What is copyright? How is it different from using proper attribution and avoiding plagiarism?
Copyright is a set of rights that the government grants to authors of original works such as novels, plays, essays, movies, and software. For a limited time (currently the life of the author plus 70 years, in most cases), copyright gives the author control over who can copy, distribute, publicly perform or display, or create derivative works (such as sequels or translations) based on their work. The purpose of copyright protection in the U.S. is to encourage the creation and dissemination of new works for the benefit of the public.
Copyright is therefore much broader than the norms against plagiarism. Plagiarism is the presentation of someone else’s work as one’s own; copyright infringement can take place even where the user is honest about the work’s true author. As long as you use proper attribution, plagiarism should not be a worry for you. Copyright is more complex: unless your use satisfies one of the exceptions or limitations described in the Copyright Act, you cannot use copyright protected material without permission, even if you give proper attribution. Fair use is one of the most important limitations to copyright.
What is fair use?
Fair use is a part of copyright law that allows certain uses of copyrighted works without permission, including making and distributing copies of protected material. It evolved over time as judges made case-by-case exceptions to copyright to accommodate legitimate uses. Typical early fair uses involved criticism, commentary, and uses in an educational or scholarly context. In 1978, fair use became part of the text of the Copyright Act – it’s codified at Section 107. In recent years, fair use has been a valuable way to accommodate innovative new uses that involve technology, such as the VCR, Internet search engines, reverse engineering of software, and the like.
How do courts decide whether a use is fair?
As you can see from the text of Section 107 (where fair use is codified in the law), fair use is not a narrow exception with clearly defined borders. It continues to evolve as judges consider new cases. In every case, however, judges must consider four factors – the purpose of the use, the nature of the work used, the amount and substantiality of the original work used, and the effect on the market for the original, as well as the overall purposes of copyright. In recent decades, however, fair use decisions have placed a strong emphasis on whether a use is “transformative.” This form of analysis synthesizes the four statutory factors into two key questions:
Did you use the work in a different manner or for a different purpose than the original?
If so, did you use an amount of the original work that is appropriate to your new, transformative purpose?
When the answer to both questions is “yes,” the court is highly likely to find fair use.
Software is often sold subject to a license. How do these kinds of licenses affect fair use?
Perhaps surprisingly, there are many circumstances where fair use is available notwithstanding the existence of a license. First, a license is only binding on those who agree to it. An archive that acquires a collection of used consumer software does not automatically agree to all the licenses that governed the original purchaser(s). If the library or archive is never presented with an opportunity to read and affirmatively agree to the license, they may not be bound. Second, fair use provides a source of authority independent of any license, so even when a license is limited, e.g., to “personal use” or on a “single machine,” fair use can supplement that permission, allowing uses that exceed the license. Finally, even if an archive is bound by a license, and the license clearly forbids certain activities that fair use permits, the remedies available for breach of a license are limited. For software that is no longer commercially exploited, the provable damages from use in ways that are consistent with the Code of Best Practices in Fair Use for Software Preservation are likely to be vanishingly small, if they exist at all.
What about creative commons or open source licensed material? How does fair use interact with open licenses?
As part of her copyrights, a copyright holder can license her work for whatever specific uses she likes. Creative Commons and Free and Open Source Software licensing provides a way for authors to grant permission to the public generally to make certain uses of their work. Works under an open license can be used in whatever ways and on whatever terms the license specifies, in addition to the uses available under fair use.
What about copy protection, encryption, authentication servers, dongles, digital rights management, and the like? Will I get in trouble if I hack through these in order to preserve and provide access to software? Does fair use protect me?
If your activities are consistent with the Code of Best Practices in Fair Use for Software Preservation, then they are the kinds of fair use activities that your community considers legitimate. However, the Digital Millennium Copyright Act does ban “circumvention” of a “technological protection measure,” and that can effectively prevent even fair uses of a copyrighted work, including software. Luckily, the Software Preservation Network and the Harvard Cyberlaw Clinic teamed up with several individual software preservation professionals to secure an exemption from the DMCA that permits software preservation at least until 2022. (Exemptions are granted in 3-year cycles, so the rule could change in the next rule making cycle. However, there is a strong tendency in favor of continuing, and even expanding, existing exemptions, rather than rolling them back.) The exemption is rather detailed and technical, but the Cyberlaw Clinic has created a handy “Preservationist’s Guide” to the exemption.
Why do librarians, archivists, and other software collection stewards need fair use? Don’t we already have copyright exceptions just for us?
It’s true that the law includes several specific exceptions that benefit libraries and archives. Section 108 allows libraries and archives to make copies for preservation, interlibrary loan, and user research, among other purposes. Section 110 gives teachers special rights to use works in the classroom and online. Section 121 makes it easier for disabled library patrons to get access to works not available to them in accessible editions. Each of these provisions can be extraordinarily helpful, and they apply in some very important situations. They do not cover every situation, however. Fair use is a broad, general, flexible doctrine that can fill gaps in these specific exceptions and go further, enabling activities that might fall beyond the limits of other exceptions. Fair use also allows for new technological uses that could not have been foreseen by the drafters of specific exceptions in the Copyright Act.
