage cannot be said to have added anything new. See Folsom v. Marsh, 9 F. Cas. 342, 345 (C.C.D. Mass. 1841) (No. 4,901) (Story, J.) ("There must be real, substantial condensation of the materials, and intellectual labor and judgment bestowed thereon; and not merely the facile use of the scissors; or extracts of the essential parts, constituting the chief value of the original work."), cited with approval in Campbell, 510 U.S. at 578-79 . We concluded as much in KCAL-TV; where the defendant television station had provided its voiceover to "Beating of Reginald Denny" and aired only selected portions, 108 F.3d at 1122, 1123, but we concluded that it "did not appear to have added anything new or transformative." Id. at 1122. Here, simply extracting the clip of the Denny beating and juxtaposing it with a clip from Denny's testimony updates, but does not change, the purpose of depicting the attack on Denny -- its newsworthiness, which had been ebbing as the event grew more removed but which had rebounded as the trial of Williams and Watson hit the headlines. [5] However, the inclusion of the clip in the video montage that introduced the Prime Time Justice program, following editing for dramatic effect, has a better claim to be within the scope of "transformation." The development of the montage at least plausibly incorporates the element of creativity beyond mere republication, and it serves some purpose beyond newsworthiness. One additional consideration affecting the analysis of this factor is that Court TV's use was largely commercial, i.e., for the purpose of promoting its coverage rather than simply reporting, or reporting on, the news. That Court TV used the work with a profit-making motive is not dispositive, but is an element of the analysis. 17 U.S.C. § 107(1); Campbell, 510 U.S. at 584-85. However, the more commercially exploitative use to which Court TV put the clip was also the more transformative, which in turn reduces the importance of the commercial purpose. Specifically, the opening montage of "Prime Time Justice" was calculated to promote the program and to retain the attention of any channel-surfer who happened to see it. Indeed, as the program's regular introduction, it was used to promote the program even when the program did not cover the Williams trial. At the same time, though, the opening montage was the more "transformative" of the allegedly infringing uses, which in turn reduced the importance of the commerciality of the use. By contrast, the use of the clip in the coverage teasers was less transformative, meaning that a commercial purpose would take on greater weight -- and this use was more plausibly associated with news reporting, a favored purpose under the statute, than was the Prime Time Justice montage. Indeed, the journalistic event was coverage of the Williams and Wilson trial, in which LANS's video featured prominently. And because the newsworthy event was the trial rather than the beating, Court TV was not in direct competition with LANS. Cf. KCAL-TV, 108 F.3d at 1121. Although both the promotional use and the use in the montage have aspects that distinguish them from the pure infringement in Reuters or KCAL-TV, in neither instance can Court TV's commercial motive be completely discounted, as it would be for coverage of the courtroom when the copyrighted works were being introduced as evidence. On balance, the "purpose and character" factor weighs weakly in favor of fair use. B

[6] We have analyzed the "nature of the copyrighted work" factor extensively in prior LANS litigation. "The Denny beating tape is informational and factual and news; each characteristic strongly favors [the user]. Likewise the fact that the tape was published before its use by [the alleged infringer]. Although the Videotape is not without creative aspect in that it is the result of [Marika] Tur's skills with a camera, still this factor makes it a great deal easier to find fair use." KCAL-TV, 108 F.3d at 1122; accord Reuters, 149 F.3d at 994; see also Harper & Row, 471 U.S. at 563 ("The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy."); L.A. News Serv. v. Tullo, 973 F.2d 791, 798 (9th Cir. 1992). This factor clearly points toward fair use.

It is undisputed that the "amount" of "Beating of Reginald Denny" actually used by Court TV was quite small, both in absolute terms -- a few seconds at most -- and "in relation to the copyrighted work as a whole." The question is whether the amount of footage Court TV left on the cutting-room floor weighs in favor of fair use or against it. LANS presses heavily on our previous statements that although an alleged infringement may have used only a small portion of LANS's several minutes of footage, the portion used constituted the "heart" -- the most valuable and pertinent portion -- of the copyrighted material. See KCAL-TV, 108 F.3d at 1122 (quoting Tullo, 973 F.2d at 798) (internal quotation marks omitted); see also 4 Nimmer & Nimmer, supra, § 13.05[A][3], at 13-180 n.209 (citing cases). The same argument applies here to some degree. However, it is worth noting that the rebroadcast in this case was of only a few seconds' worth of footage. Although our opinion in KCAL-TV does not specify the amount of footage that the defendant station copied there, it appears that it was considerably more than the minor amount that Court TV used here. See KCAL-TV, 108 F.3d at 1120 (stating that there was "no dispute that KCAL used the heart of the tape"). Moreover, and more significantly, LANS asserted in the district court in its action against Reuters "that the core or 'heart' of the tape, which is the actual attack on Reginald Denny, lasts for only about forty-five seconds." L.A. News Serv. v. Reuters Television Int'l, Ltd., 942 F. Supp. 1265, 1273 (C.D. Cal. 1996), aff'd in part and rev'd in part on other grounds, 149 F.3d 987, 994-95 (9th Cir. 1998). LANS's current assertion that the single shot that Court TV employed was the "heart" of the tape seems less plausible in light of its earlier contention. See also Nimmer & Nimmer, supra, § 13.05[A][3], at 13-180 ("To avoid circular reasoning, the plaintiff manifestly should not be heard to argue that the defendant's copying of brief passages vouchsafes their qualitative significance." (footnote omitted)). But cf. Hustler Magazine, Inc. v. Moral Majority, Inc., 796 F.2d 1148, 1155 (9th Cir. 1986) ("A creative work does not deserve less copyright protection just because it is part of a composite work. Therefore, in this case, we view the Defendants as having copied an entire work."). Case law interpreting this factor, though of course faithful to the statute's direction that we are to consider the substantiality of the portion used "in relation to the copyrighted work as a whole," has also noted that the purpose and character of the use are relevant in evaluating the denominator. See, e.g., Kelly v. Arriba Soft Corp., 280 F.3d 934, 943 (9th Cir. 2002) ("If the secondary user only copies as much as is necessary for his or her intended use, then this factor will not weigh against him or her."). The frames that Court TV used for promotional purposes were closely tied to the subject matter it was promoting, i.e., the Williams and Watson trial. And by choosing the most recognizable frames, it eliminated the need to take more; viewers who did not recognize the caption "California v. Williams" required only a short glimpse at the footage to be reminded of just who Williams was and why they might be interested in his trial. [7] To be sure, selectivity can cut both ways: Court TV urges that it took only what was necessary, while LANS accuses it of having deliberately lifted the choicest morsels. Our case law admonishes that poaching only the most significant part of a work makes the use no fairer than wholesale copying, but such reasoning is not without limit; at some point the selective extraction of significant footage can no longer be characterized as lifting the "heart." We do not gainsay the importance of the frames that Court TV used -- if not the heart, they amount at least to a ventricle -- but we think that this factor weighs less in LANS's favor than in the previous cases involving the Denny video. Indeed, weighing the brevity of the portion copied against its significance, this factor appears neutral.

In evaluating the fourth factor, the effect of Court TV's unauthorized use on the potential market for licenses to rebroadcast the work, we "consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also 'whether unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market' for the original." Campbell, 510 U.S. at 590 (quoting 4 Nimmer & Nimmer, supra, § 13.05[A][4], at 13-102 .61 (1993) (footnote omitted)). [8] The transformative use of the clip in the "Prime Time Justice" montage was quite unlikely to affect the relevant market. See, e.g., Kelly, 280 F.3d at 943 ("A transformative work is less likely to have an adverse impact on the market of the original than a work that merely supersedes the copyrighted work."). [9] The rebroadcast of the clip to promote trial coverage, which (as we note above) was a less transformative use, is more problematic. LANS argues that at precisely that time interest in the riots, and in the footage that had come to symbolize them, had been renewed by the trial, and that Court TV therefore made its unauthorized use at a time when a genuine market existed for its Denny footage. However, the relevant question is whether the use of three-second clips of LANS's work -- that is, the uncompensated use, not merely an additional use in an already saturated market -- would substantially affect the market for licenses to show "Beating of Reginald Denny" in its entirety. It might indeed have had such an effect if the only reason, or even the most significant reason, to license the footage were to obtain the privilege of using the specific clip at issue here. However, LANS's position in the previous lawsuit was that a full forty-five seconds' worth of the video deserved to be called the "heart" of the work. Additionally, many -- though not all -- of LANS's licensees used the clip in the course of news and current affairs programming that more likely than not required more than just a few seconds of video. We are mindful of Campbell's admonishment that we must imagine the effect on the market if the allegedly fair use were universalized. However, Court TV operated in a significantly different market than did LANS. As we have noted, Court TV was not competing with LANS to show riot coverage, or even breaking news of the same general type; the courtroom setting is, after all, singularly unsuited to helicopter coverage. Moreover, this incident presented no apparent effort to evade licensing outright. See KCAL-TV, 108 F.3d at 1122-23 . This is not a case of unrestrained piracy. [10] Although we are somewhat concerned with the state of the record regarding this factor in particular, neither party has urged us to hold that there is a factual dispute. As fair use is a mixed question of fact and law, so long as the record is "sufficient to evaluate each of the statutory factors," we may reweigh on appeal the inferences to be drawn from that record. Harper & Row, 471 U.S. at 560. Having done so in the context of the market-effect factor, we conclude that it points in favor of fair use.

The Supreme Court has rejected the proposition that a commercial use of the copyrighted material ends the inquiry under this factor. Instead, [t]he central purpose of this investigation is to see . . . whether the new work merely supersede[s] the objects of the original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message; it asks, in other words, whether and to what extent the new work is transformative. The more transformative the new work, the less important the other factors, including commercialism, become. There is no dispute that Arriba operates its web site for commercial purposes and that Kelly's images were part of Arriba's search engine database. As the district court found, while such use of Kelly's images was commercial, it was more incidental and less exploitative in nature than more traditional types of commercial use. Arriba was neither using Kelly's images to directly promote its web site nor trying to profit by selling Kelly's images. Instead, Kelly's images were among thousands of images in Arriba's search engine data-base. Because the use of Kelly's images was not highly exploitative, the commercial nature of the use weighs only slightly against a finding of fair use. The second part of the inquiry as to this factor involves the transformative nature of the use. We must determine if Arriba's use of the images merely superseded the object of the originals or instead added a further purpose or different character. We find that Arriba's use of Kelly's images for its thumbnails was transformative. Although Arriba made exact replications of Kelly's images, the thumbnails were much smaller, lower-resolution images that served an entirely different function than Kelly's original images. Kelly's images are artistic works intended to inform and to engage the viewer in an aesthetic experience. His images are used to portray scenes from the American West in an aesthetic manner. Arriba's use of Kelly's images in the thumbnails is unrelated to any aesthetic purpose. Arriba's search engine functions as a tool to help index and improve access to images on the internet and their related web sites. In fact, users are unlikely to enlarge the thumbnails and use them for artistic purposes because the thumbnails are of much lower-resolution than the originals; any enlargement results in a significant loss of clarity of the image, making them inappropriate as display material. Kelly asserts that because Arriba reproduced his exact images and added nothing to them, Arriba's use cannot be transformative. Courts have been reluctant to find fair use when an original work is merely retransmitted in a different medium. Those cases are inapposite, however, because the resulting use of the copyrighted work in those cases was the same as the original use. For instance, reproducing music CDs in computer MP3 format does not change the fact that both formats are used for entertainment purposes. Likewise, reproducing news footage into a different format does not change the ultimate purpose of informing the public about current affairs. Even in Infinity Broadcast Corp. v. Kirkwood, where the retransmission of radio broadcasts over telephone lines was for the purpose of allowing advertisers and radio stations to check on the broadcast of commercials or on-air talent, there was nothing preventing listeners from subscribing to the service for entertainment purposes. Even though the intended purpose of the retransmission may have been different from the purpose of the original transmission, the result was that people could use both types of transmissions for the same purpose. This case involves more than merely a retransmission of Kelly's images in a different medium. Arriba's use of the images serves a different function than Kelly's use -- improving access to information on the internet versus artistic expression. Furthermore, it would be unlikely that anyone would use Arriba's thumbnails for illustrative or aesthetic purposes because enlarging them sacrifices their clarity. Because Arriba's use is not superseding Kelly's use but, rather, has created a different purpose for the images, Arriba's use is transformative. Comparing this case to two recent cases in the Ninth and First Circuits reemphasizes the functionality distinction. In Worldwide Church of God v. Philadelphia Church of God, Inc., we held that copying a religious book to create a new book for use by a different church was not transformative. The second church's use of the book was merely to make use of the same book for another church audience. The court noted that "where the use is for the same intrinsic purpose as [the copyright holder's] . . . such use seriously weakens a claimed fair use." On the other hand, in Nunez v. Caribbean International News Corp., the First Circuit found that copying a photo-graph that was intended to be used in a modeling portfolio and using it instead in a news article was a transformative use. By putting a copy of the photograph in the newspaper, the work was transformed into news, creating a new meaning or purpose for the work. The use of Kelly's images in Arriba's search engine is more analogous to the situation in Nunez because Arriba has created a new purpose for the images and is not simply superseding Kelly's purpose. The Copyright Act was intended to promote creativity, thereby benefitting the artist and the public alike. To preserve the potential future use of artistic works for purposes of teaching, research, criticism, and news reporting, Congress created the fair use exception. Arriba's use of Kelly's images pro-motes the goals of the Copyright Act and the fair use exception. The thumbnails do not stifle artistic creativity because they are not used for illustrative or artistic purposes and there-fore do not supplant the need for the originals. In addition, they benefit the public by enhancing information-gathering techniques on the internet. In Sony Computer Entertainment America, Inc. v. Bleem, we held that when Bleem copied "screen shots" from Sony computer games and used them in its own advertising, it was a fair use. In finding that the first factor weighed in favor of Bleem, we noted that "comparative advertising redounds greatly to the purchasing public's benefit with very little corresponding loss to the integrity of Sony's copyrighted material." Similarly, this first factor weighs in favor of Arriba due to the public benefit of the search engine and the minimal loss of integrity to Kelly's images.

"Works that are creative in nature are closer to the core of intended copyright protection than are more fact-based works." Photographs that are meant to be viewed by the public for informative and aesthetic purposes, such as Kelly's, are generally creative in nature. The fact that a work is published or unpublished also is a critical element of its nature. Published works are more likely to qualify as fair use because the first appearance of the artist's expression has already occurred. Kelly's images appeared on the internet before Arriba used them in its search image. When considering both of these elements, we find that this factor weighs only slightly in favor of Kelly.

While wholesale copying does not preclude fair use per se, copying an entire work militates against a finding of fair use." However, the extent of permissible copying varies with the purpose and character of the use. If the secondary user only copies as much as is necessary for his or her intended use, then this factor will not weigh against him or her. This factor neither weighs for nor against either party because, although Arriba did copy each of Kelly's images as a whole, it was reasonable to do so in light of Arriba's use of the images. It was necessary for Arriba to copy the entire image to allow users to recognize the image and decide whether to pursue more information about the image or the originating web site. If Arriba only copied part of the image, it would be more difficult to identify it, thereby reducing the usefulness of the visual search engine.

This last factor requires courts to consider "not only the extent of market harm caused by the particular actions of the alleged infringer, but also 'whether unrestricted and wide-spread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market for the original.' " A transformative work is less likely to have an adverse impact on the market of the original than a work that merely supersedes the copyrighted work. Kelly's images are related to several potential markets. One purpose of the photographs is to attract internet users to his web site, where he sells advertising space as well as books and travel packages. In addition, Kelly could sell or license his photographs to other web sites or to a stock photo data-base, which then could offer the images to its customers. Arriba's use of Kelly's images in its thumbnails does not harm the market for Kelly's images or the value of his images. By showing the thumbnails on its results page when users entered terms related to Kelly's images, the search engine would guide users to Kelly's web site rather than away from it. Even if users were more interested in the image itself rather than the information on the web page, they would still have to go to Kelly's site to see the full-sized image. The thumbnails would not be a substitute for the full-sized images because the thumbnails lose their clarity when enlarged. If a user wanted to view or download a quality image, he or she would have to visit Kelly's web site. This would hold true whether the thumbnails are solely in Arriba's database or are more widespread and found in other search engine databases. Arriba's use of Kelly's images also would not harm Kelly's ability to sell or license his full-sized images. Arriba does not sell or license its thumbnails to other parties. Anyone who downloaded the thumbnails would not be successful selling full-sized images enlarged from the thumbnails because of the low resolution of the thumbnails. There would be no way to view, create, or sell a clear, full-sized image without going to Kelly's web sites. Therefore, Arriba's creation and use of the thumbnails does not harm the market for or value of Kelly's images. This factor weighs in favor of Arriba. Having considered the four fair use factors and found that two weigh in favor of Arriba, one is neutral, and one weighs slightly in favor of Kelly, we conclude that Arriba's use of Kelly's images as thumbnails in its search engine is a fair use. B. As mentioned above, the district court granted summary judgment to Arriba as to the full-size images as well. How-ever, because the court broadened the scope of both the par-ties' motions for partial summary judgment and Arriba's concession on the prima facie case, we must reverse this portion of the court's opinion. With limited exceptions that do not apply here, a district court may not grant summary judgment on a claim when the party has not requested it. The parties did not move for summary judgment as to copyright infringement of the full-size images. Further, Arriba had no opportunity to contest the prima facie case for infringement as to the full-size images. Accordingly, we reverse this portion of the district court's opinion and remand for further proceedings.

We first address the purpose and character of Passport's use of Plaintiffs' copyrighted materials. Although not controlling, the fact that a new use is commercial as opposed to non-profit weighs against a finding of fair use. Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 562, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985). And the degree to which the new user exploits the copyright for commercial gain — as opposed to incidental use as part of a commercial enterprise — affects the weight we afford commercial nature as a factor. See e.g., Kelly 336 F.3d at 818; see also Harper & Row, 471 U.S. at 562, 105 S. Ct. 2218("The crux of the profit/nonprofit distinction is not whether the sole motive of the use is monetary gain but whether the user stands to profit from exploitation of the copyrighted material without paying the customary price."). More importantly for the first fair-use factor, however, is the "transformative" nature of the new work. Campbell, 510 U.S. at 579, 114 S. Ct. 1164; CBS Broadcasting, 305 F.3d at 938. Specifically, we ask "whether the new work... merely supersedes the objects of the original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message...." Campbell, 510 U.S. at 579, 114 S. Ct. 1164 (citation and quotation marks omitted). The more transformative a new work, the less significant other inquiries, such as commercialism, become. Id. Two district courts have found that the use of film clips in biographies is transformative. In Monster Communications, Inc. v. Turner Broadcasting System, Inc., 935 F. Supp. 490, 491 (S.D.N.Y. 1996), the district court considered whether a movie biography about Muhammad Ali violated the plaintiff's copyrights in video footage that was used in the boxer's biography for less then two minutes. The court found that the biography, while commercial, "constitutes a combination of comment, criticism, scholarship and research" concerning "a figure of legitimate public concern" and thus the purpose and character of the biography weighed in favor of fair use. Id. at 493-94 (citation and quotation marks omitted). In Hofheinz v. A & E Television Networks, 146 F. Supp. 2d 442, 444 (S.D.N.Y. 2001), the court considered whether A & E's use of copyrighted film clips for a biography of actor Peter Graves was fair use. The court found that the biography was transformative because use of a movie trailer clip for a "B" movie "was not shown to recreate the creative expression reposing in plaintiff's film." Id. at 446. The biography narrator introduced the movie clip as outdated and "campy." Id. at 444. Its purpose was to "enabl [e] the viewer to understand the actor's modest beginnings in the film business." Id. at 446-47. The district court below found that the purpose and character of The Definitive Elvis will likely weigh against a finding of fair use. We cannot say, based on this record, that the district court abused its discretion. irst, Passport's use, while a biography, is clearly commercial in nature. But more significantly, Passport seeks to profit directly from the copyrights it uses without a license. One of the most salient selling points on the box of The Definitive Elvis is that "Every Film and Television Appearance is represented." Passport is not advertising a scholarly critique or historical analysis, but instead seeks to profit at least in part from the inherent entertainment value of Elvis' appearances on such shows as The Steve Allen Show, The Ed Sullivan Show, and The 1968 Comeback Special. Passport's claim that this is scholarly research containing biographical comments on the life of Elvis is not dispositive of the fair use inquiry. Second, Passport's use of Plaintiffs' copyrights is not consistently transformative. True, Passport's use of many of the television clips is transformative because the clips play for only a few seconds and are used for reference purposes while a narrator talks over them or interviewees explain their context in Elvis' career. But voice-overs do not necessarily transform a work. See L.A. News Serv. v. KCAL-TV Channel 9, 108 F.3d 1119, 1122 (9th Cir. 1997). "`There must be real, substantial condensation of the materials ... and not merely the facile use of scissors; or extracts of the essential parts, constituting the chief value of the original work.'" CBS Broad., Inc., 305 F.3d at 939(quoting Folsom v. Marsh, 9 F. Cas. 342, 345 (C.C.D. Mass. 1841) (Story, J.)). It would be impossible to produce a biography of Elvis without showing some of his most famous television appearances for reference purposes. But some of the clips are played without much interruption, if any. The purpose of showing these clips likely goes beyond merely making a reference for a biography, but instead serves the same intrinsic entertainment value that is protected by Plaintiffs' copyrights. This first factor is a close issue. Courts have described new works as "transformative" when the works use copyrighted material for purposes distinct from the purpose of the original material. Here, Passport's use of many of the television clips is transformative because they are cited as historical reference points in the life of a remarkable entertainer. The Definitive Elvis "nature as a biography transforms the purpose of showing these clips from pure entertainment to telling part of the story of Elvis" life. But many of the film clips seem to be used in excess of this benign purpose, and instead are simply rebroadcast for entertainment purposes that Plaintiffs rightfully own. This comes as no surprise to the viewer since The Definitive Elvis advertises as much on its external packaging. We need not decide how we would resolve this factor were we to review it de novo. For our inquiry is simply whether the district court abused its discretion. The district court's decision that the first factor weighs against fair use was not based on an erroneous legal standard or clearly erroneous factual finding. See A & M Records, Inc., 239 F.3d at 1015.

We next examine the nature of Plaintiffs' copyrighted works. "The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy." Harper & Row, 471 U.S. at 563, 105 S. Ct. 2218. In other words, "this factor calls for recognition that some works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied." Campbell, 510 U.S. at 586, 114 S. Ct. 1164. Additionally, published works are more likely to qualify for fair use by subsequent users. Kelly, 336 F.3d at 820. For example, works such as original songs, motion pictures, and photographs taken for aesthetic purposes, are creative in nature and thus fit squarely within the core of copyright protection. See Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 455 n. 40, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984); Kelly, 336 F.3d at 820. But works such as news broadcasts and news video footage are more factual in nature and thus are more conducive to fair use. See Sony, 464 U.S. at 455 n. 40, 104 S. Ct. 774; KCAL-TV, 108 F.3d at 1122. The district court found that in total this factor weighed in Plaintiffs' favor. We cannot say that the district court abused its discretion in reaching that conclusion.

The third factor is the amount and substantiality of the portion used in relation to the copyrighted work as a whole. This factor evaluates both the quantity of the work taken and the quality and importance of the portion taken. Campbell, 510 U.S. at 586, 114 S. Ct. 1164. Regarding the quantity, copying "may not be excused merely because it is insubstantial with respect to the infringing work." Harper & Row, 471 U.S. at 565, 105 S. Ct. 2218(emphasis in original). But if the amount used is substantial with respect to the infringing work, it is evidence of the value of the copyrighted work. Id. Regarding the qualitative nature of the work used, we look to see whether "the heart" of the copyrighted work is taken — in other words, whether the portion taken is the "most likely to be newsworthy and important in licensing serialization." Campbell, 510 U.S. at 586, 114 S. Ct. 1164; see also CBS Broad., 305 F.3d at 941. Finally, if the new user only copies as much as necessary for his or her intended use, this factor will not weigh against the new user. Kelly, 336 F.3d at 820-21. The district court found that this factor also weighs in Plaintiffs' favor. This conclusion was not an abuse of discretion. Passport's use of clips from television appearances, although in most cases of short duration, were repeated numerous times throughout the tapes. While using a small number of clips to reference an event for biographical purposes seems fair, using a clip over and over will likely no longer serve a biographical purpose. Additionally, some of the clips were not short in length. Passport's use of Elvis' appearance on The Steve Allen Show plays for over a minute and many more clips play for more than just a few seconds. Additionally, although the clips are relatively short when compared to the entire shows that are copyrighted, they are in many instances the heart of the work. What makes these copyrighted works valuable is Elvis' appearance on the shows, in many cases singing the most familiar passages of his most popular songs. Plaintiffs are in the business of licensing these copyrights. Taking key portions extracts the most valuable part of Plaintiffs' copyrighted works. With respect to the photographs, the entire picture is often used. The music, admittedly, is usually played only for a few seconds. But when we consider all these facts together, we cannot say that the district court abused its discretion in finding that this factor weighed in Plaintiffs' favor.

The last, and "undoubtedly the single most important" of all the factors, is the effect the use will have on the potential market for and value of the copyrighted works. Harper & Row, 471 U.S. at 566, 105 S. Ct. 2218. We must "consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also whether unrestricted and widespread conduct of the sort engaged in by the defendant ... would result in a substantially adverse impact on the potential market for the original." Campbell, 510 U.S. at 590, 114 S. Ct. 1164(quotation marks omitted). The more transformative the new work, the less likely the new work's use of copyrighted materials will affect the market for the materials. See CBS Broad., 305 F.3d at 941. Finally, if the purpose of the new work is commercial in nature, "the likelihood [of market harm] may be presumed." A & M Records, 239 F.3d at 1016(quoting Sony, 464 U.S. at 451, 104 S. Ct. 774). The district court found that Passport's use of Plaintiffs' copyrighted materials likely does affect the market for those materials. This conclusion was not clearly erroneous. First, Passport's use is commercial in nature, and thus we can assume market harm. See id. Second, Passport has expressly advertised that The Definitive Elvis contains the television appearances for which Plaintiffs normally charge a licensing fee. If this type of use became wide-spread, it would likely undermine the market for selling Plaintiffs' copyrighted material. This conclusion, however, does not apply to the music and still photographs. It seems unlikely that someone in the market for these materials would purchase The Definitive Elvis instead of a properly licensed product. Third, Passport's use of the television appearances was, in some instances, not transformative, and therefore these uses are likely to affect the market because they serve the same purpose as Plaintiffs' original works. We do not think this factor weighs strongly in either side's favor. But, for the reasons stated above that support the district court's decision, we cannot say that the district court abused its discretion in analyzing this factor. Furthermore, because we do not see any legal error or clear error in the district court's factual findings underlying any of the fair-use factors, we hold that the district court did not abuse its discretion in granting the preliminary injunction.

A. Purpose and Character of Use The "purpose and character of use" factor in the fair use inquiry asks "to what extent the new work is transformative" and does not simply "supplant[]" the original work and whether the work's purpose was for- or not-for-profit. Campbell, 510 U.S. at 579, 584. A work must add "something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Id. at 579. The Supreme Court has recognized that parodic works, like other works that comment and criticize, are by their nature often sufficiently transformative to fit clearly under the fair use exception. Id. (recognizing that parody "has an obvious claim to transformative value"). In our circuit, a "parodist is permitted a fair use of a copy-righted work if it takes no more than is necessary to 'recall' or 'conjure up' the object of his parody." Dr. Seuss, 109 F.3d at 1400. A parodic work, however, like other potential fair uses, has to "work its way through the relevant factors, and be judged case by case, in light of the ends of copyright law." Campbell, 510 U.S. at 581. "The threshold question [in the analysis of this first factor] . . . is whether a parodic character may reasonably be perceived." Id. at 582. See also Dr. Seuss, 109 F.3d at 1400. Mattel argues that the district court erred in finding parody because a reasonable jury could conclude that Forsythe's works do not parody Mattel's Barbie. In support of this argument, Mattel offered into evidence a survey in which they presented individuals from the general public in a shopping mall with color photocopies of Forsythe's photographs and asked them what meaning they perceived. Relying on this survey, Mattel asserts that only some individuals may perceive parodic character. The issue of whether a work is a parody is a question of law, not a matter of public majority opinion. See Campbell, 510 U.S. at 582-83; Dr. Seuss, 109 F.3d at 1400-01 ("Unless the plaintiff's copyrighted work is at least in part the target of the defendant's satire, then the defendant's work is not a 'parody' in the legal sense. . . ." (emphasis added)). Forsythe correctly points out that Mattel presents no case law in support of its contention that the parodic nature of a defendant's work should be assessed using surveys and opinion testimony. Forsythe is further correct that every court to address the issue whether a defendant's work qualifies as a parody has treated this question as one of law to be decided by the court. E.g., Campbell, 510 U.S. at 582-83; Leibovitz v. Paramount Pictures Corp., 137 F.3d 109, 114-15 (2d Cir. 1998); Dr. Seuss, 109 F.3d at 1400-01. We decline to consider Mattel's survey in assessing whether Forsythe's work can be reasonably perceived as a parody. Parody is an objectively defined rhetorical device. Further, because parody is "a form of social and literary criticism," it has "socially significant value as free speech under the First Amendment." Dr. Seuss, 109 F.3d at 1400. While individuals may disagree on the success or extent of a parody, parodic elements in a work will often justify fair use protection. See, e.g., Yankee Publ'g, Inc. v. News Am. Publ'g, Inc., 809 F. Supp. 267, 280 (S.D.N.Y. 1992) ("First Amendment protections do not apply only to those who speak clearly, whose jokes are funny, and whose parodies succeed."). Use of surveys in assessing parody would allow majorities to determine the parodic nature of a work and possibly silence artistic creativity. Allowing majorities to determine whether a work is a parody would be greatly at odds with the purpose of the fair use exception and the Copyright Act. See generally Campbell, 510 U.S. at 583. A parody is a "literary or artistic work that imitates the characteristic style of an author or a work for comic effect or ridicule." Id. at 580 (quoting AMERICAN HERITAGE DICTIONARY 1317 (3d. 1992)). For the purposes of copyright law, a parodist may claim fair use where he or she uses some of the "elements of a prior author's composition to create a new one that, at least in part, comments on that author's works." Id. The original work need not be the sole subject of the parody; the parody "may loosely target an original" as long as the parody "reasonably could be perceived as commenting on the original or criticizing it, to some degree." Id. at 580-81, 583. That a parody is in bad taste is not relevant to whether it constitutes fair use; "it would be a dangerous undertaking for persons trained only to the law to constitute themselves final judges of the worth of [a work]." Id. at 582-83 (quoting Bleistein v. Donaldson Lithographing Co., 188 U.S. 239, 251, 47 L. Ed. 460, 23 S. Ct. 298, 1903 Dec. Comm'r Pat. 650 (1903)). In assessing whether Forsythe's photographs parody Barbie, Mattel urges us to ignore context -- both the social context of Forsythe's work and the actual context in which Mattel's copyrighted works are placed in Forsythe's photographs. However, "in parody, as in news reporting, context is everything." 510 U.S. at 588 (citations omitted). We conclude that Forsythe's work may reasonably be perceived as a parody of Barbie. Mattel, through impressive marketing, has established Barbie as "the ideal American woman" and a "symbol of American girlhood" for many. Mattel, Inc. v. MCA Records, Inc. ("MCA"), 296 F.3d 894, 898 (9th Cir. 2002), cert. denied, 537 U.S. 1171, 123 S. Ct. 993, 154 L. Ed. 2d 912 (2003). As abundantly evidenced in the record, Mattel's advertisements show these plastic dolls dressed in various outfits, leading glamorous lifestyles and engaged in exciting activities. To sell its product, Mattel uses associations of beauty, wealth, and glamour. Forsythe turns this image on its head, so to speak, by displaying carefully positioned, nude, and sometimes frazzled looking Barbies in often ridiculous and apparently dangerous situations. His lighting, background, props, and camera angles all serve to create a context for Mattel's copyrighted work that transform Barbie's meaning. Forsythe presents the viewer with a different set of associations and a different context for this plastic figure. In some of Forsythe's photos, Barbie is about to be destroyed or harmed by domestic life in the form of kitchen appliances, yet continues displaying her well-known smile, disturbingly oblivious to her predicament. As portrayed in some of Forsythe's photographs, the appliances are substantial and overwhelming, while Barbie looks defenseless. In other photographs, Forsythe conveys a sexualized perspective of Barbie by showing the nude doll in sexually suggestive contexts. It is not difficult to see the commentary that Forsythe intended or the harm that he perceived in Barbie's influence on gender roles and the position of women in society. However one may feel about his message -- whether he is wrong or right, whether his methods are powerful or banal -- his photographs parody Barbie and everything Mattel's doll has come to signify. Undoubtedly, one could make similar statements through other means about society, gender roles, sexuality, and perhaps even social class. But Barbie, and all the associations she has acquired through Mattel's impressive marketing success, conveys these messages in a particular way that is ripe for social comment. Parody emerges from this "joinder of reference and ridicule." Campbell, 510 U.S. at 583; cf. Dr. Seuss, 109 F.3d at 1401 (holding that defendants who wrote a poem titled "Cat NOT in the HAT" about the O.J. Simpson trial were not parodying Dr. Suess' original work because the stanzas had "no critical bearing on the substance or style of" the original). By developing and transforming associations with Mattel's Barbie doll, Forsythe has created the sort of social criticism and parodic speech protected by the First Amendment and promoted by the Copyright Act. We find that this factor weighs heavily in favor of Forsythe. Another element of the first factor analysis is whether the work's "purpose" was commercial or had a non-profit aim. Campbell, 510 U.S. at 584. Clearly, Forsythe had a commercial expectation and presumably hoped to find a market for his art. However, as the Supreme Court noted in Campbell, even works involving comment and criticism "are generally conducted for profit in this country." Id. (quoting Harper & Row, 471 U.S. at 592.) On balance, Forsythe's commercial expectation does not weigh much against him. Given the extremely transformative nature and parodic quality of Forsythe's work, its commercial qualities become less important. 510 U.S. at 579 (recognizing that the more "transformative the new work, the less will be the significance of the other factors").

B. Nature of the copyrighted work The second factor in the fair use analysis "recognizes that creative works are 'closer to the core of intended copyright protection' than informational and functional works." Dr. Seuss, 109 F.3d at 1402 (quoting Campbell, 510 U.S. at 586). Mattel's copyrighted Barbie figure and face can fairly be said to be a creative work. However, the creativity of Mattel's copyrighted Barbie is typical of cases where there are infringing parodies. Campbell, 510 U.S. at 586 ("Parodies almost invariably copy publicly known, expressive works."). As we have recognized in the past, "this [nature of the copyrighted work] factor typically has not been terribly significant in the overall fair use balancing." Dr. Seuss, 109 F.3d at 1402. In any event, it may weigh slightly against Forsythe.

C. Amount and substantiality of the portion used. The third factor in the fair use analysis asks whether " 'the amount and substantiality of the portion used in relation to the copyrighted work as a whole,' are reasonable in relation to the purpose of copying." Id. (quoting 17 U.S.C. § 107(3)). We assess the "persuasiveness of a parodist's justification for the particular copying done," recognizing that the "extent of permissible copying varies with the purpose and character of the use." Campbell, 510 U.S. at 586-87. Mattel argues that Forsythe used the entirety of its copyrighted work and that this factor weighs against him. Mattel contends that Forsythe could have used less of the Barbie figure by, for example, limiting his photos to the Barbie heads. First, Forsythe did not simply copy the work "verbatim" with "little added or changed." Id. at 587-88. A verbatim copy of Barbie would be an exact three dimensional reproduction of the doll. Forsythe did not display the entire Barbie head and body in his photographs. Parts of the Barbie figure are obscured or omitted depending on the angle at which the photos were taken and whether other objects obstructed a view of the Barbie figure. Second, Mattel's argument that Forsythe could have taken a lesser portion of its work attempts to benefit from the somewhat unique nature of the copyrighted work in this case. Copyright infringement actions generally involve songs, video, or written works. See, e.g., Elvis Presley Enters., Inc. v. Passport Video, 349 F.3d 622 (9th Cir. 2003) (use of copy-righted Elvis Presley-related video clips, photographs, and music); Los Angeles News Serv., 305 F.3d at 924 (use of a few seconds of a copyrighted video footage by a news service); Worldwide Church of God v. Phila. Church of God, Inc., 227 F.3d 1110 (9th Cir. 2000) (reproduction and distribution by nonprofit organization of an entire copyrighted work), cert. denied, 532 U.S. 958, 149 L. Ed. 2d 373, 121 S. Ct. 1486 (2001); Dr. Seuss, 109 F.3d at 1394 (use of Dr. Suess's "Cat in the Hat" format in written work about the O.J. Simpson trial). Because parts of these works are naturally severable, the new work can easily choose portions of the original work and add to it. Here because the copyrighted material is a doll design and the infringing work is a photograph containing that doll, Forsythe, short of severing the doll, must add to it by creating a context around it and capturing that context in a photograph. For our purposes, Forsythe's use is no different from that of a parodist taking a basic melody and adding elements that transform the work. See Campbell, 510 U.S. at 589 (noting that 2 Live Crew's rendition of "Pretty Woman" did not approach verbatim copying because, even though 2 Live Crew may have taken the most recognizable portion of the work, it had added "scraper" noises and overlays to the music). In both Forsythe's use of the entire doll and his use of dismembered parts of the doll, portions of the old work are incorporated into the new work but emerge imbued with a different character. Moreover, Forsythe was justified in the amount of Mattel's copyrighted work that he used in his photographs. Mattel's argument that Forsythe could have used a lesser portion of the Barbie doll is completely without merit and would lead to absurd results. We do not require parodic works to take the absolute minimum amount of the copyrighted work possible. As the Supreme Court stated in Campbell, "once enough has been taken to assure identification, how much more is reasonable will depend, say, on the extent to which the [work's] overriding purpose and character is to parody the original or, in contrast, the likelihood that the parody may serve as a market substitute for the original." Id. at 587. We conclude that the extent of Forsythe's copying of the Barbie figure and head was justifiable in light of his parodic purpose and medium used. This factor also weighs in his favor.

D. Effect of the use upon potential market The fourth factor asks whether actual market harm resulted from the defendant's use of plaintiff's protected material and whether "unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market" for the original or its derivatives. Id. at 590 (quoting 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.05[A](4), at 13-102.61 (1993)). This inquiry attempts to strike a balance between the benefit the public will derive if the use is permitted and the personal gain the copyright owner will receive if the use is denied. The less adverse effect that an alleged infringing use has on the copyright owner's expectation of gain, the less public benefit need be shown to justify the use. Dr. Seuss, 109 F.3d at 1403 (quoting MCA, Inc. v. Wilson, 677 F.2d 180, 183 (2d Cir. 1981)). Mattel argues that Forsythe's work could lead to market harm by impairing the value of Barbie itself, Barbie derivatives, and licenses for use of the Barbie name and/or likeness to non-Mattel entities. Because of the parodic nature of Forsythe's work, however, it is highly unlikely that it will substitute for products in Mattel's markets or the markets of Mattel's licensees. In Campbell, the Court clearly stated, "as to parody pure and simple, it is more likely that the new work will not affect the market for the original in a way cognizable under this factor." Campbell, 510 U.S. at 591. Nor is it likely that Mattel would license an artist to create a work that is so critical of Barbie. "The unlikelihood that creators of imaginative works will license critical reviews or lampoons of their own productions removes such uses from the very notion of a potential licensing market." Id. at 592. As to Mattel's claim that Forsythe has impaired Barbie's value, this fourth factor does not recognize a decrease in value of a copyrighted work that may result from a particularly powerful critical work. Id. at 593 ("The fact that a parody may impair the market for derivative uses by the very effectiveness of its critical commentary is no more relevant under copyright than the like threat to the original market . . . ."). We recognize, however, that critical works may have another dimension beyond their critical aspects that may have effects on potential markets for the copyrighted work. Id. at 592 (recognizing that the new work "may have a more complex character, with effects not only in the arena of criticism but also in protectable markets for derivative works"). Thus, we look more generally, not only to the critical aspects of a work, but to the type of work itself in determining market harm. Id. at 593 (looking beyond the critical aspect of 2 Live Crew's rap rendition of "Pretty Woman" to the derivative market for rap music). Given the nature of Forsythe's photographs, we decline Mattel's invitation to look to the licensing market for art in general. Forsythe's photographs depict nude and often sexualized figures, a category of artistic photography that Mattel is highly unlikely to license. "The existence of this potential market cannot be presumed." Lewis Galoob Toys, Inc. v. Nintendo of Am., Inc., 964 F.2d 965, 972 (9th Cir. 1992), cert. denied, 507 U.S. 985, 123 L. Ed. 2d 149, 113 S. Ct. 1582 (1993). In a case almost identical to this one, Mattel, Inc. v. Pitt ("Pitt"), 229 F. Supp. 2d 315, 321-22 (S.D.N.Y. 2002), the Southern District Court of New York found no danger of potential market harm to derivative uses. In Pitt, Mattel brought a copyright infringement suit against Susanne Pitt, an artist who sold and designed a line of figures called "Dungeon Dolls." These dolls were essentially Barbie dolls, physically altered, clothed in sadomasochistic attire, and placed in contexts with like themes. Id. Having found the works sufficiently transformative, the Pitt court concluded that potential market harm was improbable because Mattel was unlikely to develop or license others to develop a product in the "adult" doll market. Id. at 324. Forsythe's work could only reasonably substitute for a work in the market for adult-oriented artistic photographs of Barbie. We think it safe to assume that Mattel will not enter such a market or license others to do so. As the Court noted in Campbell, "the market for potential derivative uses includes only those that creators of original works would in general develop or license others to develop." 510 U.S. at 592. Finally, the public benefit in allowing artistic creativity and social criticism to flourish is great. The fair use exception recognizes this important limitation on the rights of the owners of copyrights. No doubt, Mattel would be less likely to grant a license to an artist that intends to create art that criticizes and reflects negatively on Barbie's image. It is not in the public's interest to allow Mattel complete control over the kinds of artistic works that use Barbie as a reference for criticism and comment.

Generally speaking, "lock-out" codes fall on the functional-idea rather than the original-expression side of the copyright line. Manufacturers of interoperable devices such as computers and software, game consoles and video games, printers and toner cartridges, or automobiles and replacement parts may employ a security system to bar the use of unauthorized components. To "unlock" and permit operation of the primary device (i.e., the computer, the game console, the printer, the car), the component must contain either a certain code sequence or be able to respond appropriately to an authentication process. To the extent compatibility requires that a particular code sequence be included in the component device to permit its use, the merger and scènes à faire doctrines generally preclude the code sequence from obtaining copyright protection. See Sega Enters., 977 F.2d at 1524 ("When specific instructions, even though previously copyrighted, are the only and essential means of accomplishing a given task, their later use by another will not amount to infringement.") (quoting National Commission on New Technological Uses of Copyrighted Works, Final Report 20 (1979)) (emphasis added); Atari Games Corp. v. Nintendo of Am., Inc., Nos. 88-4805 & 89-0027, 1993 WL 207548, at *1 (N.D. Cal. May 18, 1993) ("Atari III") ("Program code that is strictly necessary to achieve current compatibility presents a merger problem, almost by definition, and is thus excluded from the scope of any copyright.").

To the extent these alternatives suggest any originality in the Toner Loading Program, at any rate, the quantum of originality may well be de minimis and accordingly insufficient to support the validity of Lexmark's copyright in the work. See Mitel, Inc. v. Iqtel, Inc., 124 F.3d 1366, 1373-74 (10th Cir. 1997)(determining that plaintiff's "arbitrary selection" of several numbers required only "de minimis creative effort" and did not "evince enough originality to distinguish authorship") (quotation omitted); cf. Sega Enters., 977 F.2d at 1524 n. 7 (noting that 20-byte code is of de minimis length and therefore likely a "word" or "short phrase" that is not protected by copyright law); Murray Hill Publ'ns, Inc. v. ABC Communs., Inc., 264 F.3d 622, 633 (6th Cir. 2001) (noting that short movie line was "a phrase or slogan not worthy of copyright protection in its own right"). Because the district court initially looked at these issues and this evidence through the wrong frame of reference, its conclusion that the Toner Loading Program had sufficient originality to obtain copyright protection does not support the preliminary injunction. At the permanent injunction stage of this dispute, we leave it to the district court in the first instance to decide whether the Toner Loading Program has sufficient originality to warrant copyright protection.

The difficulty of deriving the proper checksum value and the corresponding degree to which the checksum operation acts as a constraint on the content of the bytes comprising the Toner Loading Program may be an open question at the permanent injunction phase. But for purposes of the preliminary injunction, Dr. Goldberg's unchallenged testimony that it would be "computationally impossible" to modify the checksum value without contextual information suffices to establish that the checksum operation imposes a compatibility constraint in the most literal sense possible: if any single byte of the Toner Loading Programis altered, the printer will not function. On this record, pure compatibility requirements justified SCC's copying of the Toner Loading Program.

Where “admitted literal copying of a discrete, easily-conceptualized portion of a work” is at issue—as with Oracle’s declaring code—a court “need not perform a complete abstraction-filtration-comparison analysis” and may focus the protectability analysis on the filtration stage, with attendant reference to standard copyright principles. Mitel, 124 F.3d at 1372-73. ORACLE AMERICA, INC. v. GOOGLE INC. 25 (finding that the district court erred in assessing principles of merger and scenes a faire in the infringement analysis, rather than as a component of copyrightability), with Kregos, 937 F.2d at 705 (noting that the Second Circuit has considered the merger doctrine “in determining whether actionable infringement has occurred, rather than whether a copyright is valid”); see also Lexmark, 387 F.3d at 557 (Feikens, J., dissenting-in-part) (noting the circuit split and concluding that, where a court is assessing merger of an expression with a method of operation, “I would find the merger doctrine can operate only as a defense to infringement in that context, and as such has no bearing on the question of copyrightability.”).

We initially consider Lexmark's DMCA claim concerning the Printer Engine Program, which (the parties agree) is protected by the general copyright statute. In deciding that Lexmark's authentication sequence "effectively controls access to a work protected under [the copyright provisions]," the district court relied on a definition in the DMCA saying that a measure "effectively controls access to a work" if, "in the ordinary course of operation," it "requires the application of information, or a process or treatment, with the authority of the copyright owner, to gain access to the work." 17 U.S.C. § 1201(a) (3). Because Congress did not explain what it means to "gain access to the work," the district court relied on the "ordinary, customary meaning" of "access": "the ability to enter, to obtain, or to make use of," D. Ct. Op. at 41 (quoting Merriam-Webster's Collegiate Dictionary 6 (10th ed.1999)). Based on this definition, the court concluded that "Lexmark's authentication sequence effectively `controls access' to the Printer Engine Program because it controls the consumer's ability to make use of these programs." D. Ct. Op. at 41 (emphasis added). We disagree. It is not Lexmark's authentication sequence that "controls access" to the Printer Engine Program. See 17 U.S.C. § 1201(a) (2). It is the purchase of a Lexmark printer that allows "access" to the program. Anyone who buys a Lexmark printer may read the literal code of the Printer Engine Program directly from the printer memory, with or without the benefit of the authentication sequence, and the data from the program may be translated into readable source code after which copies may be freely distributed. Maggs Hr'g Test., JA 928. No security device, in other words, protects access to the Printer Engine Program Code and no security device accordingly must be circumvented to obtain access to that program code. The authentication sequence, it is true, may well block one form of "access" — the "ability to ... make use of" the Printer Engine Program by preventing the printer from functioning. But it does not block another relevant form of "access" — the "ability to [] obtain" a copy of the work or to "make use of" the literal elements of the program (its code). Because the statute refers to "control [ling] access to a work protected under this title," it does not naturally apply when the "work protected under this title" is otherwise accessible. Just as one would not say that a lock on the back door of a house "controls access" to a house whose front door does not contain a lock and just as one would not say that a lock on any door of a house "controls access" to the house after its purchaser receives the key to the lock, it does not make sense to say that this provision of the DMCA applies to otherwise-readily-accessible copyrighted works. Add to this the fact that the DMCA not only requires the technological measure to "control [] access" but also requires the measure to control that access "effectively," 17 U.S.C. § 1201(a) (2), and it seems clear that this provision does not naturally extend to a technological measure that restricts one form of access but leaves another route wide open. See also id. § 1201(a) (3) (technological measure must "require [] the application of information, or a process or a treatment ... to gain access to the work") (emphasis added). See The Chamberlain Group, Inc. v. Skylink Techs., Inc., 381 F.3d 1178, at *1183, 2004 U.S.App. LEXIS 18513, at *52 (Fed. Cir. Aug. 31, 2004) ("Chamberlain's proposed construction of the DMCA ignores the significant differences between defendants whose accused products enable copying and those, like Skylink, whose accused products enable only legitimate uses of copyrighted software."). Nor are we aware of any cases that have applied this provision of the DMCA to a situation where the access-control measure left the literal code or text of the computer program or data freely readable. And several cases apply the provision in what seems to us its most natural sense. See, e.g., 321 Studios v. Metro Goldwyn Mayer Studios, Inc., 307 F. Supp. 2d 1085, 1095 (N.D. Cal. 2004) (deciding that the "CSS" encryption program, which prevents viewing of DVD movies and copying of the data encoded on the DVD, effectively controls access to copyrighted DVD movies); Universal City Studios, Inc. v. Reimerdes, 111 F. Supp. 2d 294, 318 (S.D.N.Y. 2000), aff'd sub nom., Corley, 273 F.3d 429; Sony Computer Entm't Am. Inc. v. Gamemasters, 87 F. Supp. 2d 976, 987 (N.D. Cal. 1999) (deciding that technological measure on PlayStation game console, which prevented unauthorized games from being played, effectively controlled access to copyrighted CD-ROM video games, which the facts of the case do not describe as either encrypted or unencrypted); see also RealNetworks, 2000 WL 127311, at *3 (noting that the technological measure at issue was a "successful means of protecting against unauthorized duplication and distribution" of copyrighted digital works); Pearl Investments, LLC v. Standard I/O, Inc., 257 F. Supp. 2d 326, 349-50 (D. Me. 2003) (determining that plaintiff's "encrypted, password-protected virtual private network," which blocked access to data including plaintiff's copyrighted computer software, was a technological measure that effectively controlled access to that work). Lexmark defends the district court's contrary ruling on several grounds. First, it contends that SCC waived this argument by failing to raise it in the district court. The premise of this argument remains unclear. Below, SCC indeed claimed that the DMCA by its terms did not cover its conduct. While SCC may not have anticipated the district court's specific reliance on the "to make use of" definition of "access" in its preliminary injunction ruling, the district court's ruling also does not say that SCC conceded the point. Under these circumstances, it is well within our discretion to allow SCC to explain why the district court's resolution of this purely legal question — an interpretation of a statute — is mistaken. See McFarland v. Henderson, 307 F.3d 402, 407 (6th Cir. 2002). Second, Lexmark counters that several cases have embraced a "to make use of" definition of "access" in applying the DMCA. While Lexmark is partially correct, these cases (and others as well) ultimately illustrate the liability line that the statute draws and in the end explain why access to the Printer Engine Program is not covered. In the essential setting where the DMCA applies, the copyright protection operates on two planes: in the literal code governing the work and in the visual or audio manifestation generated by the code's execution. For example, the encoded data on CDs translates into music and on DVDs into motion pictures, while the program commands in software for video games or computers translate into some other visual and audio manifestation. In the cases upon which Lexmark relies, restricting "use" of the work means restricting consumers from making use of the copyrightable expression in the work. See 321 Studios, 307 F. Supp. 2d at 1095 (movies contained on DVDs protected by an encryption algorithm cannot be watched without a player that contains an access key); Reimerdes, 111 F. Supp. 2d at 303 (same); Gamemasters, 87 F. Supp. 2d at 981 (Sony's game console prevented operation of unauthorized video games). As shown above, the DMCA applies in these settings when the product manufacturer prevents all access to the copyrightable material and the alleged infringer responds by marketing a device that circumvents the technological measure designed to guard access to the copyrightable material. The copyrightable expression in the Printer Engine Program, by contrast, operates on only one plane: in the literal elements of the program, its source and object code. Unlike the code underlying video games or DVDs, "using" or executing the Printer Engine Program does not in turn create any protected expression. Instead, the program's output is purely functional: the Printer Engine Program "controls a number of operations" in the Lexmark printer such as "paper feed [,] paper movement [,] [and] motor control." Lexmark Br. at 9; cf. Lotus Dev., 49 F.3d at 815 (determining that menu command hierarchy is an "uncopyrightable method of operation"). And unlike the code underlying video games or DVDs, no encryption or other technological measure prevents access to the Printer Engine Program. Presumably, it is precisely because the Printer Engine Program is not a conduit to protectable expression that explains why Lexmark (or any other printer company) would not block access to the computer software that makes the printer work. Because Lexmark's authentication sequence does not restrict access to this literal code, the DMCA does not apply. Lexmark next argues that access-control measures may "effectively control access" to a copyrighted work within the meaning of the DMCA even though the measure may be evaded by an "`enterprising end-user.'" Lexmark Br. at 46 (quoting RealNetworks, 2000 WL 127311, at *9). Doubtless, Lexmark is correct that a precondition for DMCA liability is not the creation of an impervious shield to the copyrighted work. See RealNetworks, 2000 WL 127311, at *9; Reimerdes, 111 F. Supp. 2d at 317-18 (rejecting argument that an encryption measure does not "effectively control access" because it is only a "weak cipher"); see also 17 U.S.C. § 1201(a) (3). Otherwise, the DMCA would apply only when it is not needed. But our reasoning does not turn on the degree to which a measure controls access to a work. It turns on the textual requirement that the challenged circumvention device must indeed circumvent something, which did not happen with the Printer Engine Program. Because Lexmark has not directed any of its security efforts, through its authentication sequence or otherwise, to ensuring that its copyrighted work (the Printer Engine Program) cannot be read and copied, it cannot lay claim to having put in place a "technological measure that effectively controls access to a work protected under [the copyright statute]." 17 U.S.C. § 1201(a) (2) (B). Nor can Lexmark tenably claim that this reading of the statute fails to respect Congress's purpose in enacting it. Congress enacted the DMCA to implement the Copyright Treaty of the World Intellectual Property Organization, and in doing so expressed concerns about the threat of "massive piracy" of digital works due to "the ease with which [they] can be copied and distributed worldwide virtually instantaneously." S.Rep. No. 105-190, at 8 (1998). As Congress saw it, "copyrighted works will most likely be encrypted and made available to consumers once payment is made for access to a copy of the work. [People] will try to profit from the works of others by decoding the encrypted codes protecting copyrighted works, or engaging in the business of providing devices or services to enable others to do so." H.R.Rep. No. 105-551, pt. 1, at 10. Backing with legal sanctions "the efforts of copyright owners to protect their works from piracy behind digital walls such as encryption codes or password protections," Corley, 273 F.3d at 435, Congress noted, would encourage copyright owners to make digital works more readily available, see S.Rep. No. 105-190, at 8. See also Nimmer § 12 A. 02 [B] [1]. Nowhere in its deliberations over the DMCA did Congress express an interest in creating liability for the circumvention of technological measures designed to prevent consumers from using consumer goods while leaving the copyrightable content of a work unprotected. In fact, Congress added the interoperability provision in part to ensure that the DMCA would not diminish the benefit to consumers of interoperable devices "in the consumer electronics environment." 144 Cong. Rec. E2136 (daily ed. Oct. 13, 1998) (remarks of Rep. Bliley). See generally Anti-Circumvention Rulemaking Hearing, at 44-56, at http://www.copyright.gov/1201/2003/hearings/transcript-may9.pdf (testimony of Professor Jane Ginsburg) (Section 1201(a) does not "cover [] the circumvention of a technological measure that controls access to a work not protected under [the Copyright] title. And if we're talking about ball point pen cartridges, printer cartridges, garage doors and so forth, we're talking about works not protected under this title."). In view of our conclusion regarding the Printer Engine Program, we can dispose quickly of Lexmark's DMCA claim regarding the Toner Loading Program. The SCC chip does not provide "access" to the Toner Loading Program but replaces the program. And to the extent a copy of the Toner Loading Program appears on the Printer Engine Program, Lexmark fails to overcome the same problem that undermines its DMCA claim with respect to the Printer Engine Program: Namely, it is not the SCC chip that permits access to the Printer Engine Program but the consumer's purchase of the printer. One other point deserves mention. All three liability provisions of this section of the DMCA require the claimant to show that the "technological measure" at issue "controls access to a work protected under this title," see 17 U.S.C. § 1201(a) (2) (A)-(C), which is to say a work protected under the general copyright statute, id. § 102(a). To the extent the Toner Loading Program is not a "work protected under [the copyright statute]," which the district court will consider on remand, the DMCA necessarily would not protect it. The district court also rejected SCC's interoperability defense — that its replication of the Toner Loading Program data is a "technological means" that SCC may make "available to others" "solely for the purpose of enabling interoperability of an independently created computer program with other programs." 17 U.S.C. § 1201(f) (3). In rejecting this defense, the district court said that "SCC's SMARTEK microchips cannot be considered independently created computer programs. [They] serve no legitimate purpose other than to circumvent Lexmark's authentication sequence and ... cannot qualify as independently created when they contain exact copies of Lexmark's Toner Loading Programs." D. Ct. Op. ¶ 94, at 47. Because the issue could become relevant at the permanent injunction stage of this dispute, we briefly explain our disagreement with this conclusion. In particular, the court did not explain why it rejected SCC's testimony that the SMARTEK chips do contain other functional computer programs beyond the copied Toner Loading Program data. The affidavit of Lynn Burchette, an SCC manager, states that " [the SMARTEK] chip has a microprocessor, with software routines we developed that control its operation and function. Our chip supports additional functionality performed by our software beyond that of [the chip on Lexmark's toner cartridges]." JA 103. And Dr. Goldberg testified that "Static Control has written a substantial amount of software for managing this chip; for not only providing the int [er]operability features, but also for managing the additional functionality that the [chip manufacturer] provides and which the remanufacturers may want." JA 1023. Instead of showing why these statements are wrong, Lexmark contends that this is not "credible evidence" that "independently created computer programs" exist on the SMARTEK chip. Yet Lexmark bears the burden of establishing its likelihood of success on the merits of the DMCA claims. See Leary v. Daeschner, 228 F.3d 729, 739 (6th Cir. 2000). Because Lexmark has offered no reason why the testimony of SCC's experts is not "credible evidence" on this point and has offered no evidence of its own to dispute or even overcome the statements of Burchette and Goldberg, SCC also has satisfied the "independently created computer programs" requirement and may benefit from the interoperability defense, at least in the preliminary injunction context. Lexmark argues alternatively that if independently created programs do exist, (1) they must have existed prior to the "reverse engineering" of Lexmark's Toner Loading Program, and (2) the technological means must be "necessary or absolutely needed" to enable interoperability of SCC's SMARTEK chip with Lexmark's Printer Engine Program. As to the first argument, nothing in the statute precludes simultaneous creation of an interoperability device and another computer program; it just must be "independently" created. As to the second argument, the statute is silent about the degree to which the "technological means" must be necessary, if indeed they must be necessary at all, for interoperability. The Toner Loading Program copy satisfies any such requirement, however, because without that program the checksum operation precludes operation of the printer (and, accordingly, operation of the Printer Engine Program), unless the checksum value located elsewhere on the chip is modified — which appears to be a computational impossibility without the contextual information that Lexmark does not disclose. See supra. Also unavailing is Lexmark's final argument that the interoperability defense in § 1201(f) (3) does not apply because distributing the SMARTEK chip constitutes infringement and violates other "applicable law" (including tortious interference with prospective economic relations or contractual relations). Because the chip contains only a copy of the thus-far unprotected Toner Loading Program and does not contain a copy of the Printer Engine Program, infringement is not an issue. And Lexmark has offered no independent, let alone persuasive, reason why SCC's SMARTEK chip violates any state tort or other state law.

a. First Fair Use Factor: Purpose and Character of the Allegedly Infringing Use We first consider the nature of the work: whether the new work is transformative or simply supplants the original work, and whether the work is commercial in nature. See Campbell, 510 U.S. at 579. Of primary concern is whether the Sheriff's Department's use was transformative; the more "transformative the new work, the less will be the significance of the other factors." Id. A use is considered transformative only where a defendant changes a plaintiff's copyrighted work or uses the plaintiff's copyrighted work in a different context such that the plaintiff's work is transformed into a new creation. See, e.g., id. at 578-79. The Sheriff's Department created exact copies of RUMBA's software. It then put those copies to the identical purpose as the original software. Such a use cannot be considered transformative. See Kelly v. Arriba Soft Corp., 336 F.3d 811, 819 (9th Cir. 2003)(finding that reproducing music CDs in computer MP3 format is not a transformative use because the resulting use of the copyrighted work was the same as the original use, i.e., entertainment). In cases where "'use is for the same intrinsic purpose as [the copyright holder's] . . . such use seriously weakens a claimed fair use.' " Worldwide Church of God v. Philadelphia Church of God, Inc., 227 F.3d 1110, 1117 (9th Cir. 2000) (quoting Weissmann v. Freeman, 868 F.2d 1313, 1324 (2d Cir. 1989)). Next, the Sheriff's Department's use of hard drive imaging to copy RUMBA Office did not promote the advancement of knowledge and the arts -- goals that copyright intends to secure. "Copyright law ultimately serves the purpose of enriching the general public through access to creative works," Fogerty v. Fantasy, Inc., 510 U.S. 517, 527, 114 S. Ct. 1023, 127 L. Ed. 2d 455 (1994). But the Sheriff's Department did not "provide the marketplace with new creative works," nor was there any advancement of public knowledge in this case. Triad Sys. Corp. v. Southeastern Express Co., 64 F.3d 1330,1336 (9th Cir. 1995). Thus, allowing such a "fair use" would not further the ultimate goals of copyright law. Finally, the Sheriff's Department use was commercial in nature. The Sheriff's Department alleges that its use was "commercially insignificant" because not all of the copies of Wall Data's computer software were actually used, and because the Sheriff's Department is a government agency that does not compete with Wall Data. We disagree. As we explained in Worldwide Church of God, "repeated and exploitative copying of copyrighted works, even if the copies are not offered for sale, may constitute a commercial use." 227 F.3d at 1118. A commercial use "is demonstrated by a showing that repeated and exploitative unauthorized copies of copyrighted works were made to save the expense of purchasing authorized copies." A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1015 (9th Cir. 2001); see also Worldwide Church of God, 227 F.3d at 1118 (concluding that a user is "commercial" where the infringer profited from its infringement because "it gained an 'advantage' or 'benefit' from its distribution and use of [the plaintiff's product] without having to account to the copyright holder"). The terms of Wall Data's RUMBA license clearly restricted the Sheriff's Department's use to a "single designated computer" and prohibited the Sheriff's Department from using "the Software in any other multiple computer or multiple user arrangement." Despite this condition, the Sheriff's Department loaded an entire suite of software, including RUMBA Office, onto nearly all of the computers in the Twin Towers. By using hard drive imaging, the Sheriff's Department saved man-hours and eliminated possible errors associated with separately installing the individual software packages onto each computer in the Twin Tower facility. Hard drive imaging also ensured that those users who needed to use RUMBA would be able to access the software at whatever computer they were assigned to work. Such flexibility could only have been achieved by purchasing licenses for each of the computers on which the software was loaded, or by negotiating with Wall Data for a less restrictive license. Accordingly, we conclude that "the purpose and character" of the Sheriff's Department's use was commercial, because the copies "were made to save the expense of purchasing authorized copies," Napster, 239 F.3d at 1015, or at least the expense of purchasing a more flexible license. Thus, the district court did not err when it decided that the Sheriff's Department's use was not "for a legitimate, essentially non-exploitative purpose," and that the commercial aspect of the Sheriff's Department's use was not "of minimal significance." Sega, 977 F.2d at 1522-23. The Sheriff's Department's installation of the RUMBA Office software onto nearly all of its computers in the Twin Towers was not transformative, did not promote an advancement of the arts, and was commercial in nature. The first factor therefore weighs against a finding of fair use.

b. Second Fair Use Factor: Nature of the Copyrighted Work In analyzing the second fair use factor, we look at the nature of the copyrighted work, creative works being "'closer to the core of intended copyright protection' than informational and functional works." Dr. Seuss, 109 F.3d at 1402 (quoting Campbell, 510 U.S. at 586). Our sister circuits have also considered, under this factor, whether the copyrighted work represents "substantial investment of time and labor . . . in anticipation of a financial return." MCA, Inc. v. Wilson, 677 F.2d 180, 182 (2d Cir. 1981). Although the RUMBA software products are not purely creative works, copyright law nonetheless protects computer software. See Sega, 977 F.2d at 1519 ("The 1980 amendments to the Copyright Act unambiguously extended copyright protection to computer programs."). In addition, Wall Data presented undisputed evidence that RUMBA software products were developed over several years, and required a multi-million dollar investment on Wall Data's part. We therefore conclude that the nature of the copyrighted work weighs against a finding of fair use.

c. Third Fair Use Factor: Amount and Substantiality of the Portion Used Next, we consider whether the "amount and substantiality of the portion used in relation to the copyright work as a whole," 17 U.S.C. § 107(3), is "reasonable in relation to the purpose of copying," Dr. Seuss, 109 F.3d at 1402. The Sheriff's Department copied RUMBA Office in its entirety to ensure that all of the computers in the Twin Towers had identical copies of the software. And although "entire verbatim reproductions are justifiable where the purpose of the work differs from the original," Mattel, 353 F.3d at 804, the Sheriff's Department put its copies to the exact purpose for which the original software licenses were purchased. Consequently, the Sheriff Department's "verbatim" copying of the entire copyrighted work also weighs against a finding of fair use. See Worldwide Church of God, 227 F.3d at 1118 ("While wholesale copying does not preclude fair use per se, copying an entire work militates against a finding of fair use.") (internal quotation marks omitted).

d. Fourth Fair Use Factor: Effect of the Use Upon the Potential Market In addressing the final fair use factor, we focus on "the normal market for the copyrighted work" and whether the allegedly infringing use threatens the potential market for, or value of, a copyrighted work. See Harper & Row, 471 U.S. at 568 (noting that the fourth fair use factor is concerned with "use that supplants any part of the normal market for a copyrighted work") (quoting S. Rep. No. 473, 94th Cong., 1st Sess. 65 (1975)). We have said: this inquiry attempts to strike a balance between the benefit the public will derive if the use is permitted and the personal gain the copyright owner will receive if the use is denied. The less adverse effect that an alleged infringing use has on the copyright owner's expectation of gain, the less public benefit need be shown to justify the use. Mattel, 353 F.3d at 804-05. The Sheriff's Department contends that its copying was solely an attempt to use efficiently its licensed copies of the RUMBA software products, and accordingly, there was no negative impact on Wall Data's market. It points to its own statements that it would not have purchased additional copies of the license had it known that its configuration went beyond its license. We are not persuaded. The Sheriff's Department bought a few licenses and found a way to install the program onto all of its computers without paying the fee required for each installation. The Sheriff's Department could have bargained for the flexibility it desired, but it did not. Whenever a user puts copyrighted software to uses beyond the uses it bargained for, it affects the legitimate market for the product. Thus, although hard drive imaging might be an efficient and effective way to install computer software, we conclude that "unrestricted and widespread conduct of the sort engaged in by the defendant" would nonetheless lead to over-use of the software. Campbell, 510 U.S. at 590 (quoting 3 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.05[A](4), at 13-102.61(1993)). Equally important, ghost copies of the software lay dormant and were unuseable only until a Sheriff's Department network administrator decided to activate RUMBA Office on that computer. The Sheriff's Department thus created its own "sub-licensing" system where it granted users permission to use the software and, in essence, asked Wall Data to "trust" that it was not using RUMBA in excess of its authorization under the license. We recognize that computer licensing is generally an "honor system," in that there is little to stop a person with physical possession of software from installing it on multiple computers. But in this case, the Sheriff's Department's system made tracking infringement almost impossible, because Wall Data could not independently verify which of the computers had been used to access RUMBA and which ones had not -- it had to trust the Sheriff's Department that its system was not allowing over-use. In fact, after Wall Data brought the over-use of RUMBA to the Sheriff's Department's attention, a Sheriff's Department employee admitted, in an email, that he was not sure how to tell which computers had accessed RUMBA. This system therefore made copyright infringement easier (because no physical installation was necessary) and made detection of over-use more difficult. In recognition of the ease with which software can be over-used, courts have been cautious to extend protection to methods that would make copyright infringement of software any easier: Software fundamentally differs from more traditional forms of medium, such as print or phonographic materials, in that software can be both, more readily and easily copied on a mass scale in an extraordinarily short amount of time and relatively inexpensively. One of the primary advantages of software, its ability to record, concentrate and convey information with unprecedented ease and speed, makes it extraordinarily vulnerable to illegal copying and piracy. [Thus,] it is important to acknowledge these special characteristics of the software industry and provide enhanced copyright protection for its inventors and developers. Adobe Sys., Inc. v. Stargate Software Inc., 216 F. Supp. 2d 1051, 1059 (N.D. Cal. 2002). We believe that "widespread use" of hard drive imaging in excess of one's licenses could seriously impact the market for Wall Data's product. See Campbell, 510 U.S. at 590 (citations omitted). Therefore, we conclude that the fourth factor also weighs against a finding of fair use. * * *

Mr. Hutchins' charge of infringement relates to Zoll's use of words and phrases that are included on his copyrighted List. Both the Hutchins and the Zoll systems guide the rescuer through the CPR process by way of a series of computer-generated instructions presented in words and phrases. Mr. Hutchins asserts that Zoll's system uses twenty-seven phrases from the copyrighted list. The district court found that Zoll's instructions contain two phrases that are identical to those on Hutchins' list, viz., "call for help" and "check breathing." The court found that three more phrases are similar, viz., Hutchins' "stay calm" (Zoll's "remain calm"); "if no pulse, start CPR" ("if no pulse, continue"); and "give two breaths" ("start with two breaths"). However, the court concluded that these phrases are entirely functional, that they are not original with Hutchins but are standard CPR instructions, and that they are not subject to copyright. Copyright does not protect individual words and "fragmentary" phrases when removed from their form of presentation and compilation. Although the compilation of public information may be subject to copyright in the form in which it is presented, the copyright does not bar use by others of the information in the compilation. See Feist, 499 U.S. at 348-49, 111 S.Ct. 1282 (no matter how much original authorship is embodied, the facts and ideas are not barred from use by others). The district court found that the words and phrases on Mr. Hutchins' "Script and Word List" are standard CPR instructions devoid of "creative expression that somehow transcend the functional core of the directions," the court quoting National Nonwovens, Inc. v. Consumer Products Enterprises, Inc., 397 F.Supp.2d 245, 256 (D.Mass.2005).

Also weighing against Mr. Hutchins' charge of infringement is the pragmatic doctrine of "merger" of idea and expression, applying the " scenes à faire" principle that originated for literary works. As explained in Atari Games Corp. v. Oman, 888 F.2d 878, 886 (D.C. Cir.1989), "[t]he term scenes a faire refers to stereotyped expressions, `incidents, characters or settings which are as a practical matter indispensable, or at least standard, in the treatment of a given topic'" (citations omitted). The standard instructions for performing CPR are indispensable for applying CPR, and remain in the public domain. See John G. Danielson, Inc. v. Winchester-Conant Props., Inc., 322 F.3d 26, 43 (1st Cir.2003) (when the terms at issue are the only available forms of expression, these expressions are not subject to copyright). Summary judgment of no copyright infringement was appropriately granted.

As a preliminary matter, Leadsinger does not allege that its display or printing of BMG's copyrighted song lyrics are for the purpose of "criticism, comment, news reporting, teaching . . ., scholarship, or research . . . ." See 17 U.S.C. § 107. While Leadsinger argued on appeal that karaoke teaches singing, that allegation is not set forth in its complaint. Even if the court could infer that a karaoke device has the potential to teach singing because the device allows consumers to sing along with recorded music, it is not reasonable to infer that teaching is actually the purpose of Leadsinger's use of the copyrighted lyrics. With regard to the first factor under § 107, Leadsinger's allegations support only a commercial use. The complaint does not allege that Leadsinger's use of copyrighted lyrics is transformative. See Mattel, 353 F.3d at 800 (explaining that an analysis of "the purpose and character of use" asks to what extent the new work transforms, rather than simply supplants, the original work, and stating that "a work must add 'something new, with a further purpose or different character, altering the first with new expression, meaning, or message'" (quoting Campbell, 510 U.S. at 579)). It is reasonable to infer that Leadsinger does not add to or alter the copyrighted lyrics, which would undermine the device's ability to enable consumers to sing along with the recorded music. Though Leadsinger alleges that its use of lyrics helps consumers to understand the song lyrics and that "the words facilitate parental control over objectionable song words," "the ultimate use to which the customer puts [a copyrighted work] is irrelevant . . . ." L.A. News Serv. v. Tullo, 973 F.2d 791, 797 (9th Cir. 1992); see Harper & Row Publishers, 471 U.S. at 562 ("The crux of the profit/non-profit distinction is . . . whether the [vendor] stands to profit from exploitation of the copyrighted material without paying the customary price."); Zomba Enters., 491 F.3d at 582-83 ("[T]he end-user's utilization of the product is largely irrelevant . . . ."). Leadsinger's basic purpose remains a commercial one--to sell its karaoke device for profit. And commercial use of copyrighted material is "presumptively an unfair exploitation of the monopoly privilege that belongs to the owner of the copyright." Sony Corp. of Am., 464 U.S. at 451. The only other circuit court to address fair use in the context of karaoke reached the same conclusion. In Zomba Enterprises, the defendant, a karaoke company, argued that its use of copyrighted musical compositions was fair. 491 F.3d at 581. The Sixth Circuit held that the defendant's use of copyrighted musical compositions did not constitute "fair use." Id. at 584. In assessing the first § 107 factor, "the purpose and character of the use of the work at issue," the Sixth Circuit characterized the defendant's use as "commercial" and rejected the argument that karaoke is a teaching tool. Id. at 582-83.

The second factor under § 107, "the nature of the copyrighted work," also weighs in BMG's favor. Original song lyrics are a work of creative expression, as opposed to an informational work, which is precisely the sort of expression that the copyright law aims to protect. See Abend v. MCA, Inc., 863 F.2d 1465, 1481 (9th Cir. 1988) (fictional short story is "a quintessentially creative product"); see also Campbell, 510 U.S. at 586 (concluding that musical composition "Oh, Pretty Woman" fell "within the core of the copyright's protective purposes"); 4 Nimmer on Copyright § 13.05[A][2][a] ("[T]he more creative a work, the more protection it should be accorded from copying; correlatively, the more informational or functional the plaintiff's work, the broader should be the scope of the fair use defense.")

The third factor, "the amount and substantiality of the portion used in relation to the copyrighted work as a whole," also militates against a finding of fair use. See Worldwide Church of God v. Phila. Church of God, 227 F.3d 1110, 1118 (9th Cir. 2000) (copying an entire work "militates against a finding of fair use" (quoting Hustler Magazine, Inc. v. Moral Majority, Inc., 796 F.2d 1148, 1155 (9th Cir. 1986)). Leadsinger alleges that it has the right to print or display the song lyrics to the songs it records, and nothing in the complaint indicates that Leadsinger uses only a portion of the copyrighted lyrics. It is, therefore, reasonable to assume that Leadsinger uses copyrighted lyrics in their entirety.

On appeal, Leadsinger devotes much of its attention to the fourth factor under § 107, "the effect of use upon the potential market for or value of the copyrighted work." The district court concluded that the allegations in Leadsinger's complaint did not permit it to analyze the effect that the sale of Leadsinger's device would have on the market. We reach the same conclusion. In arguing that its use will not affect the potential market, Leadsinger contends that there is no market for song lyrics standing alone. Leadsinger failed to allege this in its complaint and we are not willing to assume that there is no such market. Leadsinger did allege that music publishers "have never (or rarely) required a print license" for the use of lyrics by record companies. But, Leadsinger's complaint specifies that this practice is for non-karaoke recordings and that music publishers have demanded "lyric reprint" fees for karaoke use. We do not find it reasonable to infer that because copyright holders allow record companies to reprint lyrics in conjunction with non-karaoke recordings, no harm will result from the reprinting and display of song lyrics in the distinctly different context of karaoke. Thus, as the district court observed, Leadsinger has failed to set forth allegations on the effect of its device on the potential market for or value of the copyrighted work. We have, however, concluded that Leadsinger's use is intended for commercial gain, and it is well accepted that when "the intended use is for commercial gain," the likelihood of market harm "may be presumed." Sony, 464 U.S. at 451. We have not hesitated to apply this presumption in the past, see, e.g., Elvis Presley Enters., Inc. v. Passport Video, 349 F.3d 622, 631 (9th Cir. 2003), and we are not reluctant to apply it here. Moreover, "the importance of [the market effect] factor [varies], not only with the amount of harm, but also with the relative strength of the showing on the other factors." See Campbell, 510 U.S. at 591 n.21. The showing on all other factors under § 107 is strong: the purpose and character of Leadsinger's use is commercial; song lyrics fall within the core of copyright protection; and Leadsinger uses song lyrics in their entirety. On this basis, we affirm the district court's dismissal of Leadsinger's request for a declaration based on the fair use doctrine.

The second “traditional contour,” the fair use defense, is codified at 17 U.S.C. §107: “[T]he fair use of a copyrighted work, including such use by reproduction in copies . . . , for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright.” This limitation on exclusivity “allows the public to use not only facts and ideas contained in a copyrighted work, but also [the author's] expression itself in certain circumstances.” Eldred, 537 U.S., at 219, 123 S. Ct. 769, 154 L. Ed. 2d 683; see id., at 220, 123 S. Ct. 769, 154 L. Ed. 2d 683 (“fair use defense affords considerable latitude for scholarship and comment, . . . even for parody” (internal quotation marks omitted)). Given the “speech-protective purposes and safeguards” embraced by copyright law, see id., at 219, 123 S. Ct. 769, 154 L. Ed. 2d 683, we concluded in Eldred that there was no call for the heightened review petitioners sought in that case.We reach the same conclusion here.Section 514 leaves [1310] undisturbed the “idea/expression” distinction and the “fair use” defense. Moreover, Congress adopted measures to ease the transition from a national scheme to an international copyright regime: It deferred the date from which enforcement runs, and it cushioned the impact of restoration on “reliance parties” who exploited foreign works denied protection before §514 took effect. See supra, at 315-316, 181 L. Ed. 2d, at 848 (describing 17 U.S.C. §104A(c), (d), and (h)). See also Eldred, 537 U.S., at 220, 123 S. Ct. 769, 154 L. Ed. 2d 683 (describing supplemental allowances and exemptions available to certain users to mitigate the CTEA's impact).

(d) Considered against this backdrop, §514 falls comfortably within Congress’ Copyright Clause authority. Congress had reason to believe that a well-functioning international copyright system would encourage the dissemination of existing and future works. And testimony informed Congress that full compliance with Berne would expand the foreign markets available to U. S. authors and invigorate protection against piracy of U. S. works abroad, thus benefitting copyright-intensive industries stateside and inducing greater investment in the creative process. This Court has no warrant to reject Congress’ rational judgment that exemplary adherence to Berne would serve the objectives of the Copyright Clause. Pp. 22–23. First Amendment - ) The pathmarking Eldred decision is again instructive. There, the Court held that the CTEA’s enlargement of a copyright’s duration did not offend the First Amendment’s freedom of expression guarantee. Recognizing that some restriction on expression is the inherent and intended effect of every grant of copyright, the Court observed that the Framers regarded copyright protection not simply as a limit on the manner in which expressive works may be used, but also as an “engine of free expression.” 537 U. S., at 219. The “traditional contours” of copyright protection, i.e., the “idea/expression dichotomy” and the “fair use” defense, moreover, serve as “built-in First Amendment accommodations.” Ibid. Given the speech-protective purposes and safeguards embraced by copyright law, there was no call for the heightened review sought in Eldred. The Court reaches the same conclusion here. Section 514 leaves undisturbed the idea/expression distinction and the fair use defense. Moreover, Congress adopted measures to ease the transition from a national scheme to an international copyright regime. Pp. 23–26. (b) Petitioners claim that First Amendment interests of a higher order are at stake because they—unlike their Eldred counterparts—enjoyed “vested rights” in works that had already entered the public domain. Their contentions depend on an argument already considered and rejected, namely, that the Constitution renders the public domain largely untouchable by Congress. Nothing in the historical record, subsequent congressional practice, or this Court’s jurisprudence warrants exceptional First Amendment solicitude for copyrighted works that were once in the public domain. Congress has several times adjusted copyright law to protect new categories of works as well as works previously in the public domain. Section 514, moreover, does not impose a blanket prohibition on public access. The question is whether would-be users of certain foreign works must pay for their desired use of the author’s expression, or else limit their exploitation to “fair use” of those works. By fully implementing Berne, Congress ensured that these works, like domestic and most other foreign works, would be governed by the same legal regime. Section 514 simply placed foreign works in the position they would have occupied if the current copyright regime had been in effect when those works were created and first published. Pp. 26–30.

The landscape changed in 1994. The Uruguay round of multilateral trade negotiations produced the World Trade Organization (WTO) and the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS). [ 7 ] The United States joined both. TRIPS mandates, on pain of WTO enforcement, implementation of Berne’s first 21 articles. TRIPS, Art. 9.1, 33 I. L. M. 1197, 1201 (requiring adherence to all but the “moral rights” provisions of Article 6bis). The WTO gave teeth to the Convention’s requirements: Noncompliance with a WTO ruling could subject member countries to tariffs or cross-sector retaliation. See Gervais, supra, at 213; 7 W. Patry, Copyright §24:1, pp. 24–8 to 24–9 (2011). The specter of WTO enforcement proceedings bolstered the credibility of our trading partners’ threats to challenge the United States for inadequate compliance with Article 18. See URAA Joint Hearing 137 (statement of Shapiro, USTR) (“It is likely that other WTO members would challenge the current U. S. implementation of Berne Article 18 under [WTO] procedures.”). [ 8 ] Congress’ response to the Uruguay agreements put to rest any questions concerning U. S. compliance with Article 18. Section 514 of the URAA, 108Stat. 4976 (codified at 17 U. S. C. §104A, 109(a)), [ 9 ] extended copyright to works that garnered protection in their countries of origin, [ 10 ] but had no right to exclusivity in the United States for any of three reasons: lack of copyright relations between the country of origin and the United States at the time of publication; lack of subject-matter protection for sound recordings fixed before 1972; and failure to comply with U. S. statutory formalities (e.g., failure to provide notice of copyright status, or to register and renew a copyright). See §104A(h)(6)(B)–(C). [ 11 ]

The central inquiry under the first factor is whether the new work is "transformative." Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (1994). Transformative works "add[] something new" to an existing work, endowing the first with "new expression, meaning, or message," rather than "merely supersed[ing] the objects of the original creation." Id. (internal quotation marks omitted). By using it as a biographical anchor, Dodger put the clip to its own transformative ends. See Elvis Presley Enters., Inc. v. Passport Video, 349 F.3d 622, 629 (9th Cir. 2003) (recognizing that the defendant's "use of many of the television clips [of Elvis's performances] is transformative because they are cited as historical reference points"), overruled on other grounds as stated in Flexible Lifeline Sys., Inc. v. Precision Lift, Inc., 654 F.3d 989, 995 (9th Cir. 2011) (per curiam); see also Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 609 (2d Cir. 2006) (concluding that the use of Grateful Dead concert posters to comment on and commemorate the performances they were designed to promote was transformative). SOFA's argument that the clip was used for its own entertainment value is not supported by the record. Moreover, because Dodger's use of the clip is transformative, the fact that Jersey Boys is a commercial production is of little significance. Campbell, 510 U.S. at 579. Therefore, the first fair use factor heavily favors Dodger.

The second fair use factor recognizes that some works — generally creative works, like fictional stories — "are closer to the core of intended copyright protection than others." Id. at 586. An alleged infringer will have a more difficult time establishing fair use when he appropriates a work of that nature. Id. While the entire episode of The Ed Sullivan Show or the individual performances may be near to the core of copyright, the clip conveys mainly factual information — who was about to perform. Therefore, the second factor also favors Dodger.

The third factor looks to the quantitative amount and qualitative value of the original work used in relation to the defendant's justification for the use. Id. 586-88. SOFA does not challenge the conclusion that the seven-second clip is quantitatively insignificant, but argues that Dodger "attempted to capitalize on the central and most beloved part of The Ed Sullivan Show, namely, Ed Sullivan's introduction of popular new rock and roll acts," by incorporating the clip into the play. SOFA's argument is flawed in two respects. First, the seven-second introduction is hardly qualitatively significant. Sullivan simply identifies the group that is about to perform and the section of his audience to whom the Four Seasons would appeal. It is doubtful that the clip on its own qualifies for copyright protection, much less as a qualitatively significant segment of the overall episode. See Murray Hill Publ'ns, Inc. v. ABC Commc'n, Inc., 264 F.3d 622, 633-34 (6th Cir. 2001) (holding that the line "J.P. on J.R. in the A.M." served only a functional purpose, i.e., identifying the radio program, the radio station, and the broadcast time, and was not subject to copyright protection), overruled in part on other grounds by Reed Elsevier, Inc. v. Muchnick, 559 U.S. 154, 130 S. Ct. 1237, 176 L. Ed. 2d 18 (2010). Second, SOFA contorts the Supreme Court's use of the phrase "distinctive expression" in Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 565, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985), to give the false impression that Mr. Sullivan's "trademark gesticulation and style" is copyrightable. Copyright only attaches to an original work fixed in a tangible medium of expression, never in the underlying ideas or facts. Id. at 547. The Court used the words "distinctive expression" to explain that defendant had copied sections of President Ford's memoirs that contained Mr. Ford's writing, as opposed to the events he was discussing. Id. at 565. Certainly movement and intonation are elements in an original performance, but SOFA's argument is not limited to Sullivan's performance in the clip. It is Sullivan's charismatic personality that SOFA seeks to protect. Charisma, however, is not copyrightable. See Downing v. Abercrombie & Fitch, 265 F.3d 994, 1003-04 (9th Cir. 2001) (holding that a person's name and likeness are outside the scope of copyright); 1 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 1.01[B][1][c] at 1-30 (2012) ("A persona can hardly be said to constitute a 'writing' of an 'author.'") (footnotes omitted).

The fourth factor requires courts to consider the secondary use's impact on the market for the original work and the market for derivative works, including if the defendant's actions became "unrestricted and widespread." Campbell, 510 U.S. at 590 (internal quotation marks omitted). Where the secondary use is not a substitute for the original and does not deprive the copyright holder of a derivative use, the fourth factor weighs in favor of fair use. See id. at 591 ("[W]hen . . . the second use is transformative, market substitution is at least less certain . . . ."); Ty, Inc. v. Publ'ns Int'l Ltd., 292 F.3d 512, 517-18 (7th Cir. 2002) (stating that complementary uses, in the sense that a book review is complementary to the book it discusses, support a finding of fair use). Jersey Boys is not a substitute for The Ed Sullivan Show. The clip is seven seconds long and only appears once in the play. Dodger does not reproduce Jersey Boys on videotape or DVD, which would allow for repeated viewing of the clip. Dodger's use of the clip advances its own original creation without any reasonable threat to SOFA's business model. Therefore the fourth factor also favors a finding of fair use.

The first factor in the fair use inquiry is "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." § 107(1). The Supreme Court has stated that the "central purpose" of this factor is to see "whether and to what extent the new work is transformative." Campbell, 510 U.S. at 579. Works of this type "lie at the heart of the fair use doctrine's guarantee of breathing space within the confines of copyright, and the more 'transformative' the new work, the less will be the significance of other factors." Id. (internal citations omitted). Although transformation is a key factor in fair use, whether a work is transformative is a often highly contentious topic. See, e.g., Cariou v. Prince, 714 F.3d 694, 713-14 (2d Cir. 2013) (Wallace, J., concurring in part and dissenting in part) (disagreeing with majority as to whether artists' use of certain copyrighted photographs was transformative; would remand for further fact finding); Monge v. Maya Magazines, Inc., 688 F.3d 1164, 1185-88 (9th Cir. 2012) (M. Smith, J., dissenting) (disagreeing with majority as to whether a magazine's publication of wedding photographs was transformative); Bouchat v. Balt. Ravens Ltd. P'ship, 619 F.3d 301, 320-21 (4th Cir. 2010) (Niemeyer, J., dissenting) (disagreeing with majority as to whether the Baltimore Ravens football team's use of a "Flying B" logo was transformative). A leading treatise on this topic has lamented the frequent misuse of the transformation test, complaining that it has become a conclusory label which is "all things to all people." Melville B. Nimmer & David Nimmer, 4 Nimmer on Copyright § 13.05[A][1][b], 13168-70 (2011) (internal quotation marks omitted); see also id. at 13-168-69 (listing cases which have "erroneous[ly] with[held]" the transformative label). The plethora of cases addressing this topic means there is no shortage of language from other courts elucidating (or obfuscating) the meaning of transformation. To navigate these treacherous waters, we turn to the most definitive formulation of the test. The Supreme Court in Campbell stated that one work transforms another when "the new work . . . adds something new, with a further purpose or different character, altering the first with new expression, meaning or message." Campbell, 510 U.S. at 579. This understanding was drawn by the Court in large part from Second Circuit Judge Pierre Leval's 1990 article in the Harvard Law Review, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105 (1990). In that article, Judge Leval further clarified his understanding of "transformative" works: The use must be productive and must employ the quoted matter in a different manner or for a different purpose from the original. A quotation of copyrighted material that merely repackages or republishes the original is unlikely to pass the test; in Justice Story's words, it would merely "supersede the objects" of the original. If, on the other hand, the secondary use adds value to the original—if the quoted matter is used as raw material, transformed in the creation of new information, new aesthetics, new insights and understandings—this is the very type of activity that the fair use doctrine intends to protect for the enrichment of society. Id. at 1111. Applying this understanding, Green Day's use of Scream Icon is transformative. Green Day used the original as "raw material" in the construction of the four-minute video backdrop. It is not simply a quotation or a republication; although Scream Icon is prominent, it remains only a component of what is essentially a street-art focused music video about religion and especially about Christianity (images of Jesus Christ appear —and are defaced—several times during the course of the video). The message and meaning of the original Scream Icon is debatable. To us, it appears to be a directionless anguished screaming face. Seltzer himself testified to his view of the meaning of the original piece: "It addresses themes of youth culture, skateboard culture, insider/outsider culture, . . . it's an iconic reference to a culture and time in Los Angeles when the image was made." But regardless of the meaning of the original, it clearly says nothing about religion. With the spray-painted cross, in the context of a song about the hypocrisy of religion, surrounded by religious iconography, Staub's video backdrop using Scream Icon conveys "new information, new aesthetics, new insights and understandings" that are plainly distinct from those of the original piece. At his deposition, Seltzer seemed to acknowledge as much, when he stated that Staub's backdrop "tainted the original message of the image and [] made it now synonymous with lyrics, a video, and concert tour that it was not originally intended to be used with." Although the law in this area is splintered, as discussed above, our conclusion on transformation is generally in line with other appellate authority on transformative use. In the typical "non-transformative" case, the use is one which makes no alteration to the expressive content or message of the original work. See, e.g., Monge, 688 F.3d at 1176 (magazine publication of photos of secret wedding of celebrity was not transformative because it did not "alter[] the first [work] with new expression, meaning or message") (second alteration in original); Elvis Presley Enters., Inc. v. Passport Video, 349 F.3d 622, 629 (9th Cir. 2003) (use of copyrighted clips of Elvis's television appearances was not transformative when the clips were "played without much interruption, if any, . . . [and] serve[d] the same intrinsic entertainment value that is protected by Plaintiffs' copyrights."), overruled on other grounds as stated in Flexible Lifeline Sys., Inc. v. Precision Lift, Inc., 654 F.3d 989, 995 (9th Cir.2011) (per curiam); L.A. News Serv. v. CBS Broad., Inc., 305 F.3d 924, 938 (9th Cir. 2002) ("Merely plucking the most visually arresting excerpt from LANS's nine minutes of footage cannot be said to have added anything new."), as amended 313 F.3d 1093 (9th Cir. 2002); Ringgold v. Black Entm't Television, Inc., 126 F.3d 70, 79 (2d Cir. 1997) (use of a poster as decoration on a TV show not transformative because it was used for "precisely a central purpose for which it was created" and defendants had done nothing with the poster to add anything new). In contrast, an allegedly infringing work is typically viewed as transformative as long as new expressive content or message is apparent. This is so even where—as here—the allegedly infringing work makes few physical changes to the original or fails to comment on the original. See, e.g., Cariou, 714 F.3d at 708 (artist who altered and incorporated several copyrighted photographs into a series of paintings and collages engaged in transformative use as to most of the paintings because the images were presented with "fundamentally different aesthetic"); Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 608-09 (2d Cir. 2006) (use of concert posters in a timeline in a book on the history of the Grateful Dead was transformative; their use was as "historical artifacts" rather than for "artistic expression and promotion"); Blanch v. Koons, 467 F.3d 244, 252-53 (2d Cir. 2006) (artist who incorporated and altered copyrighted fashion photograph of a pair of women's legs as part of a larger work of art engaged in a transformative use); Kelly v. Arriba Soft Corp., 336 F.3d 811, 818-20 (9th Cir. 2003) (use of exact replicas of artist's photographs as "thumbnail images" in a search engine was transformative because their purpose was completely transformed from their original use as fine art); L.A. News Serv., 305 F.3d at 938-39 (inclusion of copyrighted clip in video montage, using editing to increase dramatic effect, was transformative). We conclude, therefore, that Green Day's use of Scream Icon was transformative. Furthermore, although the statute instructs us to consider the "commercial nature" of a work (and Green Day's concert was undoubtedly commercial in nature) "the degree to which the new user exploits the copyright for commercial gain—as opposed to incidental use as part of a commercial enterprise—affects the weight we afford commercial nature as a factor." Elvis Presley Enters., 349 F.3d at 627. Green Day's use of Scream Icon was only incidentally commercial; the band never used it to market the concert, CDs, or merchandise. Under these circumstances, the first fair use factor weighs in Green Day's favor.

The second factor that § 107 instructs us to consider is "the nature of the copyrighted work" which recognizes the fact that "some works are closer to the core of intended copyright protection than others." Campbell, 510 U.S. at 586. Scream Icon is a creative work, meriting strong protection under this factor. Mitigating this factor in favor of Green Day is that we are instructed to consider the extent to which a work has been published. Harper & Row Publishers, Inc., 471 U.S. at 564 ("The fact that a work is unpublished is a critical element of its 'nature.'"); Kelly, 336 F.3d at 820. "Published works are more likely to qualify as fair use because the first appearance of the artist's expression has already occurred." Kelly, 336 F.3d at 820. Here, as in Kelly, Scream Icon was widely disseminated, both on the internet and on the streets of Los Angeles before Green Day used it in their concerts. Accordingly, Seltzer controlled the "first public appearance" of his work. See Harper & Row Publishers, Inc., 471 U.S. at 564. This tends to weigh in favor of the fair use of that work. Considering this factor as a whole, it weighs only slightly in Seltzer's favor.

The third factor looks to the quantitative amount and qualitative value of the original work used in relation to the justification for that use. SOFA Entm't, 709 F.3d at 1279; see also Campbell, 510 U.S. at 586. This factor captures the fact that an allegedly infringing work that copies little of the original is likely to be a fair use. See, e.g., SOFA Entm't, 709 F.3d at 1279 (holding that Ed Sullivan's seven-second introduction of the Four Seasons band was both qualitatively and quantitatively insignificant). Here, Green Day copied most of Scream Icon, both quantitatively and qualitatively. However, unlike an episode of the Ed Sullivan show or a book manuscript, Scream Icon is not meaningfully divisible. Given that fact, this court has acknowledged that this factor will not weigh against an alleged infringer, even when he copies the whole work, if he takes no more than is necessary for his intended use. Kelly, 336 F.3d at 820-21. As the Supreme Court has recognized, this factor necessarily overlaps somewhat with the first factor—the "extent of permissible copying varies with the purpose and character of the use." Campbell, 510 U.S. at 586-87. Here, as in Kelly, the use of the entire work was necessary to achieve Green Day's "new expression, meaning or message." Id. at 579; see also Kelly, 336 F.3d at 821. Thus, this factor does not weigh against Green Day.

The fourth factor asks what effect the allegedly infringing use has on the "potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). This factor should consider "the extent of market harm caused by the particular actions of the alleged infringer [and] also whether unrestricted and widespread conduct of the sort engaged in by the defendant would result in a substantially adverse impact on the potential market for the original." Campbell, 510 U.S. at 590 (internal quotation marks and citations omitted). Where the allegedly infringing use does not substitute for the original and serves a "different market function," such factor weighs in favor of fair use. Id. at 591; SOFA Entm't, 709 F.3d at 1280. At Seltzer's deposition, he repeatedly testified that the value of his work was unchanged, but that he subjectively did not care for Green Day's use of his art. He admitted that no one had ever told him that he would not buy his work as a result of Green Day's use; instead, he claimed that Scream Icon was "tarnished" for him personally, but he did not view the piece as having lost any value. Additionally, Green Daypresented evidence that its video backdrop did not perform the same "market function" as the original. The original, created six years before Green Day's use, was primarily intended as street art. Green Day's allegedly infringing use, on the other hand, was never placed on merchandise, albums, or promotional material and was used for only one song in the middle of a three hour touring show. In this context, there is no reasonable argument that conduct of the sort engaged in by Green Day is a substitute for the primary market for Seltzer's art. This factor also considers any impact on "traditional, reasonable, or likely to be developed markets." Ringgold, 126 F.3d at 81. At some point, according to Seltzer's declaration, Scream Icon was used in a music video by a band named "People." Seltzer provides no additional information about this licensing, including how much revenue he earned as a result, how the music video was used by the band, or how the music video used Scream Icon. Without further context, this fact does not suffice to show that Green Day's use harmed any existing market or a market that Seltzer was likely to develop. Thus, this factor weighs in Green Day's favor as well.

We turn first to "the purpose and character of the use." 17 U.S.C. § 107(1). Below, the district court found that this factor favored fair use because "[Bloomberg]'s work as a prominent gatherer and publisher of business and financial information serves an important public interest, for the public is served by the full, timely and accurate dissemination of business and financial news." Swatch II, 861 F. Supp. 2d at 340. Swatch argues that this conclusion was error for several reasons. First, Swatch contends that the district court improperly accepted Bloomberg's unsubstantiated claim that it had engaged in "news reporting." Swatch notes that Bloomberg itself has characterized its Bloomberg Professional service as delivering both financial "news" and "data," and argues that the district court erred in denying Swatch the chance to develop facts in discovery to show that the sound recording at issue here is the latter and not the former. Similarly, Swatch argues that the district court improperly denied Swatch the chance to develop facts relevant to Bloomberg's state of mind. Swatch acknowledges that the district court "credited [Swatch]'s allegations that [Bloomberg] was not authorized to access the Earnings Call and that [Bloomberg]'s publication of the Infringing Work violated [Swatch Group's] directive," Swatch II, 861 F. Supp. 2d at 343, but argues that Swatch should have been able to take discovery into whether Bloomberg knew at the time that obtaining and publishing the recording violated Swatch Group's directive. Swatch also argues that it should have been permitted to take discovery into whether Bloomberg Professional subscribers actually choose to access information about earnings calls by listening to recordings, or instead choose to read written transcripts or articles. More broadly, Swatch argues that the district court gave insufficient weight to the fact that Bloomberg's use was commercial and did not transform the underlying recording. We find these arguments unpersuasive and hold that the first statutory factor favors fair use here. To begin with, whether one describes Bloomberg's activities as "news reporting," "data delivery," or any other turn of phrase, there can be no doubt that Bloomberg's purpose in obtaining and disseminating the recording at issue was to make important financial information about Swatch Group available to investors and analysts. That kind of information is of critical importance to securities markets. Indeed, as Bloomberg points out, the Securities and Exchange Commission ("SEC") has mandated that when American companies disclose this kind of material nonpublic information, they must make it available to the public immediately. See Regulation FD, 17 C.F.R. § 243.100. At a minimum, such public dissemination of financial information serves this public purpose in the nature of news reporting. See Harper & Row, 471 U.S. at 561 ("News reporting is one of the examples enumerated in § 107 to 'give some idea of the sort of activities the courts might regard as fair use under the circumstances.'" (quoting S. Rep. No. 94-473, at 61 (1975)). Seizing on Bloomberg's citation to Regulation FD, Swatch protests that in crafting that regulation, the SEC expressly exempted "foreign private issuer[s]" like Swatch Group that are "incorporated or organized under the laws of [a] foreign country." 17 C.F.R. §§ 243.101(b), 230.405. In fact, as initially proposed, Regulation FD would have applied to such issuers, see Selective Disclosure and Insider Trading, 64 Fed. Reg. 72,590, 72,597 (Dec. 28, 1999), but the SEC ultimately "determined to exempt foreign private issuers . . . as it has in the past exempted them from certain U.S. reporting requirements such as Forms 10-Q and 8-K," Selective Disclosure and Insider Trading, 65 Fed. Reg. 51,716, 51,724 (Aug. 24, 2000). Swatch thus argues that giving weight to a public interest in the dissemination of important financial information in this case would in effect erase foreign issuers' exemption from Regulation FD and set up organizations like Bloomberg as private enforcers of U.S. public disclosure rules. This argument, however, misattributes to Regulation FD a role in the law of copyright. That regulation is relevant here only insofar as it provides additional support for a proposition that would be clear in any event: Investors and analysts have an interest in obtaining important financial information about companies whose securities are traded in American and other markets. The fact that the SEC has chosen not to require foreign issuers to follow certain disclosure rules imposed on domestic issuers in no way implies that information about foreign issuers is irrelevant. Accordingly, contrary to Swatch's suggestion, nothing in our decision today subjects Swatch Group or any other foreign issuer to the requirements of Regulation FD. Nor do we hold that a foreign issuer's failure to follow Regulation FD prevents it from enforcing its copyrights in the United States. We merely hold that where a financial research service obtains and disseminates important financial information about a foreign company in order to make that information available to investors and analysts, that purpose lends support to a finding of fair use. Swatch also stresses the commercial nature of Bloomberg's use. Section 107 expressly directs courts to consider whether the use "is of a commercial nature or is for nonprofit educational purposes," 17 U.S.C. § 107(1), and we have said that "[t]he greater the private economic rewards reaped by the secondary user (to the exclusion of broader public benefits), the more likely the first factor will favor the copyright holder and the less likely the use will be considered fair." Am. Geophysical Union v. Texaco Inc., 60 F.3d 913, 922 (2d Cir. 1994). It is undisputed here that Bloomberg is a commercial enterprise and that Bloomberg Professional is a subscription service available to paying users. At the same time, we have recognized that "[a]lmost all newspapers, books and magazines are published by commercial enterprises that seek a profit," Consumers Union of U.S., Inc. v. Gen. Signal Corp., 724 F.2d 1044, 1049 (2d Cir. 1983), and have discounted this consideration where "the link between [the defendant]'s commercial gain and its copying is . . . attenuated" such that it would be misleading to characterize the use as "commercial exploitation." Am. Geophysical Union, 60 F.3d at 922; see also Campbell, 510 U.S. at 594 (holding that "[i]t was error for the Court of Appeals to conclude that the commercial nature of [a secondary work] rendered it presumptively unfair"). Here, Swatch does not contest that Bloomberg Professional is a multifaceted research service, of which disseminating sound recordings of earnings calls is but one small part. Moreover, it would strain credulity to suggest that providing access to Swatch Group's earnings call more than trivially affected the value of that service. So while we will not ignore the commercial nature of Bloomberg's use, we assign it relatively little weight. Swatch also contends that Bloomberg acted in bad faith and that this should count against it. Regardless of what role good or bad faith plays in fair use analysis, see Blanch, 467 F.3d at 255-56, we need not tarry over it here. Even assuming that Bloomberg was fully aware that its use was contrary to Swatch Group's instructions, Bloomberg's overriding purpose here was not to "scoop[]" Swatch or "supplant the copyright holder's commercially valuable right of first publication," Harper & Row, 471 U.S. at 562, but rather simply to deliver newsworthy financial information to investors and analysts. That kind of activity, whose protection lies at the core of the First Amendment, would be crippled if the news media and similar organizations were limited to sources of information that authorize disclosure. See generally New York Times Co. v. United States, 403 U.S. 713, 91 S. Ct. 2140, 29 L. Ed. 2d 822 (1971). The Supreme Court has also instructed courts analyzing the first fair use factor to consider the transformativeness of the use—that is, whether "the new work merely supersedes the objects of the original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Campbell, 510 U.S. at 579 (internal citations, quotation marks, and alterations omitted). While a transformative use generally is more likely to qualify as fair use, "transformative use is not absolutely necessary for a finding of fair use." Id.; see also Sony Corp. of Am. v. Universal City Studios, Inc ., 464 U.S. 417, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984) (finding a non-transformative use to be a fair use). In the context of news reporting and analogous activities, moreover, the need to convey information to the public accurately may in some instances make it desirable and consonant with copyright law for a defendant to faithfully reproduce an original work without alteration. Courts often find such uses transformative by emphasizing the altered purpose or context of the work, as evidenced by surrounding commentary or criticism. See, e.g., Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 609-610 (2d Cir. 2006); Nunez v. Caribbean Int'l News Corp., 235 F.3d 18, 22-23 (1st Cir. 2000). Here, Bloomberg provided no additional commentary or analysis of Swatch Group's earnings call. But by disseminating not just a written transcript or article but an actual sound recording, Bloomberg was able to convey with precision not only the raw data of the Swatch Group executives' words, but also more subtle indications of meaning inferable from their hesitation, emphasis, tone of voice, and other such aspects of their delivery. This latter type of information may be just as valuable to investors and analysts as the former, since a speaker's demeanor, tone, and cadence can often elucidate his or her true beliefs far beyond what a stale transcript or summary can show. As courts have long recognized in the context of witness testimony, "'a cold transcript contains only the dead body of the evidence, without its spirit,'" and "cannot reveal . . . '[the speaker's] hesitation, his doubts, his variations of language, his confidence or precipitancy, his calmness or consideration.'" Zhou Yun Zhang v. INS, 386 F.3d 66, 73-74 (2d Cir. 2004) (quoting Regina v. Bertrand, L.R. [1867] 1 L.R.P.C. 520, 535), overruled on other grounds by Shi Liang Lin v. United States DOJ, 494 F.3d 296, 305 (2d Cir. 2007). Furthermore, a secondary work "can be transformative in function or purpose without altering or actually adding to the original work." A.V. ex rel. Vanderhye v. iParadigms, LLC, 562 F.3d 630, 639 (4th Cir. 2009) (holding that making an exact digital copy of a student's thesis for the purpose of determining whether it included plagiarism is a fair use); see also Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1165 (9th Cir. 2007) (holding that a search engine's publication of low-resolution, thumbnail copies of copyrighted images was "highly transformative" because the thumbnails were "incorporate[ed] . . . into a new work, namely, an electronic reference tool"). Here, notwithstanding that the data disseminated by Bloomberg was identical to what Swatch Group had disseminated, the two works had different messages and purposes. To begin with, while Swatch Group purported to convey true answers to the analysts' questions and to justify the propriety and reliability of its published earnings statement, Bloomberg made no representation one way or another as to whether the answers given by Swatch Group executives were true or reliable. Nor did Bloomberg purport to support the propriety or reliability of Swatch Group's earnings statement. Bloomberg was simply revealing the newsworthy information of what Swatch Group executives had said. Bloomberg's message—"This is what they said"—is a very different message from Swatch Group's—"This is what you should believe." Moreover, Swatch Group intended to exclude members of the press and to restrict the information supplied by its executives to a relatively small group of analysts who had identified themselves to the company in advance. Bloomberg's objective in rebroadcasting the call, by contrast, was to make this information public, defeating Swatch Group's effort to restrict access. Bloomberg's purpose, in other words, was to publish this factual information to an audience from which Swatch Group's purpose was to withhold it. These differences give Bloomberg's use at least an arguably transformative character. In any event, regardless of how transformative the use is, we conclude that the first fair use factor, focusing on the purpose and character of the secondary use, favors fair use. We of course recognize that a news reporting purpose by no means guarantees a finding of fair use. See Harper & Row, 471 U.S. at 557. After all, "[t]he promise of copyright would be an empty one if it could be avoided merely by dubbing the infringement a fair use 'news report.'" Id. A news organization thus may not freely copy creative expression solely because the expression itself is newsworthy. Nevertheless, we agree with the district court's conclusion that, under the unusual circumstances of this case, the purpose and character of Bloomberg's unaltered dissemination of Swatch Group's expression weighs in favor of fair use, for two reasons. First, as noted above, by disseminating a full, unadulterated recording of the earnings call, Bloomberg was able to convey valuable factual information that would have been impaired if Bloomberg had undertaken to alter the speech of the Swatch Group executives by interjecting its own interpretations. As we explained in a fair use case involving verbatim copying of a written work, "[w]here an evaluation or description is being made, copying the exact words may be the only valid way precisely to report the evaluation." Consumers Union, 724 F.2d at 1049-50; see also Harper & Row, 471 U.S. at 563 (noting that direct copying may in some instances be "necessary adequately to convey the facts"). So too here, copying the exact spoken performance of Swatch Group's executives was reasonably necessary to convey their full meaning. Bloomberg's faithful reproduction thus served "the interest of accuracy, not piracy." Consumers Union, 724 F.2d at 1049. Second, Bloomberg's use did no harm to the legitimate copyright interests of the original author. Importantly, Swatch has admitted that it "did not seek to profit from the publication of the February 8, 2011 Earnings Call in audio or written format." J.A. 294. The copyright-protected aspects of the earnings call—that is, the manner by which the facts were expressed—thus were of no value to Swatch or Swatch Group except insofar as they served to convey important information to the analysts in attendance. But Bloomberg's copying of the Swatch Group executives' words, as needed to communicate factual information about the company's earnings report, in no way diminished Swatch Group's ability to communicate with analysts, and thus caused no harm to Swatch's copyright interests. In this way, the case at bar stands in stark contrast to a case like Harper & Row, where a magazine disseminated an unpublished excerpt of President Ford's memoirs. See Harper & Row, 471 U.S. at 542. This kind of gun-jumping, which scooped the publication of the copyrighted work and, in doing so, did considerable harm to the value of the original author's copyright, is not present here. Our prior decisions in Nihon Keizai Shimbun, Inc. v. Comline Business Data, Inc., 166 F.3d 65 (2d Cir. 1999), Wainwright Securities, Inc. v. Wall Street Transcript Corp., 558 F.2d 91 (2d Cir. 1977), and Financial Information, Inc. v. Moody's Investors Service, Inc. ("FII"), 751 F.2d 501 (2d Cir. 1984), on which Swatch relies, are not to the contrary. In those cases, we rejected fair use arguments pressed by defendants who purported to be serving the public by providing access to important financial information. In Nihon and Wainwright, we stressed that the defendants had not supplemented or otherwise transformed the plaintiffs' works. Instead, they had simply translated Japanese business articles into English, Nihon, 166 F.3d at 69, or recounted the critical conclusions from research reports about major industrial and financial corporations, Wainwright, 558 F.2d at 93 & n.1. In FII, we rejected a fair use defense by a ratings agency that had copied information about municipal bond redemptions compiled by a competing financial publisher. FII, 751 F.2d at 502-03. Criticizing the district court's conclusion that the defendant's use served a "public function," we stated that to so hold "would, it seems to us, state a rule that whenever there is a market for information, the paid delivery of goods to that market rises to a public function." Id. at 509. We rejected such a rule, finding that it would "distort" proper fair use analysis. Id. In all three of those cases, however, the defendants attempted to use the banner of newsworthiness to supersede the core objects of original works whose production critically depended upon copyright protection. Finding fair use in those cases would have severely impeded the ability of news and research organizations to obtain payment for their expression, imperiling the economic foundation of vital industries. But unlike the arguments we rejected in Nihon, Wainwright, and FII, our decision today does not rest upon the newsworthiness of the original expression alone. To the contrary, we also place great weight on the absence of harm to the original author's legitimate copyright interests. Swatch's reliance on our prior cases is thus misplaced. The discovery Swatch seeks would not alter our analysis. With respect to the request for discovery into whether Bloomberg delivered "news" or "data" to its subscribers, such a distinction raises a semantic rather than factual dispute. It is undisputed that Bloomberg gave subscribers access to the full, unaltered sound recording of Swatch Group's earnings call as part of its paid financial research service. That is sufficient for present purposes. There is likewise no need for further discovery into Bloomberg's good or bad faith, since resolving that issue in Swatch's favor would not affect the outcome of our analysis. We also see no need to resolve how many of Bloomberg's subscribers chose to listen to the sound recording in question rather than read a written transcript or article. As we have explained, because the sound recording conveys information that a transcript or article cannot, the recording has independent value, regardless of how many Bloomberg subscribers chose to avail themselves of that independent value in this instance. This first factor accordingly favors fair use.

The second statutory fair use factor concerns "the nature of the copyrighted work." 17 U.S.C. § 107(2). This factor accounts for the fact that "some works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied." Campbell, 510 U.S. at 586. As relevant here, this factor requires us to consider the extent of Swatch's copyright in the recording—the "thickness" or "thinness" of Swatch's exclusive rights—as well as whether or not the recording had been published at the time of Bloomberg's use. See id. (citing cases). The district court determined that this factor favored fair use because Swatch's copyright was "at best . . . 'thin'" and because "the first publication of Swatch Group's expression occurred prior to [Bloomberg]'s publication of the Infringing Work." Swatch II, 861 F. Supp. 2d at 341. Swatch argues that the district court erred in concluding that the recording had been published. Swatch points out that the Copyright Act contemplates two methods of publishing an audio recording: "the distribution of . . . phonorecords of a work to the public by sale or other transfer of ownership, or by rental, lease, or lending," or "offering to distribute . . . phonorecords to a group of persons for purposes of further distribution, public performance, or public display." 17 U.S.C. § 101 (defining "publication"). "Phonorecords," in turn, are defined as "material objects in which sounds . . . are fixed . . . and from which the sounds can be perceived, reproduced, or otherwise communicated." Id. Applying these definitions, Swatch contends that the sound recording of the earnings call has never been published. Simply put, Swatch has never, before or after Bloomberg's use, "distribut[ed]" a CD or other "material object" embodying the spoken commentary on the earnings call "to the public," nor has it ever "offer[ed] to distribute" a phonorecord of the call to any "group of persons for purposes of further distribution, public performance, or public display." Swatch is unquestionably correct that the earnings call is unpublished under the definition of "publication" set forth in § 101. But that technical definition does not control our analysis of this aspect of the second fair use factor. While we will consider the statutory definition, we also will not blind ourselves to the fact that Swatch Group invited over three hundred investment analysts from around the globe to the earnings call, out of which over a hundred actually attended. Thus, even though the sound recording remains statutorily unpublished, it is clear that Swatch was not deprived of the ability to "control the first public appearance of [its] expression," including "when, where, and in what form" it appeared. Harper & Row, 471 U.S. at 564. Swatch insists that because the definitions in § 101 by their terms apply for all purposes under the Copyright Act "[e]xcept as otherwise provided in this title," 17 U.S.C. § 101, the statutory definition of "publication" must control. Not so. While in general, "[s]tatutory definitions control the meaning of statutory words," Burgess v. United States, 553 U.S. 124, 129, 128 S. Ct. 1572, 170 L. Ed. 2d 478 (2008) (quoting Lawson v. Suwanee Fruit & S.S. Co., 336 U.S. 198, 201, 69 S. Ct. 503, 93 L. Ed. 611 (1949)), in this case, no variant of the word "publish" appears in the text of the second fair use factor in § 107. Whether or not a work was published thus enters into our analysis of this factor as a judicial gloss on "the nature of the copyrighted work." That gloss, of course, is firmly grounded in fair use's common law origins and the legislative history of the 1976 Copyright Act. See Harper & Row, 471 U.S. at 552-54. To the extent the text of § 107 mentions publication, it is only in a closing proviso cautioning that "[t]he fact that a work is unpublished shall not itself bar a finding of fair use if such finding is made upon consideration of all the above factors." Congress added this proviso to § 107 in 1992, see Pub. L. No. 102-492, 106 Stat. 3145 (1992), to clarify, in response to certain decisions of this Court, that there is no "per se rule barring any fair use of unpublished works." H.R. Rep. No. 102-836, at 4, 9 (1992) (discussing New Era Publs. Int'l v. Henry Holt & Co.., 873 F.2d 576 (2d Cir. 1989), and Salinger v. Random House, Inc., 811 F.2d 90 (2d Cir. 1987)). This proviso in no way limits our consideration of a work's publication status to the statutory definition of "publication" in § 101. To the contrary, the proviso directs that if we find a work to be "unpublished," however that term is understood, our analysis of the four statutory factors, including "the nature of the copyrighted work," cannot end there. Limiting our consideration of a work's publication status to the statutory definition, moreover, would obscure the different purposes served by the statutory definition and the judicial gloss on "the nature of the copyrighted work" in the context of fair use. The statutory concept of "publication" serves numerous purposes, such as triggering the requirement to deposit a copy with the Library of Congress, see 17 U.S.C. § 407, measuring the copyright term for certain categories of works, see id. § 302(c)-(e), setting the circumstances under which works by foreign authors are protected, see id. § 104(b), and determining the legal effect of registration, see id. §§ 410(c), 412. See also 1 Nimmer on Copyright § 4.01 (explaining the significance of publication). Publication as a judicial gloss on "the nature of the copyrighted work," by contrast, aims to take account of "the author's right to control the first public appearance of his expression," Harper & Row, 471 U.S. at 564, which in turn forms part of our "open-ended and context-sensitive inquiry" into whether allowing the use in question would serve the goals of copyright, Blanch, 467 F.3d at 251. This is not the first time that we have found that the second statutory factor favors fair use even though the work in question was technically unpublished under the statutory definition, see Diamond v. Am-Law Pub. Corp., 745 F.2d 142, 144, 148 (2d Cir. 1984), and courts in fact commonly look past the statutory definition when considering this issue, see, e.g., Rotbart v. J.R. O'Dwyer Co., Inc., No. 94 Civ. 2091 (JSM), 1995 U.S. Dist. LEXIS 1315, 1995 WL 46625, at *4 (S.D.N.Y. Feb. 7, 1995) (finding that an unfixed, undisseminated talk, delivered publicly, had been "de facto published" for purposes of fair use); see also 4 Nimmer on Copyright § 13.05 [A][2][b][ii] ("If the author does not seek confidentiality, fair use is not necessarily precluded even as to an unpublished work.").We accordingly agree with the district court that although the sound recording is statutorily unpublished, because Swatch Group publicly disseminated the spoken performance embodied in the recording before Bloomberg's use, the publication status of the work favors fair use. Swatch does not challenge the district court's determination that Swatch's copyright in the earnings call is "at best . . . 'thin,'" Swatch II, 861 F. Supp. 2d at 341, nor could it. It is well established that "the scope of fair use is greater with respect to factual than non-factual works." New Era Publ'ns, 904 F.2d at 157. Moreover, [e]ven within the field of fact works, there are gradations as to the relative proportion of fact and fancy. One may move from sparsely embellished maps and directories to elegantly written biography. The extent to which one must permit expressive language to be copied, in order to assure dissemination of the underlying facts, will thus vary from case to case. Harper & Row, 471 U.S. at 563 (quoting Robert A. Gorman, Fact or Fancy? The Implications for Copyright, 29 J. Copyright Soc'y 560, 561 (1982)). There can be no doubt as to the manifestly factual character of the earnings call in this case. The entire copyrighted portion of the call consists of Swatch Group executives explaining the company's financial performance and outlook to a group of investment analysts. And while we assume without deciding in this appeal that the call contained sufficient original expression—in the form of the executives' tone, cadence, accents, and particular choice of words—to be copyrightable, the purpose of the call was not in any sense to showcase those forms of expression. Rather, the call's sole purpose was to convey financial information about the company to investors and analysts. In light of the thinness of Swatch's copyright, as well as Swatch Group's prior dissemination of its executives' expression, we find that the second statutory factor favors fair use.

We turn now to "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." 17 U.S.C. § 107(3). This factor asks whether "the quantity and value of the materials used are reasonable in relation to the purpose of the copying." Campbell, 510 U.S. at 586 (internal citations and quotation marks omitted). In general, "the more of a copyrighted work that is taken, the less likely the use is to be fair." Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 109 (2d Cir. 1998). It is undisputed here that Bloomberg used the entire work. The district court acknowledged that "this generally weighs against fair use," but found that the public interest in the information contained in the recording "is better served by the dissemination of that information in its entirety, including the incidents of oral speech that do not translate onto the page but color the purely factual content." Swatch II, 861 F. Supp. 2d at 342. Swatch argues that the district court improperly resolved this factor in Bloomberg's favor because, as it also argued with respect to the first fair use factor, there are genuine disputes of material fact regarding whether Bloomberg subscribers glean information about earnings calls by listening to audio recordings or instead by reading a written transcript or article. We are unpersuaded. As an initial matter, we do not understand the district court to have affirmatively weighed the third statutory fair use factor in Bloomberg's favor. Such a holding would have been novel, as "[n]either our court nor any of our sister circuits has ever ruled that the copying of an entire work favors fair use." Bill Graham Archives, 448 F.3d at 613. Rather, we believe that the district court found this factor neutral, refusing to weigh it in Swatch's favor despite Bloomberg's use of the entire recording because of the public interest in the information embodied in the recording. That holding is entirely consistent with our case law. As we have recognized, a number of courts "have concluded that such copying does not necessarily weigh against fair use because copying the entirety of a work is sometimes necessary to make a fair use." Id. (citing cases); see also A.V. ex rel. Vanderhye, 562 F.3d at 642, 645 (finding copying of an entire work to be fair use); Perfect 10, 508 F.3d at 1167-68 (same). For the reasons already explained in our discussion of the first fair use factor, we agree with the district court that Bloomberg's use of the entire recording was reasonable in light of its purpose of disseminating important financial information to investors and analysts. The recording has independent informational value over and above the value of a written transcript or article, regardless of how many Bloomberg subscribers took advantage of that value in this instance. Like the district court, we accordingly weigh this factor in neither party's favor.

The final fair use factor considers "the effect of the use upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). In Harper & Row, the Supreme Court described this factor as "undoubtedly the single most important element of fair use." 471 U.S. at 566. This factor "requires courts to consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also whether unrestricted and widespread conduct of the sort engaged in by the defendant would result in a substantially adverse impact on the potential market for the original." Campbell, 510 U.S. at 590 (quotation marks and alterations omitted). We have described this factor as "requir[ing] a balancing of 'the benefit the public will derive if the use is permitted and the personal gain the copyright owner will receive if the use is denied.'" Bill Graham Archives, 448 F.3d at 613 (quoting MCA, Inc. v. Wilson, 677 F.2d 180, 183 (2d Cir. 1981)). The district court weighed this factor in favor of fair use, noting that "the relevant market effect is that which stems from [Bloomberg]'s use of the original expression of Swatch Group's senior officers." Swatch II, 861 F. Supp. 2d at 342. The district court found "[n]othing in the record [that] suggests any possible market effect stemming from [Bloomberg]'s use." Id. We agree, especially in view of the obvious and furthermore conceded fact that Swatch had no interest in the exploitation of the copyright-protected aspects of the call. Swatch argues that the district court's analysis was erroneous because it again assumed that affording investors and analysts access to the recording, as opposed to a written transcript or article, served the public interest. As we have already explained, we see nothing mistaken in that finding. Swatch also contends that it was improperly denied the opportunity to take discovery into the existence of a market for audio recordings of earnings calls conducted by foreign companies that, like Swatch Group, are exempt from Regulation FD. Swatch has admitted that it acquired the recording without intending to profit from publishing and selling it. Any discovery thus would concern a potential market, as yet untapped by Swatch, for recordings of exempt earnings calls. While the loss of a potential yet untapped market can be cognizable under the fourth fair use factor, the potential market here is defined so narrowly that it begins to partake of circular reasoning. As the Nimmer treatise has observed, "it is a given in every fair use case that plaintiff suffers a loss of a potential market if that potential is defined as the theoretical market for licensing the very use at bar." 4 Nimmer on Copyright § 13.05[A][4]. To guard against this "vice of circular reasoning," our case law limits our consideration to a use's "impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets." Am. Geophysical Union, 60 F.3d at 930-31. The hypothesized market for audio recordings of earnings calls convened by foreign companies that are exempt from Regulation FD cannot meet this standard. Moreover, to the extent that a financial news or research organization might be willing to pay to obtain such recordings, we must bear in mind that while "[t]he immediate effect of our copyright law is to secure a fair return for an 'author's' creative labor," the "ultimate aim is, by this incentive, to stimulate creativity for the general public good." Fogerty v. Fantasy, Inc., 510 U.S. 517, 526-27, 114 S. Ct. 1023, 127 L. Ed. 2d 455 (1994) (quoting Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 156, 95 S. Ct. 2040, 45 L. Ed. 2d 84 (1975)). Here, the possibility of receiving licensing royalties played no role in stimulating the creation of the earnings call. Indeed, Swatch affirmatively argues that it does not know whether there is a potential market for this kind of recording, and cannot know without obtaining discovery from Bloomberg. Moreover, as we previously noted in relation to the first statutory fair use factor, the context of the earnings call makes perfectly plain that its purpose was to enable Swatch Group executives to disseminate financial information about the company to particular analysts in a way that they believed would be advantageous. It is that calculation of advantageousness, and not the possibility of receiving royalties, that induces Swatch Group and other similarly situated companies to hold earnings calls. Cf. New York Mercantile Exch., Inc. v. IntercontinentalExchange, Inc., 497 F.3d 109, 118 (2d Cir. 2007). Put differently, the "value" of the copyrighted expression for Swatch Group in this case lay not in its capacity to generate licensing royalties, but rather in its capacity to convey important information about the company to the investment analysts in attendance. Bloomberg's use in no way diminishes that value. At most, Bloomberg's use had the effect of depriving Swatch Group of the ability to know and control precisely who heard the call. But whatever cognizable interest Swatch Group has in maintaining that ability, it merits little weight here. As we recently observed in a related context, "a [f]irm's ability to make news—by issuing a [report] that is likely to affect the market price of a security—does not give rise to a right for it to control who breaks that news and how." Barclays Capital Inc. v. Theflyonthewall.com, Inc., 650 F.3d 876, 907 (2d Cir. 2011). We accordingly agree with the district court that the fourth statutory factor weighs in favor of fair use.

