What If We Asked the Librarians? Or, How The Librarians' Code Is Different

Today, with help from our partners at the Center for Social Media at American University, and the Law School at AU, and with support from a grant from The Andrew W. Mellon Foundation, ARL is proud to unveil the Code of Best Practices in Fair Use for Academic and Research Libraries. Based on 36 hours of focus group deliberation with 90 academic and research librarians representing 64 institutions in meetings held all over the country, the Code is comprised of eight Principles that describe general circumstances where the groups found library uses to be fair, followed by Limitations that describe the outer bounds of the consensus and Enhancements that the groups thought represented salutary but not necessary steps to protect the interests of other stakeholders.

Topics covered include:

  • Electronic reserves and course management systems
  • Preservation
  • Mass digitization for “non-consumptive” uses
  • Facilitating access for the disabled
  • Creating digital exhibits of library collections materials

That’s the overview, and you can learn more at our Code web page. In this blog post, though, I want to focus on what I think is the first question that most research and academic librarians are going to ask when they hear about this Code: Oh, no, not another set of guidelines! How is this thing any different than all the other stuff out there?

External image

Like a lot of documents, websites and resources out there, our Code tries to apply the plain statutory text and go beyond the facts of existing cases to talk about fair use in a way that feels closer to home for academic and research librarians. So why should you care about this new document? How is it different from all the other resources you can find online? And how does it fit into your institution’s policymaking process?

Any librarian who has waded into the troubled waters of copyright has encountered a dizzying array of guidelines, rules of thumb, nutshells, and checklists. As Columbia University copyright guru Kenny Crews has pointed out in his definitive study of fair use ‘guidelines’, the dangers that attend some of these tools far outweigh the promised benefit of greater certainty. Examples that were supposed to represent minimum “safe harbors” came to be treated, all too often, as the outer limits of legitimate action. Private voluntary standards negotiated among diverse (and even adverse) groups have gotten a patina of legal authority over time and are substituted for a flexible and dynamic application of the actual law.

We set out to avoid these pitfalls by providing a new kind of resource that is more narrowly targeted and more flexible than the tools that have come before it. And because judges care what communities of practice think and do, the librarians’ Code can even help improve the law by revealing and shoring up the values of academic and research librarians. So the two things that make this Code unique are where it comes from, and how it works.

The Code is different because of where it comes from. As I said at the beginning, it represents a consensus of a large number of academic and research librarians, deliberating deeply together about their own problems. This advice does not come from outside experts or consultants at ARL speaking to you from our ivory towers in Washington, nor does it come from a company or trade association with a financial interest in discouraging fair use. It comes from your peers. And this consensus isn’t a “lowest common denominator” based on the status quo or a survey of which practices are most common. It’s based on what the academic and research librarians we spoke with came to consensus on when they had the time and opportunity to deliberate together about which practices are truly “best.”

The Code is also different from many of its predecessors because of how it works: it gives academic and research librarians and library policymakers flexible principles to guide their thinking in their particular circumstances, rather than one-size-fits-all rules that ignore context. Librarians using the Code will come to different conclusions in seemingly similar cases depending on the nature of the works they are using, the goals of the users they are serving, the ways that works are distributed or copied, and on and on. In short, the Code doesn’t provide answers; it provides tools that help librarians ask the right questions.

The Code does not purport to tell everyone how to apply fair use across all situations for all time. Instead, the Code focuses on just eight concrete, recurrent scenarios where academic and research librarians are employing fair use right now. The broad cross-section of the academic and research library community that came together to work on this Code agreed that new situations may arise where fair use could come in handy for libraries, and the community will have to confront those situations afresh. They also agreed that practice in the situations they discussed may evolve over time, and no one should be bound forever by current ideas. The Code is a snapshot, then: it captures the considered opinions of academic and research librarians at this moment about these issues, rather than purporting to speak to all issues or for all time.

The Code is not intended to short circuit or preempt the legal and risk management analysis that each institution will need to undertake in deciding on its own fair use policies. While it has a solid grounding in fair use case law, the Code is not a legal memo. It was not created by lawyers (though there were lawyers involved, of course—most importantly, the panel of outside legal reviewers who ensured that the Code principles are reasonable readings of fair use), and it certainly does not pretend to be a disinterested analysis of fair use case law. No such analysis could give definitive answers to the challenges libraries face, because the case law simply is not determinative, and is not likely to become so any time soon. (Even if the Georgia State case is decided tomorrow, that will just be the start of a long process: there may be an appeal, there can always be differences across judicial circuits, and libraries with significantly different policies may get little guidance from the GSU experience.) Although it is increasingly clear that courts look to the issue of “transformative use” in deciding these cases, reasonable people can disagree about how the general principles of fair use that emerge from the cases should apply to particular situations.

And yet, there are tremendous opportunities for academic and research libraries who have the courage to act in the face of this inevitable uncertainty. The Code is designed to be one input into the risk management process, but that process has other inputs, too. General counsels’ offices will come with their own views about the law (hopefully they’ll check out some recent scholarship that has helped shape the thinking behind the Code), and administrators will have their own ideas about which projects are important enough to warrant rational risk-taking, and which are less compelling. Knowing the consensus of academic and research librarians from institutions all over the country around fair use in some key areas will help this process, but it cannot and should not override local conditions and priorities.

To tweak an old cliché, all policy is local, and that is as it should be. The Code of Best Practices in Fair Use for Academic and Research Libraries is here to help, and not to dictate, that local policymaking process.

What has developed in the content industries is a sense that copyright exists to support their businesses, so any new way they find to extract a little extra money from the rights they hold should be endorsed and protected by the courts. If you start from that premise, it makes sense to sue libraries for providing digital copies to blind people and professors for giving students access to short excerpts from a scholarly book because you believe you are acting from within the core purpose of copyright. But the premise is wrong.
—  Kevin Smith, in today’s LJ article Why Are Some Publishers So Wrong About Fair Use?
Judge Baer channels #librarianscode

In what can only be described as a total victory for libraries, Judge Harold Baer of the Southern District of New York held in an opinion published yesterday that the HathiTrust’s mass digitization project is protected fair use.

This isn’t news to academic and research librarians, who spoke loud and clear in the Code of Best Practices in Fair Use for Academic and Research Libraries. Judge Baer’s opinion should sound delightfully familiar to anyone who’s read Principles 3, 5, and 7 of the Code, which describe the consensus of academic and research librarians around preservation, accessibility, and non-consumptive uses (like search and text mining). Like the librarians, Judge Baer recognizes that these activities are “transformative,” especially the search and accessibility aspects. (For preservation he refers to the Sony decision, suggesting that non-commercial copying of this kind should be favored under the first factor.) He also recognizes that fair use generally favors these library uses as hugely valuable to the public, and particularly to “progress in science and the useful arts.” In short, Judge Baer fundamentally gets the bottom line assumption that also underlines the Code - that fair use can and does provide space for bold action by libraries in service of their public service mission, because that mission is itself in service of the same goals as copyright.

There’s a lot more to say and to celebrate in this opinion, and I’ll certainly be writing more here and elsewhere, but for now I just wanted to point out this wonderful affirmation of the logic of the Code.

I cannot imagine a definition of fair use that would not encompass the transformative uses made by Defendants’ MDP and would require that I terminate this invaluable contribution to the progress of science and cultivation of the arts that at the same time effectuates the ideals espoused by the ADA.
—  Judge Harold Baer, in his wonderful opinion in the HathiTrust case.
In fact, just a few hours ago I had the opportunity in my job to use the Code—principle two, to be precise—and it was truly a liberating feeling to know that the choice a colleague and I made (a choice we normally would have at best delayed making and at worst, and most likely, never made at all) had been informed by the best practice of our colleagues across the country and the lawyers who helped shape and vet this document.
—  When it comes to validation of all our work on the #librarianscode, it doesn’t get much better than this write up from Josh at Northwestern.
Should Libraries Fret Over Mischievous Users?

In the process of developing the Code of Best Practices in Fair use for Academic and Research Libraries, there was one worry that we encountered over and over again, and that may be doing more damage to library practice than any other myth, concern, or misapprehension: the fear of library liability for nefarious uses of copyrighted library collections material by library users.

External image

The level of fear, uncertainty, and doubt about this issue is way, way out of proportion to the actual risk of liability for any library or librarian. Put simply, it is almost impossible that you or your library, when acting on the basis of your good faith beliefs about fair use, could be held responsible for the bad acts of your patrons who abuse the access you provide. And yet every day, in libraries all around the country, decisions are being influenced by fear of liability for users’ bad behavior. It’s time to nip that fear in the bud.

I’ve written this up as a simple FAQ on the Code FAQ for Librarians page, if you want the short version. Here’s the longer version (which includes the FAQ text):

Are libraries who make good faith fair uses responsible for what their users do with works provided in the context of that fair use?

Almost certainly not. For example, what if a library makes a good faith judgment (based on Principle Two of the Code) that displaying an image as part of an online exhibit is fair use, and a user of the website downloads the image and prints it for sale on t-shirts?

In that case, only the user is responsible, not the library. It is always the responsibility of the user to make her own decisions about the limits of her own rights. In the course of making their own fair uses, libraries make content accessible to users who are then free to re-use it in other ways. And libraries can and do provide useful information about copyright to their users; indeed, the Code reflects in several places a consensus in the academic and research library community that such efforts are worthwhile and important. But neither providing content for fair use purposes nor providing advice in good faith puts the library at risk of liability for what users do with that content or that information. The responsibility for infringement lies with the user, not the library.

You may notice that in several places the Librarians Code recommends that libraries limit access to materials, or use limited amounts of copyrighted works, or (as an enhancement) use technological protection measures, depending on the purpose of their proposed fair use. Principle One, for example, includes limitations that access to materials on e-reserve be limited to eligible students and to the duration of the course. While some libraries use these strategies out of fear that they will be responsible for any third-party abuse of access to materials, that is not the justification for the limitations and enhancements in the Code. Instead, they are rooted in the academic and research librarians’ consensus that these strategies help to tailor the library’s use so that it is appropriate to a legitimate fair use purpose. Based on both the statute and the cases that have been decided, librarians know that the amount of the work they use (which includes the breadth of access they facilitate) should be tailored in this way. So, access to an e-reserve system is limited to currently enrolled students because it is only those students whose access is justified by the professor’s legitimate fair use purpose. The professor is not teaching the class to everyone on the Internet, so the Code suggests that providing unfettered access to any Internet user goes beyond the latitude that fair use provides him for teaching. Technological measures, including lower-resolution image files or streaming rather than downloadable media, serve a similar purpose. The core concern is not that users will turn around and abuse their privileges, but that the library itself provides a level of access that is justified by its fair use purpose. There is also another, related reason for these admonitions. Although most copyright lawyers agree that, technically, a user’s “good faith” shouldn’t be a consideration in fair use analysis, there doesn’t seem to be any downside to making reasonable efforts, above and beyond what the law requires, to assure that material is used appropriately.

External image

Is anyone ever responsible for the infringing acts of third parties?

Yes, in certain narrow cases. Although the Copyright Act doesn’t explicitly create secondary liability (the technical term for holding one person responsible for the acts of another), courts have recognized at least two situations where one person can be punished for the infringing acts of another:

  • If you control the activities of an infringer, and you receive a direct financial benefit from their infringement, you can be held liable for that activity. This is called “vicarious liability,” and generally involves liability by employers for the infringements of their employees. In cases of downstream use of library materials, the library cannot control the actions of its patrons, nor does it receive a direct financial benefit from the infringement. In our example, the library can’t stop the student from making t-shirts, and it doesn’t share in the profits. So, no vicarious liability.

  • If you know that someone is infringing copyright, and you induce, cause, or contribute materially to their infringement, you can be held contributorily liable for that activity. Libraries that provide access to materials for a fair use purpose do not know that their users will abuse that material, and certainly do not induce or cause them to infringe. In our example, the library will have no way of knowing that the user is selling t-shirts, and it certainly did not induce or contribute materially to the infringement. Merely providing the material that an infringer uses, especially when you provide that material in the context of your own legitimate use, does not rise to the level of inducing infringement.

The Supreme Court addressed the question of secondary liability for copyright infringement in the Grokster case, which found the file-sharing network liable for the infringements of its users. In that case, the Supreme Court said that the movie studios had to show that Grokster acted “with the object of promoting its use to infringe copyright, as shown by clear expression or other affirmative steps taken to foster infringement.” Grokster showed its intent to promote infringement by advertising designed to attract users interested in sharing copyrighted music. Libraries are obviously not engaged in any such activity.

The limits of liability for libraries can be seen in the preliminary decisions that we have seen in the lawsuit against Georgia State over its ereserves and course management systems. Way back in September 2010, Judge Evans dismissed the vicarious liability claim, saying that even if the reserves and CMS system were infringing, there is no evidence that GSU profited directly from their use. Then, in May 2011, Judge Evans dismissed the publishers’ contributory liability claim. While the one sentence dismissal makes it difficult to know the judge’s thinking, it’s clear from her earlier decisions that she was impressed by the GSU library’s good faith efforts to educate users about fair use and to encourage responsible use of the ereserves and CMS platforms. Such good faith acts cannot amount to knowingly contributing to infringement. The only remaining theory of liability in the GSU case is respondeat superior, which says that GSU is responsible for the infringing acts of its employees done in the course of their duties, regardless of knowledge or direct financial benefit. This theory only applies to employees, so again, a library could not be held liable for the acts of its users on this theory.

New Infographic: Good News in Fair Use for Libraries

A new infographic released today tells the story of library fair use and the Code of Best Practices in a clear and compelling way. There’s an embeddable PNG for your own blogs, and there’s also a print-ready 8.5" x 11" version in case you need hardcopies to hand out at events.

External image

The ARL Code of Best Practices in Fair Use for Academic and Research Libraries has been around for a year and a half, now, and we’ve seen how it’s changing practice around the country. The Code team has been on a whirlwind tour doing in-person events and webcasts to promote the Code all over the country. Now it’s time for library fair use enthusiasts to help spread the word, and this infographic is a powerful tool that can help pique interest in the Code itself and the overall story of how fair use has evolved to become a powerful users’ right. Share the link, embed the image on your site, print some copies for your next event, and help us keep moving libraries forward on this vital issue!

Vital Stats on the Code So Far...

So it’s been a little over a month since we released the Code of Best Practices in Fair Use for Academic and Research Libraries, and I’m almost used to typing out that monster title! I’m also taking stock of how things are going, and I’m really, really excited. Take a look at these vital statistics.

In the first month since the Code came out, we have had:

  • 48,000 downloads of the Code PDF (13,000 on the first day),

  • 26,000 pageviews for the ARL code web pages (7,000 views on release day),

  • 1,500 visitors to the ARL Policy Notes blog entries on the Code

  • 1,300 people reached directly either via in-person events or live webcasts, including over

  • 680 participants in the National Association of College and University Attorneys webcast last week (at least 140 of which were member attorneys)

  • 750 print copies of the Code distributed to libraries

  • 7 major in-person events for librarians across the country: Los Angeles, Berkeley, Atlanta, Dallas, Baltimore, and Washington, DC, with standing-room-only events in many of these locations

  • tailored webcast events for library directors and for general counsels’ offices, as well as a general interest webcast event that attracted nearly 500 people on the day of the Code’s launch

We couldn’t ask for a much better reception, and we’re just getting started.

The #Librarianscode Comes to Life at UCLA, UC Berkeley, U. of Florida, Texas A&M, and Yale

Today we are very happy to debut a collection of videos with directors of ARL member libraries talking about how the Code of Best Practices in Fair Use for Academic and Research Libraries has influenced the way they and their teams approach copyright. We are so excited to see the Code bearing fruit in library policy and practice!

Go here to see the new ARL webpage that collects interviews with:

  • David Carlson, Texas A&M University
  • Susan Gibbons, Yale University
  • Tom Leonard, University Librarian at UC Berkeley
  • Judith Russell, University of Florida, and
  • Gary Strong, University Librarian at UCLA.

These videos are also collected at the ARL YouTube channel, where you’ll also find archived videos of our ARL webcasts about the Code and other hot copyright topics.

We are extremely grateful to the directors who participated in this interview project, and to Bryan Bello and Angeli Gabriel (young filmmakers working with our partners at the Center for Social Media) for their outstanding work filming and editing these clips.

We will be rolling out several more blog posts over the coming week telling some of the success stories we have heard from all over the country. We hope these will provide some inspiration and encouragement to the libraries who may be watching these developments and wondering whether and when they should take steps of their own to exercise the full extent of their fair use rights. Watch this space!

Power to the People: Five Reasons Fair Use Best Practices Are Changing the World

by guest bloggers Patricia Aufderheide, University Professor, American University School of Communication; Brandon Butler, Practitioner-in-Residence at the Glushko Samuelson IP Clinic, American University Washington College of Law; and Peter Jaszi, Professor of Law and Faculty Director of the Glushko-Samuelson Intellectual Property Clinic, American University Washington College of Law

Copyright Week is the perfect occasion to celebrate fair use, certainly the most dynamic and arguably the most important doctrine in copyright law. The last 15 or 20 years have seen a remarkable series of developments that make fair use, now more than ever, the most vital protection of the public interest in the Copyright Act. For Copyright Week, we wanted to highlight a part of the fair use landscape that, perhaps more than any other, puts fair use in the hands of practitioners who need it most: the Fair Use Best Practices movement.

With a little help from a team of researchers at American University, an ever-growing cadre of communities has identified where their work necessarily encounters copyright and the kinds of fair uses that are essential to the communities’ continued flourishing and success. Each code (read them all here) contains a short but powerful description of fair use’s broad history and meaning, followed by a set of principles that describe situations where fair use may apply accompanied by limitations that describe the outer bounds of community consensus. The effects of these documents can be dramatic. Documentary filmmakers came first, and had great success, but they’ve been joined by educators, scholars, poets, online video makers, journalists, and (most importantly for this blog) librarians. As more and more people need fair use to continue getting things done, best practices are an idea whose time has come.

So, without further ado, we give you five reasons fair use best practices are changing the world:

  1. They’re based on solid legal footing. Specifically, path-breaking research by copyright scholar Michael Madison. Madison surveyed over a century of fair use decision making and found that, over and over again, courts determining whether a use was fair inquired into the mission and values of the communities standing before them to vindicate their fair use rights. Uses firmly grounded in the socially beneficial mission of a practice community were much more likely to be blessed as fair.

    Each code starts from that insight, together with the dominant paradigm of “transformative use” that informs court decisions in fair use. The community norms developed on this foundation are then further shored up by a legal review by five independent experts from diverse backgrounds who certify that the Code represents a reasonable application of fair use law to the practice area. Practices consistent with the Documentary Filmmakers Code have been blessed by federal courts, as have practices identified as fair in the #Librarianscode. Indeed, between Georgia State and HathiTrust, the practices described in four of the eight principles in the #librarianscode have been blessed by federal courts.

  2. They clear away the crud. Anyone who engages with copyrighted material for more than a few minutes will encounter a dizzying array of so-called ‘guidelines,’ rules of thumb, ‘negotiated’ agreements, and urban myths and legends that proliferate around copyright. The goal of best practices is to identify the best approaches to recurring fair use scenarios, rather than to measure the lowest common denominator of the status quo and freeze it in amber forever. Therefore, developing best practices is an opportunity for communities to step back and question current practice in light of the latest developments in fair use law and the broadest, deepest understanding of the mission of the community. Librarians, for example, categorically rejected the arbitrary numerical limits in the 1976 Classroom Photocopying Guidelines. On reflection, they were simply impossible to justify in light of the actual needs of librarians and the contours of modern fair use law.

  3. They make the law less alien, and rights less scary. By grounding fair use choices in practices and norms that are native to a community, best practices change attitudes toward fair use. People with a Code go from a kind of grudging, fearful “compliance” with an alien copyright law imposed from above to a unified exercise of core First Amendment rights that emerges from their own values. Teachers, librarians, filmmakers, and poets who used to feel like they were acting alone in the face of an intimidating body of law come to understand that they are actually engaged collectively in legitimate, lawful acts that are normal, indeed essential, for their profession.

  4. They help you get things done. The bottom line for any group with a shared mission and goals is whether they are able to advance mission and achieve goals. Where myth, misinformation, fear, uncertainty, and doubt dominate, any number of important projects and practices can be suppressed, driven underground, or stymied altogether. Films don’t get made, or they don’t get distribution; poems aren’t written or published; works languish in archives inaccessible to remote or print-disabled researchers. Best practices are relentlessly pragmatic and mission-centered; through them, practitioners articulate fair use solutions to real, live problems. When the community takes best practices seriously, real work gets done—work that might otherwise have been inconceivable.

  5. They help you get management on board. Almost everyone has a supervisor, counselor, or other gatekeeper who decides what projects they can pursue, whether their work will see the light of day, and so on. Whether it’s a Dean, a TV producer, or a publishing agent, sooner or later you’ve got to convince someone else that what you’re doing is legit. Understandably, gatekeepers are often expected (forced, even) to play the role of “copyright cop,” saying “no” to any project that looks like it might raise an eyebrow.

    Before best practices, each practitioner would face these folks alone, often as non-lawyers, and try to convince them to take a risk based on, well, who knows what. But with best practices in hand, practitioners can go to their Deans, their publishers, their producers, whomever, and say, “What I’m doing is normal. It’s grounded in the values of my community. And it’s in line with a document that’s been vetted by experts and endorsed by leading organizations in my field.” That’s powerful stuff! No wonder the Documentary Filmmakers code has been so transformative, as has the #Librarianscode, and many many others.

So, there you have it. As Copyright Week winds down and we contemplate the copyright system we have, and the opportunities for change and improvement, we submit that fair use best practices are, by far and away, the most accessible, effective, and powerful tool in the hands of the public.

For more information, check out the full roster of best practices at the Center for Media and Social Impact and check out Pat and Peter’s book, Reclaiming Fair Use.

What does the GSU decision mean for the #librarianscode?

We’ve just added the following set of questions and answers to our FAQ for librarians, which is part of the rich package of resource pages that we’re maintaining to support users of the Code. As you’ll see, the impact of Judge Evans’ decision in the Georgia State University course reserves case on libraries following the Code should be fairly limited. The decision speaks directly to just one application of one principle out of the eight. In the narrow space where the decision and the Code overlap, they are arguably consonant in practice, though it seems Judge Evans is out of step with the library community on the theoretical question of “transformativeness.”

We hope this resource will help academic and research libraries as they determine whether and how the Georgia State course reserves decision should effect their own daily practice.

In a nutshell, what is the Georgia State decision about and how might it affect how librarians use the Code of Best Practices in Fair Use for Academic and Research Libraries?

The recent decision in Cambridge U. Press v. Becker addresses the scope of fair use when professors post excerpts from scholarly books to electronic reserves and course site platforms. The decision does not affect most of the Code, but it may have some impact on how libraries choose to apply Principle One to scholarly books. A short summary of the opinion, prepared by EDUCAUSE, is available here (3 page PDF). An extended briefing on the opinion from the Association of Research Libraries is here (8-page PDF).

How binding on libraries is the Georgia State decision? Is this the new “rule” for fair use and course reserves?

The decision does not legally bind other courts or libraries, and it does not address a wide variety of important use cases, so libraries will still need to look to the norms of their community for guidance when determining the scope of their fair use rights. Although Judge Evans’ opinion is thoughtful and meticulous, and certainly provides a helpful data point for libraries working through these issues, it does not legally bind any other library, and it may not even be the last word in this particular lawsuit. Either party may appeal, and other courts are free to diverge from what Judge Evans decided. Unlike a Supreme Court decision, the wider impact of a district court decision depends mostly on how persuasive it is to other courts and to practitioners. Libraries should take some comfort in the broad safe harbor that Judge Evans describes for using scholarly books, but they should not see the decision as settling this family of issues once and for all. Even if it were the final word on the topics that it does address, the decision does not speak directly to the use of material beyond scholarly books, and its framework is particularly inapposite to uses of non-textual materials (such as video and audio). Judge Evans’ use of chapters and percentages, and her inquiry into educational licensing markets, will be little help for libraries interested in making films or songs available online to students enrolled in relevant courses, for example.

How likely is an appeal?

It is difficult to speculate about this. The judge has not decided on what she will order GSU to do in light of her decision, instead asking the publishers to propose an order by May 31, and then giving GSU an additional 15 days to respond. After that, the judge will order her remedy, which may impact the publishers’ decision whether to appeal. It’s fairly certain that GSU will not appeal, as their policy was largely vindicated by the decision.

How does Judge Evans’ decision differ from the Code in principle?

The library community’s preferred approach to fair use eschewed bright lines of the kind that Judge Evans uses in her decision. Librarians recognized that fair use determinations are context sensitive, and avoided the temptation to use quantitative rather than qualitative limitations to describe the boundaries of fair use. The judge’s approach is likely attributable to the nature of the GSU policy, which was based on a checklist. Judge Evans framed her analysis in terms of formulating a workable checklist, which is a solution that some institutions favor, but it inevitably places an emphasis on the quantitative rather than the qualitative aspects of fair use.

Another difference in the analysis has to do with whether uses of course reserves can be “transformative.” With respect to chapters from scholarly books, Judge Evans finds that copying excerpts into course reserves is not transformative. She gives no reason for this finding, simply repeating without analysis the publishers’ claim that the use is “nontransformative” because the excerpts are “mirror images of parts of books.” The implied reasoning, here, is very strange, as every transformative use involves “mirror images” - copies! - of copyrighted work. Internet search engines copy “mirror images” of websites and display “mirror images” of parts of websites in order to facilitate search, and yet these uses have been found transformative. The essential question is the context for the copying - how the “mirror image” is being used. Judge Evans doesn’t seem to have considered, much less answered, the key question: is educational use a transformative activity? Even though she awarded the first fair use factor to GSU on other grounds, Judge Evans got it wrong by giving such short shrift to the transformativeness analysis.

Does Judge Evans’ decision mean that use of course reserves is not transformative?

No, it means that the library community as a whole–comprised of people who serve the scholarly and learning communities that need this work–has been thinking about this in a very different way than Judge Evans. Her ruling doesn’t change the fact that librarians across the nation and across a range of institutions have developed a transformative reasoning applied to this activity, and because she says very little about why she does not share their understanding, she gives the community little reason to change its view.

The library community feels strongly that many uses of course reserves are transformative. As the Description of Principle One points out, using works in an educational context that were originally conceived and marketed to a different audience presents a transformative argument of the kind that has been endorsed repeatedly by federal courts. Instructors add value and create new insights from the raw material they share with their students, precisely the kind of social benefit that fair use is meant to enable and encourage.

This seeming inconsistency may be due to the library community considering uses of all media, including popular film or music, poetry, and the like, in its deliberations; the transformative argument may be more obvious for these works, for which an educational audience is marginal at best, than for scholarly books. However, such arguments could well apply to scholarly books as well, depending on the circumstances.

Librarians can continue to apply the fair use reasoning that comports with their understandings of both the law and library community values in order to meet the needs of their patrons. They might consider how their use matches with Judge Evans’ recommendations when it comes to excerpts from scholarly books, but they should also consider whether their own values and the educational rationale of the instructor make a compelling case for fair use that exceeds the boundaries Judge Evans describes.

Does the decision differ much from the Code in practical terms? Could the Code be applied to reach essentially the same outcome in this case?

While there are very real differences between Judge Evans’ theory of fair use and the theory favored by libraries, the practical outcomes may not actually be that different in the situations at issue in this case. There is some very significant overlap between Judge Evans’ decision and Principle One.

  • Like the Code, the GSU decision emphasizes the legitimacy and centrality of the non-profit educational mission, and distinguishes libraries from for-profit coursepack printers. Judge Evans also rejects the idea that fair use somehow ‘expires’ after a single semester, saying it is legitimate to post the same excerpts from semester to semester under the same fair use rationale.

  • Judge Evans agrees with the library community that limiting access to enrolled students in the relevant course is an essential part of the fair use case for course reserves. Judge Evans recognizes that use of an entire chapter may be what is reasonably necessary for a legitimate educational purpose, and so refuses to set her third factor ceiling any lower than one chapter.

  • The court argues that factor four will not favor libraries if the publisher relies on licensing digital excerpts to educational users as a significant part of the value of the copyright, which is very similar to the Code’s advice (in the first Limitation to Principle One) that users exercise caution in posting excerpts from works that are designed specifically for use in the relevant course. Indeed, it could be said that Judge Evans’ framework is more liberal than the Code in that it allows posting up to one chapter of any work, even a textbook, whereas the Code counsels generally against using more than “a brief excerpt” of works designed for classroom use. (Alternatively, Judge Evans’ insistence that excerpts from scholarly books be “decidedly small” could be seen as analogous to the Code’s use of “brief excerpt” for putatively non-transformative uses. In that case, the opinion is entirely consistent with the Code, though we could argue about which uses qualify as transformative.)

  • Finally, although Judge Evans does include some bright lines in the context of individual fair use factors, the fact that no single factor or threshold trumps all the others is consistent with the Code’s rejection of outdated “guidelines” that give dispositive weight to numerical limitations. In sum, although there is room for multiple interpretations and applications of Principle One, it is certainly feasible that a library applying Principle One to scholarly books would arrive at roughly the same outcomes on a case-by-case basis as Judge Evans did, though for different reasons. Of course, the Judge’s concern about the harm to licensing markets might have been lessened if she had taken the transformativeness argument seriously.

Did Judge Evans know about the Code? Did she say anything about it directly?

The Code was not published until after the trial was over and Judge Evans was already well into her deliberations on this case. She did take a dismissive attitude toward evidence of current community practice around fair use, saying that in the absence of a judicial determination libraries are just “guessing” at what fair use allows. It seems that Judge Evans did not follow the general trend of inquiring into community practice, but it may also be that no one presented her with a compelling argument for why she should. Perhaps if she knew that academic and research librarians had worked carefully together through a 2-year process of identifying and solving recurrent fair use problems, she would have given their opinions more weight. In any event, the decision says nothing at all specifically about the Code.

Lessons From the Siege

At last year’s Fall Membership Meeting of the Association of Research Libraries here in Washington, D.C., the refrain was, “libraries are under siege.” Copyright litigation against libraries was proceeding on several fronts, and the rejection of the Google Books settlement had library directors worrying about the next shoe to drop.

External image

What a difference a year makes. In May the federal district court in Atlanta gave a decisive victory to Georgia State University (GSU) in the lawsuit over its electronic course reserves. (Read my summary of the holdings in the case here (PDF).) On October 3, a court in California dismissed for the second time a complaint against UCLA over streaming video. Then, on the opening night of this year’s ARL Fall Meeting, came the coup de grace: Judge Harold Baer of the Southern District of New York handed down a sweeping victory for the HathiTrust and its partner libraries, which had been sued for their mass digitization of library books. (ARL is hosting a free webcast about the case tomorrow, featuring four attorneys directly involved in the case.)

It seems that, in literally every corner of the US, courts are siding with libraries as they work in good faith to bring their collections and services into the 21st century. The siege appears to be lifting, or at least on hold pending any (misguided) appeals. This is a good time to pause with the advantage of hindsight and ask what was really going on in these cases, and what lessons we can learn from the trials.

The most important lesson for the antagonists in these cases is that it’s time to stop looking to libraries for a copyright windfall. In every one of these cases the plaintiffs were looking to profit on content that libraries had already purchased, and from library programs that did not require a license. It’s time to give up on what seems to be an obstructionist, backward-looking strategy and focus instead on adding value for libraries and their users.

