 Hello, Wiki Workshop 2023. I'm Kayleigh Champion. I'm excited today to tell you about a project I've been working on with my advisor, Benjamin Mako Hill. We're at University of Washington, where we're part of the Community Data Science Collective. So this project, Taboo and Otherwise, is focused on telling the epistemic life histories of knowledge resources, aka Wikipedia articles. This project was inspired by some findings of some prior work that we conducted. It's forthcoming at CSCW 2023. And what we found is that despite our expectations, Taboo articles on Wikipedia seem to be thriving. We might expect that due to knowledge gaps, censorship, misinformation, vulnerability of contributors, that Taboo articles might end up relatively neglected. Instead, often they're thriving. And this opened for us a little bit of a puzzle, how is it that they grow and develop? How do they compare to these other articles in their growth patterns? So we conducted a series of qualitative and quantitative analyses matching some Taboo articles here on the left, clitoris and menstruation with some non-Taboo subjects, cell membrane and philip Pullman. We matched them by the date the article was born, as well as how many revisions that the article received. So we read through every revision of these articles, taking down kind of open codes and notes developing memos to observe how contributors were interacting with one another, what challenges they seem to be facing, what kinds of content or content challenges that they were handling throughout the whole history of the sort of 20 year development. We also did some quantitative analysis. So this is the quality of each revision of these articles through time. Here at the top left in red, cell membrane, we see some quality jumps as well as some quality dips. