Fair Use: A Legal Loophole... or Is It?

Hello NELA! Thank you for having me. This is the first library talk I've done in person since 2019. It's good to be somewhere where I can hear people's pens clicking and watch people check their phones or gently nap while I talk, truly. I've assembled a collection of links because it's late in the day and I talk quickly and if you want to read more about any of this, I've got the receipts (citations) because I am a librarian. URL is right here.

I'm a Vermont librarian in a state where most librarians do not share my attributes. Here are some of them. I grew up with a dad who was a technology guy (yes back in the 80s, I KNOW) so I grew up thinking computers were normal, that they often came bundled with video games, and that they were good at helping solve problems. Many other people of my generation had to use computers mainly for work, had to learn stuff on the fly in not the best of circumstances, and grew to feel that computers were the hurdle standing in their way. I get it. I do some digital divide research and I am a "qualifying authority" for the Internet Archive, helping print-disabled people get access to legally-for-them scanned books. Yes THAT Internet Archive, the one with the lawsuit. I am frequently looking for legal loopholes to help me do what I want to do, what I feel I should be able to do. This talk is about one such legal loophole: Fair Use. Here's some data but mostly some anecdata about how this stuff appears to me, I Am Not a Lawyer but I sure do know a lot of them. This is a talk for non-lawyers but it talks about legal stuff.

Ginsburg is quoted a lot in Fair Use essays, both because people like her a lot but also because of her written opinion in Eldred v Ashcroft in 2003. I certainly can't summarize the case but this quote gets tossed around a lot. Fair use is free speech... kind of. And as institutions that are governed by the 1st Amendment to some degree, this is relevant to our interests.

In case you don't know what fair use is, this is a crash course on one slide. It's a doctrine enshrined in copyright law. In short, copyright gives you the right to limit access to your intellectual property for a fixed period of time so that content creators can make a living. However, also there are some exceptions to that (the TEACH act is one I am not going into, libraries' right of first sale is another) and Fair Use is one. If you want to use someone else's copyrighted content you need to think about four things, these four things. Sometimes people think if you're just some schmo with a website you can't get sued. On the contrary it's often the schmoes that get sued! I'll zip through each of these....

Sometimes people sum it up more simply with something like this. Because, look, the answer to "Is this fair use?" is nearly always going to be "It depends," but that "It depends" can really sound more like no or more like yes depending on how you say it.

The Michael Pollan version of this is "Don't use too much, especially in proportion to the new work, mostly transformative."

This guy is a bit of a hero of mine. He worked as Director of Copyright and Scholarly Communications at the Duke University Libraries where he was always willing to say "We should be able to use more, do more, as scholars and libraries." In 2016 he became the Dean of Libraries at the University of Kansas. He talks, quite sensibly, about how making decisions around copyright is often treated as all (we don't care, sue us!) or nothing (never do anything that might get people mad) and neither is a valid strategy. Rather, institutions need to weigh risks and think about their liability in terms of those risks.

And this guy I don't know so well but he was the 9th Circuit Judge in Lenz v. Universal Music, you know that 30 second video of the baby rocking out to Prince's song Let's Go Crazy. His assertion is that fair use shouldn't just be an affirmative defense (i.e. it's there for you if someone sues you) but really needs to be considered by a lot of these algorithmic-based DMCA takedown machines that we've gotten in whatever bad simulation version we're currently living through. DMCA is the Digital Millennium Copyright Act and it's a law passed in 1998 that includes, among other things, a way for people to contact websites that are sharing their copyrighted content and get it taken down, with strong penalties if the websites don't comply.

So I'm going to talk about a few examples of how this works out for me, extremely online librarian, while I am trying to live my life. First off, robots.

The Tallman quotation is about the YouTube algo specifically overreaching. That was in 2007. I don't know if any of you have YouTube channels but if you put up a video with copyrighted music, there's a Content ID system that can catch you and give you a few options, depending on a secret number of things. It might put ads on your video, or it might make you take it down (or it will take it down for you) or it might give you a strike and you might risk losing your channel. But more and more stuff is LIVE lately. I saw things get a bit more sophisticated last year when I was watching a Facebook Live stream that abruptly terminated because the man doing the show played a short clip on his phone (yes, held up to his laptop screen) of a video with some (apparently) copyrighted music. Was it fair use? Almost certainly. Did the robot care? It did not. I'm sure this kind of thing will eventually have its day in court, it hasn't yet.

And just a few stats about the Content ID system. It's a moneymaker for rights-holders.

And it's robots all the way down. You can appeal a ruling (and probably should) but that takes time and often money. Has this happened to anyone in this room? Gotten a takedown request?

In 2011 I wrote a book. I signed a bunch of contracts including one that said I was legally responsible for copyright violations. My book was about how to teach people to use computers. Maybe you read it.

That book included a number of screenshots, assuring it would be out of date nearly immediately. My publisher, being very risk averse (and blaming their lawyers) told me I needed permissions for all of my screenshots. I dutifully wasted a bunch of time writing away for these. I often got permission, but some people never wrote me back at all including, no surprise, facebook. My publisher didn't want to go ahead with the publication. I said I was out of ideas, we were at a stalemate and I was prepared for my book to never be published. They eventually saw reason.

I also edit Wikipedia, it's a mess of an encyclopedia except for all the worse ones. I am generally someone who enjoys making things better so I help out there.

[these images are all photos of librarians of color that I've added to Wikipedia biography articles] I think Wikipedia articles are better when you can see the people they are about. And there's a provision in Wikipedia's 10,000+ word Non-Free-Content policy that says fair use images are acceptable in a very limited set of circumstances (only on Wikipedia, not Wikimedia Commons, low res, only if you fill out a fair use justification). Great! I decided that 2022 was going to be my year of making sure all deceased people had a fair use image added to their articles. I'd keep an eye on the page that listed the people who had recently died and go track down images. Easy peasy and a LOT easier than finding old images of long-dead never-famous librarians. But! There's a huge nerd war on Wikipedia about this because Wikipedia wants all their content to be FREE and so fair use is too limiting. So while you technically can do this for recently-dead people, there's kind of a gentleman's agreement to not do this before about six months have passed or a nerd might revert you. So I went back to finding pictures of long dead librarians instead. Not as simple but the nerds leave you alone.

In other parts of the internet, the rules are a little more serious business. You are probably familiar with copyright trolls but if not, here's a brief explanation. These are legal firms which use Content ID-like tools (or paralegals, I don't know) to track down websites that are using copyrighted imagery, usually from photographers or artists who they represent. You know how Google Image Search or TinEye works? Kind of like that except for their clients' copyrighted images. Who is using them? How can we sue them?

This is a longer story than I'm relating but... I run a very small image-sharing site called MLTSHP. It's a closed community, you need a login to post or comment. However we had a "best of" twitter/X account that tweets images that are popular there. It's mostly dumb memes, pictures of cats, and dinner, a kind of old people instagram. I love it. I took over the site from its previous owners and one of the things I didn't do (didn't know to do) was register as the DMCA agent for the site, that's the person who gets the takedown requests that I referred to earlier. So, in 2019, when one of the images made the "best of" twitter feed and was found by the copyright troll's robot (or paralegal), they sent me a letter demanding $15,000 for infringement and "damages." There's a lot of handwavey stuff here about how we handled it but basically we took the image down within 15 minutes, registered as DMCA agent immediately, and told the trolls we didn't have any money to pay them (mostly true) and they harassed us for MONTHS asking for money, threatening to take us to court, all the stuff. And later they found another image (by then I was registered as DMCA agent) and got more threatening letters. I finally got the fine folks at Public Citizen involved pro bono (after paying a lawyer for the early fight with them) and they did a lot of fancy footwork and got the copyright troll to cave in and give the site a retroactive license, and then piss off which I appreciated. I have not heard from them since.

But this isn't all bad news and terrible lawyers. Some of my favorite lawyers, like Kevin Smith who I mentioned earlier, have been doing some great work helping people, especially librarians, use fair use intelligently and usefully.

One of the things that they found, through examining many many fair use lawsuits, is that having a rubric or code for determining fair use within your professional context, and demonstrating that you were applying that code, was one of the better determiners of winning a fair use challenge. So the ARL (Association of Research Libraries) and the Center for Media and Social Policy helped put together these codes of best practices. They have them for poetry, dance, filmmaking, visual arts, open educational resources, journalism, media literacy education, open courseware, software preservation, dealing with orphan works, and online video. It's the single best tool (along with Creative Commons licensing) I can give to people, as a librarian, that will help them make educated and informed choices about how they use and share and reuse content.

[did you know that in many cases people can't take screenshots of Netflix shows? But that this is circumnavigable? I learned that THIS WEEK. And now you know it too] Letting people do more with more, especially giving them smart, informed guidance for working with all of the cultural content that we have access to is a complex and interesting part of the work we do in 2023. And understanding the legal landscape and when it's okay to say it's okay is a really important part of that.

When I first started talking about Fair Use, I'd be a little cheeky "Hey use what you think is fair. Who is going to sue a library???" but the last few years have shown us that a lot of people might. Or a robot might. So instead I'll tell you, don't be like those people who sue the library, be those people encouraging people to use stuff, be an access maximalist, a meme enabler.

Because hey, there's a great big world of remixing. As they say, "Adventure is out there!"

Thank you, again my slides are at this web address.