Cambridge University Press, et al v. J.L. Albert, et al

Filing 89

Amicus Brief as of right or by consent of the parties filed by Mitchell L. Stoltz representing Amici Curiae American Library Association, Association of College and Research Libraries, and Association of Research Libraries. Service date: 04/25/2013 email - District Judge Evans; Originating Clerk/Ag Hatten; Attorney for Amicus Curium: Aistars, Dove, Lerner, Rasenberger, Steinman, Tonsager, Wasoff; Attorney for Appellants: Bloom, Krugman, Larson, Rains, Rich, Singer; Attorney for Appellees: Ashby, Askew, Bates, Eskow, Gentry, Harbin, Levie, Lynn, Miller, Moffitt, Quicker, Schaetzel, Volkert, Warenzak, Whiting-Pack; Attorney for Not Party: Tenny. (ECF: Mtichell Stoltz)

Download PDF
12-14676-FF & 12-15147-FF (Consolidated Appeals) UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT Cambridge University Press, Oxford University Press, Inc., and Sage Publications, Inc., Plaintiffs-Appellants, v. Mark P. Becker, et al., Defendants-Appellees. __________________________________________________________________ On Appeal From the United States District Court for the Northern District of Georgia, Atlanta Division, D.C. No. 1:08-cv-1425 ODE (Evans, J.) __________________________________________________________________ BRIEF OF AMICI CURIAE AMERICAN LIBRARY ASSOCIATION, ASSOCIATION OF COLLEGE AND RESEARCH LIBRARIES, AND ASSOCIATION OF RESEARCH LIBRARIES IN SUPPORT OF AFFIRMANCE Mitchell Stoltz Counsel of Record ELECTRONIC FRONTIER FOUNDATION 815 Eddy Street San Francisco, CA 94109 Tel: (415) 436-9333 Email: mitch@eff.org April 25, 2013   Jonathan Band (on the brief) JONATHAN BAND PLLC 21 Dupont Circle NW, 8th Floor Washington, DC 20036 Tel: (202) 296-5675 Email: jband@policybandwidth.com Counsel for Amici Curiae CERTIFICATE OF INTERESTED PERSONS AND CORPORATE DISCLOSURE STATEMENT In addition to those identified in Appellee’s brief pursuant to Rule 26.1-1 of the Eleventh Circuit Rules, amici curiae disclose the following trial judges, attorneys, persons, associations of persons, firms, partnerships, and corporations as having an interest in the outcome of this case: American Library Association Association of College and Research Libraries Association of Research Libraries Mitchell L. Stoltz, counsel for amici curiae Corynne McSherry, counsel for amici curiae Electronic Frontier Foundation, counsel for amici curiae Jonathan Band, counsel for amici curiae Pursuant to Fed. R. App. P. 26.1, amici curiae makes the following disclosures: The American Library Association, Association of College and Research Libraries, and Association of Research Libraries are nonprofit associations or organizations that have no parent corporation, and no publicly held corporation owns 10 percent or more of their respective stock. Pursuant to Fed. R. App. P. 29(c)(5), amici curiae states that no party’s counsel authored the brief in whole or in part, no party’s counsel contributed   i money that was intended to fund preparing or submitting the brief, and no person other than amici curiae, their members, or their counsel contributed money to fund the preparation or submission of this brief. Dated: April 25, 2013 Respectfully submitted, s/ Mitchell Stoltz Mitchell Stoltz ELECTRONIC FRONTIER FOUNDATION 454 Shotwell Street San Francisco, CA 94110 Tel: (415) 436-9333 Email: mitch@eff.org Jonathan Band (on the brief) JONATHAN BAND PLLC 21 Dupont Circle NW, 8th Floor Washington, D.C., 20036 Tel: (202) 296-5675 Email: jband@policybandwidth.com Counsel for Amici Curiae   ii TABLE OF CONTENTS CERTIFICATE OF INTERESTED PERSONS AND CORPORATE DISCLOSURE STATEMENT ...................................................................................i INTEREST OF AMICI ............................................................................................. 1 INTRODUCTION ..................................................................................................... 2 ARGUMENT ............................................................................................................ 6 I. GSU’S Electronic Reserves Policy Embodies Widespread and WellEstablished Best Practices for Fair Use ................................................ 6 II. The Fair Use Factors Favor a Fair Use Finding for All the Excerpts Used By GSU ...................................................................................... 11 A. B. Second Factor – Nature of the Work ........................................ 16 C. Third Factor – Amount Used .................................................... 18 D. III. First Factor – Purpose and Character of the Use ...................... 12 Fourth Factor – Effect Upon Market ........................................ 20 Placing Licensing in Context: Library Budgets and the Public Interest................................................................................................. 22 CONCLUSION ....................................................................................................... 29   iii TABLE OF AUTHORITIES   Federal Cases A.V. v. iParadigms, LLC, 562 F.3d 630 (4th Cir. 2009) ........................................................................ 12   Ass’n for Info. Media and Equip. v. Regents of the Univ. of California, No. 2:10-CV-09378-CBM, 2012 WL 7683452 (C.D. Cal. Nov. 20, 2012) 22   Authors Guild, Inc. v. HathiTrust, 11 CV 6351 HB, 2012 WL 4808939 (S.D.N.Y. Oct. 10, 2012) .................. 13   Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605 (2d Cir. 2006) ............................................................. 13, 20, 21   Cambridge U. Press v. Becker, 1:08-CV-1425-ODE (ND GA) .............................................................. passim   Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569 (1994) ........................................................................... 6, 12, 13   Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539 (1985) ....................................................................................... 6   Kelly v. Arriba Soft Corp., 336 F.3d 811 (9th Cir. 2003) ........................................................................ 13   Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146 (9th Cir. 2007) ............................................................ 6, 12, 13   Sony Ass’n for Info. and Media Equip. v. Regents, 2012 WL 7683452, (C.D. Cal. Nov. 20, 2012) .............................................. 9     Federal Statutes 17 U.S.C. § 101 ...................................................................................................... 16   17 U.S.C. § 103 ...................................................................................................... 16         iv 17 U.S.C. § 107 ............................................................................................ 2, 12, 19   17 U.S.C. § 505 ........................................................................................................ 4     Constitutional Provisions U.S. CONST. Art. I ................................................................................................... 26     Other Authorities “Additional Guidelines for Electronic Reserves,” USG Copyright Policy (Oct. 21, 2009) ......................................................................................... 9, 10   Association of Research Libraries, Expenditure Trends in ARL Libraries, 1986-2010 .................................................................................................... 24   Association of Research Libraries’ Code of Best Practices in Fair Use for Academic and Research Libraries ........................................................ passim   Bernard Knight, USPTO General Counsel, USPTO Position on Fair Use Copies of NPL Made in Patent Examination, January 19, 2012 .................. 13   Caralee Adams, State Funding for Higher Ed. Drops by 7.6 Percent in a Year, Education Week (June 27, 2012) ................................................................. 24   Claire Redhead, Growth in the use of the CC-BY license (Mar. 8, 2013) .............. 28 Directory of Open Access Books ........................................................................... 28   Exemption to Prohibition on Circumvention of Copyright Protection Systems for Access Control Technologies, 71 FR 68472 (2006)..................................... 16   Glenn S. McGuigan & Robert D. Russell, The Business of Academic Publishing: A Strategic Analysis of the Academic Journal Publishing Industry and its Impact on the Future of Scholarly Publishing, ELECTRONIC J. ACAD. SPECIAL LIBRARIANSHIP (2008) .................................................................... 24   J. Christopher Holobar & Andrew Marshall, E-Reserves Permissions and the Copyright Clearance Center: Process, Efficiency, and Cost, PORTAL: LIBRARIES AND THE ACADEMY ..................................................................... 21     v John Quinterno & Viany Orozco, The Great Cost Shift: How Higher Education Costs Undermine the Future Middle Class, Demos (Apr. 3, 2012) ............. 24   M. Laakso, et al., The Development of Open Access Journal Publishing from 1993 to 2009, PLoS ONE (2011) ................................................................. 28   Marlie Wasserman, Reprint: How Much Does It Cost to Publish A Monograph and Why?, J. of Electronic Pub., 1998 ......................................................... 14   Patricia Aufderheide and Peter Jaszi, Reclaiming Fair Use (2011) ......................... 7   Paul Basken, Lawmakers Offer Struggling Research Universities Sympathy, Not Cash, CHRON. OF HIGHER EDUC. (June 27, 2012) ................................. 24   Peter Hirtle, Why You Might Want to Avoid the CCC’s Annual License, LibraryLaw Blog (July 5, 2007) ................................................................... 21   “Policy on the Use of Copyrighted Works in Education and Research,” USG Copyright Policy (Oct. 21, 2009) ........................................................ 10   Press Release, “Cornell University and Publishers announce new copyright guidelines governing the use of digital course materials,” (Sept. 19, 2006) .. 3   Richard Posner, Do patent and copyright law restrict competition and creativity excessively?, The Becker-Posner Blog (Sept. 30, 2012) ........ 17, 18   Robert Kasunic, Is That All There Is? Reflections on the Nature of the Second Fair Use Factor, COLUM. J. L. & ARTS (2008) ............................................ 18   Sara Hebel, State Cuts Are Pushing Public Colleges into Peril, CHRON. OF HIGHER EDUC. (Mar. 14, 2010) .................................................................... 23   Steve Kolowich, Mending Fences, Inside Higher Ed (June 21, 2012)................... 25   University System of Georgia, Fair Use Checklist .................................................. 9   vi INTEREST OF AMICI1 The American Library Association (“ALA”), established in 1876, is a nonprofit professional organization of more than 60,000 librarians, library trustees, and other friends of libraries dedicated to providing and improving library services and promoting the public interest in a free and open information society. The Association of College and Research Libraries (“ACRL”), the largest division of the ALA, is a professional association of academic and research librarians and other interested individuals. It is dedicated to enhancing the ability of academic library and information professionals to serve the information needs of the higher education community and to improve learning, teaching, and research. The Association of Research Libraries (“ARL”) is an association of 125 research libraries in North America. ARL’s members include university libraries, public libraries, government and national libraries. ARL programs and services promote equitable access to and effective use of recorded knowledge in support of teaching and research. Collectively, these three library associations represent over 100,000 libraries and 350,000 librarians and other personnel that serve the needs of their patrons in the digital age. As a result, the associations share a strong interest in the balanced application of copyright law to new digital dissemination technologies.                                                                                                                 1   All parties consent to the filing of this brief. 1 Many of the libraries represented by amici library associations offer electronic reserves systems similar to the one maintained by Georgia State University (“GSU”).2 Librarians represented by amici library associations operate these electronic reserves systems. Accordingly, the people and entities amici represent would be adversely affected by a reversal of the district court’s decision that the fair use right under 17 U.S.C. § 107 permitted GSU to include in its electronic reserves system the vast majority of the excerpts at issue in this case. INTRODUCTION Appellants’ dogged pursuit of this litigation is, in a word, perplexing. When they commenced this litigation in 2008, GSU had a copyright policy in effect that was understood to allow the copying of up to 20% of a work. Cambridge U. Press v. Becker, 1:08-CV-1425-ODE (ND GA), Dkt. No. 235, (“Slip op.”) at 38. In response to the litigation, GSU adopted a new copyright policy in 2009. That new policy was modeled on the practices of peer institutions.3 It is very similar to guidelines jointly drafted in 2006 by Cornell University and the Association of American Publishers (“AAP”) 4 for the use of materials on the                                                                                                                 2 In this brief, “GSU” is used to refer collectively to all Appellees. Appellees Br. at 9. 4 Although they are not named plaintiffs, the district court found at trial that the AAP, together with the Copyright Clearance Center (CCC), had “organized the litigation and recruited the three plaintiffs to participate.” Slip op. at 25. AAP and CCC are also funding the cost of the litigation. 3   2 Cornell library’s electronic course reserves system5 after publishers represented by AAP threatened to sue Cornell for copyright infringement. The checklist GSU adopted is in use at Cornell and other institutions whose policies AAP has praised,6 and the CCC has endorsed it as “an important means for recording your fair use analysis.”7 The district court found that “the 2009 Copyright Policy significantly reduced the unlicensed copying of Plaintiffs’ works . . . at Georgia State.” Slip op. at 38. Nonetheless, rather than declare victory, Appellants continued to prosecute this litigation. In essence, Appellants’ position was that notwithstanding the new policy based on widely-used guidelines, GSU exceeded fair use if it did not pay a                                                                                                                 Press Release, “Cornell University and Publishers announce new copyright guidelines governing the use of digital course materials,” (Sept. 19, 2006), http://www.pressoffice.cornell.edu/releases/release.cfm?r=15899&y=2006&m=9 6 In 2008, the AAP announced that it had reached agreement with several universities concerning copyright guidelines for electronic reserves. The AAP press release stated that it had worked with the universities to develop these guidelines, which were “similar to those adopted by Cornell University,” and which the AAP hoped would “serve as a model for other colleges and universities.” See http://digital-scholarship.org/digitalkoans/2008/01/17/aapreaches-agreement-with-three-academic-libraries-about-e-reserves-guidelines/; Cornell University, Checklist for Conducting a Fair Use Analysis Before Using Copyrighted Materials, http://copyright.cornell.edu/policies/docs/Fair_Use_Checklist.pdf; Hofstra University, Fair Use Checklist, http://www.hofstra.edu/pdf/library/lib_fair_use_checklist.pdf. 7 See The Campus Guide to Copyright Compliance, http://www.copyright.com/Services/copyrightoncampus/basics/fairuse_list.html (last visited April 22, 2013). 