A. Factor One (1) Transformative purpose. Campbell's explanation of the first factor's inquiry into the "purpose and character" of the secondary use focuses on whether the new work, "in Justice Story's words, . . . merely 'supersede[s] the objects' of the original creation, . . . or instead adds something new, with a further purpose . . . . [I]t asks, in other words, whether and to what extent the new work is 'transformative.'" 510 U.S. at 578-579 (citations omitted). While recognizing that a transformative use is "not absolutely necessary for a finding of fair use," the opinion further explains that the "goal of copyright, to promote science and the arts, is generally furthered by the creation of transformative works" and that "[s]uch works thus lie at the heart of the fair use doctrine's guarantee of breathing space within the confines of copyright." Id. at 579. In other words, transformative uses tend to favor a fair use finding because a transformative use is one that communicates something new and different from the original or expands its utility, thus serving copyright's overall objective of contributing to public knowledge. The word "transformative" cannot be taken too literally as a sufficient key to understanding the elements of fair use. It is rather a suggestive symbol for a complex thought, and does not mean that any and all changes made to an author's original text will necessarily support a finding of fair use. The Supreme Court's discussion in Campbell gave important guidance on assessing when a transformative use tends to support a conclusion of fair use. The defendant in that case defended on the ground that its work was a parody of the original and that parody is a time-honored category of fair use. Explaining why parody makes a stronger, or in any event more obvious, claim of fair use than satire, the Court stated, [T]he heart of any parodist's claim to quote from existing material . . . is the use of . . . a prior author's composition to . . .comment[] on that author's works. . . . If, on the contrary, the commentary has no critical bearing on the substance or style of the original composition, which the alleged infringer merely uses to get attention or to avoid the drudgery in working up something fresh, the claim to fairness in borrowing from another's work diminishes accordingly (if it does not vanish). . . . Parody needs to mimic an original to make its point, and so has some claim to use the creation of its victim's . . . imagination, whereas satire can stand on its own two feet and so requires justification for the very act of borrowing. Id. at 580-81 (emphasis added). In other words, the would-be fair user of another's work must have justification for the taking. A secondary author is not necessarily at liberty to make wholesale takings of the original author's expression merely because of how well the original author's expression would convey the secondary author's different message. Among the best recognized justifications for copying from another's work is to provide comment on it or criticism of it. A taking from another author's work for the purpose of making points that have no bearing on the original may well be fair use, but the taker would need to show a justification. This part of the Supreme Court's discussion is significant in assessing Google's claim of fair use because, as discussed extensively below, Google's claim of transformative purpose for copying from the works of others is to provide otherwise unavailable information about the originals. A further complication that can result from oversimplified reliance on whether the copying involves transformation is that the word "transform" also plays a role in defining "derivative works," over which the original rights holder retains exclusive control. Section 106 of the Act specifies the "exclusive right[]" of the copyright owner "(2) to prepare derivative works based upon the copyrighted work." See 17 U.S.C. § 106. The statute defines derivative works largely by example, rather than explanation. The examples include "translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgement, condensation," to which list the statute adds "any other form in which a work may be . . . transformed." 17 U.S.C. § 101 (emphasis added).As we noted in Authors Guild, Inc. v. HathiTrust, "[p]aradigmatic examples of derivative works include the translation of a novel into another language, the adaptation of a novel into a movie or play, or the recasting of a novel as an e-book or an audiobook." 755 F.3d 87, 95 (2d Cir. 2014). While such changes can be described as transformations, they do not involve the kind of transformative purpose that favors a fair use finding. The statutory definition suggests that derivative works generally involve transformations in the nature of changes of form . 17 U.S.C. § 101. By contrast, copying from an original for the purpose of criticism or commentary on the originalor provision of information about it, tends most clearly to satisfy Campbell's notion of the "transformative" purpose involved in the analysis of Factor One. With these considerations in mind, we first consider whether Google's search and snippet views functions satisfy the first fair use factor with respect to Plaintiffs' rights in their books. (The question whether these functions might infringe upon Plaintiffs' derivative rights is discussed in the next Part.) (2) Search Function. We have no difficulty concluding that Google's making of a digital copy of Plaintiffs' books for the purpose of enabling a search for identification of books containing a term of interest to the searcher involves a highly transformative purpose, in the sense intended by Campbell. Our court's exemplary discussion in HathiTrust informs our ruling. That case involved a dispute that is closely related, although not identical, to this one. Authors brought claims of copyright infringement against HathiTrust, an entity formed by libraries participating in the Google Library Project to pool the digital copies of their books created for them by Google. The suit challenged various usages HathiTrust made of the digital copies. Among the challenged uses was HathiTrust's offer to its patrons of "full-text searches," which, very much like the search offered by Google Books to Internet users, permitted patrons of the libraries to locate in which of the digitized books specific words or phrases appeared. 755 F.3d at 98. (HathiTrust's search facility did not include the snippet view function, or any other display of text.) We concluded that both the making of the digital copies and the use of those copies to offer the search tool were fair uses. Id. at 105. Notwithstanding that the libraries had downloaded and stored complete digital copies of entire books, we noted that such copying was essential to permit searchers to identify and locate the books in which words or phrases of interest to them appeared. Id. at 97. We concluded "that the creation of a full-text searchable database is a quintessentially transformative use . . . [as] the result of a word search is different in purpose, character, expression, meaning, and message from the page (and the book) from which it is drawn." Id. We cited A.V. ex rel. Vanderhye v. iParadigms, LLC, 562 F.3d 630, 639-40 (4th Cir. 2009), Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1165 (9th Cir. 2007), and Kelly v. Arriba Soft Corp., 336 F.3d 811, 819 (9th Cir. 2003) as examples of cases in which courts had similarly found the creation of complete digital copies of copyrighted works to be transformative fair uses when the copies "served a different function from the original." HathiTrust, 755 F.3d at 97. As with HathiTrust (and iParadigms), the purpose of Google's copying of the original copyrighted books is to make available significant information about those books, permitting a searcher to identify those that contain a word or term of interest, as well as those that do not include reference to it. In addition, through the ngrams tool, Google allows readers to learn the frequency of usage of selected words in the aggregate corpus of published books in different historical periods. We have no doubt that the purpose of this copying is the sort of transformative purpose described in Campbell as strongly favoring satisfaction of the first factor. We recognize that our case differs from HathiTrust in two potentially significant respects. First, HathiTrust did not "display to the user any text from the underlying copyrighted work," 755 F.3d at 91, whereas Google Books provides the searcher with snippets containing the word that is the subject of the search. Second, HathiTrust was a nonprofit educational entity, while Google is a profit-motivated commercial corporation. We discuss those differences below. (3) Snippet View. Plaintiffs correctly point out that this case is significantly different from HathiTrust in that the Google Books search function allows searchers to read snippets from the book searched, whereas HathiTrust did not allow searchers to view any part of the book. Snippet view adds important value to the basic transformative search function, which tells only whether and how often the searched term appears in the book. Merely knowing that a term of interest appears in a book does not necessarily tell the searcher whether she needs to obtain the book, because it does not reveal whether the term is discussed in a manner or context falling within the scope of the searcher's interest. For example, a searcher seeking books that explore Einstein's theories, who finds that a particular book includes 39 usages of "Einstein," will nonetheless conclude she can skip that book if the snippets reveal that the book speaks of "Einstein" because that is the name of the author's cat. In contrast, the snippet will tell the searcher that this is a book she needs to obtain if the snippet shows that the author is engaging with Einstein's theories. Google's division of the page into tiny snippets is designed to show the searcher just enough context surrounding the searched term to help her evaluate whether the book falls within the scope of her interest (without revealing so much as to threaten the author's copyright interests). Snippet view thus adds importantly to the highly transformative purpose of identifying books of interest to the searcher. With respect to the first factor test, it favors a finding of fair use (unless the value of its transformative purpose is overcome by its providing text in a manner that offers a competing substitute for Plaintiffs' books, which we discuss under factors three and four below). (4) Google's Commercial Motivation. Plaintiffs also contend that Google's commercial motivation weighs in their favor under the first factor. Google's commercial motivation distinguishes this case from HathiTrust, as the defendant in that case was a non-profit entity founded by, and acting as the representative of, libraries. Although Google has no revenues flowing directly from its operation of the Google Books functions, Plaintiffs stress that Google is profit-motivated and seeks to use its dominance of book search to fortify its overall dominance of the Internet search market, and that thereby Google indirectly reaps profits from the Google Books functions. For these arguments Plaintiffs rely primarily on two sources. First is Congress's specification in spelling out the first fair use factor in the text of § 107 that consideration of the "purpose and character of the [secondary] use" should "include[e] whether such use is of a commercial nature or is for nonprofit educational purposes." Second is the Supreme Court's assertion in dictum in Sony Corporation of America v. Universal City Studios, Inc, that "every commercial use of copyrighted material is presumptively . . . unfair." 464 U.S. 417, 451, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984). If that were the extent of precedential authority on the relevance of commercial motivation, Plaintiffs' arguments would muster impressive support. However, while the commercial motivation of the secondary use can undoubtedly weigh against a finding of fair use in some circumstances, the Supreme Court, our court, and others have eventually recognized that the Sony dictum was enormously overstated. The Sixth Circuit took the Sony dictum at its word in Acuff-Rose Music, Inc. v. Campbell, concluding that, because the defendant rap music group's spoof of the plaintiff's ballad was done for profit, it could not be fair use. 972 F.2d 1429, 1436-1437 (6th Cir. 1992). The Supreme Court reversed on this very point, observing that "Congress could not have intended" such a broad presumption against commercial fair uses, as "nearly all of the illustrative uses listed in the preamble paragraph of § 107 . . . are generally conducted for profit in this country." Campbell, 510 U.S. at 584 (internal quotation marks and citations omitted). The Court emphasized Congress's statement in the House Report to the effect that the commercial or nonprofit character of a work is "not conclusive" but merely "a fact to be 'weighed along with other[s] in fair use decisions." Id. at 585 (quoting H.R. Rep. No. 94-1476, at 66 (1976)). In explaining the first fair use factor, the Court clarified that "the more transformative the [secondary] work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Id. at 579. Our court has since repeatedly rejected the contention that commercial motivation should outweigh a convincing transformative purpose and absence of significant substitutive competition with the original. See Cariou v. Prince, 714 F.3d 694, 708 (2d Cir. 2013), cert. denied, 134 S. Ct. 618, 187 L. Ed. 2d 411 (2013) ("The commercial/nonprofit dichotomy concerns the unfairness that arises when a secondary user makes unauthorized use of copyrighted material to capture significant revenues as a direct consequence of copying the original work. This factor must be applied with caution because, as the Supreme Court has recognized, Congress could not have intended a rule that commercial uses are presumptively unfair. Instead, the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use.") (internal quotation marks, citations, and alterations omitted); Castle Rock Entm't, Inc. v. Carol Pub. Grp., Inc., 150 F.3d 132, 141-42 (2d Cir. 1998) ("We . . . do not give much weight to the fact that the secondary use was for commercial gain. The more critical inquiry under the first factor and in fair use analysis generally is whether the allegedly infringing work merely supersedes the original work or instead adds something new, with a further purpose or different character, altering the first with new meaning or message, in other words whether and to what extent the new work is transformative.") (internal quotation marks, citations, and alterations omitted). While we recognize that in some circumstances, a commercial motivation on the part of the secondary user will weigh against her, especially, as the Supreme Court suggested, when a persuasive transformative purpose is lacking, Campbell, 510 U.S. at 579, we see no reason in this case why Google's overall profit motivation should prevail as a reason for denying fair use over its highly convincing transformative purpose, together with the absence of significant substitutive competition, as reasons for granting fair use. Many of the most universally accepted forms of fair use, such as news reporting and commentary, quotation in historical or analytic books, reviews of books, and performances, as well as parody, are all normally done commercially for profit.20

B. Factor Two The second fair use factor directs consideration of the "nature of the copyrighted work." While the "transformative purpose" inquiry discussed above is conventionally treated as a part of first factor analysis, it inevitably involves the second factor as well. One cannot assess whether the copying work has an objective that differs from the original without considering both works, and their respective objectives. The second factor has rarely played a significant role in the determination of a fair use dispute. See William F. Patry, Patry on Fair Use § 4.1 (2015). The Supreme Court in Harper & Row made a passing observation in dictum that, "[t]he law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy." 471 U.S. 539, 563, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985). Courts have sometimes speculated that this might mean that a finding of fair use is more favored when the copying is of factual works than when copying is from works of fiction. However, while the copyright does not protect facts or ideas set forth in a work, it does protect that author's manner of expressing those facts and ideas. At least unless a persuasive fair use justification is involved, authors of factual works, like authors of fiction, should be entitled to copyright protection of their protected expression. The mere fact that the original is a factual work therefore should not imply that others may freely copy it. Those who report the news undoubtedly create factual works. It cannot seriously be argued that, for that reason, others may freely copy and re-disseminate news reports. In considering the second factor in HathiTrust, we concluded that it was "not dispositive," 755 F.3d at 98, commenting that courts have hardly ever found that the second factor in isolation played a large role in explaining a fair use decision. The same is true here. While each of the three Plaintiffs' books in this case is factual, we do not consider that as a boost to Google's claim of fair use. If one (or all) of the plaintiff works were fiction, we do not think that would change in any way our appraisal. Nothing in this case influences us one way or the other with respect to the second factor considered in isolation. To the extent that the "nature" of the original copyrighted work necessarily combines with the "purpose and character" of the secondary work to permit assessment of whether the secondary work uses the original in a "transformative" manner, as the term is used in Campbell, the second factor favors fair use not because Plaintiffs' works are factual, but because the secondary use transformatively provides valuable information about the original, rather than replicating protected expression in a manner that provides a meaningful substitute for the original.

C. Factor Three The third statutory factor instructs us to consider "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." The clear implication of the third factor is that a finding of fair use is more likely when small amounts, or less important passages, are copied than when the copying is extensive, or encompasses the most important parts of the original.The obvious reason for this lies in the relationship between the third and the fourth factors. The larger the amount, or the more important the part, of the original that is copied, the greater the likelihood that the secondary work might serve as an effectively competing substitute for the original, and might therefore diminish the original rights holder's sales and profits. (1) Search Function. The Google Books program has made a digital copy of the entirety of each of Plaintiffs' books. Notwithstanding the reasonable implication of Factor Three that fair use is more likely to be favored by the copying of smaller, rather than larger, portions of the original, courts have rejected any categorical rule that a copying of the entirety cannot be a fair use.Complete unchanged copying has repeatedly been found justified as fair use when the copying was reasonably appropriate to achieve the copier's transformative purpose and was done in such a manner that it did not offer a competing substitute for the original.The Supreme Court said in Campbell that "the extent of permissible copying varies with the purpose and character of the use" and characterized the relevant questions as whether "the amount and substantiality of the portion used . . . are reasonable in relation to the purpose of the copying," Campbell, 510 U.S. at 586-587, noting that the answer to that question will be affected by "the degree to which the [copying work] may serve as a market substitute for the original or potentially licensed derivatives," id. at 587-588 (finding that, in the case of a parodic song, "how much . . . is reasonable will depend, say, on the extent to which the song's overriding purpose and character is to parody the original or, in contrast, the likelihood that the parody may serve as a market substitute for the original"). In HathiTrust, our court concluded in its discussion of the third factor that "[b]ecause it was reasonably necessary for the [HathiTrust Digital Library] to make use of the entirety of the works in order to enable the full-text search function, we do not believe the copying was excessive." 755 F.3d at 98. As with HathiTrust, not only is the copying of the totality of the original reasonably appropriate to Google's transformative purpose, it is literally necessary to achieve that purpose. If Google copied less than the totality of the originals, its search function could not advise searchers reliably whether their searched term appears in a book (or how many times). While Google makes an unauthorized digital copy of the entire book, it does not reveal that digital copy to the public. The copy is made to enable the search functions to reveal limited, important information about the books. With respect to the search function, Google satisfies the third factor test, as illuminated by the Supreme Court in Campbell. (2) Snippet View. Google's provision of snippet view makes our third factor inquiry different from that inquiry in HathiTrust. What matters in such cases is not so much "the amount and substantiality of the portion used" in making a copy, but rather the amount and substantiality of what is thereby made accessible to a public for which it may serve as a competing substitute. In HathiTrust, notwithstanding the defendant's full-text copying, the search function revealed virtually nothing of the text of the originals to the public. Here, through the snippet view, more is revealed to searchers than in HathiTrust. Without doubt, enabling searchers to see portions of the copied texts could have determinative effect on the fair use analysis. The larger the quantity of the copyrighted text the searcher can see and the more control the searcher can exercise over what part of the text she sees, the greater the likelihood that those revelations could serve her as an effective, free substitute for the purchase of the plaintiff's book. We nonetheless conclude that, at least as presently structured by Google, the snippet view does not reveal matter that offers the marketplace a significantly competing substitute for the copyrighted work. Google has constructed the snippet feature in a manner that substantially protects against its serving as an effectively competing substitute for Plaintiffs' books. In the Background section of this opinion, we describe a variety of limitations Google imposes on the snippet function. These include the small size of the snippets (normally one eighth of a page), the blacklisting of one snippet per page and of one page in every ten, the fact that no more than three snippets are shown—and no more than one per page—for each term searched, and the fact that the same snippets are shown for a searched term no matter how many times, or from how many different computers, the term is searched. In addition, Google does not provide snippet view for types of books, such as dictionaries and cookbooks, for which viewing a small segment is likely to satisfy the searcher's need. The result of these restrictions is, so far as the record demonstrates, that a searcher cannot succeed, even after long extended effort to multiply what can be revealed, in revealing through a snippet search what could usefully serve as a competing substitute for the original. The blacklisting, which permanently blocks about 22% of a book's text from snippet view, is by no means the most important of the obstacles Google has designed. While it is true that the blacklisting of 22% leaves 78% of a book theoretically accessible to a searcher, it does not follow that any large part of that 78% is in fact accessible. The other restrictions built into the program work together to ensure that, even after protracted effort over a substantial period of time, only small and randomly scattered portions of a book will be accessible. In an effort to show what large portions of text searchers can read through persistently augmented snippet searches, Plaintiffs' counsel employed researchers over a period of weeks to do multiple word searches on Plaintiffs' books. In no case were they able to access as much as 16% of the text, and the snippets collected were usually not sequential but scattered randomly throughout the book. Because Google's snippets are arbitrarily and uniformly divided by lines of text, and not by complete sentences, paragraphs, or any measure dictated by content, a searcher would have great difficulty constructing a search so as to provide any extensive information about the book's use of that term. As snippet view never reveals more than one snippet per page in response to repeated searches for the same term, it is at least difficult, and often impossible, for a searcher to gain access to more than a single snippet's worth of an extended, continuous discussion of the term. The fact that Plaintiffs' searchers managed to reveal nearly 16% of the text of Plaintiffs' books overstates the degree to which snippet view can provide a meaningful substitute. At least as important as the percentage of words of a book that are revealed is the manner and order in which they are revealed. Even if the search function revealed 100% of the words of the copyrighted book, this would be of little substitutive value if the words were revealed in alphabetical order, or any order other than the order they follow in the original book. It cannot be said that a revelation is "substantial" in the sense intended by the statute's third factor if the revelation is in a form that communicates little of the sense of the original. The fragmentary and scattered nature of the snippets revealed, even after a determined, assiduous, time-consuming search, results in a revelation that is not "substantial," even if it includes an aggregate 16% of the text of the book. If snippet view could be used to reveal a coherent block amounting to 16% of a book, that would raise a very different question beyond the scope of our inquiry.

D. Factor Four The fourth fair use factor, "the effect of the [copying] use upon the potential market for or value of the copyrighted work," focuses on whether the copy brings to the marketplace a competing substitute for the original, or its derivative, so as to deprive the rights holder of significant revenues because of the likelihood that potential purchasers may opt to acquire the copy in preference to the original. Because copyright is a commercial doctrine whose objective is to stimulate creativity among potential authors by enabling them to earn money from their creations, the fourth factor is of great importance in making a fair use assessment. See Harper & Row, 471 U.S. at 566 (describing the fourth factor as "undoubtedly the single most important element of fair use"). Campbell stressed the close linkage between the first and fourth factors, in that the more the copying is done to achieve a purpose that differs from the purpose of the original, the less likely it is that the copy will serve as a satisfactory substitute for the original. 510 U.S. at 591. Consistent with that observation, the HathiTrust court found that the fourth factor favored the defendant and supported a finding of fair use because the ability to search the text of the book to determine whether it includes selected words "does not serve as a substitute for the books that are being searched." 755 F.3d at 100. However, Campbell's observation as to the likelihood of a secondary use serving as an effective substitute goes only so far. Even if the purpose of the copying is for a valuably transformative purpose, such copying might nonetheless harm the value of the copyrighted original if done in a manner that results in widespread revelation of sufficiently significant portions of the original as to make available a significantly competing substitute. The question for us is whether snippet view, notwithstanding its transformative purpose, does that. We conclude that, at least as snippet view is presently constructed, it does not. Especially in view of the fact that the normal purchase price of a book is relatively low in relation to the cost of manpower needed to secure an arbitrary assortment of randomly scattered snippets, we conclude that the snippet function does not give searchers access to effectively competing substitutes. Snippet view, at best and after a large commitment of manpower, produces discontinuous, tiny fragments, amounting in the aggregate to no more than 16% of a book. This does not threaten the rights holders with any significant harm to the value of their copyrights or diminish their harvest of copyright revenue. We recognize that the snippet function can cause some loss of sales. There are surely instances in which a searcher's need for access to a text will be satisfied by the snippet view, resulting in either the loss of a sale to that searcher, or reduction of demand on libraries for that title, which might have resulted in libraries purchasing additional copies. But the possibility, or even the probability or certainty, of some loss of sales does not suffice to make the copy an effectively competing substitute that would tilt the weighty fourth factor in favor of the rights holder in the original. There must be a meaningful or significant effect "upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). Furthermore, the type of loss of sale envisioned above will generally occur in relation to interests that are not protected by the copyright. A snippet's capacity to satisfy a searcher's need for access to a copyrighted book will at times be because the snippet conveys a historical fact that the searcher needs to ascertain. For example, a student writing a paper on Franklin D. Roosevelt might need to learn the year Roosevelt was stricken with polio. By entering "Roosevelt polio" in a Google Books search, the student would be taken to (among numerous sites) a snippet from page 31 of Richard Thayer Goldberg's The Making of Franklin D. Roosevelt (1981), telling that the polio attack occurred in 1921. This would satisfy the searcher's need for the book, eliminating any need to purchase it or acquire it from a library. But what the searcher derived from the snippet was a historical fact. Author Goldberg's copyright does not extend to the facts communicated by his book. It protects only the author's manner of expression. Hoehling v. Universal City Studios, Inc., 618 F.2d 972, 974 (2d Cir. 1980) ( "A grant of copyright in a published work secures for its author a limited monopoly over the expression it contains.") (emphasis added). Google would be entitled, without infringement of Goldberg's copyright, to answer the student's query about the year Roosevelt was afflicted, taking the information from Goldberg's book. The fact that, in the case of the student's snippet search, the information came embedded in three lines of Goldberg's writing, which were superfluous to the searcher's needs, would not change the taking of an unprotected fact into a copyright infringement. Even if the snippet reveals some authorial expression, because of the brevity of a single snippet and the cumbersome, disjointed, and incomplete nature of the aggregation of snippets made available through snippet view, we think it would be a rare case in which the searcher's interest in the protected aspect of the author's work would be satisfied by what is available from snippet view, and rarer still—because of the cumbersome, disjointed, and incomplete nature of the aggregation of snippets made available through snippet view—that snippet view could provide a significant substitute for the purchase of the author's book. Accordingly, considering the four fair use factors in light of the goals of copyright, we conclude that Google's making of a complete digital copy of Plaintiffs' works for the purpose of providing the public with its search and snippet view functions (at least as snippet view is presently designed) is a fair use and does not infringe Plaintiffs' copyrights in their books.

a. Purpose and Character of the Use (1) Transformative Use The first statutory fair use factor considers the purpose and character of the secondary use. In this regard, the uses identified by Congress in the preamble to § 107—criticism, comment, news reporting, teaching, scholarship, and research—might be deemed "most appropriate" for a purpose or character finding indicative of fair use. Nimmer § 13.05[A][1][a], at 13-162; see Authors Guild v. Google, Inc., 804 F.3d 202, 215 (2d Cir. 2015) (noting that providing commentary or criticism on another's work is "[a]mong the best recognized justifications for copying"). The challenged use here does not appear to fit within any of these statutory categories. Nevertheless, the district court concluded that defendants' use was "transformative," indeed, so "highly transformative" as to be "determinative" of fair use. TCA Television Corp. v. McCollum, 151 F. Supp. 3d at 434-35. The district court explained that by having a single character perform the Routine, the Play's authors were able to contrast "Jason's seemingly soft-spoken personality and the actual outrageousness of his inner nature, which he expresses through the sock puppet." Id. at 436. This contrast was "a darkly comedic critique of the social norms governing a small town in the Bible Belt." Id. This reasoning is flawed in that what it identifies are the general artistic and critical purpose and character of the Play. The district court did not explain how defendants' extensive copying of a famous comedy routine was necessary to this purpose, much less how the character of the Routine was transformed by defendants' use. The Supreme Court has stated that "the goal of copyright . . . is generally furthered by the creation of transformative works." Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 579.But how does a court decide "whether and to what extent the new work is 'transformative'"? Id. Campbell instructs that a court properly considers "whether the new work merely supersedes the objects of the original creation, or instead adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Id. (emphases added) (alterations, citations, and internal quotation marks omitted). As the highlighted language indicates, the focus of inquiry is not simply on the new work, i.e., on whether that work serves a purpose or conveys an overall expression, meaning, or message different from the copyrighted material it appropriates. Rather, the critical inquiry is whether the new work uses the copyrighted material itself for a purpose, or imbues it with a character, different from that for which it was created. See id. Otherwise, any play that needed a character to sing a song, tell a joke, or recite a poem could use unaltered copyrighted material with impunity, so long as the purpose or message of the play was different from that of the appropriated material. In sum, even if, as the district court concluded, Hand to God is a "darkly comedic critique of the social norms governing a small town in the Bible Belt," TCA Television Corp. v. McCollum, 151 F. Supp. 3d at 436, and even if the Play's purpose and character are completely different from the vaudevillian humor originally animating Who's on First?, that, by itself, does not demonstrate that defendants' use of the Routine in the Play was transformative of the original work. We made this point in Cariou v. Prince, 714 F.3d 694. There, the defendant, a self-styled "appropriation artist," id. at 699, had taken plaintiff's copyrighted photographs—"serene and deliberately composed" portraits of Rastafarian men—and altered them to create "crude and jarring" collages, id. at 706. Defendant acknowledged that he had not used the photographs to "comment on" the original works. Id. at 707. Instead, both works had an underlying artistic purpose, but defendant stated that he had sought to change the original material "into something that's completely different." Id. (internal quotation marks omitted). Reversing a district court award of summary judgment in favor of plaintiff, this court ruled that, although commentary frequently constitutes fair use, it is not essential that a new creative work comment on an incorporated copyrighted work to be transformative. See id. at 706.Rather, "to qualify as a fair use" in the absence of such a different purpose, the new work "generally must alter the original with 'new expression, meaning, or message.'" Id. (quoting Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 579). Cariou concluded that the challenged artworks there satisfied this standard because they not only strove for "new aesthetics with creative and communicative results distinct from" that of the copyrighted material, but also gave the incorporated photographs "new expression," thereby admitting a transformative purpose. Id. at 708. Indeed, where the defendant's use so "heavily obscured and altered" the original photographs as to make them "barely recognizable" within the new work, the court ruled that transformative purpose (and ultimately fair use) was established as a matter of law. Id. at 710. But where lesser changes retained certain of the original work's aesthetics, the court could not say "for sure" that their incorporation into the defendant's works had "transformed [the original] work enough to render it transformative." Id. at 711. As to those works, Cariou remanded for further proceedings. Id. Insofar as Cariou might be thought to represent the high-water mark of our court's recognition of transformative works, it has drawn some criticism. See Kienitz v. Sconnie Nation LLC, 766 F.3d 756, 758 (7th Cir. 2014) (expressing skepticism as to Cariou's approach and criticizing reliance on transformativeness as substitute for the statutory factors, which threatens to override the copyright owner's exclusive right to prepare derivative works); see also Nimmer § 13.05[B][6], at 13.224.20 (stating with respect to Cariou: "It would seem that the pendulum has swung too far in the direction of recognizing any alteration as transformative, such that this doctrine now threatens to swallow fair use. It is respectfully submitted that a correction is needed in the law."). We need not defend Cariou here, however, because our point is that even scrupulous adherence to that decision does not permit defendants' use of Who's on First? in Hand to God to be held transformative. Far from altering Who's on First? to the point where it is "barely recognizable" within the Play, Cariou v. Prince, 714 F.3d at 710, defendants' use appears not to have altered the Routine at all. The Play may convey a dark critique of society, but it does not transform Abbott and Costello's Routine so that it conveys that message. To the contrary, it appears that the Play specifically has its characters perform Who's on First? without alteration so that the audience will readily recognize both the famous Routine and the boy's false claim to having created it. Indeed, it is only after Who's on First? is performed—at some length, almost verbatim, and with the Play's characters mimicking the original timing, tone, and delivery of Abbott and Costello—that the boy's lie about creating the classic Routine—no part of the Routine—becomes the triggering event for the puppet to assume an independent persona. Defendants nevertheless maintain that using the Routine for such a "dramatic," rather than comedic, purpose was transformative. Appellees' Br. 18 (stating that Play's use of Routine was "far cry" from original "comedy schtick"). The argument will not bear close scrutiny. The "dramatic" purpose served by the Routine in the Play appears to be as a "McGuffin," that is, as a theatrical device that sets up the plot, but is of little or no significance in itself.To advance the plot of the Play, specifically, to have the puppet Tyrone take on a persona distinct from that of Jason, defendants needed Jason to lie about something and for Tyrone to call him on it. But the particular subject of the lie—the Routine—appears irrelevant to that purpose. Such unaltered use of an allegedly copyrighted work, having no bearing on the original work, requires justification to qualify for a fair use defense. See Authors Guild v. Google, Inc., 804 F.3d at 215 (stating that "taking from another author's work for the purpose of making points that have no bearing on the original may well be fair use, but the taker would need to show a justification"). More than the Routine's ability to capture audience attention is necessary to provide such justification. As the Supreme Court has cautioned, where a secondary use "has no critical bearing on the substance or style of the original composition, which the alleged infringer merely uses to get attention . . . , the claim to fairness in borrowing from another's work diminishes accordingly (if it does not vanish)." Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 580. Nor is a different conclusion warranted because defendants here place the unaltered Routine in a sharply different context from its original authors. See id. at 598 (Kennedy, J., concurring) (observing that courts should not afford fair use protection to persons who merely place characters from familiar copyrighted works into novel or eccentric settings). The fact that, even as a McGuffin, the Routine is quite funny, also cannot justify its use in the Play. That humor is an achievement of the Routine's creators, not of the playwright who takes advantage of it without transforming the Routine's aesthetic. Moreover, the Play appropriates the Routine's humor not incidentally, but extensively by having the characters perform some dozen of the original exchanges on the comic ambiguity of the words "who's on first." No new dramatic purpose was served by so much copying. Cf. Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 589 (concluding that copying of song for parodic purpose was fair where it was not "excessive" in relation to that purpose as secondary user took only what was necessary to evoke original). The only purpose served by the extent of defendants' taking is identically comedic to that of the original authors, that is, to have two performers expand on a singular joke in order to generate increasing audience laughter. As this court has recognized, there is "nothing transformative" about using an original work "in the manner it was made to be" used. On Davis v. Gap, Inc., 246 F.3d 152, 174 (2d Cir. 2001); see Ringgold v. Black Entertainment TV, 126 F.3d 70, 79 (2d Cir. 1997) (concluding that purpose factor favored copyright owner where defendant used "work for precisely a central purpose for which it was created"). Defendants' use of the Routine, not briefly as the basis for a dramatic lie, but extensively for its original comedic effect, cannot be deemed transformative. In sum, nothing in the 12(b)(6) record shows that the Play imbued the Routine with any new expression, meaning, or message. Nor does any new dramatic purpose justify defendants' extensive copying of the Routine. Accordingly, the district court erred both in finding defendants' use of the Routine transformative and in concluding, on that basis, that a fair use defense was established as a matter of law.

(2) Commercial Purpose The first statutory factor specifically instructs courts to consider whether copyrighted materials are used for a commercial purpose or for a nonprofit educational purpose, the former tending "to weigh against a finding of fair use." Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 585 (internal quotation marks omitted). There is no question here that defendants' use of Who's on First? in Hand to God was for a commercial purpose. Nevertheless, the district court discounted that fact upon finding the use "highly transformative." TCA Television Corp. v. McCollum, 151 F. Supp. 3d at 434-35; see Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 579 (recognizing that "the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use"). Because we here conclude that defendants' use was not transformative, let alone "highly transformative," we conclude that the district court erred in discounting Hand to God's commercial character. See Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 579-81. This conclusion applies with particular force here where plaintiffs allege that defendants not only used an unaltered and appreciable excerpt of the Routine in a commercial play but also featured the Routine in the Play's advertising, conduct which reasonably qualifies as commercial exploitation weighing strongly against fair use. See id. at 585 (observing that use of copyrighted work "to advertise a product, even in a parody, will be entitled to less indulgence under the first factor of the fair use enquiry than the sale of a parody for its own sake"); American Geophysical Union v. Texaco Inc., 60 F.3d 913, 922 (2d Cir. 1994) (stating that fair use claim will not be sustained when secondary use can fairly be characterized as "commercial exploitation" (internal quotation marks omitted)); Consumers Union of U.S., Inc. v. Gen. Signal Corp., 724 F.2d 1044, 1049 (2d Cir. 1983) (observing that some infringement actions involve copying of creative expression for "purpose of having that precise form of expression advance someone else's commercial interests—for example, using well-known copyrighted lines to attract attention to an advertisement"). Indeed, to the extent defendants excessively copied from the Routine even within the Play, their advertising focus on the Routine's comic exchanges raises particular commercial exploitation concerns. Thus, defendants' commercial use of the Routine was not transformative. Rather, it duplicated to a significant degree the comedic purpose of the original work. As such, the first statutory factor, far from weighing in defendants' favor, weighs in favor of plaintiffs. b. Nature of Copyrighted Work The second statutory factor, "the nature of the copyrighted work," also weighs in plaintiffs' favor. As the Supreme Court has observed, certain "works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied." Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 586. Like the district court, we conclude that Who's on First?, an original comedy sketch created for public entertainment, lies at the heart of copyright's intended protection. See id. (recognizing that creative expression created for public dissemination is at core of "copyright's protective purposes"); On Davis v. Gap, Inc., 246 F.3d at 175 (concluding that second factor favored plaintiff because copyrighted work was "in the nature of an artistic creation"). Thus, while the secondary user of noncreative information can more readily claim fair use based on the law's recognition of "a greater need to disseminate factual works than works of fiction or fantasy," Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. at 563; see Authors Guild v. Google, Inc., 804 F.3d at 220 & n.21 (explaining that factual works "often present well justified fair uses" because "there is often occasion to test the accuracy of, to rely on, or to repeat their factual propositions," which "may reasonably require quotation"), the secondary user of a creative work must justify his use, usually by explaining the functional or creative rationale behind its quotation, see Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1113 (1990) (explaining that, in considering whether quotation is fair, courts must consider utility of each challenged passage). Defendants argue that their use was justified by the dramatic need to use an instantly recognizable "cultural" touchstone in the relevant scene. Appellees' Br. 15. Defendants do not explain, however, why Jason's lie had to pertain to a cultural touchstone, as opposed to any obvious tall tale—e.g., inventing the Internet, traveling to Mars, out-swimming Michael Phelps. See generally Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. at 1111 (observing that court considering fair use must determine not only if justification for copying exists, but also "how powerful, or persuasive, is the justification"). But even assuming defendants' professed dramatic need, they do not explain why the cultural touchstone had to be the Routine—or even a comedy sketch—as opposed to some other readily recognizable exchange, including those already in the public domain. Most troubling, even if defendants could justify their dramatic need to use a small, identifiable segment of the Routine, that does not justify having their characters perform, verbatim, some dozen variations on the Routine's singular joke. As already noted, the purpose of such extensive use was to provoke audience laughter in exactly the same way as the Routine's creators had done. In sum, because defendants' use of the Routine cannot be deemed transformative, and because the record is devoid of any persuasive justification for the extent of defendants' use, the creative nature of the Routine weighs strongly against a fair use defense.

c. Amount and Substantiality of Use The third statutory factor asks whether "'the amount and substantiality of the portion used in relation to the copyrighted work as a whole' . . . are reasonable in relation to the purpose of the copying." Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 586 (quoting 17 U.S.C. § 107(3)). In assessing this factor, we consider not only "the quantity of the materials used" but also "their quality and importance." Id. at 587; see Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. at 565 (stating that "fact that a substantial portion of the infringing work was copied verbatim is evidence of the qualitative value of the copied material, both to the originator and to the plagiarist who seeks to profit from marketing someone else's copyrighted expression"). While acknowledging that the portion of the Routine used by defendants was "instantly recognizable" and "more than merely the 'introductory premise,'" the district court decided—without explanation—that this factor tipped only "slightly" in plaintiffs' favor in light of defendants' transformative use. TCA Television Corp. v. McCollum, 151 F. Supp. 3d at 434. We reject the district court's transformative use determination for reasons already explained. On de novo review, we further conclude that the third statutory factor weighs strongly in favor of plaintiffs. While the portion of the Routine copied by defendants takes less than two minutes to perform, it plainly reveals the singular joke underlying the entire Routine: that words understood by one person as a question can be understood by another as an answer. Moreover, defendants repeatedly exploit that joke through a dozen variations. This manifests substantial copying. See Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. at 548, 565-66 (concluding that third factor favored plaintiffs where defendants copied approximately 300 words verbatim in light of "expressive value of the excerpts"); Castle Rock Entm't v. Carol Publ'g Grp., Inc., 955 F. Supp. 260, 269 (S.D.N.Y. 1997) (citing Harper & Row as support for proposition that copying even few words of challenged work can constitute substantial taking if it amounts to taking heart of original work), aff'd, 150 F.3d 132 (2d Cir. 1998). Even a substantial taking, however, can constitute fair use if justified. See Authors Guild, Inc. v. HathiTrust, 755 F.3d 87, 98 (2d Cir. 2014) (acknowledging that some purposes require copying entirety of copyrighted work); see also Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. at 1123 (explaining that, under third factor, "an important inquiry is whether the selection and quantity of the material taken are reasonable in relation to the purported justification"). But, as already explained, defendants offer no persuasive justification for their extensive use of the Routine. Cf. Authors Guild, Inc. v. HathiTrust, 755 F.3d at 98 (stating that "crux" of third factor inquiry is whether "no more was taken than necessary" (internal quotation marks omitted)).

d. Effect on Potential Market for Copyrighted Work The final statutory factor considers "the effect of the use upon the potential market for or value of the copyrighted work," 17 U.S.C. § 107(4), focusing on whether the secondary use usurps demand for the protected work by serving as a market substitute, see Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 592 (stating that role of courts is to distinguish between "biting criticism that merely suppresses demand and copyright infringement, which usurps it" (alterations and internal quotation marks omitted)). In weighing this factor, a court properly looks to "not only the extent of market harm caused by the particular actions of the alleged infringer, but also whether unrestricted and widespread conduct of the sort engaged in by the defendant would result in a substantially adverse impact on the potential market for the original." Id. at 590 (alteration and internal quotation marks omitted). The district court weighed this factor in defendants' favor, concluding that the Play's use of the Routine could not reasonably be expected to usurp the market for Abbott and Costello's original performance. See TCA Television Corp. v. McCollum, 151 F. Supp. 3d at 434-35. In so doing, however, the district court disregarded the possibility of defendants' use adversely affecting the licensing market for the Routine. See id. at 434 (citing Cariou v. Prince, 714 F.3d at 708 (stating that fourth factor "does not focus principally on the question of damage to [a] derivative market")). While derivative markets are not the principal focus of the fourth inquiry, that does not mean that they are irrelevant. See Campbell v. Acuff-Rose Music, Inc., 510 U.S. at 593 (recognizing that evidence of substantial harm to derivative market would weigh against fair use). A court considering fair use properly identifies and weighs relevant harm to the derivative market for a copyrighted work, which market includes uses that creators of original works might "license others to develop." Id. at 592; see American Geophysical Union v. Texaco Inc., 60 F.3d at 929 ("[T]he impact on potential licensing revenues is a proper subject for consideration in assessing the fourth factor."). To be clear, in assessing harm posed to a licensing market, a court's focus is not on possible lost licensing fees from defendants' challenged use. See American Geophysical Union v. Texaco Inc., 60 F.3d at 929 n.17 (explaining that fourth factor would always favor copyright owner if courts focused on loss of potential licensing fees from alleged infringer); see also Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. at 1124 ("By definition every fair use involves some loss of royalty revenue because the secondary user has not paid royalties."). Rather, a court properly considers the challenged use's "impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets." American Geophysical Union v. Texaco Inc., 60 F.3d at 930; accord Swatch Grp. Mgmt. Servs. Ltd. v. Bloomberg L.P., 756 F.3d 73, 91 (2d Cir. 2014). Plaintiffs here alleged the existence of a traditional—and active—derivative market for licensing the Routine. See Am. Compl. ¶¶ 40, 81-83 (alleging that plaintiffs receive "regular" requests to license Who's on First?, which they frequently grant). Further, they alleged market harm. See id. at ¶ 83 (alleging that defendants' unlicensed use of Who's on First? negatively affected commercial market for future licensing). Accepting these allegations as true at this stage of the litigation, we conclude that this factor weighs in favor of plaintiffs. See On Davis v. Gap, Inc., 246 F.3d at 175-76 (concluding that fourth factor favored copyright owner where defendants' taking caused both loss of royalty revenue and "diminution of [owner's] opportunity to license to others who might regard [owner's] design as preempted by [defendant's] ad").

In addressing the first factor, the court asks "whether the new work merely supersedes the objects of the original creation, or instead adds something new, with a further purpose or different character . . . [;] in other words, whether and to what extent the new work is 'transformative.'" Id. at 579 (citations omitted, alterations incorporated); see § 107(1). VidAngel concedes its use is commercial, and thus "presumptively . . . unfair." Leadsinger, Inc. v. BMG Music Publ'g, 512 F.3d 522, 530 (9th Cir. 2008) (citation omitted). But, it argues that its use is "profoundly transformative" because "omissions can transform a work," affirming "[r]eligious convictions and parental views." The district court found, however, that "VidAngel's service does not add anything to Plaintiff's works. It simply omits portions that viewers find objectionable," and transmits them for the "same intrinsic entertainment value" as the originals. This factual finding was not clearly erroneous. Although removing objectionable content may permit a viewer to enjoy a film, this does not necessarily "add[] something new" or change the "expression, meaning, or message" of the film. Campbell, 510 U.S. at 579. Nor does reproducing the films' discs in digital streaming format, because "both formats are used for entertainment purposes." Kelly v. Arriba Soft Corp., 336 F.3d 811, 819 (9th Cir. 2003). Star Wars is still Star Wars, even without Princess Leia's bikini scene. Moreover, VidAngel's service does not require removing a crucial plot element—it requires the use of only one filter, which can be an audio filter temporarily silencing a portion of a scene without removing imagery, or skipping a gratuitous scene. Indeed, the FMA sanctions only making "limited portions" of a work imperceptible. 17 U.S.C. § 110(11). The district court did not abuse its discretion in finding that VidAngel's use is not transformative. See Kelly, 336 F.3d at 819 ("Courts have been reluctant to find fair use when an original work is merely retransmitted in a different medium . . . . for entertainment purposes." (footnote omitted)).