Do specific exceptions for libraries, archives, teachers, and others preempt fair use?
Not at all. Uses not explicitly covered by other exceptions can still be protected by fair use. Indeed, when Congress wrote some of the exceptions (for example, Section 110(2), also known as the TEACH Act), it specifically intended for fair use to be available to cover ‘near-miss’ cases. Courts have applied fair use to library preservation, rejecting the argument that libraries must rely only on Section 108.
Isn’t fair use pretty vague? I need to have clear guidelines, not just for me but for my staff.
It’s true that fair use is intentionally broad and flexible, and it will apply differently to different users in different situations. That may seem frustrating, but it can also be liberating, especially for communities who have a code of best practices. It means the law can adapt to local norms and conditions – it will apply differently to libraries than it does to for-profit vendors, for example.
Of course, every institution will have to adopt its own policies based on its own values, resources, and priorities. The code of best practices is an input for that process, not a substitute for it, and it will not replace more specific rules and guidance tailored to your institution.
Fair use is a “case by case” analysis—none of us have time to keep making judgment calls all day long!
Fair use is a case-by-case decision, but many cases are very similar to each other–you may even encounter them daily or weekly. It gets much easier to make these calls when you are evaluating the same situations on a regular basis. People make such decisions in other free speech areas with such speed and confidence that they don’t even think about it. That is because in those other areas, they understand what is normal and reasonable. Codes of best practices, like those created by filmmakers, visual arts scholars, and folks who preserve software, make it much easier to find the normal and reasonable approach to recurring situations. Once you get comfortable with these norms, you can apply them readily.
Is fair use risky? I don’t want to put my institution at risk.
Taking advantage of fair use doesn’t have to be ‘risky.’ Indeed, individuals, institutions, and even huge corporations rely on fair use every day–because they know how to assess risk, by understanding what is the normal, acceptable practice. If you are coloring within the lines of existing case law, and especially if you are following established norms in your practice community, taking advantage of fair use should be a very low risk. Indeed, the law even creates a special safe haven for employees of non-profit educational institutions who act with a good faith belief that they are within their fair use rights, protecting you from paying damages for some uses.
You should also remember that fair use creates risks for rights holders, too. It is costly to file a lawsuit, and even if a rightsholder believes your use isn’t fair, the judge may disagree. Judges have dismissed cases at a very early stage where the fair use case is strong, and even forced rights holders to pay defendants’ legal fees. The vast, vast majority of rightsholder concerns over library and archival uses are addressed amicably without any need for resort to the courts.
It’s also important to realize that risk is a part of our daily lives in most areas. But in most other areas, we’ve built our risk calculations into the background of our thinking. (How dangerous will it be to cross in the middle of the street? Will the clothing I choose for work this morning be appropriate?) When we know how to assess the level of risk, and balance that risk against other risks, we’re able to make the calls that make us, and our institutions, comfortable.
Finally, you need to look at all the risks, not just one. For software preservationists, any low level of risk from employing your rights also has to be weighed in terms of the risks incurred by not doing important work, which is core to your field.
How can fair use help me if a rightsholder objects to my use?
While nothing can guarantee that someone who objects to your use will not file a lawsuit or take other negative action, knowing your fair use rights can help manage an objection if it arises. If you have a strong, good-faith story to tell about why what you are doing is fair (for example, that your use is transformative and within the bounds of community norms), that can be a powerful deterrent to a reasonable rights holder. Any copyright holder knows they are taking a risk if they file a lawsuit, and their legal costs can be significant, too. The Code of Best Practices in Fair Use for Software Preservation can help you by giving you examples of practices that your peers have thought deeply about and found to be fair. It tells someone unfamiliar with software preservation that the community of practice has endorsed these norms.
Some people say “Fair use is just a defense; it’s not a right.” What does that mean?
This is a meaningless technicality. Yes, if you are actually sued, the issue of fair use will be raised as a “defense” to the charge of infringement. This is also true for libel – if you are accused of libeling someone, the fact that what you said was true is a “mere defense.” But, of course, it’s really the heart of the matter! If you know that what you’re saying is true, you can know with confidence that you aren’t guilty of libel. The same is true for fair use. The statute is clear, in multiple places. The Copyright Act says that any use that is fair is “not an infringement,” and refers to “the right of fair use.” It also says that the copyright holder’s rights are “subject to” fair use and the rest of the limitations and exceptions in the law. The law does not play favorites between rights holders and fair users.
How can I depend on a code of best practices in fair use created by my peers? Don’t I need a document that’s been negotiated with big copyright holders?