These lawsuits were wasteful because the plaintiffs generally had little or nothing to offer to libraries as an acceptable substitute for the supposedly infringing uses they sued over. Every time the trial courts looked at the market realities in these cases, they found that libraries were using technology along with lawfully purchased collections in order to fill a need that markets could not, would not, or should not meet. The plaintiffs expected libraries to do all the work of making valuable new uses of collections materials, and then pay for the privilege. As Duke University Scholarly Communications Officer Kevin Smith has explained persuasively, these plaintiffs have fundamentally misunderstood the Copyright Act as a blanket entitlement to profit from any activity that involves books or videos, without regard to the consequences for society at large.

For example, information disclosed at trial in the GSU case shows that publishers often withhold licenses to use digital excerpts, partly in hopes that students might have no choice but to purchase an entire book in order to read a few relevant pages. One of the reasons many of the GSU uses were found to be fair was that the excerpts assigned were often not available for licensed use in the appropriate format. Even where licenses are available, Judge Evans observed in her decision that licensing agents like the Copyright Clearance Center (CCC) don’t sell actual content or technological services; they only sell permissions. Libraries who pay CCC’s fees still have to find and digitize material from their own collections, and mount the digital excerpts on their own platforms. The ultimate irony is that it was made clear at the GSU trial that if course reserves were to require a fee, faculty would simply stop using them. But Oxford University Press, Cambridge University Press, and SAGE (with funding from CCC and the Association of American Publishers) were still willing to sue.

Similarly, the court in the HathiTrust case found that there simply was no workable market for the mass digitization uses that the libraries had made, and that, in any event, fair use would not allow the formation of a market to extract payment for such socially valuable, transformative uses. And, again, the Authors Guild and the other plaintiffs had made no effort to enable libraries to embark on mass digitization projects. It is unlikely the plaintiffs have the ability to authorize such a project, as they represent a tiny minority of authors of library collections materials, and they may not even possess the necessary rights. (Once they get past these lawsuits over digitization, there will surely be more litigation over whether authors or publishers are the owners of the right to authorize ebook publication.) These plaintiffs don’t seem interested in providing libraries with any kind of service or solution to make books searchable or accessible to the print-disabled, so why should they profit when libraries take on these transformative tasks?

The UCLA case is a bit of an outlier, here, in that one of the plaintiffs, Ambrose, claimed to offer a streaming video product that would be cheaper and more useful than UCLA’s homegrown solution. However, it turned out that Ambrose had already sold UCLA the right to stream the films when it sold UCLA the DVDs long before it began to offer its own streaming service. That fact makes this an especially vivid case of double dipping. In any case, the market is a better judge of value than the courts; if Ambrose has a fantastic and affordable streaming service, libraries who need it are sure to sign up. If the product has disappointing sales, they should invest in improving it, rather than investing in legal fees to sue libraries.

In general, rather than sue to shut down or monetize library services they cannot or will not provide, the plaintiffs should refocus their energy on publishing and selling useful new content and services that libraries will want to buy. Libraries spend many millions of dollars every year acquiring content and services for their users. Collectively, ARL libraries spent over $1.4 billion on library materials in 2010-2011, about $12 million per library on average. That money is significant, but as Penn State Associate Dean for Research and Scholarly Communications Mike Furlough pointed out at a recent meeting of the AAUP, it is not infinite: every dollar libraries are forced to spend defending frivolous lawsuits and paying for fair uses of works they already own is a dollar they cannot spend on new books, serials, videos, and services. The courts in these cases recognized that imposing such a needless tax on educational use is contrary to the purpose of copyright. Everyone would be much better off if those suing libraries would return to copyright’s constitutional design: profiting by promoting, rather than halting, progress in science and the useful arts.

The lesson for libraries is equally important: there is a critical need to stand up, and be united, in defense of your rights. For years rightsholders relied on the uncertainty around copyright to force individual libraries into settlement agreements without having to actually win a court case. Libraries were worried that even if their work is fair and socially beneficial, the Copyright Act might not give them the tools they need to defend themselves. Multiply that uncertainty by the potentially dramatic statutory damages in the law and the cost of defending a lawsuit, and you can see why even a library that believed its activities were perfectly reasonable might nevertheless agree to curb them in deference to an assertive plaintiff. Many libraries were effectively intimidated before GSU, UCLA, and HathiTrust (together with several HathiTrust members) tested their claims in court.

External image

In addition to those three brave stands, the community collectively stood up for its rights with the publication earlier this year of the ARL Code of Best Practices in Fair Use for Academic and Research Libraries. That document has since been endorsed by a wide variety of library and academic groups, including the American Library Association, the Association of College and Research Libraries, the Music Library Association, the Visual Resources Association, the Art Libraries Society of North America, and the College Art Association. As I’ve noted on this blog, the GSU and HathiTrust cases both offer further confirmation of the consensus in that document. Academic and research libraries around the country have already taken the Code as a new standard for their practices. I believe libraries have happily learned the lessons of this exciting year. It remains to be seen whether the plaintiffs will get the message.