The fourth fair use factor evaluates "the extent of market harm caused by" the infringing activity and "whether unrestricted and widespread conduct of the sort engaged by the defendant . . . would result in a substantially adverse impact on the potential market for the original." Campbell, 510 U.S. at 590 (citation omitted, alteration incorporated); see § 107(4). Because the district court concluded that VidAngel's use was commercial and not transformative, it was not error to presume likely market harm. Leadsinger, 512 F.3d at 531 . VidAngel argues that its service actually benefits the Studios because it purchases discs and expands the audience for the copyrighted works to viewers who would not watch without filtering. But, the district court found that "VidAngel's service [is] an effective substitute for Plaintiff's unfiltered works," because surveys suggested that 49% of its customers would watch the movies without filters. This finding was not clearly erroneous. VidAngel's purchases of discs also do not excuse its infringement. See A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1017 (9th Cir. 2001) ("Any allegedly positive impact of defendant's activities on plaintiffs' prior market in no way frees defendant to usurp a further market that directly derives from reproduction of the plaintiffs' copyrighted works.") (quoting UMG Recordings, Inc. v. MP3.com, Inc., 92 F. Supp. 2d 349, 352 (S.D.N.Y. 2000)). And, the market factor is less important when none of the other factors favor VidAngel. See Leadsinger, 512 F.3d at 532.12

A In considering the first statutory factor--the "purpose and character" of the secondary use, 17 U.S.C. § 107(1)--the primary inquiry is whether the use "communicates something new and different from the original or [otherwise] expands its utility," that is, whether the use is "transformative." Google Books, 804 F.3d at 214. To be transformative, a use must "do[] something more than repackage or republish the original copyrighted work"; it must "'add[] something new, with a further purpose or different character, altering the first with new expression, meaning or message . . . .'" Authors Guild, Inc. v. HathiTrust, 755 F.3d 87, 96 (2d Cir. 2014) (quoting Campbell, 510 U.S. at 579). "Although . . . transformative use is not absolutely necessary for a finding of fair use, . . . [transformative] works . . . lie at the heart of the fair use doctrine," Campbell, 510 U.S. at 579, and "a use of copyrighted material that 'merely repackages or republishes the original' is unlikely to be deemed a fair use," Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 108 (2d Cir. 1998) (quoting Pierre N. Leval, Toward a Fair Use Standard, 103 Harv. L. Rev. 1105, 1111 (1990)). Precedent is helpful. Both parties rely most heavily on Google Books, which provides the starting point for analysis. In Google Books, a consortium of libraries collaborated to make digital copies of millions of books, many of them under copyright. Google pooled these digital copies into a text-searchable database. 804 F.3d at 207. Anyone could search the database free. When a user entered a search term, Google returned a list of books that included the term, and, for each responsive book, Google provided a few "snippets" that contained the term. Id. We held that Google's copying served a transformative purpose because it created a text-searchable database that "communicate[d] something new and different from the original." Id. at 214. "[T]he result of a word search is different in purpose, character, expression, meaning, and message from the page (and the book) from which it is drawn." Id. at 217 (quoting HathiTrust, 755 F.3d at 97). We also held that the "snippet view" of unaltered, copyrighted text "add[ed] important value to the basic transformative search function" by allowing users to verify that the list of books returned by the database was responsive to the user's search. Id. Thus, a user searching for the term "Hindenburg" could infer from snippets whether the book was referencing the Weimar president or the exploded zeppelin. See id. at 217-18. TVEyes's copying of Fox's content for use in the Watch function is similarly transformative insofar as it enables users to isolate, from an ocean of programming, material that is responsive to their interests and needs, and to access that material with targeted precision. It enables nearly instant access to a subset of material--and to information about the material--that would otherwise be irretrievable, or else retrievable only through prohibitively inconvenient or inefficient means. Sony Corporation of America vs. Universal City Studios, Inc. is instructive. See 464 U.S. 417, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984). In Sony, a television customer, who (by virtue of owning a television set) had acquired authorization to watch a program when it was broadcast, recorded it in order to watch it instead at a later, more convenient time. That was held to be a fair use. While Sony was decided before "transformative" became a term of art, the apparent reasoning was that a secondary use may be a fair use if it utilizes technology to achieve the transformative purpose of improving the efficiency of delivering content without unreasonably encroaching on the commercial entitlements of the rights holder. The Watch function certainly qualifies as technology that achieves the transformative purpose of enhancing efficiency: it enables TVEyes's clients to view all of the Fox programming that (over the prior thirty-two days) discussed a particular topic of interest to them, without having to monitor thirty-two days of programming in order to catch each relevant discussion; and it eliminates the clients' need even to view entire programs, because the ten most relevant minutes are presented to them. Much like the television customer in Sony, TVEyes clients can view the Fox programming they want at a time and place that is convenient to them, rather than at the time and place of broadcast. For these reasons, TVEyes's Watch function is at least somewhat transformative. The first statutory factor also implicates considerations distinct from whether the secondary use is transformative. In particular, Fox argues that the "commercial nature" of TVEyes's copying (its sale of access to Fox's content) weighs against a finding of fair use. 17 U.S.C. § 107(1). The commercial nature of a secondary use weighs against a finding of fair use. See Campbell, 510 U.S. at 585. And it does so especially when, as here, the transformative character of the secondary use is modest. See id. at 579 ("[T]he [less] transformative the new work, the [more] will be the significance of other factors, like commercialism . . . ."). The Watch function has only a modest transformative character because, notwithstanding the transformative manner in which it delivers content, it essentially republishes that content unaltered from its original form, with no "new expression, meaning or message." HathiTrust, 755 F.3d at 96 (quoting Campbell, 510 U.S. at 579); cf. Kirkwood, 150 F.3d at 106 (service that transmits unaltered radio broadcasts in real time over telephone lines is not transformative); Video Pipeline, Inc. v. Buena Vista Home Entm't, Inc., 342 F.3d 191, 199-200 (3d Cir. 2003) (service that streams short previews of movies without commentary is not transformative). The clients of TVEyes use Fox's news broadcasts for the same purpose that authorized Fox viewers use those broadcasts--the purpose of learning the information reported. The first statutory factor therefore favors TVEyes, albeit slightly.

B The second statutory factor is "the nature of the copyrighted work." 17 U.S.C. § 107(2). This factor "has rarely played a significant role in the determination of a fair use dispute," and it plays no significant role here. Google Books, 804 F.3d at 220. TVEyes presses the argument that, since facts are not copyrightable, the factual nature of Fox's content militates in favor of a finding of fair use. We have rejected this argument: "Those who report the news undoubtedly create factual works. It cannot seriously be argued that, for that reason, others may freely copy and re-disseminate news reports." Id. at 220.

C The third statutory factor is "the amount and substantiality of the portion used in relation to the copyrighted work as a whole." 17 U.S.C. § 107(3). The relevant consideration is the amount of copyrighted material made available to the public rather than the amount of material used by the copier. Google Books, 804 F.3d at 222. This factor clearly favors Fox because TVEyes makes available virtually the entirety of the Fox programming that TVEyes users want to see and hear. While "courts have rejected any categorical rule that a copying of the entirety cannot be a fair use," "a finding of fair use is [less] likely . . . when the copying is extensive, or encompasses the most important parts of the original." Id. at 221. In this respect, the TVEyes Watch function is radically dissimilar to the service at issue in Google Books. Google's snippet function was designed to ensure that users could see only a very small piece of a book's contents. Each snippet was three lines of text, constituting approximately one-eighth of a page; a viewer could see at most three snippets per book for any searched term, and no more than one per page. Users were prevented from performing repeated searches to find multiple snippets that could be compiled into a coherent block of text. Approximately 22% of a book's text was "blacklist[ed]": no snippet could be shown from those pages. Id. at 222. And snippets were not available at all for such books as dictionaries or cookbooks, in which a snippet might convey all the information that a searcher was likely to need. While the snippets allowed a user to judge whether a book was responsive to the user's needs, they were abbreviated to ensure that it would be nearly impossible for a user to see a meaningful exposition of what the author originally intended to convey to readers. TVEyes redistributes Fox's news programming in ten-minute clips, which--given the brevity of the average news segment on a particular topic--likely provide TVEyes's users with all of the Fox programming that they seek and the entirety of the message conveyed by Fox to authorized viewers of the original. Cf. Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 564-65, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985) (finding no fair use when the copying involved only about 300 words, but the portion copied was "the heart of the book"). TVEyes's use of Fox's content is therefore both "extensive" and inclusive of all that is "important" from the copyrighted work. Google Books, 804 F.3d at 221.

D The fourth statutory factor is "the effect of the [secondary] use upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). This factor is "undoubtedly the single most important element of fair use." Harper & Row Publishers, Inc. v. Nation Enters., 471 U.S. 539, 566, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985). It "focuses on whether the copy brings to the marketplace a competing substitute for the original, or its derivative, so as to deprive the rights holder of significant revenues because of the likelihood that potential purchasers may opt to acquire the copy in preference to the original." Google Books, 804 F.3d at 223. Critically, it requires consideration of "not only the . . . market harm caused by the particular actions of the alleged infringer," but also the market harm that would result from "unrestricted and widespread conduct of the [same] sort." Campbell, 510 U.S. at 590 (internal quotation marks and alteration omitted). TVEyes argues that its service poses little risk of being a "competing substitute" for Fox's offerings. Google Books, 804 F.3d at 223. Fox argues that TVEyes undercuts Fox's ability to profit from licensing searchable access to its copyrighted content to third parties. Fox has much the stronger point. "It is indisputable that, as a general matter, a copyright holder is entitled to demand a royalty for licensing others to use its copyrighted work, and that the impact on potential licensing revenues is a proper subject for consideration in assessing the fourth factor." Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 614 (2d Cir. 2006) (quoting Texaco, 60 F.3d at 929). However, "not every effect on potential licensing revenues enters the analysis under the fourth factor." Texaco, 60 F.3d at 929. A copyright owner has no right to demand that users take a license unless the use that would be made is one that would otherwise infringe an exclusive right. See Bill Graham Archives, 448 F.3d at 615. Even if a use does infringe an exclusive right, "[o]nly an impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets should be legally cognizable when evaluating a secondary use's effect upon the potential market for or value of the copyrighted work." Texaco, 60 F.3d at 930 (internal quotation marks omitted). That limitation does not restrict our analysis here. The success of the TVEyes business model demonstrates that deep-pocketed consumers are willing to pay well for a service that allows them to search for and view selected television clips, and that this market is worth millions of dollars in the aggregate. Consequently, there is a plausibly exploitable market for such access to televised content, and it is proper to consider whether TVEyes displaces potential Fox revenues when TVEyes allows its clients to watch Fox's copyrighted content without Fox's permission. Such displacement does occur. Since the ability to re-distribute Fox's content in the manner that TVEyes does is clearly of value to TVEyes, it (or a similar service) should be willing to pay Fox for the right to offer the content. By providing Fox's content to TVEyes clients without payment to Fox, TVEyes is in effect depriving Fox of licensing revenues from TVEyes or from similar entities. And Fox itself might wish to exploit the market for such a service rather than license it to others. TVEyes has thus "usurp[ed] a market that properly belongs to the copyright-holder." Kirkwood, 150 F.3d at 110. It is of no moment that TVEyes allegedly approached Fox for a license but was rebuffed: the failure to strike a deal satisfactory to both parties does not give TVEyes the right to copy Fox's copyrighted material without payment. In short, by selling access to Fox's audiovisual content without a license, TVEyes deprives Fox of revenues to which Fox is entitled as the copyright holder. Therefore, the fourth factor favors Fox.

I. Copyright infringement claim A. Copyrightability of the pairings as compilations The boundaries of copyright protection are by now well-settled. Facts are not copyrightable and original works are. See Feist Publ'ns, Inc. v. Rural Tel. Serv., Co., 499 U.S. 340, 344-45, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (1991) (Feist). In between, however, are to be found a variety of works, including lists, compilations, directories, and guides that include facts, but demonstrate varying degrees of creativity in their selection, arrangement, or coordination. This case is about the area in between. Our Constitution provides the basis for copyright protection, and authorizes Congress to "secur[e] for limited [t]imes to [a]uthors . . . the exclusive [r]ight to their respective [w]ritings." U.S. Const. art. I, § 8, cl. 8. Under this authority, Congress enacted the Copyright Act to protect "original works of authorship." 17 U.S.C. § 102(a). The term "original" indicates "that the work was independently created by the author (as opposed to copied from other works), and that it possesses at least some minimal degree of creativity." Feist, 499 U.S. at 345 (citing 1 M. Nimmer & D. Nimmer, Copyright §§ 2.01[A], [B] (1990)). Originality is thus not an exacting standard, since it requires only minimal creativity. "[I]t does not require that facts be presented in an innovative or surprising way." Id. at 362. Facts are not copyrightable, because they lack any degree of creativity. This is so whether facts stand alone or as part of a compilation. Id. at 350; see also 17 U.S.C. §§ 101-03. Facts exist and are not created. Thus, there is a distinction between creating a work and discovering a fact. As the Supreme Court put it in Feist, "[t]he first person to find and report a particular fact has not created the fact; he or she has merely discovered its existence." 499 U.S. at 347. Even though facts themselves are not copyrightable, the Copyright Act recognizes that collections or compilations of facts may possess the originality necessary for copyright protection. 17 U.S.C. §§ 101-03; see also Feist, 499 U.S. at 348. A "compilation" is defined under the Copyright Act as "[1] a work formed by the collection and assembling of preexisting materials or of data [2] that are selected, coordinated, or arranged in such a way that [3] the resulting work as a whole constitutes an original work of authorship." 17 U.S.C. § 101; see also Feist, 499 U.S. at 357. The Copyright Act makes clear, however, that when a collection of facts is copyrightable, the underlying facts themselves are not protected. Copyright protection "extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material." 17 U.S.C. § 103(b). The Supreme Court stated this principle in Feist as follows: "[C]hoices as to selection and arrangement, so long as they are made independently by the compiler and entail a minimal degree of creativity, are sufficiently original that Congress may protect such compilations through the copyright laws." 499 U.S. at 348 (citations omitted). The Supreme Court authoritatively set the boundaries in Feist when it held that a telephone book fell below the creative limit. Id. at 362. The Court held that a telephone company's alphabetical list of the names, addresses, and telephone numbers of its subscribers in a directory was not copyrightable. Id. at 361-62. There was no creativity in the selection or arrangement of the listings, because the telephone company's subscribers were persons desiring telephone service in the company's service area. Id. at 362. When the subscribers filled out an application, the company issued them a telephone number. Id. The company therefore did not select its subscribers or the data listed in its telephone book. Id. In preparing the directory, the company simply took the data that its subscribers provided to it and listed the subscribers alphabetically by surname. Id. The Supreme Court in Feist went on to recognize that a little creativity in the selection or arrangement of facts would make a big difference. The compilation of facts could be copyrightable if the company "selected, coordinated, or arranged these uncopyrightable facts in an original way." Id. The company's mere replication of the names, addresses, and telephone numbers that each subscriber provided was not enough. It lacked "the modicum of creativity necessary to transform mere selection into copyrightable expression." Id. The arrangement of names did not help, because the company did "nothing more than list [its] subscribers in alphabetical order," which is an "age-old practice, firmly rooted in tradition and so commonplace that it has come to be expected as a matter of course." Id. at 363. The Court also recognized that hard work is not necessarily creative, and so the investment of time and effort in a project does not necessarily make it copyrightable. Feist rejected the "sweat of the brow" or "industrious collection" doctrine, which the Court described as the notion that the Copyright Act should "hand[] out proprietary interests in facts" simply because some authors spend considerable time and effort compiling facts in their works. Id. at 352-60. Instead, the Supreme Court held that whether a factual compilation is an original work depends upon "the manner in which the collected facts have been selected, coordinated, and arranged." Id. at 358. The selection or arrangement need not be novel; rather, the selection or arrangement need only be made independently and "display some minimal level of creativity." Id. After Feist, there has been a considerable amount of federal litigation concerning when compilations of facts are copyrightable. Such litigation, not surprisingly, has served to illustrate that not a great deal of creativity in selection or arrangement is required. We briefly review the principal circuit court decisions. Our Court has decided two significant cases since Feist. In CDN Inc. v. Kapes, 197 F.3d 1256, 1257, 1262 (9th Cir. 1999), we held that published lists of wholesale prices for collectible coins contained sufficient originality to sustain copyright protection. This was because the prices were not facts, but were estimates based on empirical research. They were "wholly the product of [CDN's] creativity." CDN, 197 F.3d at 1260. They were therefore "not mere listings of actual prices paid; rather, they [were] CDN's best estimate of the fair value of [each] coin." Id. For that reason, the lists were copyrightable. More recently, we decided Bikram's Yoga College of India, L.P. v. Evolation Yoga, LLC, 803 F.3d 1032 (9th Cir. 2015) (Bikram's Yoga). There, we considered a book containing depictions of a sequence of yoga poses and breathing exercises. The sequence itself was not copyrightable, because it was an idea or process under the Copyright Act. Bikram's Yoga, 803 F.3d at 1042, 1044; see also 17 U.S.C. § 102(b) ("In no case does copyright protection for an original work of authorship extend to any idea . . . [or] process, . . . regardless of the form in which it is described, explained, illustrated, or embodied in such work."). The depiction of the sequence in the book, however, was protected by copyright, because the depiction was the expression of the idea, i.e., the words and pictures that described the sequence. Bikram's Yoga, 803 F.3d at 1038-39. The book was thus entitled to copyright protection. Id. Other circuit decisions have more directly considered compilations of factual or other non-copyrightable elements. The Second Circuit has addressed the copyrightability of compilations in two cases in which it found fairly minimal creativity sufficient for protection. In Key Publications, Inc. v. Chinatown Today Publishing Enterprises, Inc., 945 F.2d 509, 511 (2d Cir. 1991), the court considered a directory intended for the New York City Chinese-American community. The directory consisted of business names, addresses, and phone numbers, which were arranged into separate descriptive categories. Key Publ'ns, 945 F.2d at 512-13. The court held this was a sufficiently creative factual compilation to warrant copyright protection. Id. Similar to Experian in this case, the compiler in Key Publications excluded information that she thought would not be useful for her customers. Id. at 513. She excluded businesses that would not remain open long, including "certain insurance brokers, take-out restaurants, and traditional Chinese medical practitioners." Id. The court held that this process of exclusion "indicate[d] thought and creativity in the selection of businesses," and that the work was copyrightable. Id. In Kregos v. Associated Press, 937 F.2d 700, 702 (2d Cir. 1991), the Second Circuit dealt with a form or chart that conveyed information about the past performances of the opposing baseball pitchers scheduled to start each day's game. The form was protected because, similar to Experian's compilation, there were no other pitching forms that were identical or nearly identical to the plaintiff's. Kregos, 937 F.2d at 705. The form included nine items of information about each pitcher's past performance, grouped into three categories. Id. at 702. The protection was for the form itself, not for the daily factual statistics it contained. Id. The court also cautioned that any relief would be extremely limited because the plaintiff could prevail only against those who used forms that exactly copied his selection of information. Id. at 709-10. Nevertheless, the statistical compilation was entitled to some protection. The Second Circuit reached a different result when considering publications of judicial opinions. In Matthew Bender & Co. v. West Publishing Co., 158 F.3d 674, 677 (2d Cir. 1998), the court held that West's publications of judicial opinions did not merit copyright protection. Although West claimed it had demonstrated sufficient creativity in several respects, including its arrangement of information about parties, courts, and dates of decisions and its addition of certain information concerning counsel, the court concluded that West's selection and arrangement were "obvious, typical, and lack[ed] even minimal creativity." Matthew Bender & Co., 158 F.3d at 677. Creativity was lacking because industry conventions, such as those in the legal profession, made the choices obvious. Such conventions, the court stated, "so dictate selection that any person composing a compilation of the type at issue would necessarily select the same categories of information," and that "creativity inheres in making non-obvious choices from among more than a few options." Id. at 677, 682. In this case, Experian's choices are not obvious, as illustrated by Experian's evidence that the content in its database differs materially from the content in other compilations of consumer data. Closer to our case are two Eleventh Circuit decisions involving directories. Warren Publishing, Inc. v. Microdos Data Corp., 115 F.3d 1509, 1511-12 (11th Cir. 1997) (en banc), involved a compilation of information about cable system operators. The information was printed in a directory that included "the name, address, and telephone number of the cable system operator, the number of subscribers, the channels offered, the price of service, and the types of equipment used." Warren Publ'g, 115 F.3d at 1512. The compiler used whatever information the cable companies provided in response to the compiler's request for information. Id. at 1519. The compilation was not copyrightable under Feist because there was no selectivity; unlike in this case, the compiler included the "entire relevant universe known to it" in the directory. Id. at 1518, 1520. In BellSouth Advertising & Publishing Corp. v. Donnelley Information Publishing, Inc., 999 F.2d 1436, 1439, 1442 (11th Cir. 1993) (en banc) (BellSouth), the Eleventh Circuit similarly held that the plaintiff's Miami yellow page business directory was not sufficiently original to merit copyright protection because the contents involved no more creativity than the telephone book in Feist. Unlike this case, the listed businesses in BellSouth chose whether to be listed and the information to be included. 999 F.2d at 1441. A Fifth Circuit case involved a process of data selection similar to the one in this case. The underlying information conveyed was factual, but the manner of selecting and displaying the information involved originality. Mason v. Montgomery Data, Inc., 967 F.2d 135, 141-42 (5th Cir. 1992). In Mason, the plaintiff selected real estate ownership information and displayed it on maps. Id. at 136. The court held that this was sufficiently creative in selection to warrant copyright protection, because the compiler made "choices . . . independently . . . to select information from numerous and sometimes conflicting sources," including from factual public records, and combined that information to make "an effective pictorial expression of those locations." Id. at 136 n.3, 140, 141. The same is true here. From this survey of circuit decisions since Feist, we draw three general principles. First, although facts are not entitled to copyright protection, factual compilations are entitled to some protection as long as there is creativity in the selection, arrangement, or coordination of the facts. Feist, 499 U.S. at 358; Warren Publ'g, 115 F.3d at 1515; Mason, 967 F.2d at 141. Second, the creativity that suffices to establish copyright protection in factual compilations is minimal. Feist, 499 U.S. at 348; Key Publ'ns, 945 F.2d at 512-13. For illustration, the compilation of business names, addresses, and phone numbers of interest to the New York City Chinese-American community was sufficiently creative to warrant copyright protection of a directory in Key Publications. 945 F.2d at 512-13; see also Kregos, 937 F.2d at 701-02. Third, such compilations of factual information receive only limited protection. 17 U.S.C. § 103(b); Feist, 499 U.S. at 359; Kregos, 937 F.2d at 702, 709. This means that a compiler may freely use the facts contained in a compilation when preparing a competing work, as long as the competing work does not exhibit the same selection or arrangement. See Feist, 499 U.S. at 349; see also Kregos, 937 F.2d at 702, 709. Applying these principles to this case leads to the conclusion that Experian's lists are entitled to limited protection. Experian's selection process in culling data from multiple sources and selecting the appropriate pairing of addresses with names before entering them in the database involves a process of at least minimal creativity. The listings are compiled by first collecting and comparing multiple sources, and then sorting conflicting information through the creation of business rules that Experian created to select from among the conflicts. As the Fifth Circuit said in Mason, selection is sufficiently creative when the compiler makes "choices . . . independently . . . to select information from numerous and sometimes conflicting sources." 967 F.2d at 141. Experian's process more than meets that standard. Much like the compilers in Mason and Key Publications, Experian's employees choose from multiple and sometimes conflicting sources, and they use their judgment in selecting which names and addresses to include in the database. See Mason, 967 F.2d at 141; Key Publ'ns, 945 F.2d at 512-13. Experian's employees, like the compiler in Key Publications, also exclude information they deem irrelevant to the interests of Experian's marketing clients, information such as business addresses, and the names and addresses of the very elderly and incarcerated. See 945 F.2d at 513. Such exclusions indicate some "thought and creativity in the selection" of names and addresses to include in the database, id., which indicates the "modicum of creativity necessary to transform mere selection into copyrightable expression." Feist, 499 U.S. at 362. The name and address pairings in Experian's database are also materially different from those in other databases. With respect to the baseball pitching form in Kregos, the Second Circuit observed that "[t]here is no prior form that is identical . . . nor one from which [it] varies in only a trivial degree." 937 F.2d at 705. The same observation can be made here. Natimark asserts that Experian's selectivity cannot be creative, because it is no more than a lengthy process to discover facts, which are not copyrightable. Natimark attempts to match Experian's process to the Supreme Court's discussion in Feist of "industrious collection," i.e., the hard work that a compiler exerts in gathering facts that lacks creativity to warrant copyright protection. 499 U.S. at 352-53. Experian's methods, however, do not entail simple replication of the data that Experian receives, as was the situation in Feist. 499 U.S. at 362; see also Warren Publ'g, 115 F.3d at 1517-20; BellSouth, 999 F.2d at 1441. In all those cases, the directory compilers were simply fed the data. See Feist, 499 U.S. at 362 ("In preparing its white pages, [the telephone company] simply t[ook] the data provided by its subscribers and list[ed] it alphabetically by surname."); Warren Publ'g, 115 F.3d at 1519-20 (holding that directory of cable operator information was not copyrightable when the selection was not the compiler's "own, but rather that of the cable operators"); BellSouth, 999 F.2d at 1438, 1441 (involving a yellow pages directory containing information entirely provided by others). Experian does not include the "entire relevant universe known to it," as in Warren Publishing. 115 F.3d at 1518. Rather, Experian's selection process produces different, and at least according to Experian, more reliable data than the other four largest database compilers in the United States. On the basis of our review of the decisions in this and other circuits, we must conclude that the name and address pairings in this case are copyrightable as compilations under post-Feist standards. That does not end the copyright claim inquiry, however. To establish copyright infringement, the plaintiff must prove not only ownership of a valid copyright, but copying by the alleged infringer of constituent elements of the protected work. Feist, 499 U.S. at 361; Unicolors, Inc. v. Urban Outfitters, Inc., 853 F.3d 980, 984 (9th Cir. 2017). We therefore must consider whether Experian has shown that Natimark infringed, i.e., copied the material. Although this issue was not reached by the District Court as it found the material was not copyrightable, we may affirm the District Court on any ground supported by the record. See, e.g., Cosmetic Ideas, Inc. v. IAC/Interactivecorp, 606 F.3d 612, 615 (9th Cir. 2010) (citation omitted).

2. The Java Language and Its API — Important Details. Java syntax includes separators (e.g., {, }, ;), operators (e.g., +, -, *, /, <, >), literal values (e.g., 123, 'x', "Foo"), and keywords (e.g., if, else, while, return). These elements carry precise predefined meanings. Java syntax also includes identifiers (e.g., String, java.lang.Object), which are used to name specific values, fields, methods, and classes as described below. These syntax elements are used to form statements, each statement being a single command executed by the Java compiler to take some action. Statements are run in the sequence written. Statements are commands that tell the computer to do work. A method is like a subroutine. Once declared, it can be invoked or "called on" elsewhere in the program. When a method is called on elsewhere in the program or in an application, "arguments" are usually passed to the method as inputs. The output from the method is known as the "return." An example is a method that receives two numbers as inputs and returns the greater of the two as an output. Another example is a method that receives an angle expressed in degrees and returns the cosine of that angle. Methods can be much more complicated. A method, for example, could receive the month and day and return the Earth's declination to the sun for that month and day. A method consists of the method header and the method body. A method header contains the name of the method; the number, order, type and name of the parameters used by the method; the type of value returned by the method; the checked exceptions that the method can throw; and various method modifiers that provide additional information about the method. At the trial, witnesses frequently referred to the method header as the "declaration." This discrepancy has no impact on the ultimate analysis. The main point is that this header line of code introduces the method body and specifies very precisely its inputs, name and other functionality. Anyone who wishes to supply a method with the same functionality must write this line of code in the same way and must do so no matter how different the implementation may be from someone else's implementation. The method body is a block of code that then implements the method. If a method is declared to have a return type, then the method body must have a statement and the statement must include the expression to be returned when that line of code is reached. During trial, many witnesses referred to the method body as the "implementation." It is the method body that does the heavy lifting, namely the actual work of taking the inputs, crunching them, and returning an answer. The method body can be short or long. Google came up with its own implementations for the method bodies and this accounts for 97 percent of the code for the 37 packages. Once the method is written, tested and in place, it can be called on to do its work. A method call is a line of code somewhere else, such as in a different program that calls on (or invokes) the method and specifies the arguments to be passed to the method for crunching. The method would be called on using the command format "java.package.Class.method()" where () indicates the inputs passed to the method. For example, a = java.package.Class.method() would set the field "a" to equal the return of the method called. (The words "java.package.Class.method" would in a real program be other names like "java.lang.Math.max"; "java.package.Class.method" is used here simply to explain the format.) After a method, the next higher level of syntax is the class. A class usually includes fields that hold values (such as pi = 3.141592) and methods that operate on those values. Classes are a fundamental structural element in the Java language. A Java program is written as one or more classes. More than one method can be in a class and more than one class can be in a package. All code in a Java program must be placed in a class. A class declaration (or header) is a line that includes the name of the class and other information that define the class. The body of the class includes fields and methods, and other parameters. Classes can have subclasses that "inherit" the functionality of the class itself. When a new subclass is defined, the declaration line uses the word "extends" to alert the compiler that the fields and methods of the parent class are inherited automatically into the new subclass so that only additional fields or methods for the subclass need to be declared. The Java language does not allow a class to extend (be a subclass of) more than one parent class. This restrictiveness may be problematic when one class needs to inherit fields and methods from two different non-related classes. The Java programming language alleviates this dilemma through the use of "interfaces," which refers to something different from the word "interface" in the API acronym. An interface is similar to a class. It can also contain methods. It is also in its own source code file. It can also be inherited by classes. The distinction is that a class may inherit from more than one interface whereas, as mentioned, a class can only inherit from one other class. For convenience, classes and interfaces are grouped into "packages" in the same way we all group files into folders on our computers. There is no inheritance function within packages; inheritance occurs only at the class and interface level. Here is a simple example of source code that illustrates methods, classes and packages. The italicized comments on the right are merely explanatory and are not compiled: package java.lang; // Declares package java.lang public class Math { // Declares class Math public static int max (int x, int y) { // Declares method max if (x > y) return x ; // Implementation, returns x or else return y ; // Implementation, returns y } // Closes method } // Closes class To invoke this method from another program (or class), the following call could be included in the program: int a = java.lang.Math.max (2, 3); Upon reaching this statement, the computer would go and find the max method under the Math class in the java.lang package, input "2" and "3" as arguments, and then return a "3," which would then be set as the value of "a." The above example illustrates a point critical to our first main copyright issue, namely that the declaration line beginning "public static" is entirely dictated by the rules of the language. In order to declare a particular functionality, the language demands that the method declaration take a particular form. There is no choice in how to express it. To be specific, that line reads: public static int max (int x, int y) { The word "public" means that other programs can call on it. (If this instead says "private," then it can only be accessed by other methods inside the same class.) The word "static" means that the method can be invoked without creating an instance of the class. (If this instead is an instance method, then it would always be invoked with respect to an object.) The word "int" means that an integer is returned by the method. (Other alternatives are "boolean," "char," and "String" which respectively mean "true/false," "single character," and "character string.") Each of these three parameters is drawn from a short menu of possibilities, each possibility corresponding to a very specific functionality. The word "max" is a name and while any name (other than a reserved word) could have been used, names themselves cannot be copyrighted, as will be shown. The phrase "(int x, int y)" identifies the arguments that must be passed into the method, stating that they will be in integer form. The "x" and the "y" could be "a" and "b" or "arg1" and "arg2," so there is a degree of creativity in naming the arguments. Again, names cannot be copyrighted. (Android did not copy all of the particular argument names used in Java but did so as to some arguments.) Finally, "{" is the beginning marker that tells the compiler that the method body is about to follow. The marker is mandatory. The foregoing description concerns the rules for the language itself. Again, each parameter choice other than the names has a precise functional choice. If someone wants to implement a particular function, the declaration specification can only be written in one way. Part of the declaration of a method can list any exceptions. When a program violates the semantic constraints of the Java language, the Java virtual machine will signal this error to the program as an exception for special handling. These are specified via "throw" statements appended at the end of a declaration. Android and Java are not identical in their throw designations but they are very similar as to the 37 packages at issue. A Java program must have at least one class. A typical program would have more than one method in a class. Packages are convenient folders to organize the classes. This brings us to the application programming interface. When Java was first introduced in 1996, the API included eight packages of pre-written programs. At least three of these packages were "core" packages, according to Sun, fundamental to being able to use the Java language at all. These packages were java.lang, java.io, and java.util. As a practical matter, anyone free to use the language itself (as Oracle concedes all are), must also use the three core packages in order to make any worthwhile use of the language. Contrary to Oracle, there is no bright line between the language and the API. Each package was broken into classes and those in turn broken into methods. For example, java.lang (a package) included Math (a class) which in turn included max (a method) to return the greater of two inputs, which was (and remains) callable as java.lang.Math.max with appropriate arguments (inputs) in the precise form required (see the example above). After Java's introduction in 1996, Sun and the Java Community Process, a mechanism for developing a standard specifications for Java classes and methods, wrote hundreds more programs to carry out various nifty functions and they were organized into coherent packages by Sun to become the Java application programming interface. In 2008, as stated, the Java API had grown from the original eight to 166 packages with over six hundred classes with over six thousand methods. All of it was downloadable from Sun's (now Oracle's) website and usable by anyone, including Java application developers, upon agreement to certain license restrictions. Java was particularly useful for writing programs for use via the Internet and desktop computers. Although the declarations must be the same to achieve the same functionality, the names of the methods and the way in which the methods are grouped do not have to be the same. Put differently, many different API organizations could supply the same overall range of functionality. They would not, however, be interoperable. Specifically, code written for one API would not run on an API organized differently, for the name structure itself dictates the precise form of command to call up any given method. To write a fresh program, a programmer names a new class and adds fields and methods. These methods can call upon the pre-written functions in the API. Instead of re-inventing the wheels in the API from scratch, programmers can call on the tried-and-true pre-packaged programs in the API. These are ready-made to perform a vast menu of functions. This is the whole point of the API. For example, a student in high school can write a program that can call upon java.lang.Math.max to return the greater of two numbers, or to find the cosine of an angle, as one step in a larger homework assignment. Users and developers can supplement the API with their own specialized methods and classes. The foregoing completes the facts necessary to decide the copyrightability issue but since Oracle has made much of two small items copied by Google, this order will now make findings thereon so that there will be proper context for the court of appeals.