While it can be useful to negotiate with potentially adverse groups and come to agreement where possible, there are two reasons that best practices written by your own community are still very important. First, case law shows that courts care about what a practice community has to say about its own norms of fairness and best practice. The values of software preservation are internal to the community of stewards who practice it, and to carry the authority and authenticity of the community, these best practices needed to originate in the community. A negotiation between those values and the values of other groups, especially where those values can be in tension or direct conflict, would not have the same kind of power or authority. Second, past efforts at such negotiated arrangements were simply unworkable. Unfortunately, libraries and other groups interested in facilitating access to copyrighted work have been unable to reach consensus with rights holder groups on the most basic issues regarding copyright reform. We hope that will change one day, but the missions of our institutions can’t wait.
What happens when someone makes a bad judgment using a Code of Best Practices and gets us all in trouble?
Every user has their own interpretations and applications of the codes, and some will undoubtedly make mistakes. That should not be a concern for good faith actors who apply their code conscientiously. This is where the “case-by-case” nature of fair use is actually an advantage for users, because your uses will be judged on their own merits, not on the basis of what other institutions do.
Isn’t copyright law just fundamentally broken? Shouldn’t we focus on reforming the law, instead of living with it?
While the copyright law could certainly benefit from reform, cultural stewards cannot and should not wait for that reform to take advantage of rights they already enjoy under fair use. In fact, the more you take advantage of fair use, the more you can benefit from any future copyright reform. By taking advantage of your fair use rights, you help to strengthen values and interests that will themselves support broader reform. Your work also becomes an important example of the importance of balancing features to copyright, such as fair use.
Don’t we really need a “test case” that makes it clear what we can and cannot do?
The value of particular test cases can be surprisingly limited. First, these cases take a long time to wind their way through courts, leaving libraries, archives and other preservation institutions hanging in the meantime. Second, particular parties may mount better or worse defenses, and may decide to settle claims rather than litigate. Third, and most importantly, if your use differs in a significant way from the ‘test case,’ even a final fair use determination after a well-mounted defense could be of little use to you. Ultimately, libraries, archives, and museums are much better off taking advantage of their rights now, rather than waiting for the perfect case to make its way through the system.
Fair use is all well and good for creators who are making a new and “transformative” work with copyrighted material, but it’s not going to work for cultural memory institutions, which are making copies of entire works available with little or no alteration.
Actually, “transformative” doesn’t just mean creating a new work. It means doing something different with the work than the original author. That includes creating new works, such as criticism or commentary, but it also includes making the work available in contexts and for purposes that are not a part of the author’s original intent. Search engines, for example, copy huge swaths of content on the web in order to index those pages and help users find the sites they’re looking for. Search engines use fair use to create a valuable new service, and that’s a perfectly valid, transformative project. More recently, the US Patent and Trademark Office has argued persuasively that its use of entire scholarly articles in the patent review process constitutes a transformative use. Similarly, cultural memory institutions undertake valuable projects that may not create new works, per se, but serve new and different purposes than the ones intended by the original authors. The code of best practices identifies several specific uses with strong transformative rationales. A recent article by the facilitators of the software preservation code lays out in more detail the argument that software preservation generally is transformative.
I thought fair use is only for copying small parts of a work. But I often need to make the whole thing available. What then?
This is another common myth about fair use. While it’s certainly true that one of the four statutory factors asks judges to consider how much of a work is used, that factor is supposed to be balanced against the other three factors and the overall purposes of copyright. Fair use is designed to enable important cultural work, and if you cannot do your transformative work without using the whole thing, fair use is available. It is not always the case that the less you use, the better. For software preservationists, often the whole thing is required to do the job. Judges consider whether the amount used is appropriate to a legitimate fair use purpose, and they know that in some cases it is appropriate to use entire works. Rather than following arbitrary maximums, or erring on the side of using less, focus on the legitimate purpose and tailor your use to that purpose. Don’t overuse, but don’t underuse either. Both extremes will undermine your fair use argument.
What authority does this Code have? Why should anyone pay attention to it?
This best-practices code is the latest in a series of codes developed by professionals faced with the challenge of interpreting fair use for their particular professional needs. In each case, professional associations conducted a survey of some kind with their constituencies, to determine where the needs for employing fair use occur. Then, veteran members of the profession gathered in small discussion groups to discuss recurrent practice scenarios and how fair use might apply in these scenarios. The consensus positions expressed in these discussions were then captured in the final Code. A distinguished, independent panel of copyright experts reviewed the Code and determined that the positions it expresses are reasonable ones. By articulating how these uses are fair and legitimate according to their own needs and mission, practice communities send a clear message to judges and other gatekeepers. To date, these statements have had a powerful effect on practice in the affected communities. Never has a code of best practices in fair use been used against the community of practice. Large copyright holders have not challenged these codes, either. Indeed, some even have found them useful.