CCUMC Endorses #Librarianscode, Retires Guidelines

Today the Consortium of College and University Media Centers has announced two very important moves to empower its members and the many constituencies that look to them for guidance about proper use of copyrighted media. First, they’ve endorsed the Code of Best Practices in Fair Use for Academic and Research Libraries as a vision of fair use practice that reflects where their membership currently stands on fair use. In endorsing the #librarianscode CCUMC joins a formidable group of library associations and allied groups:

  • The American Library Association

  • The Association of College and Research Libraries

  • The Music Library Association

  • The Art Libraries Society of North America

  • The College Art Association

  • The Visual Resources Association

Like any document reflecting community norms, the Code will grow and thrive as it is embraced by the practitioners on the front lines whom it is intended to help. The CCUMC’s support makes the Code that much stronger as a tool for libraries facing fair use questions.

The second thing CCUMC did was officially retire its 1996 Fair Use Guidelines for Educational Multimedia. If you work with multimedia in an educational context, you’ve probably encountered some of the fair use rules of thumb from the guidelines, like “Up to 10% or 3 minutes, whichever is less” for “motion media,” or “[u]p to 10%, but in no event more than 30 seconds, of the music and lyrics from an individual musical work.”

If you look up fair use in the Copyright Act, you’ll find an expansive provision with no mention of counting words or percentages. In the aftermath of the passage of the 1976 Copyright Act, many practitioners worried that there was perhaps not enough information in the law to guide practitioners who must apply the doctrine frequently. Over the years many communities have promulgated rules of thumb to mitigate this uncertainty, but over time those rules have not kept pace with case law. If you peruse the scholarly literature, you’ll see that fair use has blossomed into a robust protection for transformative and educational uses that can happily accommodate uses that exceed any particular numerical limits often associated with it, and that the case law and the various codes of best practice have provided reliable guidance without imposing numerical boundaries. It’s wonderful to see CCUMC’s guidance grow and evolve along with the relevant law and practice. This is big news for media professionals in higher education, and for all the students and professors who rely on their wise counsel as they work to achieve their missions.


Video from a nice chat I had about #SOPA and #librarianscode with NCompass Live, a webcast for Nebraska librarians. We also touch on the Golan decision. All opinions are my own and not (necessarily) official ARL positions!

UC Boulder Takes VHS Tapes Back to the Future With the #Librarianscode

This is the first blog post in a series highlighting some of the fair use success stories we’re beginning to hear from librarians using the Code to move past fear and uncertainty and into positive action using their fair use rights. As with every Code of Best Practices, the #librarianscode can, will, and should be applied differently by different people and institutions in different situations. It is not one-size-fits-all. Some will be more conservative than the consensus described in the Code, while others may go further, depending on local circumstances. These stories are not meant to highlight ideal or best applications of the Code, as there is really no single right way to use the document. Rather, these stories show libraries moving from inaction to action thanks to the encouragement and support that the Code provides. How will you use the Code? If you have a story to share, please email

When we were talking to academic and research librarians about the kinds of problems they encounter in dealing with copyrighted materials, one question surfaced perhaps more than any other. Here’s how Brice Austin at UC Boulder described it to us via email:

Our campus IT department recently decided it would no longer support VHS on campus either by providing or servicing machines to play this format. While the Libraries continued to provide some VHS machines, faculty and graduate students still had a need to show VHS tapes in the classroom. Faculty also preferred DVD format for course reserves, so that students could use their laptops for individual viewing.

Section 108 © of the Copyright Act addresses the issue of replacing “obsolete formats,” but that term is defined so narrowly in the law that it seems to exclude VHS tapes, at least as long as someone, somewhere, is willing to sell you a VCR. If you’re going to convert those VHS tapes to DVD, or some other modern format, it helps to know your rights under fair use.

That’s why the Code of Best Practices in Fair Use for Academic and Research Libraries includes a principle about migrating formats to preserve and facilitate access to older media. The fair use doctrine exists to enable culturally beneficial uses that would be unduly curbed by copyright protection, and this is a prime example.

External image

CC BY Rob Pearce

The Code outlines a thoughtful, common sense approach to the question of when it is fair use to reformat materials, including asking whether materials can be obtained in modern formats at a reasonable price, and taking the old format out of circulation when a surrogate is added in its place. With these ideas in mind, and the Code in hand, the folks at UC Boulder were prepared to have a productive conversation with their campus counsel:

The Libraries Management Team made a case for conversion of VHS tapes to DVD format from our collections, relying heavily upon ARL’s Code of Best Practices, Principle Three: “Digitizing to Preserve at-risk items.” After some back and forth surrounding specific details, Counsel agreed to our proposal. Specifically, Counsel stated “In addition to reviewing the Libraries’ summary, I reviewed provisions of the Copyright Act, the Code of Best Practices in Fair Use of Academic and Research Libraries issued January 2012,“ and noted that “It greatly assisted me in providing this response.”

Now the UC Boulder libraries are moving forward with a policy that allows them to migrate old VHS tapes in a way that supports faculty teaching while respecting copyright, thanks to the efforts of mission-driven librarians like Brice, and with a little help from the #librarianscode.

The results reported here suggest that creators are willing to significantly reduce the amount of money they demand to license their IP rights in exchange for the opportunity to receive attribution for their work.
—  From Valuing Attribution and Publication in Intellectual Property by Christopher Sprigman, Christopher Buccafusco, Zachary Burns. This is consistent with the #librarianscode recommendation that libraries use proper attribution wherever possible when engaging in fair use.

Part 2 of the #librarianscode event at UCLA: I talk about #fairuse in libraries and Peter and I describe the Code.