5   3 fee whenever the publisher was willing to license use of the work — even if the publisher did not offer a license for the use in question, and even if the excerpt was no longer than a chapter or 10% of the work. The district court correctly refused to apply fair use in such a restrictive manner. After carefully reviewing the fair use case law and each of the claims of infringement submitted for judgment, it determined that only five of the excerpts infringed copyright. It further determined that the 2009 Copyright Policy caused those infringements in that it did not limit the copying in those instances to “decidedly small excerpts,” and it did not provide sufficient guidance in determining the effect on the market. See id. at 37-38. Nonetheless, the district court considered GSU to be the prevailing party for purposes of the award of attorneys’ fees under 17 U.S.C. § 505 because it “prevailed on all but five of the 99 copyright claims which were at issue when the trial of the case began.” August 10, 2012 Order, Dkt. 441, at 12. In this brief, amici address three issues. First, we explain that GSU’s 2009 Copyright Policy is consistent with a code of best practices for fair use established in early 2012 by a broad consensus of libraries. Similar codes are being used by a variety of communities that rely on fair use as part of their everyday practice. These codes help them anticipate and avoid legal risk so that they can continue to serve the public, create and distribute new works, and share research without fear   4 of crushing copyright liability. Amici urge the Court to resist the publishers invitation to upend the consensus GSU’s policy reflects. Second, amici argue that the district court’s finding that five excerpts infringed copyright may have been based on an incorrect assumption. The district court inferred with little evidence that the target audience for the books at issue included students. Based on this assumption, the district court concluded that the inclusion of these excerpts in electronic reserves was not a transformative use under the first fair use factor. This, in turn, adversely affected the analysis of the third and fourth fair use factors. In particular, the district court incorrectly concluded that the third factor favors fair use only if the amount copied is the lesser of one chapter or 10% of a book. Although amici do not believe remand is necessary, this Court should clarify that inclusion of excerpts in electronic reserves could be a transformative use in appropriate circumstances. Third, amici explain how a fair use finding in this case serves the public interest. The Supreme Court has stressed that the fair use analysis must “be mindful of the extent to which a use promotes the purposes of copyright and serves the interests of the public.” Perfect 10, Inc. v. Amazon.com, Inc., 508 F.3d 1146, 1166 (9th Cir. 2007) (citing Campbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 579 (1994)); Harper & Row Publishers, Inc. v. Nation Enterprises, 471 U.S. 539, 55657 (1985). Libraries are already investing as aggressively as they can in support of   5 scholarship, and scholars are more prolific than ever, with new models emerging to support even broader access to information. A ruling against fair use in this case will create a net loss to the public by suppressing educational uses, diverting scarce resources away from valuable educational investments, or both. This loss will not be balanced by any new incentive for creative activity. Such an outcome would surely disserve the public interest. ARGUMENT I. GSU’s Electronic Reserves Policy Embodies Widespread and WellEstablished Best Practices for Fair Use Appellants assert that GSU’s electronic reserves practices fall outside the bounds of fair use. In fact, GSU’s electronic reserves practices are consistent with a widespread fair use consensus among libraries, as embodied in the Association of Research Libraries’ Code of Best Practices in Fair Use for Academic and Research Libraries.8 The development of the Code was prompted by Professor Michael Madison’s insight (following a review of numerous fair use decisions) that U.S. courts were:                                                                                                                 8 The Code has been endorsed by amici ALA and ACRL, as well as the Arts Libraries Society of North America, the College Art Association, the Visual Resources Association, and the Music Library Association. See http://www.arl.org/storage/documents/publications/code-of-best-practices-fairuse.pdf.   6 implicitly or explicitly, asking about habit, custom, and social context of the use, using what Madison termed a ‘pattern-oriented’ approach to fair use reasoning. If the use was normal in a community, and you could understand how it was different from the original market use, then judges typically decided for fair use. Patricia Aufderheide and Peter Jaszi, Reclaiming Fair Use 71 (2011). Based on this insight, numerous communities have developed codes of fair use best practices in order to make fair use analysis more predictable for their members.9 Indeed, the need for predictability in the application of fair use has grown more acute during the information revolution over the past three decades. Digital technology invariably involves making copies, and it is the fair use doctrine that has enabled the copyright law to accommodate the rapid pace of innovation. More people rely on fair use for more activities than ever before.10 To help make fair use more predictable, the Association of Research Libraries set out to “document[] the considered views of the library community about best practices in fair use, drawn from the actual practices and experience of                                                                                                                 9 Codes of fair use best practices have been developed inter alia for OpenCourseWare, Documentary Filmmakers, Journalism, and Film and Media Educators. American University Center for Social Media, Best Practices, http://www.centerforsocialmedia.org/fair-use/best-practices. 10 True to its common law origins, fair use is constantly evolving. Appellants’ amici focus on the scope of fair use in 1976, when the doctrine was codified, but the interpretation of fair use 37 years ago is primarily of historical interest. Indeed, Appellants and their amici stress the importance of transformativeness to the fair use analysis, but Judge Leval wrote his seminal article on this subject fourteen years after the enactment of the 1976 Act.   7 the library community itself.” Association of Research Libraries, et al., Code of Best Practices in Fair Use for Academic and Research Libraries (“ARL Code”) 3 (2012). The resulting Code of Best Practices identified “situations that represent the library community’s current consensus about acceptable practices for the fair use of copyrighted materials.” Id. One of the Code’s principles addresses electronic reserves directly: “It is fair use to make appropriately tailored course-related content available to enrolled students via digital networks.” ARL Code at 14. Explaining the background of this principle, the Code observes that: Academic and research libraries have a long, and largely noncontroversial, history of supporting classroom instruction by providing students with access to reading materials, especially via physical on-site reserves. Teachers, in turn, have depended on libraries to provide this important service. Today, students and teachers alike strongly prefer electronic equivalents (e-reserves for text, streaming for audio and video) to the old-media approaches to course support. Id. at 13. The Code goes on to identify several reasons why electronic reserves can be considered fair uses: (1) This form of course support occurs in a nonprofit educational environment; (2) It is a form of noncommercial “space-shifting;”11 and                                                                                                                 11 In a recent decision involving streaming video of libraries’ lawfully-owned DVDs to students enrolled in relevant courses, a district court found “compelling” the analogy to “time shifting” of television programs blessed in Sony Ass’n for Info. and Media Equip. v. Regents, 2012 WL 7683452, *6 (C.D. Cal. Nov. 20, 2012) (citing Sony Corp., 464 U.S. 417).   8 (3) It is often transformative, as discussed below in greater detail. After concluding that it is fair use to make appropriately tailored courserelated content available to enrolled students via digital networks, the Code lists steps that institutions should take to ensure the strongest possible fair use argument, each of which is reflected in GSU’s policy: • Closer scrutiny should be applied to uses of content created and marketed primarily for use in courses such as the one at issue (e.g., a textbook, workbook, or anthology designed for the course).12 • The availability of materials should be coextensive with the duration of the course or other time-limited use (e.g., a research project) for which they have been made available at an instructor’s direction.13 • Only eligible students and other qualified persons (e.g., professors’ graduate assistants) should have access to materials.14 • Materials should be made available only when, and only to the extent that, there is a clear articulable nexus between the                                                                                                                 12 University System of Georgia, Fair Use Checklist 2, http://www.usg.edu/images/copyright_docs/fair_use_checklist.pdf (On fair use checklist, “Consumable work (workbook, test)” weighs second factor against fair use). 13 “Access should be terminated as soon as the student has completed the course . . .. Library reserves staff should delete materials available on electronic reserves at the conclusion of each semester.” “Additional Guidelines for Electronic Reserves,” USG Copyright Policy (Oct. 21, 2009), http://www.usg.edu/copyright/additional_guidelines_for_electronic_reserves. 14 “Access to course materials on electronic reserves should be restricted by password to students and instructors enrolled in and responsible for the course.” Id.   9 instructor’s pedagogical purpose and the kind and amount of content involved.15 • Libraries should provide instructors with useful information about the nature and the scope of fair use, in order to help them make informed requests.16 • Students should also be given information about their rights and responsibilities regarding their own use of course materials.17 • Full attribution, in a form satisfactory to scholars in the field, should be provided for each work included or excerpted.18 Id. at 14. GSU has thus made every effort to ensure that its e-reserve activities fall squarely within the mainstream of practice at educational institutions in the United States, as reflected in the ARL Code.19 That practice reflects, in turn, widespread understanding of the contours of fair use for libraries and educational institutions, upon which these institutions rely to provide                                                                                                                 15 “Instructors are responsible for evaluating, on a case by case basis, whether the use of a copyrighted work on electronic reserves requires permission or qualifies as a fair use . . .. Inclusion of materials on electronic reserves will be at the request of the instructor for his or her educational needs.” Id. 16 See “Policy on the Use of Copyrighted Works in Education and Research,” USG Copyright Policy (Oct. 21, 2009), http://www.usg.edu/copyright. 17 Id. 18 “Materials made available on electronic reserves should include a citation to the original source of publication and a form of copyright notice.” Additional Guidelines. 19 Indeed, the district court found that “many schools’ copyright policies allow more liberal unlicensed copying that does Georgia State’s 2009 Copyright Policy.” Slip op. at 42.   10 services that benefit the public. A finding by this Court that GSU nonetheless violated copyright would mean upending well-established practice and expectations, thereby thwarting the public interest. Amici urge the Court not to take that step. II. The Fair Use Factors Favor a Fair Use Finding for All the Excerpts Used By GSU Appellees have explained in detail why the district court correctly concluded that fair use shelters GSU’s activities, and amici will not replicate that argument here. Rather, we focus on an incorrect assumption that underlay the district court’s finding that five of the excerpts did not qualify as fair use. The district court held, with little analysis, that the inclusion of verbatim copies of complete chapters was not “transformative” as that term is understood in the context of the first fair use factor. This mistake had a cascading impact on the analysis of the third and fourth factor. Nonetheless, we do not believe it is necessary to remand for the district court to reevaluate whether the use of these five excerpts is transformative. Rather this Court should simply clarify for future courts that the inclusion in electronic reserves of a verbatim copy of a book chapter may be transformative where the copying serves a function that is distinct from the purpose of the original work.   11 A. First Factor – Purpose and Character of the Use The district court correctly found that the first factor favored GSU because its use was for a nonprofit, educational purpose.20 However, amici disagree with the district court’s assumption that GSU’s use was not “transformative” within the meaning of fair use jurisprudence. To be sure, as the district court observed, the Supreme Court has suggested that “the straight reproduction of multiple copies for classroom distribution” would not be transformative. Campbell, 510 U.S. at 579 n.11. However, not only was that suggestion sheer dicta, it also falls well short of establishing that “mirror images of parts of the books,” slip op. at 49, are never transformative. In fact, courts in several circuits have treated as transformative ‘mirror image’ copies where the purpose of the copies differed from that of the original. A.V. v. iParadigms, LLC, 562 F.3d 630, 640 (4th Cir. 2009) (using student papers to create a plagiarism detection database); Perfect 10, 508 F.3d at 1165 (using website images to create a search index); Bill Graham Archives v. Dorling Kindersley Ltd., 448 F.3d 605, 609 (2d Cir. 2006) (using concert posters for historical and educational, rather than advertising and informational, purposes);                                                                                                                 20 Education is directly referenced twice in Section 107: once in the list of favored purposes (“teaching (including multiple copies for classroom use”) and once in the first factor (“nonprofit educational purposes”). Four of the other favored purposes in Section 107 are integral to the educational enterprise: “criticism,” “comment,” “scholarship,” and “research.”   12 Kelly v. Arriba Soft Corp., 336 F.3d 811, 819 (9th Cir. 2003) (using website images to create a search index). As the Ninth Circuit stated, “even making an exact copy of a work may be transformative so long as the copy serves a different function than the original work.” Perfect 10, 508 F.3d at 1164.21 Based on this line of cases, the court in Authors Guild v. HathiTrust recently found the making of digital copies of millions of books to facilitate access for print disabled individuals to be transformative. Authors Guild, Inc. v. HathiTrust, 11 CV 6351 HB, 2012 WL 4808939 (S.D.N.Y. Oct. 10, 2012). The HathiTrust court noted that publishers and authors did not treat print-disabled individuals as a significant or potential audience. “As a result,” the court continued, “the provision of access for them was not the intended use of the original work (enjoyment and use by sighted persons) and thus the use is transformative.” Many excerpts included in an electronic course reserve system could similarly serve a different function from the original work. As the ARL Code explains:                                                                                                                 21 See also Bernard Knight, USPTO General Counsel, USPTO Position on Fair Use Copies of NPL Made in Patent Examination, January 19, 2012, http://www.uspto.gov/about/offices/ogc/USPTOPositiononFairUse_of_CopiesofN PLMadeinPatentExamination.pdf (copying of complete articles was transformative because the copies were made for the purpose of documenting that certain features of an applicant’s claim were in the prior art). Thus, the “straight reproduction” in the dicta from Campbell, 510 U.S. at 579, could be understood to refer to copying without repurposing or recontextualizing, e.g., the photocopying of pages from a workbook.   13 Most of the information objects made available to students, in whatever format, are not originally intended for educational use. For example, works intended for consumption as popular entertainment present a case for transformative repurposing when an instructor uses them (or excerpts from them) as the objects of commentary and criticism, or for purposes of illustration. ARL Code at 13. The same argument can be made with respect to scholarly monographs. Scholars write book-length works on narrow topics for the small market of other specialists in the field. These monographs have limited print runs – often as few as three or four hundred copies. Students typically are not part of the audience for these works.22 The district court acknowledged that neither party sought at trial to establish “the target market for the particular books that are involved in this case.” Slip Op. at 22 n.15. Nonetheless, the district court proceeded to draw “inferences” from the books themselves. Id. The district court inferred that the “target market” for Sage’s                                                                                                                 22 The director of Rutgers University Press has said that, by definition, “a monograph is a book intended for specialists that has no significant course adoption potential at the undergrad level, and that about 200 libraries will buy.” Marlie Wasserman, Reprint: How Much Does It Cost to Publish A Monograph and Why?, 4 J. Electronic Pub., no. 1, 1998, available at http://quod.lib.umich.edu/cgi/t/text/textidx?c=jep;view=text;rgn=main;idno=3336451.0004.104. Amici Authors Guild, et al., cite Wasserman’s presentation for the proposition that increased licensing revenue is necessary to support publication of scholarly monographs (and, thereby, to support tenure), but Wasserman’s concern necessarily is limited to books published with no expectation of income from course adoption, i.e., books whose primary purpose (and market) is to contribute to the scholarly discourse, not to facilitate education of students.   14 books was “educators who teach upper level undergraduate and graduate students in colleges and universities and, derivatively, their students.” Id. Further, the district court speculated that Oxford’s and Cambridge’s books “probably” were marketed to professors “and derivatively their students,” as well as the broader academic community. Id. Amici question the reasonableness of the district court’s inferences, particularly treating students “derivatively” as part of the target professorial market. Nonetheless, even if this Court decides to accept the district court’s inferences here concerning the target market, it should make clear that future courts should make their own assessments of the target or likely audience and purpose, rather than applying an erroneous blanket rule that “mirror image” copying is per se non-transformative. Such clarity will ensure that future courts will be free to determine that such copying, in appropriate circumstances, may be transformative.23                                                                                                                 23 Appellants themselves provide yet another basis for treating GSU’s use of excerpts as transformative. Appellants argue repeatedly that GSU compiled the excerpts into “digital anthologies” and “digital course reading compilations.” Appellants Brief at 5. Assuming arguendo that professors at GSU did create digital compilations, the assembly of excerpts of preexisting works into a digital compilation can be a quintessentially transformative use. Indeed, the Copyright Act contains specific provisions defining the scope of protection in compilations, acknowledging the originality in selection and arrangement of constituent excerpts. See 17 U.S.C. § 101; 17 U.S.C. § 103. The Librarian of Congress has specifically recognized that creating “compilations of portions of [audiovisual     15 B. Second Factor – Nature of the Work The district court correctly found that the second statutory fair use factor, the nature of the copyrighted work, tilted towards fair use: “the books involved in this case are properly classified as informational in nature, within the spectrum of factual materials and hence favoring fair use.” Id. at 52. The scholarly nature of the books actually weighs this factor even more strongly in favor of GSU than indicated by the district court. Judge Richard Posner of the Seventh Circuit, a prolific author on intellectual property matters, recently wrote a blog post arguing that scholarly works require little to no copyright protection from a policy perspective. Judge Posner suggested that “modern action movies often costing hundreds of millions of dollars to make, yet copiable almost instantaneously and able to be both copied and distributed almost costlessly,” may require strong copyright protection to ensure their creation. At the other end of the spectrum, Judge Posner observed, are:                                                                                                                                                                                                                                                                                                                                                                     works from a university library’s collection] for educational use” is a fair use. Exemption to Prohibition on Circumvention of Copyright Protection Systems for Access Control Technologies, 71 FR 68472 (2006). Perhaps not all compilations qualify as transformative fair uses, but carefully selected educational compilations certainly do.   16 academic books and articles (apart from textbooks), which are produced as a byproduct of academic research that the author must conduct in order to preserve his professional reputation and that would continue to be produced even if not copyrightable at all. It is doubtful that there is any social benefit to the copyrighting of academic work other than textbooks . . .. Richard Posner, Do patent and copyright law restrict competition and creativity excessively?, The Becker-Posner Blog (Sept. 30, 2012), http://www.becker-posnerblog.com/2012/09/do-patent-and-copyright-law-restrict-competition-andcreativity-excessively-posner.html. Amici do not suggest that scholarly works should receive no copyright protection.24 But they do agree with Judge Posner that academic authors do not need the economic incentive afforded by copyright to motivate them to write scholarly works.25 Because scholarly works require “thinner” copyright protection                                                                                                                 24 The district court rightly distinguished between the threshold question of copyrightability and the further question of whether the second factor should favor fair use. Appellants and their amici often confuse the two, suggesting that an adverse finding under the second factor amounts to wholesale denial of copyright protection. 25 See also slip op. at 82 (“[A]cademic authors as a group value publication as an enhancement to professional reputation and achievement and . . . as a contribution to academic knowledge. [ . . .] There is no reason to believe that allowing unpaid, nonprofit academic use of small excerpts in controlled circumstances would diminish creation of academic works.”) The “publish or perish” system of advancement in higher education provides academics with ample incentive to create scholarly works.   17 to ensure their production, the second factor strongly favors the fair use of scholarly works.26 At the same time, Judge Posner properly distinguished between textbooks and other forms of academic writing. Academic authors require copyright’s economic incentive to produce textbooks because textbooks “require a lot of work and generally do not enhance the author’s academic reputation and may undermine it.” Id. The district court drew the same distinction between textbooks and other scholarly writings – as does the ARL Code of Best Practices. C. Third Factor – Amount Used The district court held that “[f]actor three requires consideration of both the quantity and the value of the amount taken in relation to the overall book.” Slip op. at 55. Amici agree with many aspects of the district court’s factor three analysis. The district court correctly found that the 1976 Classroom Guidelines, “which would preclude a use from falling within the safe harbor solely on the basis of the number of words copied, is not compatible with the language and intent of § 107.”                                                                                                                 26 See also Robert Kasunic, Is That All There Is? Reflections on the Nature of the Second Fair Use Factor, 31 COLUM. J. L. & ARTS 529, 565 (2008). Appellants’ amici complain that the district court did not appreciate the creativity and originality reflected in academic writings. Amici, however, confuse originality of expression, which warrants copyright protection, with originality of ideas, which does not. See, e.g., AAUP Brief at 4 (“scholars’ complete—and complex— intellectual analysis”); Authors Guild Brief (“interpretive insights”); Copyright Alliance Brief (“analysis of theory and practice”).   18 Id. at 59. The district court properly determined that for purposes of the factor three analysis, the work at issue was the entire book, rather than a chapter within the book, even when each chapter has a different author. The district court perceptively understood that selection of a whole chapter of a book “likely will serve a more valuable educational purpose than an excerpt containing a few isolated paragraphs.” Id. at 68. Further, the district court recognized that prohibiting a professor from using the same chapter from one semester to the next was “an impractical, unnecessary limitation.” Id. at 71. However, amici disagree with the district court’s conclusion that the third factor favors fair use only if the amount copied is the lesser of one chapter or 10% of a book. While the court provided ample reason for allowing the use of chapterlength excerpts rather than isolated paragraphs, it did not explain why the excerpts should be restricted to one chapter. It simply stated that “excerpts which fall within these limits are decidedly small, and allowable as such under factor three.” Id. at 88. The “decidedly small” standard appears to derive from the district court’s earlier assumption that the use here was not transformative: “Taking into account the fact that this case involves only mirror-image, nontransformative uses, the amount must be decidedly small to qualify as fair use.” Id. at 65. But, as demonstrated above, many of the uses may well have been transformative, to the extent that excerpting portions of the work for use in instruction was not an   19 intended use of the work. And if the use is transformative, the amount used need not be “decidedly small.” Rather, the user may copy the amount appropriate to achieve her transformative purpose. See Bill Graham Archives, 448 F.3d at 613. D. Fourth Factor – Effect Upon Market The district court’s assumption that the uses at issue were nontransformative also adversely affected its analysis of the fourth fair use factor. For purposes of the fourth factor, courts have found that when a use is transformative, a copyright owner “does not suffer market harm due to the loss of license fees.” Id. at 615. This is because a copyright owner may not preempt a transformative market. Id. at 614. By assuming that inclusion of excerpts in electronic reserves was never transformative, the district court incorrectly gave undue weight to a publisher’s theoretical loss of licensing fees in the rare case that it offered a digital license for those excerpts. While the availability of a license need not be determinative (especially where the use is transformative), the district court did not err in considering the absence of an available license relevant to the fourth factor analysis. The record below established that obtaining a license for use of digital excerpts is often difficult or impossible, sometimes by design.27 The gaps in blanket licenses like the                                                                                                                 27 See Slip op. at 28-29, 77 (noting that “sometimes publishers, for whatever reason, simply prefer limiting sales to the whole book,” that “Cambridge did not and does     20 one offered by the Copyright Clearance Center are also well known