ANALYSIS AND CONCLUSIONS OF LAW 1. Names and Short Phrases. To start with a clear-cut rule, names, titles and short phrases are not copyrightable, according to the United States Copyright Office, whose rule thereon states as follows: Copyright law does not protect names, titles, or short phrases or expressions. Even if a name, title, or short phrase is novel or distinctive or lends itself to a play on words, it cannot be protected by copyright. The Copyright Office cannot register claims to exclusive rights in brief combinations of words such as: • Names of products or services. • Names of business organizations, or groups (including the names of performing groups). • Pseudonyms of individuals (including pen or stage names). • Titles of works. • Catchwords, catchphrases, mottoes, slogans, or short advertising expressions. • Listings of ingredients, as in recipes, labels, or formulas. When a recipe or formula is accompanied by an explanation or directions, the text directions may be copyrightable, but the recipe or formula itself remains uncopyrightable. U.S. Copyright Office, Circular 34; see 37 C.F.R. 202.1(a). This rule is followed in the Ninth Circuit. Sega Enters., Ltd. v. Accolade, Inc., 977 F.2d 1510, 1524 n.7 (9th Cir. 1992). This has relevance to Oracle's claim of copyright ownership over names of methods, classes and packages. 2. The Development of Law on the Copyrightability Of Computer Programs and Their Structure, Sequence and Organization. Turning now to the more difficult question, this trial showcases a distinction between copyright protection and patent protection. It is an important distinction, for copyright exclusivity lasts 95 years whereas patent exclusivity lasts twenty years. And, the Patent and Trademark Office examines applications for anticipation and obviousness before allowance whereas the Copyright Office does not. This distinction looms large where, as here, the vast majority of the code was not copied and the copyright owner must resort to alleging that the accused stole the "structure, sequence and organization" of the work. This phrase — structure, sequence and organization — does not appear in the Act or its legislative history. It is a phrase that crept into use to describe a residual property right where literal copying was absent. A question then arises whether the copyright holder is more appropriately asserting an exclusive right to a functional system, process, or method of operation that belongs in the realm of patents, not copyrights. A. Baker v. Seldon. The general question predates computers. In the Supreme Court's decision in Baker v. Selden, 101 U.S. 99, 25 L. Ed. 841, 1880 Dec. Comm'r Pat. 422 (1879), the work at issue was a book on a new system of double-entry bookkeeping. It included blank forms, consisting of ruled lines, and headings, illustrating the system. The accused infringer copied the method of bookkeeping but used different forms. The Supreme Court framed the issue as follows: The evidence of the complainant is principally directed to the object of showing that Baker uses the same system as that which is explained and illustrated in Selden's books. It becomes important, therefore, to determine whether, in obtaining the copyright of his books, he secured the exclusive right to the use of the system or method of book-keeping which the said books are intended to illustrate and explain. Id. at 101. Baker held that using the same accounting system would not constitute copyright infringement. The Supreme Court explained that only patent law can give an exclusive right to a method: To give to the author of the book an exclusive property in the art described therein, when no examination of its novelty has ever been officially made, would be a surprise and a fraud upon the public. That is the province of letters-patent, not of copyright. The claim to an invention or discovery of an art or manufacture must be subjected to the examination of the Patent Office before an exclusive right therein can be obtained; and it can only be secured by a patent from the government. Id. at 102. The Supreme Court went on to explain that protecting the method under copyright law would frustrate the very purpose of publication: The copyright of a work on mathematical science cannot give to the author an exclusive right to the methods of operation which he propounds, or to the diagrams which he employs to explain them, so as to prevent an engineer from using them whenever occasion requires. The very object of publishing a book on science or the useful arts is to communicate to the world the useful knowledge which it contains. But this object would be frustrated if the knowledge could not be used without incurring the guilt of piracy of the book. Id. at 103. Baker also established the "merger" doctrine for systems and methods intermingled with the texts or diagrams illustrating them: And where the art it teaches cannot be used without employing the methods and diagrams used to illustrate the book, or such as are similar to them, such methods and diagrams are to be considered as necessary incidents to the art, and given therewith to the public; not given for the purpose of publication in other works explanatory of the art, but for the purpose of practical application. Ibid. It is true that Baker is aged but it is not passé. To the contrary, even in our modern era, Baker continues to be followed in the appellate courts, as will be seen below. B. The Computer Age and Section 102(b) of the 1976 Act. Almost a century later, Congress revamped the Copyright Act in 1976. By then, software for computers was just emerging as a copyright issue. Congress decided in the 1976 Act that computer programs would be copyrightable as "literary works." See H.R. Rep. No. 94-1476, at 54 (1976). There was, however, no express definition of a computer program until an amendment in 1980. The 1976 Act also codified a Baker-like limitation on the scope of copyright protection in Section 102(b). See Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1443 n.11 (9th Cir. 1994). Section 102(b) stated (and still states): In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work. The House Report that accompanied Section 102(b) of the Copyright Act explained: Copyright does not preclude others from using the ideas or information revealed by the author's work. It pertains to the literary, musical, graphic, or artistic form in which the author expressed intellectual concepts. Section 102(b) makes clear that copyright protection does not extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work. Some concern has been expressed lest copyright in computer programs should extend protection to the methodology or processes adopted by the programmer, rather than merely to the 'writing' expressing his ideas. Section 102(b) is intended, among other things, to make clear that the expression adopted by the programmer is the copyrightable element in a computer program, and that the actual processes or methods embodied in the program are not within the scope of the copyright law. Section 102(b) in no way enlarges or contracts the scope of copyright protection under the present law. Its purpose is to restate, in the context of the new single Federal system of copyright, that the basic dichotomy between expression and idea remains unchanged. H.R. Rep. No. 94-1476, at 56-57 (1976) (emphasis added). Recognizing that computer programs posed novel copyright issues, Congress established the National Commission on New Technological Uses of Copyrighted Works (referred to as CONTU) to recommend the extent of copyright protection for software. The Commission consisted of twelve members with Judge Stanley Fuld as chairman and Professor Melville Nimmer as vice-chairman. The Commission recommended that a definition of "computer program" be added to the copyright statutes. This definition was adopted in 1980 and remains in the current statute: A "computer program" is a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result. 17 U.S.C. 101. Moreover, the CONTU report stated that Section 102(b)'s preclusion of copyright protection for "procedure, process, system, method of operation" was reconcilable with the new definition of "computer program." The Commission explained the dichotomy between copyrightability and non-copyrightability as follows: Copyright, therefore, protects the program so long as it remains fixed in a tangible medium of expression but does not protect the electromechanical functioning of a machine. The way copyright affects games and game-playing is closely analogous: one may not adopt and republish or redistribute copyrighted game rules, but the copyright owner has no power to prevent others from playing the game. Thus, one is always free to make a machine perform any conceivable process (in the absence of a patent), but one is not free to take another's program. Nat'l Comm'n on New Technological Uses of Copyrighted Works, Final Report 20 (1979) (emphasis added). The Commission also recognized the "merger" doctrine, a rule of importance a few pages below in this order (emphasis added): The "idea-expression identity" exception provides that copyrighted language may be copied without infringing when there is but a limited number of ways to express a given idea. This rule is the logical extension of the fundamental principle that copyright cannot protect ideas. In the computer context this means that when specific instructions, even though previously copyrighted, are the only and essential means of accomplishing a given task, their later use by another will not amount to an infringement . . . . [C]opyright protection for programs does not threaten to block the use of ideas or program language previously developed by others when that use is necessary to achieve a certain result. When other language is available, programmers are free to read copyrighted programs and use the ideas embodied in them in preparing their own works. Ibid. The Commission realized that differentiating between the copyrightable form of a program and the uncopyrightable process was difficult, and expressly decided to leave the line drawing to federal courts: [T]he many ways in which programs are now used and the new applications which advancing technology will supply may make drawing the line of demarcation more and more difficult. To attempt to establish such a line in this report written in 1978 would be futile. . . . Should a line need to be drawn to exclude certain manifestations of programs from copyright, that line should be drawn on a case-by-case basis by the institution designed to make fine distinctions — the federal judiciary. Id. at 22-23. Congress prepared no legislative reports discussing the CONTU comments regarding Section 102(b). See H.R. Rep. No. 96-1307, at 23-24 (1980). Nevertheless, Congress followed CONTU's recommendations by adding the definition of computer programs to the statute and amending a section of the Act not relevant to this order. See Apple Computer, Inc. v. Formula Intern. Inc., 725 F.2d 521, 522-25 (9th Cir. 1984). Everyone agrees that no one can copy line-for-line someone else's copyrighted computer program. When the line-by-line listings are different, however, some copyright owners have nonetheless accused others of stealing the "structure, sequence and organization" of the copyrighted work. That is the claim here. C. Decisions Outside the Ninth Circuit. No court of appeals has addressed the copyrightability of APIs, much less their structure, sequence and organization. Nor has any district court. Nevertheless, a review of the case law regarding non-literal copying of software provides guidance. Circuit decisions outside the Ninth Circuit will be considered first. The Third Circuit led off in Whelan Associates, Inc. v. Jaslow Dental Laboratory, Inc., 797 F.2d 1222 (3d Cir. 1986). In that case, the claimant owned a program, Dentalab, that handled the administrative and bookkeeping tasks of dental prosthetics businesses. The accused infringer developed another program, Dentcom, using a different programming language. The Dentcom program handled the same tasks as the Dentalab program and had the following similarities: The programs were similar in three significant respects . . . most of the file structures, and the screen outputs, of the programs were virtually identical . . . five particularly important "subroutines" within both programs — order entry, invoicing, accounts receivable, end of day procedure, and end of month procedure — performed almost identically in both programs. Id. at 1228. On these facts, the district court had found, after a bench trial, that the accused infringer copied the claimant's software program. Id. at 1228-29. On appeal, the accused infringer argued that the structure of the claimant's program was not protectable under copyright. In rejecting this argument, the court of appeals created the following framework to deal with non-literal copying of software: [T]he line between idea and expression may be drawn with reference to the end sought to be achieved by the work in question. In other words, the purpose or function of a utilitarian work would be the work's idea, and everything that is not necessary to that purpose or function would be part of the expression of the idea. Id. at 1236 (emphasis in original). Applying this test, Whelan found that the structure of Dentalab was copyrightable because there were many different ways to structure a program that managed a dental laboratory: [T]he idea of the Dentalab program was the efficient management of a dental laboratory (which presumably has significantly different requirements from those of other businesses). Because that idea could be accomplished in a number of different ways with a number of different structures, the structure of the Dentalab program is part of the program's expression, not its idea. Id. at 1236 n.28. The phrase "structure, sequence and organization" originated in a passage in Whelan explaining that the opinion used those words interchangeably and that, although not themselves part of the Act, they were intended to capture the thought that "sequence and order could be parts of the expression, not the idea, of a work." Id. at 1239, 1248. To summarize, in affirming the district court's final judgment of infringement, Whelan held that the structure of the Dentalab program was copyrightable because there were many other ways to perform the same function of handling the administrative and bookkeeping tasks of dental prosthetics businesses with different structures and designs. Id. at 1238. Others were free to come up with their own version but could not appropriate the Dentalab structure. This decision plainly seems to have been the high-water mark of copyright protection for the structure, sequence and organization of computer programs. It was also the only appellate decision found by the undersigned judge that affirmed (or directed) a final judgment of copyrightability on a structure, sequence and organization theory. Perhaps because it was the first appellate decision to wade into this problem, Whelan has since been criticized by subsequent treatises, articles, and courts, including our own court of appeals. See Sega Enters., Ltd. v. Accolade, Inc., 977 F.2d 1510, 1524-25 (9th Cir. 1992). Instead, most circuits, including ours, have adopted some variation of an approach taken later by the Second Circuit. See Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1445 (9th Cir. 1994). In Computer Associates International, Inc. v. Altai, Inc., 982 F.2d 693 (2d Cir. 1992), the claimant owned a program designed to translate the language of another program into the particular language that the computer's operating system would be able to understand. The accused infringer developed its own program with substantially similar structure but different source code (using the same programming language). The Second Circuit criticized Whelan for taking too narrow a view of the "idea" of a program. The Second Circuit adopted instead an "abstract-filtration-comparison" test. The test first dissected the copyrighted program into its structural components: In ascertaining substantial similarity under [the abstract-filtration-comparison test], a court would first break down the allegedly infringed program into its constituent structural parts. Then, by examining each of these parts for such things as incorporated ideas, expression that is necessarily incidental to those ideas, and elements that are taken from the public domain, a court would then be able to sift out all non-protectable material. Id. at 706. Then, the test filtered out structures that were not copyrightable. For this filtration step, the court of appeals relied on the premise that programmers fashioned structures "to maximize the program's speed, efficiency, as well as simplicity for user operation, while taking into consideration certain externalities such as the memory constraints of the computer upon which the program will be run." Id. at 698. Because these were "practical considerations," the court held that structures based on these considerations were not copyrightable expressions. Thus, for the filtration step, the court of appeals outlined three types of structures that should be precluded from copyright protection. First, copyright protection did not extend to structures dictated by efficiency. A court must inquire whether the use of this particular set of modules [is] necessary efficiently to implement that part of the program's process being implemented. If the answer is yes, then the expression represented by the programmer's choice of a specific module or group of modules has merged with their underlying idea and is unprotected. Id. at 708 (emphasis in original). Paradoxically, this meant that non-efficient structures might be copyrightable while efficient structures may not be. Nevertheless, the Second Circuit explained its reasoning as follows: In the context of computer program design, the concept of efficiency is akin to deriving the most concise logical proof or formulating the most succinct mathematical computation. Thus, the more efficient a set of modules are, the more closely they approximate the idea or process embodied in that particular aspect of the program's structure. While, hypothetically, there might be a myriad of ways in which a programmer may effectuate certain functions within a program — i.e., express the idea embodied in a given subroutine — efficiency concerns may so narrow the practical range of choice as to make only one or two forms of expression workable options. Ibid. Efficiency also encompassed user simplicity and ease of use. Id. at 708-09. Second, copyright protection did not extend to structures dictated by external factors. The court explained this as follows: [I]n many instances it is virtually impossible to write a program to perform particular functions in a specific computing environment without employing standard techniques. This is a result of the fact that a programmer's freedom of design choice is often circumscribed by extrinsic considerations such as (1) the mechanical specifications of the computer on which a particular program is intended to run; (2) compatibility requirements of other programs with which a program is designed to operate in conjunction; (3) computer manufacturers' design standards; (4) demands of the industry being serviced; and (5) widely accepted programming practices within the computer industry. Id. at 709-10. Third, copyright protection did not extend to structures already found in the public domain. The court reasoned that materials in the public domain, such as elements of a computer program that have been freely accessible, cannot be appropriated. Ibid. Ultimately, in the case before it, the Second Circuit held that after removing unprotectable elements using the criteria discussed above, only a few lists and macros in accused product were similar to the copied product, and their impact on the program was not large enough to declare copyright infringement. Id. at 714-15. The copyright claim, in short, failed. The Tenth Circuit elaborated on the abstract-filtration-comparison test in Gates Rubber Co. v. Bando Chemical Industries, Ltd., 9 F.3d 823 (10th Cir. 1993). There, the claimant developed a computer program that determined the proper rubber belt for a particular machine by performing complicated calculations involving numerous variables. The program used published formulas in conjunction with certain mathematical constants developed by the claimant to determine belt size. The Tenth Circuit offered the following description of a software program's structure: The program's architecture or structure is a description of how the program operates in terms of its various functions, which are performed by discrete modules, and how each of these modules interact with each other. Id. at 835. As had the Second Circuit, the Tenth Circuit held that filtration should eliminate the unprotectable elements of processes, facts, public domain information, merger material, scenes a faire material, and other unprotectable elements suggested by the particular facts of the program under examination. For Section 102(b) processes, the court gave the following description: Returning then to our levels of abstraction framework, we note that processes can be found at any level, except perhaps the main purpose level of abstraction. Most commonly, processes will be found as part of the system architecture, as operations within modules, or as algorithms. Id. at 837. The court described the scenes a faire doctrine for computer programs as follows: The scenes a faire doctrine also excludes from protection those elements of a program that have been dictated by external factors. In the area of computer programs these external factors may include: hardware standards and mechanical specifications, software standards and compatibility requirements, Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525-27 (9th Cir. 1993), computer manufacturer design standards, target industry practices and demands, and computer industry programming practices. * * * We recognize that the scenes a faire doctrine may implicate the protectability of interfacing and that this topic is very sensitive and has the potential to effect [sic] widely the law of computer copyright. This appeal does not require us to determine the scope of the scenes a faire doctrine as it relates to interfacing and accordingly we refrain from discussing the issue. Id. at 838 & n.14 (all citations omitted except Sega). Like the Second Circuit, the Tenth Circuit also listed many external considerations — such as compatibility, computer industry programming practices, and target industry practices and demands — that would exclude elements from copyright protection under the scenes a faire doctrine. Ultimately, the Tenth Circuit remanded because the district court had failed to make specific findings that fit this framework. The First Circuit weighed in with its 1995 decision Lotus Development Corp. v. Borland International, Inc., 49 F.3d 807 (1st Cir. 1995). In Lotus, the claimant owned the Lotus 1-2-3 spreadsheet program that enabled users to perform accounting functions electronically on a computer. Users manipulated and controlled the program via a series of menu commands, such as "Copy," "Print," and "Quit." In all, Lotus 1-2-3 had 469 commands arranged into more than 50 menus and submenus. Lotus 1-2-3 also allowed users to write "macros," whereby a user could designate a series of command choices (sequence of menus and submenus) with a single macro keystroke. Then, to execute that series of commands, the user only needed to type the single pre-programmed macro keystroke, causing the program to recall and perform the designated series of commands automatically. Id. at 809-10. The accused infringer Borland developed a competing spreadsheet program. Borland included the Lotus menu command hierarchy in its program to make it compatible with Lotus 1-2-3 so that spreadsheet users who were already familiar with Lotus 1-2-3 would be able to switch to the Borland program without having to learn new commands or rewrite their Lotus macros. In so doing, Borland did not copy any of Lotus's underlying source or object code. (The opinion did not say whether the programs were written in the same language.) The district court had ruled that the Lotus 1-2-3 menu command hierarchy was a copyrightable expression because there were many ways to construct a spreadsheet menu tree. Thus, the district court had concluded that the Lotus developers' choice and arrangement of command terms, reflected in the Lotus menu command hierarchy, constituted copyrightable expression. Id. at 810-11. The First Circuit, however, held that the Lotus menu command hierarchy was not copyrightable because it was a method of operation under Section 102(b). The court explained: We think that "method of operation," as that term is used in § 102(b), refers to the means by which a person operates something, whether it be a car, a food processor, or a computer. Thus a text describing how to operate something would not extend copyright protection to the method of operation itself; other people would be free to employ that method and to describe it in their own words. Similarly, if a new method of operation is used rather than described, other people would still be free to employ or describe that method. Id. at 815. The court reasoned that because the menu command hierarchy was essential to make use of the program's functional capabilities, it should be properly categorized as a "method of operation" under Section 102(b). The court explained: The Lotus menu command hierarchy does not merely explain and present Lotus 1-2-3's functional capabilities to the user; it also serves as the method by which the program is operated and controlled . . . . In other words, to offer the same capabilities as Lotus 1-2-3, Borland did not have to copy Lotus's underlying code (and indeed it did not); to allow users to operate its programs in substantially the same way, however, Borland had to copy the Lotus menu command hierarchy. Thus the Lotus 1-2-3 code is not a uncopyrightable "method of operation." Ibid. Thus, the court reasoned that although Lotus had made "expressive" choices of what to name the command terms and how to structure their hierarchy, it was nevertheless an uncopyrightable "method of operation." The Lotus decision was affirmed by an evenly divided Supreme Court (four to four). The Federal Circuit had the opportunity to apply Lotus in an appeal originating from the District of Massachusetts in Hutchins v. Zoll Medical Corp., 492 F.3d 1377 (Fed. Cir. 2007) (affirming summary judgment against copyright owner). In Hutchins, the claimant owned a program for performing CPR and argued that his copyright covered the "system of logic whereby CPR instructions are provided by computerized display, and [] the unique logic contained in [his] software program." Id. at 1384. The claimant argued that the accused program was similar because it "perform[ed] the same task in the same way, that is, by measuring heart activity and signaling the quantity and timing of CPR compressions to be performed by the rescuer." Ibid. The court of appeals rejected this argument, holding that copyright did not protect the "technologic method of treating victims by using CPR and instructing how to use CPR." Ibid. (citing Lotus). D. Decisions in the Supreme Court and in our Circuit. Our case is governed by the law in the Ninth Circuit and, of course, the Supreme Court. The Supreme Court missed the opportunity to address these issues in Lotus due to the four-to-four affirmance and has, thus, never reached the general question. Nonetheless, Baker, which is still good law, provides guidance and informs how we should read Section 102(b). Another Supreme Court decision, Feist Publications, Inc. v. Rural Telephone Services Co., Inc., 499 U.S. 340, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (1991), which dealt primarily with the copyrightability of purely factual compilations, provided some general principles. In Feist, the Supreme Court considered the copyrightability of a telephone directory comprised of names, addresses, and phone numbers organized in alphabetical order. The Supreme Court rejected the notion that copyright law was meant to reward authors for the "sweat of the brow." This meant that we should not yield to the temptation to award copyright protection merely because a lot of sweat went into the work. The Supreme Court concluded that protection only extended to the original components of an author's work. Id. at 353. The Supreme Court concluded: This inevitably means that the copyright in a factual compilation is thin. Notwithstanding a valid copyright, a subsequent compiler remains free to use the facts contained in another's publication to aid in preparing a competing work, so long as the competing work does not feature the same selection and arrangement. Id. at 349. Turning to our own Ninth Circuit, our court of appeals has recognized that non-literal components of a program, including the structure, sequence and organization and user interface, can be protectable under copyright depending on whether the structure, sequence and organization in question qualifies as an expression of an idea rather than an idea itself. Johnson Controls, Inc. v. Phoenix Control Sys., Inc., 886 F.2d 1173, 1175 (9th Cir. 1989). This decision arrived between the Third Circuit's Whelan decision and the Second Circuit's Computer Associates decision. Johnson Controls is one of Oracle's mainstays herein. In Johnson Controls, the claimant developed a system of computer programs to control wastewater treatment plants. The district court found that the structure, sequence and organization of the program was expression and granted a preliminary injunction even though the accused product did not have similar source or object code. Id. at 1174. Therefore, the standard of review on appeal was limited to abuse of discretion and clear error. Our court of appeals affirmed the preliminary injunction, stating that the claimant's program was very sophisticated and each individual application was customized to the needs of the purchaser, indicating there may have been room for individualized expression in the accomplishment of common functions. Since there was some discretion and opportunity for creativity in the structure, the structure of the program was expression rather than an idea. Id. at 1175. Johnson Controls, however, did not elaborate on which particular structures deserved copyright protection. In Brown Bag Software v. Symantec Corp., 960 F.2d 1465 (9th Cir. 1992), our court of appeals outlined a two-part test for determining similarity between computer programs: the extrinsic and intrinsic tests. This pertained to infringement, not copyrightability. The claimant, who owned a computer program for outlining, alleged that an accused infringer copied his program's non-literal features. Id. at 1472. The claimant alleged that seventeen specific features in the programs were similar. On summary judgment, the district court had found that each feature was either not protectable or not similar as a matter of law: The district court ruled that one group of features represented a claim of copyright in "concepts . . . fundamental to a host of computer programs" such as "the need to access existing files, edit the work, and print the work." As such, these features, which took the form of four options in the programs' opening menus, were held to be unprotectable under copyright. A second group of features involved "nine functions listed in the menu bar" and the fact that "virtually all of the functions of the PC-Outline program can be performed by Grandview." The district court declared that "these functions constitute the idea of the outlining program" and, furthermore, "[t]he expression of the ideas inherent in the features are . . . distinct." The court also held that "the similarity of using the main editing screen to enter and edit data . . . is essential to the very idea of a computer outlining program." The third group of features common to PC-Outline and Grandview concerned "the use of pull-down windows." Regarding these features, the district court made three separate rulings. The court first found that "[p]laintiffs may not claim copyright protection of an . . . expression that is, if not standard, then commonplace in the computer software industry" . . . . [and] that the pull-down windows of the two programs look different. Id. at 1472-73. Our court of appeals affirmed the district court's order without elaborating on the copyrightability rulings quoted above. In Atari Games Corp. v. Nintendo of America Inc., 975 F.2d 832 (Fed. Cir. 1992), the Federal Circuit had occasion to interpret Ninth Circuit copyright precedent. In Atari, the claimant Nintendo sued Atari for copying the Nintendo 10NES program, which prevented the Nintendo game console from accepting unauthorized game cartridges. Atari deciphered the 10NES program through reverse engineering and developed its own program to unlock the Nintendo game console. Atari's new program generated signals indistinguishable from 10NES but was written in a different programming language. Id. at 835-36. Applying our Ninth Circuit precedents, Johnson Controls and Brown Bag, the Federal Circuit affirmed the district court's preliminary injunction for copyright infringement. The Federal Circuit held that the 10NES program contained copyrightable expression because it had organization and sequencing unnecessary to the unlocking function: Nintendo's 10NES program contains more than an idea or expression necessarily incident to an idea. Nintendo incorporated within the 10NES program creative organization and sequencing unnecessary to the lock and key function. Nintendo chose arbitrary programming instructions and arranged them in a unique sequence to create a purely arbitrary data stream. This data stream serves as the key to unlock the NES. Nintendo may protect this creative element of the 10NES under copyright. Id. at 840 (emphasis added). The Federal Circuit stated that there were creative elements in the 10NES program beyond the literal expression used to effect the unlocking process. The district court defined the unprotectable 10NES idea or process as the generation of a data stream to unlock a console. This court discerns no clear error in the district court's conclusion. The unique arrangement of computer program expression which generates that data stream does not merge with the process so long as alternate expressions are available. In this case, Nintendo has produced expert testimony showing a multitude of different ways to generate a data stream which unlocks the NES console. Ibid. (citation omitted). Thus, the Federal Circuit held that the district court did not err in concluding that the 10NES program contained protectable expression and affirmed the preliminary injunction. Next came two decisions holding that Section 102(b) bars from copyright software interfaces necessary for interoperability. The Section 102(b) holdings arose in the context of larger holdings that it had been fair use to copy software to reverse-engineer it so as to isolate the unprotectable segments. These two decisions will now be described in detail. In Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992), the accused infringer had to copy object code in order to understand the interface procedures between the Sega game console and a game cartridge, that is, how the software in the game console interacted with the software in the game cartridge to achieve compatibility. Id. at 1515-16. After learning and documenting these interactions (interface procedures), the accused infringer wrote its own source code to mimic those same interface procedures in its own game cartridges so that its cartridges could run on the Sega console. Our court of appeals held that the copying of object code for the purpose of achieving compatibility was fair use. Notably, in its fair-use analysis, our court of appeals expressly held that the interface procedures for compatibility were functional aspects not copyrightable under Section 102(b): "Accolade copied Sega's software solely in order to discover the functional requirements for compatibility with the Genesis console — aspects of Sega's programs that are not protected by copyright. 17 U.S.C. § 102(b)." Id. at 1522. The court used the phrase "interface procedures," a term describing the interface between applications, multiple times to describe the functional aspect of the interaction between software programs and summarized its analysis of copyrightability as follows: In summary, the record clearly establishes that disassembly of the object code in Sega's video game cartridges was necessary in order to understand the functional requirements for Genesis compatibility. The interface procedures for the Genesis console are distributed for public use only in object code form, and are not visible to the user during operation of the video game program. Because object code cannot be read by humans, it must be disassembled, either by hand or by machine. Disassembly of object code necessarily entails copying. Those facts dictate our analysis of the second statutory fair use factor. If disassembly of copyrighted object code is per se an unfair use, the owner of the copyright gains a de facto monopoly over the functional aspects of his work — aspects that were expressly denied copyright protection by Congress. 17 U.S.C. § 102(b). In order to enjoy a lawful monopoly over the idea or functional principle underlying a work, the creator of the work must satisfy the more stringent standards imposed by the patent laws. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 159-64, 109 S.Ct. 971, 982-84, 103 L.Ed.2d 118 (1989). Sega does not hold a patent on the Genesis console. Sega, 977 F.2d at 1526 (emphasis added). In Sega, the interface procedure that was required for compatibility was "20 bytes of initialization code plus the letters S—E—G—A." Id. at 1524 n.7. Our court of appeals found that this interface procedure was functional and therefore not copyrightable under Section 102(b). The accused infringer Accolade was free to copy this interface procedure for use in its own games to ensure compatibility with the Sega Genesis game console. Our court of appeals distinguished the Atari decision, where the Federal Circuit had found that the Nintendo's 10NES security system was infringed, because there was only one signal that unlocked the Sega console, unlike the "multitude of different ways to unlock" the Nintendo console: We therefore reject Sega's belated suggestion that Accolade's incorporation of the code which "unlocks" the Genesis III console is not a fair use. Our decision on this point is entirely consistent with Atari v. Nintendo, 975 F.2d 832 (Fed. Cir.1992). Although Nintendo extended copyright protection to Nintendo's 10NES security system, that system consisted of an original program which generates an arbitrary data stream "key" which unlocks the NES console. Creativity and originality went into the design of that program. See id. at 840. Moreover, the federal circuit concluded that there is a "multitude of different ways to generate a data stream which unlocks the NES console." Atari, 975 F.2d at 839. The circumstances are clearly different here. Sega's key appears to be functional. It consists merely of 20 bytes of initialization code plus the letters S-E-G-A. There is no showing that there is a multitude of different ways to unlock the Genesis III console. Sega, 977 F.2d at 1524 n.7. This order reads Sega footnote seven (quoted above) as drawing a line between copying functional aspects necessary for compatibility (not copyrightable) versus copying functional aspects unnecessary for compatibility (possibly copyrightable). Our court of appeals explained that in Atari, the Nintendo game console's 10NES program had had functionality unnecessary to the lock-and-key function. See also Atari, 975 F.2d at 840. Since the accused infringer Atari had copied the entire 10NES program, it also had copied aspects of the 10NES program unnecessary for compatibility between the console and game cartridges. This was inapposite to the facts of Sega, where the accused infringer Accolade's final product duplicated only the aspect of Sega's program necessary for compatibility between the console and game cartridges. Thus, the holding of our court of appeals was that the aspect of a program necessary for compatibility was unprotectable, specifically invoking Section 102(b), but copyrightable expression could still exist for aspects unnecessary for compatibility. The Sega decision and its compatibility reasoning was followed in a subsequent reverse-engineering decision by our court of appeals, Sony Computer Entertainment, Inc., v. Connectix Corporation, 203 F.3d 596 (9th Cir. 2000). The facts were somewhat different in Sony. There, the accused infringer Connectix did not create its own games for Sony's Playstation game console; instead, the accused infringer created an emulated environment that duplicated the interface procedures of Sony's console so that games written for Sony's console could be played on a desktop computer running the emulator. In order to do this, the accused infringer copied object code for the Sony Playstation's operating software, its BIOS program, in order to discover signals sent between the BIOS and the rest of the game console. Id. at 600. After uncovering these signals (again, application interfaces), the accused infringer wrote its own source code to duplicate these interfaces in order to create its emulator for the desktop computer. Thus, games written for the Playstation console were playable on Connectix's emulator for the desktop computer. Citing Section 102(b) and Sega, our court of appeals stated that the Playstation BIOS contained "unprotected functional elements," and concluded that the accused infringer's intermediate step of copying object code was fair use because it was done for the "purpose of gaining access to the unprotected elements of Sony's software." Id. at 602-03. * * * With apology for its length, the above summary of the development of the law reveals a trajectory in which enthusiasm for protection of "structure, sequence and organization" peaked in the 1980s, most notably in the Third Circuit's Whelan decision. That phrase has not been re-used by the Ninth Circuit since Johnson Controls in 1989, a decision affirming preliminary injunction. Since then, the trend of the copyright decisions has been more cautious. This trend has been driven by fidelity to Section 102(b) and recognition of the danger of conferring a monopoly by copyright over what Congress expressly warned should be conferred only by patent. This is not to say that infringement of the structure, sequence and organization is a dead letter. To the contrary, it is not a dead letter. It is to say that the Whelan approach has given way to the Computer Associates approach, including in our own circuit. See Sega Enters., Ltd. v. Accolade, Inc., 977 F.2d 1510, 1525 (9th Cir. 1992); Apple Computer, Inc. v. Microsoft Corp., 35 F.3d 1435, 1445 (9th Cir. 1994). In this connection, since the CONTU report was issued in 1980, the number of software patents in force in the United States has dramatically increased from barely a thousand in 1980 to hundreds of thousands today. See Iain Cockburn, Patents, Tickets and the Financing of Early-Stage Firms: Evidence from the Software Industry, 18 Journal of Economics & Management Strategy 729-73 (2009). This has caused at least one noted commentator to observe: As software patents gain increasingly broad protection, whatever reasons there once were for broad copyright protection of computer programs disappear. Much of what has been considered the copyrightable "structure, sequence and organization" of a computer program will become a mere incident to the patentable idea of the program or of one of its potentially patentable subroutines. Mark Lemley, Convergence in the Law of Software Copyright?, 10 High Technology Law Journal 1, 26-27 (1995). Both Oracle and Sun have applied for and received patents that claim aspects of the Java API. See, e.g., U.S. Patents 6,598,093 and 7,006,855. (These were not asserted at trial.) * * * In view of the foregoing, this order concludes that our immediate case is controlled by these principles of copyright law: • Under the merger doctrine, when there is only one (or only a few) ways to express something, then no one can claim ownership of such expression by copyright. • Under the names doctrine, names and short phrases are not copyrightable. • Under Section 102(b), copyright protection never extends to any idea, procedure, process, system, method of operation or concept regardless of its form. Functional elements essential for interoperability are not copyrightable. • Under Feist, we should not yield to the temptation to find copyrightability merely to reward an investment made in a body of intellectual property.

In view of the foregoing, this order concludes that our immediate case is controlled by these principles of copyright law: • Under the merger doctrine, when there is only one (or only a few) ways to express something, then no one can claim ownership of such expression by copyright. • Under the names doctrine, names and short phrases are not copyrightable. • Under Section 102(b), copyright protection never extends to any idea, procedure, process, system, method of operation or concept regardless of its form. Functional elements essential for interoperability are not copyrightable. • Under Feist, we should not yield to the temptation to find copyrightability merely to reward an investment made in a body of intellectual property.

The Ninth Circuit has clarified, however, that the jury role in this context is limited to determining disputed “historical facts,” not the inferences or conclusions to be drawn from those facts. See Fisher, 794 F.2d at 436. In Fisher, for example, the court explained that “[n]o material historical facts are at issue in this case. The parties dispute only the ultimate conclusions to be drawn from the admitted facts. Because, under Harper & Row, these judgments are legal in nature, we can make them without usurping the function of the jury.” Id.; see also Seltzer v. Green Day, Inc., 725 F.3d 1170, 1175 (9th Cir. 2013) (“As in Fisher, ‘[n]o material historical facts are at issue in this case. The parties dispute only the ultimate conclusion to be drawn from the admitted facts.’” (citing Fisher, 794 F.2d at 436)); Hustler Magazine, Inc. v. Moral Majority, Inc., 606 F. Supp. 1526, 1532 (C.D. Cal. 1985) (noting that “fair use normally is a question of fact for the jury,” but concluding that “the issue of fair use, at least in the context of this case, presents primarily a question of law”). Accordingly, while inferences from the four-factor analysis and the ultimate question of fair use are “legal in nature,” in the Ninth Circuit, disputed historical facts represent questions for the jury. Fisher, 794 F.2d at 436. Where there are no disputed material historical facts, fair use can be decided by the court alone. Id. Despite this case law, all aspects of Google’s fair use defense went to the jury with neither party arguing that it should not. Thus, the jury was asked not just what the historical facts were, but what the implications of those facts were for the fair use defense. During the first appeal, Google argued to this court that there were disputed issues of material historical fact relevant to its fair use defense. As discussed below, the parties stipulated—or at least ceased to dispute—some of those facts, and presented the remaining disputed historical facts to the jury on remand. The jury returned a verdict in favor of Google on its fair use defense. Because the verdict form—though captioned as a “special verdict”—did not ask the jury to articulate its fact findings in any detail, we must assume that the jury resolved all factual issues relating to the historical facts in favor of the verdict.4 Despite the posture of the fair use finding, we must break that finding into its constituent parts. We must then review the subsidiary and controverted findings of historical fact for substantial evidence. See Seltzer, 725 F.3d at 1175; see also Brewer v. Hustler Magazine, Inc., 749 F.2d 527, 528 (9th Cir. 1984) (“We may disturb a jury verdict only if the evidence was insufficient as a matter of law.”).

Next, Google argues that the Federal Circuit’s approach violates the Seventh Amendment. The Amendment both requires that “the right of trial by jury . . . be preserved”and forbids courts to “re-examin[e]” any “fact tried by a jury.” U. S. Const., Amdt. 7; see also Gasperini v. Center for Humanities, Inc., 518 U. S. 415, 432–433 (1996). The Reexamination Clause is no bar here, however, for, as we have said, the ultimate question here is one of law, not fact. It does not violate the Reexamination Clause for a court to determine the controlling law in resolving a challenge to a jury verdict, as happens any time a court resolves a motionfor judgment as a matter of law. See, e.g., Neely v. Martin K. Eby Constr. Co., 386 U. S. 317, 322 (1967). Nor is Google correct that “the right of trial by jury” includes the right to have a jury resolve a fair use defense. That Clause is concerned with “the particular trial decision”at issue. Markman v. Westview Instruments, Inc., 517 U. S. 370, 376 (1996). Even though it is possible to find pre-Revolutionary English cases in which a judge sent related questions like fair abridgment to a jury, those questions were significantly different from the “fair use” doctrine as courts apply it today. See, e.g., Gyles v. Wilcox, 2 Atk. 141, 142– 144, 26 Eng. Rep. 489, 490–491 (Ch. 1740) (asking theCourt to resolve the narrow question whether a shortened work could be considered a new work); Sayre v. Moore, 1 East 361, n., 102 Eng. Rep. 138, 139, n. (K. B. 1785) (discussing the jury’s role in resolving whether copying constituted infringement). As far as contemporary fair use is concerned, we have described the doctrine as an “equitable,” not a “legal,” doctrine. We have found no case suggestingthat application of U. S. Bank here would fail “to preserve the substance of the common-law [jury trial] right as it existed in 1791.” Markman, 517 U. S., at 376.

In reaching this conclusion, the district court seems to have relied upon language contained in a First Circuit decision: Lotus Development Corp. v. Borland International, Inc., 49 F.3d 807 (1st Cir. 1995), aff’d without opinion by equally divided court, 516 U.S. 233 (1996) . . . On appeal, Oracle argues that the district court’s reliance on Lotus is misplaced because it is distinguishable on its facts and is inconsistent with Ninth Circuit law. We agree.

Importantly, this full analysis only applies where a copyright owner alleges infringement of the non-literal aspects of its work. Where “admitted literal copying of a discrete, easily-conceptualized portion of a work” is at issue—as with Oracle’s declaring code—a court “need not perform a complete abstraction-filtration-comparison analysis” and may focus the protectability analysis on the filtration stage, with attendant reference to standard copyright principles. Mitel, 124 F.3d at 1372-73.

With respect to Google’s cross-appeal, we affirm the district court’s decisions: (1) granting Oracle’s motion for JMOL as to the eight decompiled Java files that Google copied into Android; and (2) denying Google’s motion for JMOL with respect to the rangeCheck function. Accordingly, we affirm-in-part, reverse-in-part, and remand for further proceedings.

6. With respect to the Factor One and "transformativeness," a use is transformative if it "adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message." Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (1994). Oracle argues that no jury could reasonably find that Google's use of the declaring code and SSO from 37 Java API packages in Android imbued the copyrighted works with new expression, meaning, or message. Specifically, Oracle argues that the copied code served the same function in Android as it did in Java, inasmuch as the code served as an interface for accessing methods in both systems (see Astrachan Testimony, Tr. 1265; Bloch Testimony, Tr. 997). It should go without saying (but it must be said anyway) that, of course, the words copied will always be the same (or virtually so) in a copyright case — otherwise there can be no copyright problem in the first place. And, of course, the copied declarations serve the same function in both works, for by definition, declaring code in the Java programming language serves the specific definitional purposes explained above. If this were enough to defeat fair use, it would be impossible ever to duplicate declaring code as fair use and presumably the Federal Circuit would have disallowed this factor on the first appeal rather than remanding for a jury trial. With respect to transformativeness, our jury could reasonably have found that (i) Google's selection of 37 out of 166 Java API packages (ii) re-implemented with new implementing code adapted to the constrained operating environment of mobile smartphone devices with small batteries, and (iii) combined with brand new methods, classes, and packages written by Google for the mobile smartphone platform — all constituted a fresh context giving new expression, meaning, or message to the duplicated code.(The copyrighted works were designed and used for desktop and laptop computers.) In Campbell, the accused work (a rap parody song) used the same bass riff and an identical first line of Roy Orbison's "Oh, Pretty Woman." The parody also included exact copies of certain phrases in subsequent lines and maintained the same structure and rhyme scheme throughout. The copied elements served the same function in the accused work as in the original. Nevertheless, the Supreme Court acknowledged that the transformative purpose of parody had a "need to mimic an original to make its point," and thus, warranted copying some exact elements. Id. at 580-81. The question of the extent of the copying permissible to serve that function was the subject of the inquiry of the third statutory fair use factor. So too here. Android did not merely incorporate the copyrighted work "as part of a broader work," without any change to the purpose, message, or meaning of the underlying work (see Dkt. No. 1780). Android did not merely adopt the Java platform wholesale as part of a broader software platform without any changes. Instead, it integrated selected elements, namely declarations from 37 packages to interface with all new implementing code optimized for mobile smartphones and added entirely new Java packages written by Google itself. This enabled a purpose distinct from the desktop purpose of the copyrighted works — or so our jury could reasonably have found. In light of the foregoing, our jury could reasonably have concluded that Google's use of the declaring code and SSO of 37 API packages from the desktop platform work in a full-stack, open-source mobile operating system for smartphones was transformative.

7. With respect to Factor Two, the "nature of the copyrighted work," the final charge to the jury stated "[t]his factor recognizes that traditional literary works are closer than informational works, such as instruction manuals, to the core of intended copyright protection. Creative writing and expression lie at the very heart of copyright protection, so fair use is generally more difficult to establish for copying of traditional literary works than for copying of informational works" (Dkt. No. 1981 ¶ 28); see also Campbell, 510 U.S. at 586. The Java programming language itself requires the package-class-method hierarchy, an idea on which Oracle does not claim any copyright. Oracle instead argues that because there were countless ways to name and organize the packages in Java and because Google could have used a completely new taxonomy in Android (except as to the 62 "necessary" classes), our jury should have concluded that the process of designing APIs must have been "highly creative" and thus at the core of copyright's protection. Of course, such a conclusion would have been within the evidence, but our jury could reasonably have gone the other way and concluded that the declaring code was not highly creative. Oracle highlights Google's own witness, Joshua Bloch, who designed many of the Java APIs while working at Sun and who later worked at Google on the Android team. Bloch testified that one of the challenges he faced in designing API was "the complexity of figuring out how best to express what it is that the programmer wants done" (Tr. 1007). Oracle focuses on Bloch's use of the word "express" to demonstrate the expressive nature of API design but it ignores the fact that he addressed the challenge of expressing a particular function. Similarly, Oracle notes that Bloch described API design as "an art not a science" and cites his eloquence regarding "design principles" (Tr. 971). In citing this, Oracle resorts to the time-honored tactic of emphasizing a concession by one of the other side's witness. But other witnesses (e.g., Dr. Owen Astrachan, among others) emphasized the functional role of the declaring lines of code and their SSO and minimized the "creative" aspect. Our jury could reasonably have found that, while the declaring code and SSO were creative enough to qualify for copyright protection, functional considerations predominated in their design, and thus Factor Two was not a strong factor in favor of Oracle after all.

8. With respect to Factor Three, our jury could reasonably have found that Google duplicated the bare minimum of the 37 API packages, just enough to preserve inter-system consistency in usage, namely the declarations and their SSO only, and did not copy any of the implementing code, thus finding that Google copied only so much as was reasonably necessary for a transformative use. The number of lines of code duplicated constituted a tiny fraction of one percent of the copyrighted works (and even less of Android, for that matter).

9. With respect to Factor Four, our jury could reasonably have found that use of the declaring lines of code (including their SSO) in Android caused no harm to the market for the copyrighted works, which were for desktop and laptop computers. As to Java ME, our jury could reasonably have found that Java ME eventually declined in revenue just as predicted by Sun before Android was even released, meaning that Android had no further negative impact on Java ME beyond the tailspin already predicted within Sun. Also, before Android was released, Sun made all of the Java API available as free and open source under the name OpenJDK, subject only to the lax terms of the General Public License Version 2 with Classpath Exception. This invited anyone to subset the API. Anyone could have duplicated, for commercial purposes, the very same 37 packages as wound up in Android with the very same SSO and done so without any fee, subject only to lenient "give-back" conditions of the GPLv2+CE. Although Google didn't acquire the 37 packages via OpenJDK, our jury could reasonably have found that Android's impact on the market for the copyrighted works paralleled what Sun already expected via its OpenJDK.

4. With respect to Factor One, Oracle presses hard its view that Google copied in bad faith disregard of Sun/Oracle's property rights. As stated, there remains an ongoing debate regarding whether the "propriety of the use" is a cognizable consideration in any fair use inquiry. Nevertheless, our jury was instructed, as requested by Oracle, to consider whether Google acted in good faith or not as part of its consideration of the first statutory factor. Although mental state is a classic question reserved to the jury, and in our trial mental state was much contested, Oracle now insists that our jury could not reasonably have concluded that Google acted in good faith. Oracle cites numerous examples of internal documents and trial testimony that suggested that Google felt it needed to copy the Java API as an accelerant to bring Android to the market quicker. It points to the breakdown in negotiations between Google and Sun seeking to form a full partnership leaving Google with Java class libraries that were "half-ass at best. [It] need[ed] another half of an ass" (TX 215). In light of that breakdown, Google elected to "[d]o Java anyway and defend [its] decision, perhaps making enemies along the way" (TX 7 at 2). Oracle further notes that even after Sun's CEO at the time publicly praised Android, Andy Rubin (head of Google's Android team) instructed representatives at a trade show, "don't demonstrate [Android] to any [S]un employees or lawyers" (TX 29). Finally, Oracle points to internal communications indicating that Google believed it needed a license to use Java (TX 10; see also TX 409 (discussing the possibility of buying Sun to "solve all these lawsuits we're facing")). On the other hand, Google presented evidence that many at Google (and Sun) understood that at least the declaring code and their SSO were free to use and re-implement, both as a matter of developer practice and because the availability of independent implementations of the Java API enhanced the popularity of the Java programming language, which Sun promoted as free for all to use (Schmidt Testimony, Tr. 361; Page Testimony, Tr. 1846; Rubin Testimony, Tr. 639; Rubin Testimony, Tr. 1088-89). Sun's own CEO at the time, Jonathan Schwartz, testified on Google's behalf at trial and supported Google's view that a practice of duplicating declarations existed and that the competition was on implementations. Oracle's harsh cross-examination focused on character assassination and showing that Schwartz resented Oracle for its treatment of Schwartz after the buyout. That Oracle resorted to such impeachment underscores how fact-bound the issue was, another classic role of a jury to resolve. In light of the foregoing, our jury could reasonably have concluded that Google's use of parts of the Java API as an accelerant was undertaken based on a good faith belief that at least the declaring code and SSO were free to use (which it did use), while a license was necessary for the implementing code (which it did not use). Our jury could reasonably have concluded that Google's concern about making an enemy of Sun reflected concern about the parties' business relationship in light of the failed negotiations that would have brought Sun in as a major partner in Android, rather than concerns about litigation. Mental state was and remains a classic province of the jury.

5. With respect to the Factor One and commercialism, it is undisputed that Google's use of the declaring code and SSO from 37 Java API packages served commercial purposes and our jury was so instructed, including an instruction that a commercial use weighed against fair use. Nevertheless, our jury could reasonably have found that Google's decision to make Android available open source and free for all to use had non-commercial purposes as well (such as the general interest in sharing software innovation). Indeed, Sun itself acknowledged (before Android launched) that making OpenJDK available as open source, as Sun did, could undermine its own commercial efforts with Java SE licensing (TX 971 at 14). Thus, even though Google's use was commercial, which weighed against fair use, the jury could reasonably have found the open-source character of Android tempered Google's overall commercial goals. Of course, even a wholly commercial use may still constitute a fair use. Campbell, 510 U.S. at 585. Thus, in the alternative, our jury could reasonably have found that Google's use of the declaring code and SSO from 37 Java API packages constituted a fair use despite even a heavily commercial character of that use. It is true that in the first appeal, the following exchange occurred at oral argument between Circuit Judge Kathleen O'Malley and counsel for Google: Judge O'Malley: But for purpose and character, though, you don't dispute that it was entirely a commercial purpose. Van Nest: No. Oral Arg., Oracle Am., Inc. v. Google Inc., Nos. 2013-1021, 2013-1022 (Fed. Circ.) 1:02:54-1:03:00. On remand, Oracle sought to convert this colloquy to a judicial admission that Google's use was "entirely commercial." It is for the district court, in its discretion, to determine the extent, if any, of a judicial admission. American Title Ins. Co. v. Lacelaw Corp., 861 F.2d 224, 226 (9th Cir. 1988). As set forth in the final pretrial order (Dkt. No. 1760), the undersigned examined the colloquy (and all other statements of record on the point) and determined that the "commercial" part would be treated as a judicial admission, but the "entirely" part would not be. The word "entirely" was part of the give and take of an oral argument. In light of all statements by counsel and in light of the free and open availability of Android, the word "entirely" would have been too conclusive, inaccurate, and unfair. The district court exercised its discretion to limit the admission to "commercial" and let the jury decide for itself how commercial, according to the evidence. Accordingly, our jury was instructed that Google's use was commercial, but that it was up to the jury to determine the extent of the commerciality, as follows (Dkt. No. 1981 ¶ 21) (emphasis added): In evaluating the first statutory factor, the extent of the commercial nature of the accused use must be considered. In this case, all agree that Google's accused use was commercial in nature but disagree over the extent. Commercial use weighs against a finding of fair use, but even a commercial use may be found (or not found, as the case may be) to be sufficiently transformative that the first statutory factor, on balance, still cuts in favor of fair use. To put it differently, the more transformative an accused work, the more other factors, such as commercialism, will recede in importance. By contrast, the less transformative the accused work, the more other factors like commercialism will dominate. Our jury could reasonably have agreed with Oracle that the evidence showed the use was entirely commercial (yet still ruled for Google), but it could also have reasonably found that the use, while commercial, served non-commercial purposes as well, i.e., as part of a free and open software platform, namely Android.7

2. On fair use, Oracle's most emphatic argument remains the "propriety of the defendant's conduct," meaning the subjective awareness by Google Inc. of the copyrights and, construing its internal e-mails in a light most unfavorable to Google, its "bad faith." This, however, underscores an important point for the appeal. Although the Federal Circuit opinion omitted any reference to the "propriety of the defendant's conduct" (good faith versus bad faith) as a consideration under any part of the four-factor test for fair use, Oracle insisted on remand that the jury be told that it could consider bad faith by our accused infringer as a subfactor under Factor One. This Court acquiesced in Oracle's view and did so despite an omission — conceivably a studied omission — of any such consideration in the Federal Circuit opinion and despite the fact that there is a respectable view that good or bad faith should no longer be a consideration after the Supreme Court's decision in Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 585 n.18, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (1994). See 2 Paul Goldstein, Goldstein on Copyright § 12.2.2, at 12:44.5-12:45 (3d ed. 2016). Put differently, either a use is objectively fair or it is not and subjective worry over the issue arguably should not penalize the user. Still, Oracle is correct that the Supreme Court in Harper & Row, Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 562, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985), called out the propriety of the defendant's conduct as a consideration in that case. Footnote 18 in Campbell later questioned whether or not propriety should persist as a consideration but did not rule it out. At the district court level, we must treat Harper & Row as still the law and leave it to the appellate courts to revise (although our instructions included a modification based on Campbell). This is no small point in this case, for no Oracle jury argument received more airtime than its argument that Google "knew" it needed a license and chose in bad faith to "make enemies" instead. This leads to a reciprocal key point. Given that Oracle was allowed to try to prove Google acted in bad faith, Google was allowed to try to prove good faith. Its witnesses testified that they had understood that "re-implementing" an API library was a legitimate, recognized practice so long as all that was duplicated was the "declaring code" and so long as the duplicator supplied its own "implementing code," that is, the methods were "re-implemented." In this way, Java programmers using the Android API could call on functionalities with the same Java command statements needed to call the same functionalities in the Java API, thereby avoiding splintering of the ways that identical functionalities became invoked by Java programmers. Google asked to go a step further and asked for an instruction on "custom," citing Wall Data Inc. v. Los Angeles County Sheriff's Dept., 447 F.3d 769, 778 (9th Cir. 2006), which stated "fair use is appropriate where a 'reasonable copyright owner' would have consented to the use, i.e., where the 'custom or public policy' at the time would have defined the use as reasonable" (quoting Subcomm. on Patents, Trademarks & Copyrights of the Sen. Comm. on the Judiciary, 86th Cong., 2d Sess., Study No. 14, Fair Use of Copyrighted Works 15 (Latman) (Comm. Print 1960)). Oracle objected on the ground that custom was omitted from the Federal Circuit opinion. It was omitted from that opinion — true. But neither was there any mention in that opinion of the propriety of the accused infringer's conduct. So, that omission by itself wasn't a good reason to ignore a pertinent statement by the Ninth Circuit, the law applicable in this copyright case arising in the Ninth Circuit. Oracle also argued that "custom" had to be vastly more entrenched than the "practice" evidence Google wished to present.

