[META] An Archive of Their Own: A Case Study of Feminist HCI and Values in Design (CHI 2016)

An Archive of Their Own: A Case Study of Feminist HCI and Values in Design (CHI 2016):


In which my study about the design of AO3 not only gets into the big publication venue in my field, but also wins a major award. Fandom is amazing, and I want all the computing scholars to know it, too!  

The link in this post is to my blog post about the paper, which is the TL;DR version. But here’s an even more TL;DR version, i.e., what I think is most interesting to fans about this work. (Disclaimer: I’ve been on the legal committee of OTW since 2009, but this work was entirely independent of them! I also didn’t screen potential study participants for any particular attitudes towards OTW or the archive.)

Why did you study AO3? My dissertation was largely about social norms about copyright in online creative communities. I interviewed a lot of fan creators over a few years, and when we talked about copyright norms around things like attribution, remixing remixes, etc., a lot of people mentioned specific AO3 design features. I’ve been following AO3 closely since it’s very beginnings, and I know that there’s something really unique and amazing about it: It’s a massively successful online platform built completely by the people who needed it, built to reflect their values and norms. And the majority of those builders have been women. It’s amazing! And I thought, there is probably something that designers can learn from this. 


So for this study I (with the help of an undergraduate research assistant) interviewed a bunch of AO3 users, as well as people who worked on the development of the archive in the early days. And by the way: THANK YOU TUMBLR because I had so many volunteers that I had to turn people away. Trust me, this never happens. All my colleagues were super jealous. I was like, well, you guys really should study fans because they’re awesome.

What is feminist HCI?  First, HCI is human-computer interaction, so welcome to my discipline! And feminist HCI (here’s the paper about it!) is the idea that a lot of the central commitments of feminism – like empowerment, agency, equity, participation, identity, advocacy, social justice – are great things to integrate into interaction design. Imagine if all of these things were really important to the people building the technologies that you use!


So how does this apply to AO3? Talking to folks about AO3 made it clear that a lot of the values that were baked into the design are the same values at the core of feminist HCI. For example, participation: this is the entire reason that AO3 exists, so that fans themselves have control over their own space. And accessibility, diversity, and inclusivity – there are so many little design decisions towards these things, an attempt to try to make sure that not only does everyone have the ability to use the site, but that everyone feels welcome. One of my favorite quotes: “If you think you’re a fan, then you’re a fan, and you’re welcome here.” And also, the tagging system at AO3 is pretty amazing – not only did most of the users I talked to speak at length about how this improves over other sites they’ve used, but also the user-created folksonomy means that the archive doesn’t make content judgments. You can use any tag you want, and this actually becomes a pretty powerful thing because even a system picking categories for you to choose from (e.g., gender or relationship options on social networking sites) is an exercise of power. And the way that AO3 handles identity and pseudonyms is pretty nuanced, too. It all adds up to many small things that users really seem to appreciate. (The image below is from the Tag Wranglers page on Fanlore – this is in my paper, so it’s probably the only CHI paper to have the phrase “mermaid!sex” in it.)


And what did you learn? Besides just presenting AO3 as a case study of feminist HCI as successful, there are also some useful lessons for designing to reconcile competing values. After all, fandom doesn’t always agree on priorities! This of course is a huge problem in lots of contexts – the idea that you can’t please everyone. And of course, not all of AO3′s design or policy choices have been popular over the years. But there are a few things that AO3 does to mitigate some of these value tensions. For example, fan history is important! It sucks when archives disappear (Geocities anyone???) and all those stories you loved are just gone forever. But control is also important! If you want to wipe your fannish identity off the face of the earth, you should be able to do that. So AO3 has orphaning, which lets you erase your name/identity/footprint from fics while not erasing the fic forever. Another example is the content warning system, which was a compromise between the desire to not cast judgment on content (”your kink is not my kink but I will defend it!”) as long as it’s legal, and the desire to protect people from stuff that they don’t want to see or is triggering. Of course, this solution isn’t perfect, and some of my interview participants talked about wanting a “tag blacklist” to help even more. But in short: AO3 does some cool and thoughtful design things that are interesting to people studying HCI.

So now what!  Well, I’m a professor now, and I hope to keep studying these sorts of things. I’m interested in feminism and women in technology (remember the Barbie remix? Yeah that was me, I was Internet famous for about a day), online communities and especially fandom, and social norms and law. If you want to know if I do more studies of fans in the future, you can follow me here or on Twitter. Because can’t stop won’t stop writing!


Finally, thank you to everyone who participated, volunteered, or shared, because this kind of work isn’t possible without awesome people to talk to. And you can read the full paper at the link at the top of this post!