in the library community,28 and the difficulty of obtaining a la carte licenses for educational uses is well documented.29 Because educational uses serve the public interest, and “[t]he ultimate test of fair use is whether the copyright law’s goal of promoting the Progress of Science and useful Arts would be better served by allowing the use than by preventing it,” Bill Graham Archives, 448 F.3d at 608, it makes sense to weigh the fourth factor in favor of fairness where the rights holder has made little effort to serve the relevant market effectively. To the extent that a court does consider the impact of a use on potential licensing revenues, it should examine not only the existence of an effective mechanism for licensing the works, but also whether the license is a likely option. See Ass’n for Info. Media and Equip. v. Regents of the Univ. of California,                                                                                                                                                                                                                                                                                                                                                                     not participate” in the CCC excerpt licensing program, and that “the record affirmatively shows that Cambridge has been quite skeptical of granting licenses to create digital excerpts of its works.”). 28 See, e.g., Peter Hirtle, Why You Might Want to Avoid the CCC’s Annual License, LibraryLaw Blog (July 5, 2007), http://blog.librarylaw.com/librarylaw/2007/07/why-you-might-w.html (“The bookstore manager at Cornell has told me that CCC can provide fewer than 50% of the permissions he needs for course packs, and the annual license covers only a subset of CCC publishers.”). 29 See, e.g., J. Christopher Holobar & Andrew Marshall, E-Reserves Permissions and the Copyright Clearance Center: Process, Efficiency, and Cost, 11 PORTAL: LIBRARIES AND THE ACADEMY 517, 520 available at https://scholarsphere.psu.edu/downloads/9k41zd523 (64 percent overall success rate seeking a la carte permissions from CCC, but only 45 percent of permissions granted quickly).   21 No. 2:10-CV-09378-CBM, 2012 WL 7683452, at *6 (C.D. Cal. Nov. 20, 2012) (fourth factor weighs in favor of finding fair use because a student “is no more likely to purchase a DVD” if she could not stream the work on her computer). Evidence entered at trial suggests that licensing was not a likely option at GSU. See Appellees Br. at 22 (“Many professors testified that they would not have used any excerpt if students were required to pay a licensing fee”). Instead, a professor (with the able assistance of a research librarian) would find substitutes, such as material distributed under a Creative Commons or other open license, or articles in journals the university already licenses. Alternatively, since the excerpts are supplemental reading, the professor might leave the excerpts out of electronic reserves altogether, and just place a few photocopies on physical reserve in the library.30 III. Placing Licensing in Context: Library Budgets and the Public Interest In practical terms, there is no real licensing market for including these kinds of excerpts in electronic reserves. Appellants and their amici speculate that such a                                                                                                                 30 Further underscoring the small likelihood of GSU paying license fees for the use of these excerpts is their low utilization rate. In Fall 2009, the 1,000 excerpts posted on GSU’s electronic course system were accessed a total of 4,000 times. Appellants Brief at 23. Each excerpt was accessed just four times although the average class size is 38 students. Georgia State University Admissions, http://www.gsu.edu/admissions/at_a_glance.html (last visited Apr. 18, 2013). Thus, on average, barely ten percent of the students in any given class bothered to look at a given excerpt even once.   22 market could exist but for the decision below. However, any attempt to create such a market could only succeed, if at all, at the expense of the public interest. Academic libraries simply do not have the budget to participate in any “new” licensing market.31 Their only alternative is to divert scarce funds from some other area. Thus, Appellants are effectively asking this Court to require libraries to reorganize their budget priorities in order to benefit some rights holders at the expense of others – or simply decline to continue including excerpts in e-reserves. Amici suspect most libraries will be forced to take the latter path. Like the rest of the economy, research library budgets have contracted since the economic crisis of 2008. Sara Hebel, State Cuts Are Pushing Public Colleges into Peril, CHRON. OF HIGHER EDUC. (Mar. 14, 2010). Moreover, it is unlikely that many library budgets, particularly those in state institutions, will recover any time soon. Id. Funding per public full-time equivalent student dropped by 26.1 percent from 1990–1991 to 2009–2010. John Quinterno & Viany Orozco, The Great Cost Shift: How Higher Education Costs Undermine the Future Middle Class, Demos (Apr. 3, 2012). Total fiscal support for higher education declined by another 7.6 percent from fiscal year 2011 to fiscal year 2012, with cuts occurring in 41 states. Caralee Adams, State Funding for Higher Ed. Drops by 7.6 Percent in a Year, Education                                                                                                                 31 Unlike with course packs, where the cost of license fees can be passed on to the student purchasing the course pack, the library would have to absorb the cost of ereserve licenses because students have free access to e-reserves.   23 Week (June 27, 2012). Congress has made it clear that it will not increase its support for research universities to make up for the decline of state funding. Paul Basken, Lawmakers Offer Struggling Research Universities Sympathy, Not Cash, CHRON. OF HIGHER EDUC. (June 27, 2012). At the same time, other library costs have increased dramatically. For example, between 1986 and 2010, research library expenditures on academic journals increased 379 percent, more than double the rate of overall library expenditures. See Association of Research Libraries, Expenditure Trends in ARL Libraries, 1986-2010. In 1986, journal subscriptions represented 16.8 percent of median research library expenditures; by 2010, journal subscriptions grew to 31.1 percent of expenditures. Id. At one Big Ten university, “if the average changes in library budgets were compared to the average increase in serial costs from the years 2001-2005, the entire library budget would be consumed by journal costs by the year 2014.” Glenn S. McGuigan & Robert D. Russell, The Business of Academic Publishing: A Strategic Analysis of the Academic Journal Publishing Industry and its Impact on the Future of Scholarly Publishing, 9 ELECTRONIC J. ACAD. SPECIAL LIBRARIANSHIP 3 (2008). And, of course, journal costs are only the tip of the iceberg that is threatening library budgets. In short, libraries do not have the resources to pay additional license fees for the “right” to include excerpts in electronic reserves.   