Factor 1: The Purpose and Character of the Use The first factor in the fair use inquiry involves "the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes." 17 U.S.C. § 107(1). This factor has two primary components: (1) whether the use is commercial in nature, rather than for educational or public interest purposes; and (2) "whether the new work is transformative or simply supplants the original." Wall Data, 447 F.3d at 778 (citing Campbell, 510 U.S. at 579). As explained below, the first is a question of fact and the second is a question of law. As Oracle points out, moreover, courts sometimes also consider whether the historical facts support the conclusion that the infringer acted in bad faith. See Harper & Row, 471 U.S. at 562. We address each component in turn.

a. Commercial Use Analysis of the first factor requires inquiry into the commercial nature of the use. Use of the copyrighted work that is commercial "tends to weigh against a finding of fair use." Harper & Row, 471 U.S. at 562. Courts have recognized, however, that, "[s]ince many, if not most, secondary users seek at least some measure of commercial gain from their use, unduly emphasizing the commercial motivation of a copier will lead to an overly restrictive view of fair use." Am. Geophysical Union v. Texaco, Inc., 60 F.3d 913, 921 (2d Cir. 1994); see also Infinity Broad. Corp. v. Kirkwood, 150 F.3d 104, 109 (2d Cir. 1998) ("[N]otwithstanding its mention in the text of the statute, commerciality has only limited usefulness to a fair use inquiry; most secondary uses of copyrighted material, including nearly all of the uses listed in the statutory preamble, are commercial."). Accordingly, although the statute requires us to consider the "commercial nature" of the work, "the degree to which the new user exploits the copyright for commercial gain—as opposed to incidental use as part of a commercial enterprise—affects the weight we afford commercial nature as a factor." Elvis Presley Enters., Inc. v. Passport Video, 349 F.3d 622, 627 (9th Cir. 2003). "[I]t is undisputed that Google's use of the declaring code and SSO from 37 Java API packages served commercial purposes." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *7. Although the jury was instructed that commercial use weighed against fair use, the district court explained that the jury "could reasonably have found that Google's decision to make Android available open source and free for all to use had non-commercial purposes as well (such as the general interest in sharing software innovation)." Id. On appeal, Oracle argues that Android is "hugely profitable" and that "Google reaps billions from exploiting Java in Android." Appellant Br. 29. As such, Oracle maintains that no reasonable jury could have found Android anything but "overwhelmingly commercial." Id. Google responds that: (1) because it gives Android away for free under an open source license the jury could have concluded that Android has non-commercial purposes; and (2) the jury could have reasonably found that Google's revenue flows from the advertisements on its search engine which preexisted Android. Neither argument has merit. First, the fact that Android is free of charge does not make Google's use of the Java API packages noncommercial. Giving customers "for free something they would ordinarily have to buy" can constitute commercial use. A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004, 1015 (9th Cir. 2001) (finding that "repeated and exploitative copying of copyrighted works, even if the copies are not offered for sale, may constitute a commercial use"). That Google might also have non-commercial motives is irrelevant as a matter of law. As the Supreme Court made clear when The Nation magazine published excerpts from Harper & Row's book, partly for the purpose of providing the public newsworthy information, the question "is not whether the sole motive of the use is monetary gain but whether the user stands to profit from exploitation of the copyrighted material without paying the customary price." Harper & Row, 471 U.S. at 562. Second, although Google maintains that its revenue flows from advertisements, not from Android, commerciality does not depend on how Google earns its money. Indeed, "[d]irect economic benefit is not required to demonstrate a commercial use." A&M Records, 239 F.3d at 1015. We find, therefore, that, to the extent we must assume the jury found Google's use of the API packages to be anything other than overwhelmingly commercial, that conclusion finds no substantial evidentiary support in the record. Accordingly, Google's commercial use of the API packages weighs against a finding of fair use.

b. Transformative Use Although the Copyright Act does not use the word "transformative," the Supreme Court has stated that the "central purpose" of the first fair use factor is to determine "whether and to what extent the new work is transformative." Campbell, 510 U.S. at 579. Transformative works "lie at the heart of the fair use doctrine's guarantee of breathing space within the confines of copyright, and the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Id. (internal citation omitted). A use is "transformative" if it "adds something new, with a further purpose or different character, altering the first with new expression, meaning or message." Id. The critical question is "whether the new work merely supersede[ s] the objects of the original creation . . . or instead adds something new." Id. (citations and internal quotation marks omitted). This inquiry "may be guided by the examples given in the preamble to § 107, looking to whether the use is for criticism, or comment, or news reporting, and the like." Id. at 578-79. "The Supreme Court has recognized that parodic works, like other works that comment and criticize, are by their nature often sufficiently transformative to fit clearly under the fair use exception." Mattel Inc. v. Walking Mountain Prods., 353 F.3d 792, 800 (9th Cir. 2003) (citing Campbell, 510 U.S. at 579). "Although transformation is a key factor in fair use, whether a work is transformative is a often highly contentious topic." Seltzer, 725 F.3d at 1176. Indeed, a "leading treatise on this topic has lamented the frequent misuse of the transformation test, complaining that it has become a conclusory label which is 'all things to all people.'" Id. (quoting Melville B. Nimmer & David Nimmer, 4 Nimmer on Copyright § 13.05[A][1][b], 13168-70 (2011)). To be transformative, a secondary work must either alter the original with new expression, meaning, or message or serve a new purpose distinct from that of the original work. Campbell, 510 U.S. at 579; Elvis Presley Enters., 349 F.3d at 629. Where the use "is for the same intrinsic purpose as [the copyright holder's] . . . such use seriously weakens a claimed fair use." Worldwide Church of God v. Phila. Church of God, Inc., 227 F.3d 1110, 1117 (9th Cir. 2000) (quoting Weissmann v. Freeman, 868 F.2d 1313, 1324 (2d Cir. 1989)). Although "transformative use is not absolutely necessary for a finding of fair use, the goal of copyright, to promote science and the arts, is generally furthered by the creation of transformative works." Campbell, 510 U.S. at 579 (citation and footnote omitted). As such, "the more transformative the new work, the less will be the significance of other factors, like commercialism, that may weigh against a finding of fair use." Id. Importantly, in the Ninth Circuit, whether a work is transformative is a question of law. See Mattel, 353 F.3d at 801 (explaining that parody—a well-established species of transformative use—"is a question of law, not a matter of public majority opinion"); see also Fox News Network, LLC v. TVEyes, Inc., No. 15-3885, 883 F.3d 169, 2018 U.S. App. LEXIS 4786, 2018 WL 1057178, at *3-4 (2d Cir. Feb. 27, 2018) (reassessing whether the use in question was transformative and deciding it was as a matter of law). In denying JMOL, the district court explained that "of course, the copied declarations serve the same function in both works, for by definition, declaring code in the Java programming language serves the [same] specific definitional purposes." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *8.The court concluded, however, that the jury could reasonably have found that Google's selection of some, but not all, of the Java API packages—"with new implementing code adapted to the constrained operating environment of mobile smartphone devices," together with new "methods, classes, and packages written by Google for the mobile smartphone platform"—constituted "a fresh context giving new expression, meaning, or message to the duplicated code." 2016 U.S. Dist. LEXIS 74931, [WL] at *9. On appeal, Oracle argues that Google's use was not transformative because it did not alter the APIs with "new expression, meaning, or message." Appellant Br. 29 (quoting Campbell, 510 U.S. at 579). Because Google concedes that it uses the API packages for the same purpose, Oracle maintains that it was unreasonable for either the jury or the court to find that Google sufficiently transformed the APIs to overcome its highly commercial use. Google responds that a reasonable jury could have concluded that Google used a small portion of the Java API packages to create a new work in a new context—"Android, a platform for smartphones, not desktops and servers." Cross-Appellant Br. 37. Google argues that, although the declarations and SSO may perform the same functions in Android and Java, the jury could reasonably find that they have different purposes because the "point of Android was to create a groundbreaking platform for smartphones." Id. at 39. Google's arguments are without merit. As explained below, Google's use of the API packages is not transformative as a matter of law because: (1) it does not fit within the uses listed in the preamble to § 107; (2) the purpose of the API packages in Android is the same as the purpose of the packages in the Java platform; (3) Google made no alteration to the expressive content or message of the copyrighted material; and (4) smartphones were not a new context. First, though not dispositive, we turn to the examples given in the preamble to § 107, "looking to whether the use is for criticism, or comment, or news reporting, and the like." Campbell, 510 U.S. at 578-79. Google's use of the Java API packages does not fit within the statutory categories, and Google does not suggest otherwise. Instead, Google cites Sony Computer Entertainment, Inc. v. Connectix Corp., 203 F.3d 596 (9th Cir. 2000), for the proposition that the "Ninth Circuit has held other types of uses—specifically including uses of computer code—to be fair." Cross-Appellant Br. 41. In Sony, the court found that the defendant's reverse engineering and intermediate copying of Sony's copyrighted software system "was a fair use for the purpose of gaining access to the unprotected elements of Sony's software." 203 F.3d at 602. The court explained that Sony's software program contained unprotected functional elements and that the defendant could only access those elements through reverse engineering. Id. at 603. The defendant used that information to create a software program that let consumers play games designed for Sony's PlayStation console on their computers. The court found that the defendant's use was only "modestly transformative" where: (1) the defendant created "a wholly new product" with "entirely new . . . code," and (2) the intermediate copying was performed to "produce a product that would be compatible." Id. at 606-07. As Oracle points out, even the "modest" level of transformation at issue in Sony is more transformative than what Google did here: copy code verbatim to attract programmers to Google's "new and incompatible platform." Appellant Response Br. 21. It is undisputed that the API packages "serve the same function in both works." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *8. And, as Oracle explains, the historical facts relevant to transformative use are also undisputed: what declaring code is, what it does in Java and in Android, how the audience of computer developers perceives it, how much Google took and added, what the added code does, and why Google used the declaring code and SSO. Indeed, Google conceded that "including the declarations (and their associated SSO) was for the benefit of developers, who—familiar with the Java programming language—had certain expectations regarding the language's APIs." Google's Opp. to Oracle's Rule 50(a) Motion for JMOL at 20, Oracle Am., Inc. v. Google Inc., No. 3:10-cv-3561 (N.D. Cal. May 21, 2016), ECF No. 1935. The fact that Google created exact copies of the declaring code and SSO and used those copies for the same purpose as the original material "seriously weakens [the] claimed fair use." See Wall Data, 447 F.3d at 778 (finding that, where the "Sheriff's Department created exact copies of RUMBA's software . . . [and] put those copies to the identical purpose as the original software," the use was not transformative); see also Campbell, 510 U.S. at 580 (noting that where the alleged infringer merely seeks "to avoid the drudgery in working up something fresh," any "claim to fairness . . . diminishes accordingly"). Google argues that Android is transformative because Google selectively used the declarations and SSO of only 37 of the 166 Java SE API packages and wrote its own implementing code. But taking only select passages of a copyrighted work is, by itself, not transformative. See L.A. News Serv. v. CBS Broad., Inc., 305 F.3d 924, 938-39 (9th Cir. 2002) ("Merely plucking the most visually arresting excerpt from LANS's nine minutes of footage cannot be said to have added anything new."). While, as discussed below, the volume of work copied is relevant to the fair use inquiry generally, thought must be given to the quality and importance of the copied material, not just to its relative quantity vis-à-vis the overall work. See Campbell, 510 U.S. at 586-87. To hold otherwise would mean that verbatim copying could qualify as fair use as long as the plagiarist stops short of taking the entire work. That approach is inconsistent with settled law and is particularly troubling where, as here, the portion copied is qualitatively significant. See Harper & Row, 471 U.S. at 569 (finding that verbatim copying of 300 words from a manuscript of more than 200,000 words was not a fair use); see also Folsom v. Marsh, 9 F. Cas. 342, 345, F. Cas. No. 4901 (C.C.D. Mass 1841) (Story, J.) ("There must be real, substantial condensation of the materials, and intellectual labor and judgment bestowed thereon; and not merely the facile use of the scissors; or extracts of the essential parts, constituting the chief value of the original work."). That Google wrote its own implementing code is irrelevant to the question of whether use of the APIs was transformative. As we noted in the prior appeal, "no plagiarist can excuse the wrong by showing how much of his work he did not pirate." Oracle, 750 F.3d at 1375 (quoting Harper & Row, 471 U.S. at 565). The relevant question is whether Google altered "the expressive content or message of the original work" that it copied—not whether it rewrote the portions it did not copy. See Seltzer, 725 F.3d at 1177 (explaining that a work is not transformative where the user "makes no alteration to the expressive content or message of the original work"). That said, even where the allegedly infringing work "makes few physical changes to the original or fails to comment on the original," it will "typically [be] viewed as transformative as long as new expressive content or message is apparent." Id. Here, however, there is no suggestion that the new implementing code somehow changed the expression or message of the declaring code. While Google's use could have been transformative if it had copied the APIs for some other purpose—such as teaching how to design an API—merely copying the material and moving it from one platform to another without alteration is not transformative. Google's primary argument on appeal is that Android is transformative because Google incorporated the declarations and SSO of the 37 API packages into a new context —smartphones. But the record showed that Java SE APIs were in smartphones before Android entered the market. Specifically, Oracle presented evidence that Java SE was in SavaJe mobile phones and that Oracle licensed Java SE to other smartphone manufacturers, including Danger and Nokia. Because the Java SE was already being used in smartphones, Google did not "transform" the copyrighted material into a new context and no reasonable jury could conclude otherwise. In any event, moving material to a new context is not transformative in and of itself—even if it is a "sharply different context." TCA Television Corp. v. McCollum, 839 F.3d 168, 181-83 (2d Cir. 2016) (finding that use "at some length, almost verbatim," of the copyrighted comedy routine "Who's on First?" in a dramatic play was not transformative where the play neither "imbued the Routine with any new expression, meaning, or message," nor added "any new dramatic purpose"). As previously explained, a use becomes transformative only if it serves a different purpose or alters the "expression, meaning, or message" of the original work. Kelly, 336 F.3d at 818. As such, "[c]ourts have been reluctant to find fair use when an original work is merely retransmitted in a different medium." A&M Records, 239 F.3d at 1015. Accordingly, although a change of format may be "useful," it "is not technically a transformation." Infinity Broad., 150 F.3d at 108 n.2 (finding that retransmitting copyrighted radio transmissions over telephone lines was not transformative because there was no new expression, meaning, or message). The Ninth Circuit has stated that "[a] use is considered transformative only where a defendant changes a plaintiff's copyrighted work or uses the plaintiff's copyrighted work in a different context such that the plaintiff's work is transformed into a new creation." Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1165 (9th Cir. 2007) (quoting Wall Data, 447 F.3d at 778). In Perfect 10, for example, the court found Google's use of thumbnail versions of copyrighted images "highly transformative" because, "[a]lthough an image may have been created originally to serve an entertainment, aesthetic, or informative function, a search engine transforms the image into a pointer directing a user to a source of information." Id. Although the court discussed the change in context (moving the copyrighted images into the electronic reference tool), it emphasized that Google used the images "in a new context to serve a different purpose." Id. In reaching this conclusion, the court reiterated that "even making an exact copy of a work may be transformative so long as the copy serves a different function than the original work." Id. (citing Kelly, 336 F.3d at 818-19). It is clear, therefore, that the change in context alone was not dispositive in Perfect 10; rather, the change in context facilitated the change in purpose, which made the use transformative. To some extent, any use of copyrighted work takes place in a slightly different context than the original. And of course, there is no bright line identifying when a use becomes transformative. But where, as here, the copying is verbatim, for an identical function and purpose, and there are no changes to the expressive content or message, a mere change in format (e.g., from desktop and laptop computers to smartphones and tablets) is insufficient as a matter of law to qualify as a transformative use.8

c. Bad faith In evaluating the "purpose and character" factor, the Ninth Circuit applies "the general rule that a party claiming fair use must act in a manner generally compatible with principles of good faith and fair dealing." Perfect 10, 508 F.3d at 1164 n.8 (citing Harper & Row, 471 U.S. at 562-63). In part, this is based on the fact that, in Harper & Row, the Supreme Court expressly stated that "[f]air use presupposes 'good faith' and 'fair dealing.'" 471 U.S. at 562 (citation omitted). It is also in part true because, as the Ninth Circuit has said, one who acts in bad faith should be barred from invoking the equitable defense of fair use. Fisher, 794 F.2d at 436 (calling the principle of considering the alleged infringer's "bad conduct" as a "bar [to] his use of the equitable defense of fair use" a sound one). Consistent with this authority, and at Oracle's request, the district court instructed the jury that it could consider whether Google acted in bad faith (or not) as part of its assessment of the first fair use factor. Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *6. And, because Oracle was permitted to introduce evidence that Google acted in bad faith, the court permitted Google to try to prove its good faith. Id. At trial, Oracle introduced evidence suggesting that "Google felt it needed to copy the Java API as an accelerant to bring Android to the market quicker" and knew that it needed a license to use Java. Id. For its part, Google presented evidence that it believed that the declaring code and SSO were "free to use and re-implement, both as a matter of developer practice and because the availability of independent implementations of the Java API enhanced the popularity of the Java programming language, which Sun promoted as free for all to use." 2016 U.S. Dist. LEXIS 74931, [WL] at *7. Given this conflicting evidence, the district court found that the jury could reasonably have concluded that "Google's use of parts of the Java API as an accelerant was undertaken based on a good faith belief that at least the declaring code and SSO were free to use (which it did use), while a license was necessary for the implementing code (which it did not use)." Id. On appeal, Oracle argues that there was ample evidence that Google intentionally copied Oracle's copyrighted work and knew that it needed a license to use Java. Google responds that the jury heard sufficient evidence of Google's good faith based on industry custom and was entitled to credit that evidence. But, while bad faith may weigh against fair use, a copyist's good faith cannot weigh in favor of fair use. Indeed, the Ninth Circuit has expressly recognized that "the innocent intent of the defendant constitutes no defense to liability." Monge, 688 F.3d at 1170 (quoting 4 Melville B. Nimmer & David Nimmer, Nimmer on Copyright § 13.08[B][1] (Matthew Bender rev. ed. 2011)). If it were clear, accordingly, that the jury found fair use solely or even largely because it approved of Google's motives even if they were in bad faith, we would find such a conclusion improper. Because evidence of Google's good faith was relevant to rebut evidence of its bad faith, however, and there is no objection to the instructions to the jury on this or any other point, we must assume that the jury simply did not find the evidence of Google's bad faith persuasive.We note, moreover, that merely "being denied permission to use a work does not weigh against a finding of fair use." Campbell, 510 U.S. at 585 n.18 ("If the use is otherwise fair, then no permission need be sought or granted."). Ultimately, we find that, even assuming the jury was unpersuaded that Google acted in bad faith, the highly commercial and non-transformative nature of the use strongly support the conclusion that the first factor weighs against a finding of fair use.

Factor 2: Nature of the Copyrighted Work The second factor—the nature of the copyrighted work—"calls for recognition that some works are closer to the core of intended copyright protection than others, with the consequence that fair use is more difficult to establish when the former works are copied." Campbell, 510 U.S. at 586. This factor "turns on whether the work is informational or creative." Worldwide Church of God, 227 F.3d at 1118; see also Harper & Row, 471 U.S. at 563 ("The law generally recognizes a greater need to disseminate factual works than works of fiction or fantasy."). Creative expression "falls within the core of the copyright's protective purposes." Campbell, 510 U.S. at 586. Although "software products are not purely creative works," it is well established that copyright law protects computer software. Wall Data, 447 F.3d at 780 (citing Sega Enters. Ltd. v. Accolade, Inc., 977 F.2d 1510, 1519 (9th Cir. 1992) ("[T]he 1980 amendments to the Copyright Act unambiguously extended copyright protection to computer programs.")). Here, the district court found that the jury could have concluded that the process of designing APIs was "highly creative" and "thus at the core of copyright's protection" or it could "reasonably have gone the other way and concluded that the declaring code was not highly creative." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *10. While the jury heard testimony from Google's own expert that API design is "an art, not a science," other witnesses emphasized the functional role of the declaring code and the SSO and minimized the creative aspects. Id. Accordingly, the district court concluded that the "jury could reasonably have found that, while the declaring code and SSO were creative enough to qualify for copyright protection, functional considerations predominated in their design." Id. On appeal, Oracle emphasizes that designing the APIs was a highly creative process and that the organization of the packages was not mandated by function. Indeed, this court has already held that the declaring code and the SSO of the 37 API packages at issue were sufficiently creative and original to qualify for copyright protection. Oracle, 750 F.3d at 1356. According to Oracle, the district court erred in assuming that, because the APIs have a "functional role," they cannot be creative. As Google points out, however, all we found in the first appeal was that the declarations and SSO were sufficiently creative to provide the "minimal degree of creativity," Feist Publ'ns, Inc. v. Rural Tel. Serv. Co., 499 U.S. 340, 345, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (1991), that is required for copyrightability. We also recognized that a reasonable jury could find that "the functional aspects of the packages" are "relevant to Google's fair use defense." Oracle, 750 F.3d at 1369, 1376-77. On remand, Oracle stipulated that some of the declarations were necessary to use the Java language and presented no evidence explaining how the jury could distinguish the functionality and creativity of those declarations from the others. Google maintains that it presented evidence that the declarations and SSO were functional and the jury was entitled to credit that evidence. Although it is clear that the 37 API packages at issue involved some level of creativity—and no reasonable juror could disagree with that conclusion—reasonable jurors could have concluded that functional considerations were both substantial and important. Based on that assumed factual finding, we conclude that factor two favors a finding of fair use. The Ninth Circuit has recognized, however, that this second factor "typically has not been terribly significant in the overall fair use balancing." Dr. Seuss Enters., L.P. v. Penguin Books USA, Inc., 109 F.3d 1394, 1402 (9th Cir. 1997) (finding that the "creativity, imagination and originality embodied in The Cat in the Hat and its central character tilts the scale against fair use"); Mattel, 353 F.3d at 803 (similar). Other circuits agree. Fox News Network, 2018 U.S. App. LEXIS 4786, 2018 WL 1057178, at *5 ("This factor 'has rarely played a significant role in the determination of a fair use dispute,' and it plays no significant role here." (quoting Authors Guild v. Google, Inc., 804 F.3d 202, 220 (2d Cir. 2015))). We note, moreover, that allowing this one factor to dictate a conclusion of fair use in all cases involving copying of software could effectively negate Congress's express declaration—continuing unchanged for some forty years—that software is copyrightable. Accordingly, though the jury's assumed view of the nature of the copyrighted work weighs in favor of finding fair use, it has less significance to the overall analysis.

Factor 3: Amount and Substantiality of the Portion Used The third factor focuses on the "amount and substantiality of the portion used in . . . the context of the copyrighted work, not the infringing work." Oracle, 750 F.3d at 1375. Indeed, the statutory language makes clear that "a taking may not be excused merely because it is insubstantial with respect to the infringing work." Harper & Row, 471 U.S. at 565. "[T]he fact that a substantial portion of the infringing work was copied verbatim [from the original work] is evidence of the qualitative value of the copied material, both to the originator and to the plagiarist who seeks to profit from marketing someone else's copyrighted expression." Id. Thus, while "wholesale copying does not preclude fair use per se, copying an entire work militates against a finding of fair use." Worldwide Church of God, 227 F.3d at 1118 (citation and quotation marks omitted). But, there is no relevance to the opposite—i.e., adding substantial content to the copyrighted work is not evidence that what was copied was insubstantial or unimportant. The inquiry under this third factor "is a flexible one, rather than a simple determination of the percentage of the copyrighted work used." Monge, 688 F.3d at 1179. The Ninth Circuit has explained that this third factor looks to the quantitative amount and qualitative value of the original work used in relation to the justification for its use. Seltzer, 725 F.3d at 1178. The percentage of work copied is not dispositive where the portion copied was qualitatively significant. Harper & Row, 471 U.S. at 566 ("In view of the expressive value of the excerpts and their key role in the infringing work, we cannot agree with the Second Circuit that the 'magazine took a meager, indeed an infinitesimal amount of Ford's original language.'" (citation omitted)). Google is correct that the Ninth Circuit has said that, "this factor will not weigh against an alleged infringer, even when he copies the whole work, if he takes no more than is necessary for his intended use." Id. (citing Kelly v. Arriba Soft Corp., 336 F.3d 811, 820-21 (9th Cir. 2003)). But the Ninth Circuit has only said that is true where the intended use was a transformative one, because the "extent of permissible copying varies with the purpose and character of the use." Id. (quoting Campbell, 510 U.S. at 586-87). Here, we have found that Google's use was not transformative and Google has conceded both that it could have written its own APIs and that the purpose of its copying was to make Android attractive to programmers. "Necessary" in the context of the cases upon which Google relies does not simply mean easier. In assessing factor three, the district court explained that the "jury could reasonably have found that Google duplicated the bare minimum of the 37 API packages, just enough to preserve inter-system consistency in usage, namely the declarations and their SSO only, and did not copy any of the implementing code," such that Google "copied only so much as was reasonably necessary." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *10. In reaching this conclusion, the court noted that the jury could have found that the number of lines of code Google duplicated was a "tiny fraction of one percent of the copyrighted works (and even less of Android, for that matter)." Id. We disagree that such a conclusion would have been reasonable or sufficient on this record. On remand, the parties stipulated that only 170 lines of code were necessary to write in the Java language. It is undisputed, however, that Google copied 11,500 lines of code—11,330 more lines than necessary to write in Java. That Google copied more than necessary weighs against fair use. See Monge, 688 F.3d at 1179 (finding that, where the copyist "used far more than was necessary" of the original work, "this factor weighs against fair use"). And, although Google emphasizes that it used a small percentage of Java (11,500 lines of declarations out of roughly 2.86 million lines of code in the Java SE libraries), it copied the SSO for the 37 API packages in its entirety. The district court emphasized Google's desire to "preserve inter-system consistency" to "avoid confusion among Java programmers as between the Java system and the Android system." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *10-11. As we noted in the prior appeal, however, Google did not seek to foster any "inter-system consistency" between its platform and Oracle's Java platform. Oracle, 750 F.3d at 1371. And Google does not rely on any interoperability arguments in this appeal.Google sought "to capitalize on the fact that software developers were already trained and experienced in using the Java API packages at issue." Id. But there is no inherent right to copy in order to capitalize on the popularity of the copyrighted work or to meet the expectations of intended customers. Taking those aspects of the copyrighted material that were familiar to software developers to create a similar work designed to be popular with those same developers is not fair use. See Dr. Seuss Enters., 109 F.3d at 1401 (copying the most famous and well recognized aspects of a work "to get attention" or "to avoid the drudgery in working up something fresh" is not a fair use (quoting Campbell, 510 U.S. at 580)). Even assuming the jury accepted Google's argument that it copied only a small portion of Java, no reasonable jury could conclude that what was copied was qualitatively insignificant, particularly when the material copied was important to the creation of the Android platform. Google conceded as much when it explained to the jury the importance of the APIs to the developers it wished to attract. See Tr. of Proceedings held on 5/16/16 at 106:8-14, Oracle Am., Inc. Google Inc., No. 3:10-cv-3561 (N.D. Cal. May 20, 2016), ECF No. 1930; Id. at 134:6-11. Indeed, Google's own expert conceded that "it was a sound business practice for Google to leverage the existing community of developers, minimizing the amount of new material and maximizing existing knowledge," even though Google also conceded that it could have written the APIs differently to achieve the same functions. Id. at 144:5-10. For these reasons, we find that the third factor is, at best, neutral in the fair use inquiry, and arguably weighs against such a finding.

Factor 4: Effect Upon the Potential Market The fourth and final factor focuses on "the effect of the use upon the potential market for or value of the copyrighted work." 17 U.S.C. § 107(4). This factor reflects the idea that fair use "is limited to copying by others which does not materially impair the marketability of the work which is copied." Harper & Row, 471 U.S. at 566-67. It requires that courts "consider not only the extent of market harm caused by the particular actions of the alleged infringer, but also whether unrestricted and widespread conduct of the sort engaged in by the defendant . . . would result in a substantially adverse impact on the potential market for the original." Campbell, 510 U.S. at 590 (citation and quotation marks omitted). The Supreme Court once said that factor four is "undoubtedly the single most important element of fair use." Harper & Row, 471 U.S. at 566. In its subsequent opinion in Campbell, however, the Court emphasized that none of the four factors can be viewed in isolation and that "[a]ll are to be explored, and the results weighed together, in light of the purposes of copyright." 510 U.S. at 578; see also Infinity Broad., 150 F.3d at 110 ("Historically, the fourth factor has been seen as central to fair use analysis, although the Supreme Court appears to have backed away from this position." (internal citation omitted)). The Court has also explained that "[m]arket harm is a matter of degree, and the importance of this factor will vary, not only with the amount of harm, but also with the relative strength of the showing on the other factors." Campbell, 510 U.S. at 590 n.21. The Ninth Circuit recently indicated that likely market harm can be presumed where a use is "commercial and not transformative." Disney Enters., Inc. v. VidAngel, Inc., 869 F.3d 848, 861 (9th Cir. 2017) (citing Leadsinger, 512 F.3d at 531, for the proposition that, where a use "was commercial and not transformative, it was not error to presume likely market harm"). That presumption allegedly traces back to Sony Corp. of America v. University City Studios, Inc., 464 U.S. 417, 451, 104 S. Ct. 774, 78 L. Ed. 2d 574 (1984), where the Supreme Court stated that, "[i]f the intended use is for commercial gain, that likelihood [of future harm] may be presumed. But if it is for a noncommercial purpose, the likelihood must be demonstrated." The Supreme Court has since clarified that market impact, "no less than the other three [factors], may be addressed only through a 'sensitive balancing of interests'" and that earlier interpretations of Sony to the contrary were incorrect. Campbell, 510 U.S. at 590 n.21 (quoting Sony, 464 U.S. at 455 n.40);see also Monge, 688 F.3d at 1181 (cautioning against overemphasis on a presumption of market harm after Campbell). On this point, we must apply clear Supreme Court precedent rather than the more recent Ninth Circuit's statements to the contrary. In evaluating the fourth factor, courts consider not only harm to the actual or potential market for the copyrighted work, but also harm to the "market for potential derivative uses," including "those that creators of original works would in general develop or license others to develop." Campbell, 510 U.S. at 592; see also A&M Records, 239 F.3d at 1017 ("[L]ack of harm to an established market cannot deprive the copyright holder of the right to develop alternative markets for the works."). A court can therefore consider the challenged use's "impact on potential licensing revenues for traditional, reasonable, or likely to be developed markets." Swatch Grp. Mgmt. Servs. Ltd. v. Bloomberg L.P., 756 F.3d 73, 91 (2d Cir. 2014) (citation omitted); see also Seltzer, 725 F.3d at 1179 ("This factor also considers any impact on 'traditional, reasonable, or likely to be developed markets.'" (citation omitted)). Also relevant to the inquiry is the fact that a copyright holder has the exclusive right to determine "when, 'whether and in what form to release'" the copyrighted work into new markets, whether on its own or via a licensing agreement. Monge, 688 F.3d at 1182 (quoting Harper & Row, 471 U.S. at 553). Indeed, the Ninth Circuit has recognized that "[e]ven an author who had disavowed any intention to publish his work during his lifetime" was entitled to copyright protection because: (1) "the relevant consideration was the 'potential market'" and (2) "he has the right to change his mind." Worldwide Church, 227 F.3d at 1119 (citing Salinger v. Random House, Inc., 811 F.2d 90, 99 (2d Cir. 1987)); see also Micro Star v. Formgen Inc., 154 F.3d 1107, 1113 (9th Cir. 1998) (noting that only the copyright holder "has the right to enter that market; whether it chooses to do so is entirely its business"). Here, the district court concluded that the jury "could reasonably have found that use of the declaring lines of code (including their SSO) in Android caused no harm to the market for the copyrighted works, which were for desktop and laptop computers." Order Denying JMOL, 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, at *10. In reaching this conclusion, the district court noted that, before Android was released, Sun made all of the Java API packages available for free and open source under the name OpenJDK, subject only to the terms of a general public license. Id. According to the district court, the jury could have concluded that "Android's impact on the market for the copyrighted works paralleled what Sun already expected via its Open-JDK." Id. On appeal, Oracle argues that the evidence of actual and potential harm stemming from Google's copying was "overwhelming," and that the district court erred as a matter of law in concluding otherwise. Appellant Br. 52. We agree. First, with respect to actual market harm, the evidence showed that Java SE had been used for years in mobile devices, including early smartphones, prior to Android's release. Specifically, the jury heard testimony that Java SE was already in smartphones, including Blackberry, SavaJe, Danger, and Nokia. That Android competed directly with Java SE in the market for mobile devices is sufficient to undercut Google's market harm arguments. With respect to tablets, the evidence showed that Oracle licensed Java SE for the Amazon Kindle. After Android's release, however, Amazon was faced with two competing options—Java SE and Android—and selected Android.The jury also heard evidence that Amazon later used the fact that Android was free to negotiate a steep discount to use Java SE in its newer e-reader. In other words, the record contained substantial evidence that Android was used as a substitute for Java SE and had a direct market impact. Given this evidence of actual market harm, no reasonable jury could have concluded that there was no market harm to Oracle from Google's copying. Even if there were a dispute about whether Oracle was licensing Java SE in smartphones at the time Android launched, moreover, "fair use focuses on potential, not just actual, market harm." Monge, 688 F.3d at 1181. Accordingly, although the district court focused exclusively on the market it found that Oracle had already entered—desktops and laptops—it should have considered how Google's copying affected potential markets Oracle might enter or derivative works it might create or license others to create. See Campbell, 510 U.S. at 590. Licensing Java SE for smartphones with increased processing capabilities was one such potential new market. And the fact that Oracle and Google engaged in lengthy licensing negotiations demonstrates that Oracle was attempting to license its work for mobile devices, including smartphones.Smartphones were, therefore, a "traditional, reasonable, or likely to be developed market." See Swatch Grp., 756 F.3d at 91; see also Seltzer, 725 F.3d at 1179. Google argues that a reasonable jury could have concluded that Java SE and Android did not compete in the same market because Oracle: (1) was not a device maker; and (2) had not yet built its own smartphone platform. Neither argument has merit. That Oracle never built a smartphone device is irrelevant because potential markets include licensing others to develop derivative works. See Campbell, 510 U.S. at 592. The fact that Oracle had not yet developed a smartphone platform is likewise irrelevant as a matter of law because, as Oracle submits, a market is a potential market even where the copyright owner has no immediate plans to enter it or is unsuccessful in doing so. See Worldwide Church, 227 F.3d at 1119; Micro Star, 154 F.3d at 1113. Even assuming a reasonable jury could have found no current market harm, the undisputed evidence showed, at a minimum, that Oracle intended to license Java SE in smartphones; there was no evidence in the record to support any contrary conclusion. Because the law recognizes and protects a copyright owner's right to enter a "potential market," this fact alone is sufficient to establish market impact. Given the record evidence of actual and potential harm, we conclude that "unrestricted and widespread conduct of the sort engaged in by" Google would result in "a substantially adverse impact on the potential market for the original" and its derivatives. See Campbell, 510 U.S. at 590 (citation and quotation marks omitted). Accordingly, the fourth factor weighs heavily in favor of Oracle.

As to the likely amount of loss, the jury could have found that Android did not harm the actual or potential markets for Java SE. And it could have found that Sun itself (now Oracle) would not have been able to enter those markets successfully whether Google did, or did not, copy a part of its API. First, evidence at trial demonstrated that, regardless of Android’s smartphone technology, Sun was poorly positioned to succeed in the mobile phone market. The jury heard ample evidence that Java SE’s primary market was laptops and desktops. App. 99, 200. It also heard that Sun’s many efforts to move into the mobile phone market had proved unsuccessful. Id., at 135, 235, 671. As far back as 2006, prior to Android’s release, Sun’s executives projected declining revenue for mobile phones because of emerging smartphone technology. Id., at 240. When Sun’s former CEO was asked directly whether Sun’s failure to build a smartphone was attributable to Google’s development of Android, he answered that it was not. Id., at 650. Given the evidence showing that Sun was beset by business challenges in developing a mobile phone product, the jury was entitled to agree with that assessment. Second, the jury was repeatedly told that devices using Google’s Android platform were different in kind from those that licensed Sun’s technology. For instance, witnesses explained that the broader industry distinguished between smartphones and simpler “feature phones.” Id., at 237. As to the specific devices that used Sun-created software, the jury heard that one of these phones lacked a touchscreen, id., at 359-360, while another did not have a QWERTY keyboard, id., at 672. For other mobile devices, the evidence showed that simpler products, like the Kindle, used Java software, id., at 396, while more advanced technology, like the Kindle Fire, were built on the Android operating system, id., at 206. This record evidence demonstrates that, rather than just “repurposing [Sun’s] code from larger computers to smaller computers,” post, at 16, Google’s Android platform was part of a distinct (and more advanced) market than Java software.

It is important, however, to consider why and how Oracle might have become entitled to this money. When a new interface, like an API or a spreadsheet program, first comes on the market, it may attract new users because of its expressive qualities, such as a better visual screen or because of its superior functionality. As time passes, however, it may be valuable for a different reason, namely, because users, including programmers, are just used to it. They have already learned how to work with it. See Lotus Development Corp., 49 F. 3d, at 821 (Boudin, J., concurring). The record here is filled with evidence that this factor accounts for Google’s desire to use the Sun Java API. See, e.g., App. 169-170, 213-214. This source of Android’s profitability has much to do with third parties’ (say, programmers’) investment in Sun Java programs. It has correspondingly less to do with Sun’s investment in creating the Sun Java API. We have no reason to believe that the Copyright Act seeks to protect third parties’ investment in learning how to operate a created work. Cf. Campbell, 510 U. S., at 591-592, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (discussing the need to identify those harms that are “cognizable under the Copyright Act”). Finally, given programmers’ investment in learning the Sun Java API, to allow enforcement of Oracle’s copyright here would risk harm to the public. Given the costs and difficulties of producing alternative APIs with similar appeal to programmers, allowing enforcement here would make of the Sun Java API’s declaring code a lock limiting the future creativity of new programs. Oracle alone would hold the key. The result could well prove highly profitable to Oracle (or other firms holding a copyright in computer interfaces). But those profits could well flow from creative improvements, new applications, and new uses developed by users who have learned to work with that interface. To that extent, the lock would interfere with, not further, copyright’s basic creativity objectives. See Connectix Corp., 203 F. 3d, at 607; see also Sega Enterprises, 977 F. 2d, at 1523-1524 (“An attempt to monopolize the market by making it impossible for others to compete runs counter to the statutory purpose of promoting creative expression”); Lexmark Int’l, 387 F. 3d, at 544 (noting that where a subsequent user copied a computer program to foster functionality, it was not exploiting the programs “commercial value as a copyrighted work” (emphasis in original)). After all, “copyright supplies the economic incentive to [both] create and disseminate ideas,” Harper & Row, 471 U. S., at 558, 105 S. Ct. 2218, 85 L. Ed. 2d 588, and the reimplementation of a user interface allows creative new computer code to more easily enter the market.