24 Moreover, even assuming arguendo that libraries could pay such fees, requiring this would thwart the purpose of copyright by undermining the overall market for scholarship. Given libraries’ stagnant or shrinking budgets, any new spending for licenses must be reallocated from existing expenditures, and the most likely source of reallocated funds is the budget for collections. As one librarian pointed out at a meeting of the American Association of University Presses, “[W]e pay six figures each year to CCC, and that money is reallocated from our collections budget . . .. So that’s new content we’re not buying.”32 An excerpt license requirement thus will harm the market for new scholarly works and works by new scholars, as the works assigned for student reading are likely to be more established pieces written by well-known academics.33 Libraries’ total investment in scholarship will be the same but resources will be diverted away from new works to redundant payments for existing ones, in direct contradiction of copyright’s purpose of “promot[ing] progress.” U.S. CONST. Art. I, § 8, cl. 8.                                                                                                                 32 Steve Kolowich, Mending Fences, Inside Higher Ed (June 21, 2012), http://www.insidehighered.com/news/2012/06/21/university-presses-debate-howreconcile-libraries-wake-georgia-state-copyright#ixzz1yQpFoFhe. 33 The impact on young scholars seeking tenure, then, will be exactly the opposite of what Amici Authors Guild, et al., suggest. Authors Guild Br. 3, 12-13. Notably, membership in the Guild is limited to book authors who earn royalties and a “meaningful advance,” as well as retaining copyright in their books, three conditions that often do not apply to academic authors. Guild Membership Eligibility, http://www.authorsguild.org/members/guild-membership-eligibility/. This may explain the Guild’s lack of familiarity with the academic publishing landscape.   25 The district court rejected as “glib” Appellants’ claims that lost licensing would put them out of business, and as “speculative” and “unpersuasive on this record” the claim that lost revenue would reduce their scholarly output. Slip op. at 84-85. Amici’s broader doomsday predictions are even less credible. In fact, the scholarly communications market is undergoing a renaissance that is enabling more publications to disseminate more scholarly writings to more students and experts than ever before. This renaissance is based on open access publishing. Historically, publishers of scholarly communications performed critical and costly functions: coordination of the peer-review process, and the printing, marketing, and distribution of the copies of the journals or monographs.34 The publishers needed strong copyright protection to ensure that they would recover their investment in the production and distribution of the copies, even though they received the content itself at no cost from the academic authors. The Internet has dramatically changed the economics of scholarly communications. Email and software have reduced the cost of coordinating the peer-review process, and the Internet has cut printing and distribution costs. These reduced costs have enabled the emergence of open access business models, where                                                                                                                 34 Although publishers coordinate the peer-review process, they do not pay the peer reviewers. Members of the academic community donate their time to peer-review activities as part of their contribution to the scholarly enterprise.   26 readers can obtain online access to the writings for free. Given the restrictive licensing terms and conditions and the skyrocketing cost of science, technology, and medical journals discussed above, researchers and scientists are highly motivated to embrace these new models. Additionally, scholars are attracted to the functionality open access models permit, including the linking of databases and journal literature, and the mining and manipulation of these resources. An academic author typically grants the open access publisher a nonexclusive copyright license to distribute the writing to the public at no charge. The open access publisher covers its costs by charging the author a fee for publishing the article or monograph or by receiving funding from another source, such as a granting agency or the institution that hosts the publication.35 Over the past fifteen years, the number of open access publishers has increased dramatically, as has the number of materials they have published. Since 2000, the members of the Open Access Scholarly Publications Association (OASPA) have published over 250,000 articles under open licenses, including over 80,000 in 2012 alone. Claire Redhead, Growth in the use of the CC-BY license (Mar. 8, 2013), http://oaspa.org/growth-in-use-of-the-cc-by-license-2/. Over 20%                                                                                                                 35 Many granting agencies now include extra funds in grant awards to cover the cost of publication in an open access format. And unlike educational funding in general, state and federal funding for the creation of open educational resources has increased.   27 of all peer-reviewed articles are now published in the more than 4700 open access journals. M. Laakso, et al., The Development of Open Access Journal Publishing from 1993 to 2009, PLoS ONE 6 (2011). The Directory of Open Access Books, created in 2012, already lists 1,271 academic peer-reviewed books from 35 publishers. Directory of Open Access Books, http://www.doabooks.org/ (last visited Apr. 19, 2013). The demand for open access publishing among academic authors and readers is so strong that even highly profitable publishers such as Appellants Oxford and SAGE have open access publications and are members of OASPA. Placed in this context, it is clear that the public benefit of permitting activities such as GSU’s far outweighs any potential cost to publishers. Although some academic publishers may have difficulty adjusting to the digital environment, amici’s predictions of the devastating impact the decision below would have on the evolving scholarly communications ecosystem are complete fiction.   28 CONCLUSION For the foregoing reasons, Amici urge this Court to affirm the decision below. At the same time, this Court should clarify that the inclusion of excerpts in an electronic reserves system could be a transformative use. Dated: April 25, 2013 Respectfully submitted, s/ Mitchell Stoltz Mitchell Stoltz ELECTRONIC FRONTIER FOUNDATION 454 Shotwell Street San Francisco, CA 94110 Tel: (415) 436-9333 Email: mitch@eff.org Jonathan Band (on the brief) JONATHAN BAND PLLC 21 Dupont Circle NW, 8th Floor Washington, D.C., 20036 Tel: (202) 296-5675 Email: jband@policybandwidth.com Counsel for Amici Curiae   29 CERTIFICATE OF COMPLIANCE 1. This brief complies with the type-volume limitations of Fed. R. App. P. 32(a)(7)(B) because it contains 6,931 words, excluding the parts of the brief exempted by Fed. R. App. P. 32(a)(7)(B)(iii). 2. This brief complies with the typeface requirements of Fed. R. App. P. 32(a)(5) and the types style requirements of Fed. R. App. P. 32(a)(6) because it has been prepared in a proportionally spaced typeface using Microsoft Word in 14 point Times New Roman. s/ Mitchell Stoltz Mitchell Stoltz ELECTRONIC FRONTIER FOUNDATION   30 CERTIFICATE OF SERVICE I hereby certify, that on April 25, 2013, a true and correct copy of the foregoing Brief of Amici Curiae American Library Association, Association of College and Research Libraries, and Association of Research Libraries was timely filed in accordance with FRAP 25(a)(2)(D) and served on all counsel of record via CM/ECF pursuant to Local Rule 25.1(h). Dated: April 25, 2013   s/ Mitchell Stoltz Mitchell Stoltz ELECTRONIC FRONTIER FOUNDATION 31