IV The language of §107, the “fair use” provision, reflects its judge-made origins. It is similar to that used by Justice Story in Folsom v. Marsh, 9 F. Cas. 342, 348, F. Cas. No. 4901 (No. 4,901) (CC Mass. 1841). See Campbell, 510 U. S., at 576, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (noting how “Justice Story’s summary [of fair use considerations] is discernable” in §107). That background, as well as modern courts’ use of the doctrine, makes clear that the concept is flexible, that courts must apply it in light of the sometimes conflicting aims of copyright law, and that its application may well vary depending upon context. Thus, copyright’s protection may be stronger where the copyrighted material is fiction, not fact, where it consists of a motion picture rather than a news broadcast, or where it serves an artistic rather than a utilitarian function. See, e.g., Stewart, 495 U. S., at 237-238, 110 S. Ct. 1750, 109 L. Ed. 2d 184; Harper & Row, 471 U. S., at 563, 105 S. Ct. 2218, 85 L. Ed. 2d 588; see also 4 M. Nimmer & D. Nimmer, Copyright §13.05[A] [2][a] (2019) (hereinafter Nimmer on Copyright) (“[C]opyright protection is narrower, and the corresponding application of the fair use defense greater, in the case of factual works than in the case of works of fiction or fantasy”). Similarly, courts have held that in some circumstances, say, where copyrightable material is bound up with uncopyrightable material, copyright protection is “thin.” See Feist, 499 U. S., at 349, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (noting that “the copyright in a factual compilation is thin”); see also Experian Information Solutions, Inc. v. Nationwide Marketing Servs. Inc., 893 F. 3d 1176, 1186 (CA9 2018) (“In the context of factual compilations, . . . there can be no infringement unless the works are virtually identical” (internal quotation marks omitted)). Generically speaking, computer programs differ from books, films, and many other “literary works” in that such programs almost always serve functional purposes. These and other differences have led at least some judges to complain that “applying copyright law to computer programs is like assembling a jigsaw puzzle whose pieces do not quite fit.” Lotus Development Corp. v. Borland Int’l, Inc., 49 F. 3d 807, 820 (CA1 1995) (Boudin, J., concurring). These differences also led Congress to think long and hard about whether to grant computer programs copyright protection. In 1974, Congress established a National Commission on New Technological Uses of Copyrighted Works (CONTU) to look into the matter. §§201-208, 88 Stat. 1873-1875. After several years of research, CONTU concluded that the “availability of copyright protection for computer programs is desirable.” Final Report 11 (July 31, 1978). At the same time, it recognized that computer programs had unique features. Mindful of not “unduly burdening users of programs and the general public,” it wrote that copyright “should not grant anyone more economic power than is necessary to achieve the incentive to create.” Id., at 12. And it believed that copyright’s existing doctrines (e.g., fair use), applied by courts on a case-by-case basis, could prevent holders from using copyright to stifle innovation. Ibid. (“Relatively few changes in the Copyright Act of 1976 are required to attain these objectives”). Congress then wrote computer program protection into the law. See §10, 94 Stat. 3028. The upshot, in our view, is that fair use can play an important role in determining the lawful scope of a computer program copyright, such as the copyright at issue here. It can help to distinguish among technologies. It can distinguish between expressive and functional features of computer code where those features are mixed. It can focus on the legitimate need to provide incentives to produce copyrighted material while examining the extent to which yet further protection creates unrelated or illegitimate harms in other markets or to the development of other products. In a word, it can carry out its basic purpose of providing a context-based check that can help to keep a copyright monopoly within its lawful bounds. See H. R. Rep. No. 94-1476, pp. 65-66 (1976) (explaining that courts are to “adapt the doctrine [of fair use] to particular situations on a case-by-case basis” and in light of “rapid technological change”); see, e.g., Lexmark Int’l, Inc. v. Static Control Components, Inc., 387 F. 3d 522, 543-545 (CA6 2004) (discussing fair use in the context of copying to preserve compatibility); Sony Computer Entertainment, Inc. v. Connectix Corp., 203 F. 3d 596, 603-608 (CA9 2000) (applying fair use to intermediate copying necessary to reverse engineer access to unprotected functional elements within a program); Sega Enterprises Ltd. v. Accolade, Inc., 977 F. 2d 1510, 1521-1527 (CA9 1992) (holding that wholesale copying of copyrighted code as a preliminary step to develop a competing product was a fair use). Justice Thomas’ thoughtful dissent offers a very different view of how (and perhaps whether) fair use has any role to play for computer programs. We are told that no attempt to distinguish among computer code is tenable when considering “the nature of the work,” see post, at 10, even though there are important distinctions in the ways that programs are used and designed, post, at 18 (“The declaring code is what attracted programmers”). We are told that no reuse of code in a new program will ever have a valid “purpose and character,” post, at 16, even though the reasons for copying computer code may vary greatly and differ from those applicable to other sorts of works, ibid. (accepting that copying as part of “reverse engineer[ing] a system to ensure compatibility” could be a valid purpose). And we are told that our fair use analysis must prioritize certain factors over others, post, at 9, n. 5, even though our case law instructs that fair use depends on the context, see Campbell, 510 U. S., at 577-578, 114 S. Ct. 1164, 127 L. Ed. 2d 500. We do not understand Congress, however, to have shielded computer programs from the ordinary application of copyright’s limiting doctrines in this way. By defining computer programs in §101, Congress chose to place this subject matter within the copyright regime. Like other protected works, that means that the owners of computer programs enjoy the exclusive rights set forth in the Act, including the right to “reproduce [a] copyrighted work” or to “prepare derivative works.” 17 U. S. C. §106. But that also means that exclusive rights in computer programs are limited like any other works. Just as fair use distinguishes among books and films, which are indisputably subjects of copyright, so too must it draw lines among computer programs. And just as fair use takes account of the market in which scripts and paintings are bought and sold, so too must it consider the realities of how technological works are created and disseminated. We do not believe that an approach close to “all or nothing” would be faithful to the Copyright Act’s overall design.

V At the outset, Google argues that “fair use” is a question for a jury to decide; here the jury decided the question in Google’s favor; and we should limit our review to determining whether “substantial evidence” justified the jury’s decision. The Federal Circuit disagreed. It thought that the “fair use” question was a mixed question of fact and law; that reviewing courts should appropriately defer to the jury’s findings of underlying facts; but that the ultimate question whether those facts showed a “fair use” is a legal question for judges to decide de novo. We agree with the Federal Circuit’s answer to this question. We have said, “[f]air use is a mixed question of law and fact.” Harper & Row, 471 U. S., at 560, 105 S. Ct. 2218, 85 L. Ed. 2d 588. We have explained that a reviewing court should try to break such a question into its separate factual and legal parts, reviewing each according to the appropriate legal standard. But when a question can be reduced no further, we have added that “the standard of review for a mixed question all depends—on whether answering it entails primarily legal or factual work.” U. S. Bank N. A. v. Village at Lakeridge, LLC, 583 U. S. ___, ___, 138 S. Ct. 960, 967, 200 L. Ed. 2d 218 (2018). In this case, the ultimate “fair use” question primarily involves legal work. “Fair use” was originally a concept fashioned by judges. Folsom, 9 F. Cas., at 348. Our cases still provide legal interpretations of the fair use provision. And those interpretations provide general guidance for future cases. See, e.g., Campbell, 510 U. S., at 592-593, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (describing kinds of market harms that are not the concern of copyright); Harper & Row, 471 U. S., at 564, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (“scope of fair use is narrower with respect to unpublished works”); Sony, 464 U. S., at 451, 104 S. Ct. 774, 78 L. Ed. 2d 574 (wholesale copying aimed at creating a market substitute is presumptively unfair). This type of work is legal work. U. S. Bank, 583 U. S., at ___, 138 S. Ct. 960, 967, 200 L. Ed. 2d 218 (“When applying the law involves developing auxiliary legal principles for use in other cases[,] appellate courts should typically review a decision de novo”). Applying a legal “fair use” conclusion may, of course, involve determination of subsidiary factual questions, such as “whether there was harm to the actual or potential markets for the copyrighted work” or “how much of the copyrighted work was copied.” 886 F. 3d, at 1196; see, e.g., Peter F. Gaito Architecture, LLC v. Simone Development Corp., 602 F. 3d 57, 63 (CA2 2010) (noting that in an infringement suit “the question of substantial similarity typically presents an extremely close question of fact”). In this case the Federal Circuit carefully applied the fact/law principles we set forth in U. S. Bank, leaving factual determinations to the jury and reviewing the ultimate question, a legal question, de novo. Next, Google argues that the Federal Circuit’s approach violates the Seventh Amendment. The Amendment both requires that “the right of trial by jury . . . be preserved” and forbids courts to “re-examin[e ]” any “fact tried by a jury.” U. S. Const., Amdt. 7; see also Gasperini v. Center for Humanities, Inc., 518 U. S. 415, 432-433, 116 S. Ct. 2211, 135 L. Ed. 2d 659 (1996). The Reexamination Clause is no bar here, however, for, as we have said, the ultimate question here is one of law, not fact. It does not violate the Reexamination Clause for a court to determine the controlling law in resolving a challenge to a jury verdict, as happens any time a court resolves a motion for judgment as a matter of law. See, e.g., Neely v. Martin K. Eby Constr. Co., 386 U. S. 317, 322, 87 S. Ct. 1072, 18 L. Ed. 2d 75 (1967). Nor is Google correct that “the right of trial by jury” includes the right to have a jury resolve a fair use defense. That Clause is concerned with “the particular trial decision” at issue. Markman v. Westview Instruments, Inc., 517 U. S. 370, 376, 116 S. Ct. 1384, 134 L. Ed. 2d 577 (1996). Even though it is possible to find pre-Revolutionary English cases in which a judge sent related questions like fair abridgment to a jury, those questions were significantly different from the “fair use” doctrine as courts apply it today. See, e.g., Gyles v. Wilcox, 2 Atk. 141, 142-144, 26 Eng. Rep. 489, 490-491 (Ch. 1740) (asking the Court to resolve the narrow question whether a shortened work could be considered a new work); Sayre v. Moore, 1 East 361, n., 102 Eng. Rep. 138, 139, n. (K. B. 1785) (discussing the jury’s role in resolving whether copying constituted infringement). As far as contemporary fair use is concerned, we have described the doctrine as an “equitable,” not a “legal,” doctrine. We have found no case suggesting that application of U. S. Bank here would fail “to preserve the substance of the common-law [jury trial] right as it existed in 1791.” Markman, 517 U. S., at 376, 116 S. Ct. 1384, 134 L. Ed. 2d 577.

B. “The Purpose and Character of the Use” In the context of fair use, we have considered whether the copier’s use “adds something new, with a further purpose or different character, altering” the copyrighted work “with new expression, meaning or message.” Id., at 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Commentators have put the matter more broadly, asking whether the copier’s use “fulfill[s] the objective of copyright law to stimulate creativity for public illumination.” Leval 1111. In answering this question, we have used the word “transformative” to describe a copying use that adds something new and important. Campbell, 510 U. S., at 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500. An “‘artistic painting’” might, for example, fall within the scope of fair use even though it precisely replicates a copyrighted “‘advertising logo to make a comment about consumerism.’” 4 Nimmer on Copyright §13.05[A][1][b] (quoting Netanel, Making Sense of Fair Use, 15 Lewis & Clark L. Rev. 715, 746 (2011)). Or, as we held in Campbell, a parody can be transformative because it comments on the original or criticizes it, for “[p]arody needs to mimic an original to make its point.” 510 U. S., at 580-581, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Google copied portions of the Sun Java API precisely, and it did so in part for the same reason that Sun created those portions, namely, to enable programmers to call up implementing programs that would accomplish particular tasks. But since virtually any unauthorized use of a copyrighted computer program (say, for teaching or research) would do the same, to stop here would severely limit the scope of fair use in the functional context of computer programs. Rather, in determining whether a use is “transformative,” we must go further and examine the copying’s more specifically described “purpose[s]” and “character.” 17 U. S. C. §107(1). Here Google’s use of the Sun Java API seeks to create new products. It seeks to expand the use and usefulness of Android-based smartphones. Its new product offers programmers a highly creative and innovative tool for a smartphone environment. To the extent that Google used parts of the Sun Java API to create a new platform that could be readily used by programmers, its use was consistent with that creative “progress” that is the basic constitutional objective of copyright itself. Cf. Feist, 499 U. S., at 349-350, 111 S. Ct. 1282, 113 L. Ed. 2d 358 (“The primary objective of copyright is not to reward the labor of authors, but ‘[t]o promote the Progress of Science and useful Arts’” (quoting U. S. Const., Art. I, §8, cl. 8)). The jury heard that Google limited its use of the Sun Java API to tasks and specific programming demands related to Android. It copied the API (which Sun created for use in desktop and laptop computers) only insofar as needed to include tasks that would be useful in smartphone programs. App. 169-170. And it did so only insofar as needed to allow programmers to call upon those tasks without discarding a portion of a familiar programming language and learning a new one. Id., at 139-140. To repeat, Google, through Android, provided a new collection of tasks operating in a distinct and different computing environment. Those tasks were carried out through the use of new implementing code (that Google wrote) designed to operate within that new environment. Some of the amici refer to what Google did as “reimplementation,” defined as the “building of a system . . . that repurposes the same words and syntaxes” of an existing system—in this case so that programmers who had learned an existing system could put their basic skills to use in a new one. Brief for R Street Institute et al. as Amici Curiae 2. The record here demonstrates the numerous ways in which reimplementing an interface can further the development of computer programs. The jury heard that shared interfaces are necessary for different programs to speak to each other. App. 125 (“We have to agree on the APIs so that the application I write to show a movie runs on your device”). It heard that the reimplementation of interfaces is necessary if programmers are to be able to use their acquired skills. Id., at 191 (“If the API labels change, then either the software wouldn’t continue to work anymore or the developer . . . would have to learn a whole new language to be able to use these API labels”). It heard that the reuse of APIs is common in the industry. Id., at 115, 155, 663. It heard that Sun itself had used pre-existing interfaces in creating Java. Id., at 664. And it heard that Sun executives thought that widespread use of the Java programming language, including use on a smartphone platform, would benefit the company. Id., at 130-133. Amici supporting Google have summarized these same points—points that witnesses explained to the jury. See, e.g., Brief for Copyright Scholars as Amici Curiae 25 (“[T]he portions of Java SE that Google reimplemented may have helped preserve consistency of use within the larger Java developer community”); Brief for Microsoft Corporation as Amicus Curiae 22 (“[A]llowing reasonable fair use of functional code enables innovation that creates new opportunities for the whole market to grow”); Brief for 83 Computer Scientists as Amici Curiae 20 (“Reimplementing interfaces fueled widespread adoption of popular programming languages” (emphasis deleted)); Brief for R Street Institute et al. as Amici Curiae 15-20 (describing Oracle’s reimplementation of other APIs); see also Brief for American Antitrust Institute as Amicus Curiae 7 (“Copyright on largely functional elements of software that [have] become an industry standard gives a copyright holder anti-competitive power”). These and related facts convince us that the “purpose and character” of Google’s copying was transformative—to the point where this factor too weighs in favor of fair use. There are two other considerations that are often taken up under the first factor: commerciality and good faith. The text of §107 includes various noncommercial uses, such as teaching and scholarship, as paradigmatic examples of privileged copying. There is no doubt that a finding that copying was not commercial in nature tips the scales in favor of fair use. But the inverse is not necessarily true, as many common fair uses are indisputably commercial. For instance, the text of §107 includes examples like “news reporting,” which is often done for commercial profit. So even though Google’s use was a commercial endeavor—a fact no party disputed, see 886 F. 3d, at 1197—that is not dispositive of the first factor, particularly in light of the inherently transformative role that the reimplementation played in the new Android system. As for bad faith, our decision in Campbell expressed some skepticism about whether bad faith has any role in a fair use analysis. 510 U. S., at 585, n. 18, 114 S. Ct. 1164, 127 L. Ed. 2d 500. We find this skepticism justifiable, as “[c]opyright is not a privilege reserved for the well-behaved.” Leval 1126. We have no occasion here to say whether good faith is as a general matter a helpful inquiry. We simply note that given the strength of the other factors pointing toward fair use and the jury finding in Google’s favor on hotly contested evidence, that factbound consideration is not determinative in this context.

A. “The Nature of the Copyrighted Work” The Sun Java API is a “user interface.” It provides a way through which users (here the programmers) can “manipulate and control” task-performing computer programs “via a series of menu commands.” Lotus Development Corp., 49 F. 3d, at 809. The API reflects Sun’s division of possible tasks that a computer might perform into a set of actual tasks that certain kinds of computers actually will perform. Sun decided, for example, that its API would call up a task that compares one integer with another to see which is the larger. Sun’s API (to our knowledge) will not call up the task of determining which great Arabic scholar decided to use Arabic numerals (rather than Roman numerals) to perform that “larger integer” task. No one claims that the decisions about what counts as a task are themselves copyrightable—although one might argue about decisions as to how to label and organize such tasks (e.g., the decision to name a certain task “max” or to place it in a class called “Math.” Cf. Baker v. Selden, 101 U. S. 99, 25 L. Ed. 841, 1880 Dec. Comm'r Pat. 422 (1880)). As discussed above, supra, at 3-5, and in Appendix B, infra, we can think of the technology as having three essential parts. First, the API includes “implementing code,” which actually instructs the computer on the steps to follow to carry out each task. Google wrote its own programs (implementing programs) that would perform each one of the tasks that its API calls up. Second, the Sun Java API associates a particular command, called a “method call,” with the calling up of each task. The symbols java.lang., for example, are part of the command that will call up the program (whether written by Sun or, as here, by Google) that instructs the computer to carry out the “larger number” operation. Oracle does not here argue that the use of these commands by programmers itself violates its copyrights. Third, the Sun Java API contains computer code that will associate the writing of a method call with particular “places” in the computer that contain the needed implementing code. This is the declaring code. The declaring code both labels the particular tasks in the API and organizes those tasks, or “methods,” into “packages” and “classes.” We have referred to this organization, by way of rough analogy, as file cabinets, drawers, and files. Oracle does claim that Google’s use of the Sun Java API’s declaring code violates its copyrights. The declaring code at issue here resembles other copyrighted works in that it is part of a computer program. Congress has specified that computer programs are subjects of copyright. It differs, however, from many other kinds of copyrightable computer code. It is inextricably bound together with a general system, the division of computing tasks, that no one claims is a proper subject of copyright. It is inextricably bound up with the idea of organizing tasks into what we have called cabinets, drawers, and files, an idea that is also not copyrightable. It is inextricably bound up with the use of specific commands known to programmers, known here as method calls (such as java.lang.Math.max , etc.), that Oracle does not here contest. And it is inextricably bound up with implementing code, which is copyrightable but was not copied. Moreover, the copied declaring code and the uncopied implementing programs call for, and reflect, different kinds of capabilities. A single implementation may walk a computer through dozens of different steps. To write implementing programs, witnesses told the jury, requires balancing such considerations as how quickly a computer can execute a task or the likely size of the computer’s memory. One witness described that creativity as “magic” practiced by an API developer when he or she worries “about things like power management” for devices that “run on a battery.” App. 143; see also id., at 147, 204. This is the very creativity that was needed to develop the Android software for use not in laptops or desktops but in the very different context of smartphones. The declaring code (inseparable from the programmer’s method calls) embodies a different kind of creativity. Sun Java’s creators, for example, tried to find declaring code names that would prove intuitively easy to remember. Id., at 211. They wanted to attract programmers who would learn the system, help to develop it further, and prove reluctant to use another. See post, at 10 (“Declaring code . . . is user facing. It must be designed and organized in a way that is intuitive and understandable to developers so that they can invoke it”). Sun’s business strategy originally emphasized the importance of using the API to attract programmers. It sought to make the API “open” and “then . . . compete on implementations.” App. 124-125. The testimony at trial was replete with examples of witnesses drawing this critical line between the user-centered declaratory code and the innovative implementing code. Id., at 126-127, 159-160, 163-164, 187, 190-191. These features mean that, as part of a user interface, the declaring code differs to some degree from the mine run of computer programs. Like other computer programs, it is functional in nature. But unlike many other programs, its use is inherently bound together with uncopyrightable ideas (general task division and organization) and new creative expression (Android’s implementing code). Unlike many other programs, its value in significant part derives from the value that those who do not hold copyrights, namely, computer programmers, invest of their own time and effort to learn the API’s system. And unlike many other programs, its value lies in its efforts to encourage programmers to learn and to use that system so that they will use (and continue to use) Sun-related implementing programs that Google did not copy. Although copyrights protect many different kinds of writing, Leval 1116, we have emphasized the need to “recogni[ze] that some works are closer to the core of [copyright] than others,” Campbell, 510 U. S., at 586, 114 S. Ct. 1164, 127 L. Ed. 2d 500. In our view, for the reasons just described, the declaring code is, if copyrightable at all, further than are most computer programs (such as the implementing code) from the core of copyright. That fact diminishes the fear, expressed by both the dissent and the Federal Circuit, that application of “fair use” here would seriously undermine the general copyright protection that Congress provided for computer programs. And it means that this factor, “the nature of the copyrighted work,” points in the direction of fair use.

C. “The Amount and Substantiality of the Portion Used” If one considers the declaring code in isolation, the quantitative amount of what Google copied was large. Google copied the declaring code for 37 packages of the Sun Java API, totaling approximately 11,500 lines of code. Those lines of code amount to virtually all the declaring code needed to call up hundreds of different tasks. On the other hand, if one considers the entire set of software material in the Sun Java API, the quantitative amount copied was small. The total set of Sun Java API computer code, including implementing code, amounted to 2.86 million lines, of which the copied 11,500 lines were only 0.4 percent. App. 212. The question here is whether those 11,500 lines of code should be viewed in isolation or as one part of the considerably greater whole. We have said that even a small amount of copying may fall outside of the scope of fair use where the excerpt copied consists of the “‘heart’” of the original work’s creative expression. Harper & Row, 471 U. S., at 564-565, 105 S. Ct. 2218, 85 L. Ed. 2d 588. On the other hand, copying a larger amount of material can fall within the scope of fair use where the material copied captures little of the material’s creative expression or is central to a copier’s valid purpose. See, e.g., Campbell, 510 U. S., at 588, 114 S. Ct. 1164, 127 L. Ed. 2d 500; New Era Publications Int’l, ApS v. Carol Publishing Group, 904 F. 2d 152, 158 (CA2 1990). If a defendant had copied one sentence in a novel, that copying may well be insubstantial. But if that single sentence set forth one of the world’s shortest short stories—“When he awoke, the dinosaur was still there.”—the question looks much different, as the copied material constitutes a small part of the novel but the entire short story. See A. Monterroso, El Dinosaurio, in Complete Works & Other Stories 42 (E. Grossman transl. 1995). (In the original Spanish, the story reads: “Cuando despertó, el dinosaurio todavía estaba allí.”) Several features of Google’s copying suggest that the better way to look at the numbers is to take into account the several million lines that Google did not copy. For one thing, the Sun Java API is inseparably bound to those task-implementing lines. Its purpose is to call them up. For another, Google copied those lines not because of their creativity, their beauty, or even (in a sense) because of their purpose. It copied them because programmers had already learned to work with the Sun Java API’s system, and it would have been difficult, perhaps prohibitively so, to attract programmers to build its Android smartphone system without them. Further, Google’s basic purpose was to create a different task-related system for a different computing environment (smartphones) and to create a platform—the Android platform—that would help achieve and popularize that objective. The “substantiality” factor will generally weigh in favor of fair use where, as here, the amount of copying was tethered to a valid, and transformative, purpose. Supra, at 25-26; see Campbell, 510 U. S., at 586-587, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (explaining that the factor three “enquiry will harken back to the first of the statutory factors, for . . . the extent of permissible copying varies with the purpose and character of the use”). We do not agree with the Federal Circuit’s conclusion that Google could have achieved its Java-compatibility objective by copying only the 170 lines of code that are “necessary to write in the Java language.” 886 F. 3d, at 1206. In our view, that conclusion views Google’s legitimate objectives too narrowly. Google’s basic objective was not simply to make the Java programming language usable on its Android systems. It was to permit programmers to make use of their knowledge and experience using the Sun Java API when they wrote new programs for smartphones with the Android platform. In principle, Google might have created its own, different system of declaring code. But the jury could have found that its doing so would not have achieved that basic objective. In a sense, the declaring code was the key that it needed to unlock the programmers’ creative energies. And it needed those energies to create and to improve its own innovative Android systems. We consequently believe that this “substantiality” factor weighs in favor of fair use.

D. Market Effects The fourth statutory factor focuses upon the “effect” of the copying in the “market for or value of the copyrighted work.” 17 U. S. C. §107(4). Consideration of this factor, at least where computer programs are at issue, can prove more complex than at first it may seem. It can require a court to consider the amount of money that the copyright owner might lose. As we pointed out in Campbell, “verbatim copying of the original in its entirety for commercial purposes” may well produce a market substitute for an author’s work. 510 U. S., at 591, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Making a film of an author’s book may similarly mean potential or presumed losses to the copyright owner. Those losses normally conflict with copyright’s basic objective: providing authors with exclusive rights that will spur creative expression. But a potential loss of revenue is not the whole story. We here must consider not just the amount but also the source of the loss. As we pointed out in Campbell, a “lethal parody, like a scathing theatre review,” may “kil[l ] demand for the original.” Id., at 591-592, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Yet this kind of harm, even if directly translated into foregone dollars, is not “cognizable under the Copyright Act.” Id., at 592, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Further, we must take into account the public benefits the copying will likely produce. Are those benefits, for example, related to copyright’s concern for the creative production of new expression? Are they comparatively important, or unimportant, when compared with dollar amounts likely lost (taking into account as well the nature of the source of the loss)? Cf. MCA, INC. v. Wilson, 677 F. 2d 180, 183 (CA2 1981) (calling for a balancing of public benefits and losses to copyright owner under this factor). We do not say that these questions are always relevant to the application of fair use, not even in the world of computer programs. Nor do we say that these questions are the only questions a court might ask. But we do find them relevant here in helping to determine the likely market effects of Google’s reimplementation. As to the likely amount of loss, the jury could have found that Android did not harm the actual or potential markets for Java SE. And it could have found that Sun itself (now Oracle) would not have been able to enter those markets successfully whether Google did, or did not, copy a part of its API. First, evidence at trial demonstrated that, regardless of Android’s smartphone technology, Sun was poorly positioned to succeed in the mobile phone market. The jury heard ample evidence that Java SE’s primary market was laptops and desktops. App. 99, 200. It also heard that Sun’s many efforts to move into the mobile phone market had proved unsuccessful. Id., at 135, 235, 671. As far back as 2006, prior to Android’s release, Sun’s executives projected declining revenue for mobile phones because of emerging smartphone technology. Id., at 240. When Sun’s former CEO was asked directly whether Sun’s failure to build a smartphone was attributable to Google’s development of Android, he answered that it was not. Id., at 650. Given the evidence showing that Sun was beset by business challenges in developing a mobile phone product, the jury was entitled to agree with that assessment. Second, the jury was repeatedly told that devices using Google’s Android platform were different in kind from those that licensed Sun’s technology. For instance, witnesses explained that the broader industry distinguished between smartphones and simpler “feature phones.” Id., at 237. As to the specific devices that used Sun-created software, the jury heard that one of these phones lacked a touchscreen, id., at 359-360, while another did not have a QWERTY keyboard, id., at 672. For other mobile devices, the evidence showed that simpler products, like the Kindle, used Java software, id., at 396, while more advanced technology, like the Kindle Fire, were built on the Android operating system, id., at 206. This record evidence demonstrates that, rather than just “repurposing [Sun’s] code from larger computers to smaller computers,” post, at 16, Google’s Android platform was part of a distinct (and more advanced) market than Java software. Looking to these important differences, Google’s economic expert told the jury that Android was not a market substitute for Java’s software. As he explained, “the two products are on very different devices,” and the Android platform, which offers “an entire mobile operating stack,” is a “very different typ[e] of produc[t]” than Java SE, which is “just an applications programming framework.” App. 256; see also id., at 172-174. Taken together, the evidence showed that Sun’s mobile phone business was declining, while the market increasingly demanded a new form of smartphone technology that Sun was never able to offer. Finally, the jury also heard evidence that Sun foresaw a benefit from the broader use of the Java programming language in a new platform like Android, as it would further expand the network of Java-trained programmers. Id., at 131-133; see also id., at 153 (“Once an API starts getting reimplemented, you know it has succeeded”). In other words, the jury could have understood Android and Java SE as operating in two distinct markets. And because there are two markets at issue, programmers learning the Java language to work in one market (smartphones) are then able to bring those talents to the other market (laptops). See 4 Nimmer on Copyright §13.05[A][4] (explaining that factor four asks what the impact of “widespread conduct of the sort engaged in by the defendant” would be on the market for the present work). Oracle presented evidence to the contrary. Indeed, the Federal Circuit held that the “market effects” factor militated against fair use in part because Sun had tried to enter the Android market. 886 F. 3d, at 1209 (Sun sought licensing agreement with Google). But those licensing negotiations concerned much more than 37 packages of declaring code, covering topics like “the implementation of [Java’s] code” and “branding and cooperation” between the firms. App. 245; see also 4 Nimmer on Copyright §13.05[A][4] (cautioning against the “danger of circularity posed” by considering unrealized licensing opportunities because “it is a given in every fair use case that plaintiff suffers a loss of a potential market if that potential is defined as the theoretical market for licensing the very use at bar”). In any event, the jury’s fair use determination means that neither Sun’s effort to obtain a license nor Oracle’s conflicting evidence can overcome evidence indicating that, at a minimum, it would have been difficult for Sun to enter the smartphone market, even had Google not used portions of the Sun Java API. On the other hand, Google’s copying helped Google make a vast amount of money from its Android platform. And enforcement of the Sun Java API copyright might give Oracle a significant share of these funds. It is important, however, to consider why and how Oracle might have become entitled to this money. When a new interface, like an API or a spreadsheet program, first comes on the market, it may attract new users because of its expressive qualities, such as a better visual screen or because of its superior functionality. As time passes, however, it may be valuable for a different reason, namely, because users, including programmers, are just used to it. They have already learned how to work with it. See Lotus Development Corp., 49 F. 3d, at 821 (Boudin, J., concurring). The record here is filled with evidence that this factor accounts for Google’s desire to use the Sun Java API. See, e.g., App. 169-170, 213-214. This source of Android’s profitability has much to do with third parties’ (say, programmers’) investment in Sun Java programs. It has correspondingly less to do with Sun’s investment in creating the Sun Java API. We have no reason to believe that the Copyright Act seeks to protect third parties’ investment in learning how to operate a created work. Cf. Campbell, 510 U. S., at 591-592, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (discussing the need to identify those harms that are “cognizable under the Copyright Act”). Finally, given programmers’ investment in learning the Sun Java API, to allow enforcement of Oracle’s copyright here would risk harm to the public. Given the costs and difficulties of producing alternative APIs with similar appeal to programmers, allowing enforcement here would make of the Sun Java API’s declaring code a lock limiting the future creativity of new programs. Oracle alone would hold the key. The result could well prove highly profitable to Oracle (or other firms holding a copyright in computer interfaces). But those profits could well flow from creative improvements, new applications, and new uses developed by users who have learned to work with that interface. To that extent, the lock would interfere with, not further, copyright’s basic creativity objectives. See Connectix Corp., 203 F. 3d, at 607; see also Sega Enterprises, 977 F. 2d, at 1523-1524 (“An attempt to monopolize the market by making it impossible for others to compete runs counter to the statutory purpose of promoting creative expression”); Lexmark Int’l, 387 F. 3d, at 544 (noting that where a subsequent user copied a computer program to foster functionality, it was not exploiting the programs “commercial value as a copyrighted work” (emphasis in original)). After all, “copyright supplies the economic incentive to [both] create and disseminate ideas,” Harper & Row, 471 U. S., at 558, 105 S. Ct. 2218, 85 L. Ed. 2d 588, and the reimplementation of a user interface allows creative new computer code to more easily enter the market. The uncertain nature of Sun’s ability to compete in Android’s market place, the sources of its lost revenue, and the risk of creativity-related harms to the public, when taken together, convince that this fourth factor—market effects—also weighs in favor of fair use.

C. The Purpose and Character of the Use The second-most important factor—“the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes,” §107(1)—requires us to consider whether use was “commercial” and whether it was “transformative.” Campbell, 510 U. S., at 578-579, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Both aspects heavily favor Oracle. Begin with the overwhelming commercial nature of Google’s copying. In 2015 alone, the year before the fair-use trial, Google earned $18 billion from Android. That number has no doubt dramatically increased as Android has grown to dominate the global market share. On this scale, Google’s use of Oracle’s declaring code weighs heavily—if not decisively—against fair use. The majority attempts to dismiss this overwhelming commercial use by noting that commercial use does “not necessarily” weigh against fair use. Ante, at 1204. True enough. Commercial use sometimes can be overcome by use that is sufficiently “transformative.” Campbell, 510 U. S., at 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500. But “we cannot ignore [Google’s] intended purpose of supplanting [Oracle’s] commercially valuable” platform with its own. Harper, 471 U. S., at 562, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (emphasis in original). Even if we could, we have never found fair use for copying that reaches into the tens of billions of dollars and wrecks the copyright holder’s market. Regardless, Google fares no better on transformative use. A court generally cannot find fair use unless the copier’s use is transformative. A work is “transformative” if it “adds something new, with a further purpose or different character, altering the first with new expression, meaning, or message.” Campbell, 510 U. S., at 579, 114 S. Ct. 1164, 127 L. Ed. 2d 500. This question is “guided by the examples [of fair use] given in the preamble to §107.” Id., at 578, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Those examples include: “criticism, comment, news reporting, teaching . . ., scholarship, or research.” §107. Although these examples are not exclusive, they are illustrative, and Google’s repurposing of Java code from larger computers to smaller computers resembles none of them. Google did not use Oracle’s code to teach or reverse engineer a system to ensure compatibility. Instead, to “avoid the drudgery in working up something fresh,” id., at 580, 114 S. Ct. 1164, 127 L. Ed. 2d 500, Google used the declaring code for the same exact purpose Oracle did. As the Federal Circuit correctly determined, “[t]here is nothing fair about taking a copyrighted work verbatim and using it for the same purpose and function as the original in a competing platform.” 886 F. 3d, at 1210. The majority acknowledges that Google used the copied declaring code “for the same reason” Oracle did. Ante, at 1203. So, by turns, the majority transforms the definition of “transformative.” Now, we are told, “transformative” simply means—at least for computer code—a use that will help others “create new products.” Ibid; accord, ante, at 1203 (Google’s copying “can further the development of computer programs”). That new definition eviscerates copyright. A movie studio that converts a book into a film without permission not only creates a new product (the film) but enables others to “create products”—film reviews, merchandise, YouTube highlight reels, late night television interviews, and the like. Nearly every computer program, once copied, can be used to create new products. Surely the majority would not say that an author can pirate the next version of Microsoft Word simply because he can use it to create new manuscripts. Ultimately, the majority wrongly conflates transformative use with derivative use. To be transformative, a work must do something fundamentally different from the original. A work that simply serves the same purpose in a new context—which the majority concedes is true here—is derivative, not transformative. Congress made clear that Oracle holds “the exclusive rights . . . to prepare derivative works.” §106(2). Rather than create a transformative product, Google “profit[ed] from exploitation of the copyrighted material without paying the customary price.” Harper, 471 U. S., at 562, 105 S. Ct. 2218, 85 L. Ed. 2d 588.

A. The Nature of the Copyrighted Work This factor requires courts to assess the level of creativity or functionality in the original work. It generally favors fair use when a copyrighted work is more “informational or functional” than “creative.” 4 M. Nimmer & D. Nimmer, Copyright §13.05[A][2][a] (2019). Because code is predominantly functional, this factor will often favor copying when the original work is computer code. But because Congress determined that declaring and implementing code are copyrightable, this factor alone cannot support a finding of fair use. The majority, however, uses this factor to create a distinction between declaring and implementing code that in effect removes copyright protection from declaring code. It concludes that, unlike implementing code, declaring code is far “from the core of copyright” because it becomes valuable only when third parties (computer programmers) value it and because it is “inherently bound together with uncopyrightable ideas.” Ante, at 1202 - 1203. Congress, however, rejected this sort of categorical distinction that would make declaring code less worthy of protection. The Copyright Act protects code that operates “in a computer in order to bring about a certain result” both “directly” (implementing code) and “indirectly” (declaring code). §101. And if anything, declaring code is closer to the “core of copyright.” Ante, at 1202 - 1203. Developers cannot even see implementing code. Oracle Am. Inc., v. Google Inc., 2016 U.S. Dist. LEXIS 74931, 2016 WL 3181206, *4 (ND Cal., June 8, 2016); see also ante, at1201 - 1202 (declaring code is “user-centered”). Implementing code thus conveys no expression to developers. Declaring code, in contrast, is user facing. It must be designed and organized in a way that is intuitive and understandable to developers so that they can invoke it. Even setting those concerns aside, the majority’s distinction is untenable. True, declaring code is “inherently bound together with uncopyrightable ideas.” Ante, at1201 - 1203. Is anything not? Books are inherently bound with uncopyrightable ideas—the use of chapters, having a plot, or including dialogue or footnotes. This does not place books far “from the core of copyright.” And implementing code, which the majority concedes is copyrightable, is inherently bound up with “the division of computing tasks” that cannot be copyrighted. Ante, at 1201. We have not discounted a work of authorship simply because it is associated with noncopyrightable ideas. While ideas cannot be copyrighted, expressions of those ideas can. Golan, 565 U. S., at 328, 132 S. Ct. 873, 181 L. Ed. 2d 835. Similarly, it makes no difference that the value of declaring code depends on how much time third parties invest in learning it. Many other copyrighted works depend on the same. A Broadway musical script needs actors and singers to invest time learning and rehearsing it. But a theater cannot copy a script—the rights to which are held by a smaller theater—simply because it wants to entice actors to switch theaters and because copying the script is more efficient than requiring the actors to learn a new one. What the majority says is true of declaring code is no less true of implementing code. Declaring code is how programmers access prewritten implementing code. The value of that implementing code thus is directly proportional to how much programmers value the associated declaring code. The majority correctly recognizes that declaring code “is inextricably bound up with implementing code,” ante, at 1201 - 1202, but it overlooks the implications of its own conclusion. Only after wrongly concluding that the nature of declaring code makes that code generally unworthy of protection does the Court move on to consider the other factors. This opening mistake taints the Court’s entire analysis.

D. The Amount and Substantiality of the Portion Used The statutory fair-use factors also instruct us to consider “the amount and substantiality of the portion used in relation to the copyrighted work as a whole.” §107(3). In general, the greater the amount of use, the more likely the copying is unfair. Ibid. But even if the copier takes only a small amount, copying the “‘heart’” or “focal points” of a work weighs against fair use, Harper, 471 U. S., at 565-566, 105 S. Ct. 2218, 85 L. Ed. 2d 588, unless “‘no more was taken than necessary’” for the copier to achieve transformative use, Campbell, 510 U. S., at 589, 114 S. Ct. 1164, 127 L. Ed. 2d 500. Google does not dispute the Federal Circuit’s conclusion that it copied the heart or focal points of Oracle’s work. 886 F. 3d, at 1207. The declaring code is what attracted programmers to the Java platform and why Google was so interested in that code. And Google copied that code “verbatim,” which weighs against fair use. Harper, 471 U. S., at 565, 105 S. Ct. 2218, 85 L. Ed. 2d 588. The majority does not disagree. Instead, it concludes that Google took no more than necessary to create new products. That analysis fails because Google’s use is not transformative. Campbell, 510 U. S., at 586, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (recognizing that this fourth factor “will harken back to the [purpose-and-character] statutory facto[r]”). This factor thus weighs against Google. Even if Google’s use were transformative, the majority is wrong to conclude that Google copied only a small portion of the original work. The majority points out that the 11,500 lines of declaring code—enough to fill about 600 pages in an appendix, Tr. of Oral Arg. 57—were just a fraction of the code in the Java platform. But the proper denominator is declaring code, not all code. A copied work is quantitatively substantial if it could “serve as a market substitute for the original” work or “potentially licensed derivatives” of that work. Campbell, 510 U. S., at 587, 114 S. Ct. 1164, 127 L. Ed. 2d 500. The declaring code is what attracted programmers. And it is what made Android a “market substitute” for “potentially licensed derivatives” of Oracle’s Java platform. Google’s copying was both qualitatively and quantitatively substantial.

B. Market Effects “[U]ndoubtedly the single most important element of fair use” is the effect of Google’s copying “‘upon the potential market for or value of [Oracle’s] copyrighted work.’” Harper & Row, Publishers Inc., v. Nation Enterprises, 471 U. S. 539, 566, 105 S. Ct. 2218, 85 L. Ed. 2d 588 (1985). As the Federal Circuit correctly determined, “evidence of actual and potential harm stemming from Google’s copying was ‘overwhelming.’” 886 F. 3d 1179, 1209 (2018). By copying Oracle’s code to develop and release Android, Google ruined Oracle’s potential market in at least two ways. First, Google eliminated the reason manufacturers were willing to pay to install the Java platform. Google’s business model differed from Oracle’s. While Oracle earned revenue by charging device manufacturers to install the Java platform, Google obtained revenue primarily through ad sales. Its strategy was to release Android to device manufacturers for free and then use Android as a vehicle to collect data on consumers and deliver behavioral ads. With a free product available that included much of Oracle’s code (and thus with similar programming potential), device manufacturers no longer saw much reason to pay to embed the Java platform. For example, before Google released Android, Amazon paid for a license to embed the Java platform in Kindle devices. But after Google released Android, Amazon used the cost-free availability of Android to negotiate a 97.5% discount on its license fee with Oracle. Evidence at trial similarly showed that right after Google released Android, Samsung’s contract with Oracle dropped from $40 million to about $1 million. Google contests none of this except to say that Amazon used a different Java platform, Java Micro Edition instead of Java Standard Edition. That difference is inconsequential because the former was simply a smaller subset of the latter. Google copied code found in both platforms. The majority does not dispute—or even mention—this enormous harm. Second, Google interfered with opportunities for Oracle to license the Java platform to developers of smartphone operating systems. Before Google copied Oracle’s code, nearly every mobile phone on the market contained the Java platform. Oracle’s code was extraordinarily valuable to anybody who wanted to develop smartphones, which explains why Google tried no fewer than four times to license it. The majority’s remark that Google also sought other licenses from Oracle, ante, at 1207 - 1208, does not change this central fact. Both parties agreed that Oracle could enter Google’s current market by licensing its declaring code. But by copying the code and releasing Android, Google eliminated Oracle’s opportunity to license its code for that use. The majority writes off this harm by saying that the jury could have found that Oracle might not have been able to enter the modern smartphone market successfully. Ante, at 1206 -1207. But whether Oracle could itself enter that market is only half the picture. We look at not only the potential market “that creators of original works would in general develop” but also those potential markets the copyright holder might “license others to develop.” Campbell v. Acuff-Rose Music, Inc., 510 U. S. 569, 592, 114 S. Ct. 1164, 127 L. Ed. 2d 500 (1994). A book author need not be able to personally convert a book into a film so long as he can license someone else to do so. That Oracle could have licensed its code for use in Android is undisputed. Unable to seriously dispute that Google’s actions had a disastrous effect on Oracle’s potential market, the majority changes course and asserts that enforcing copyright protection could harm the public by giving Oracle the power to “limi[t] the future creativity” of programs on Android. Ante, at 1208. But this case concerns only versions of Android released through November 2014. Order in No. 3:10-cv-3561 (ND Cal., Feb. 5, 2016), Doc. 1479, p. 2 (identifying versions through Android Lollipop 5.0). Google has released six major versions since then. Only about 7.7% of active Android devices still run the versions at issue. The majority’s concern about a lock-in effect might carry more weight if this suit concerned versions of Android widely in use or that will be widely in use. It makes little sense in a suit about versions that are close to obsolete. The majority’s concern about a lock-in effect also is speculation belied by history. First, Oracle never had lock-in power. The majority (again) overlooks that Apple and Microsoft created mobile operating systems without using Oracle’s declaring code. Second, Oracle always made its declaring code freely available to programmers. There is little reason to suspect Oracle might harm programmers by stopping now. And third, the majority simply assumes that the jury, in a future suit over current Android versions, would give Oracle control of Android instead of just awarding damages or perpetual royalties. If the majority is going to speculate about what Oracle might do, it at least should consider what Google has done. The majority expresses concern that Oracle might abuse its copyright protection (on outdated Android versions) and “‘attempt to monopolize the market.’” Ante, at 1208 - 1209. But it is Google that recently was fined a record $5 billion for abusing Android to violate antitrust laws. Case AT.40099, Google Android, July 18, 2018 (Eur. Comm’n-Competition); European Comm’n Press Release, Commission Fines Google €4.34 Billion for Illegal Practices Regarding Android Mobile Devices to Strengthen Dominance of Google’s Search Engine, July 18, 2018. Google controls the most widely used mobile operating system in the world. And if companies may now freely copy libraries of declaring code whenever it is more convenient than writing their own, others will likely hesitate to spend the resources Oracle did to create intuitive, well-organized libraries that attract programmers and could compete with Android. If the majority is worried about monopolization, it ought to consider whether Google is the greater threat. By copying Oracle’s work, Google decimated Oracle’s market and created a mobile operating system now in over 2.5 billion actively used devices, earning tens of billions of dollars every year. If these effects on Oracle’s potential market favor Google, something is very wrong with our fair-use analysis.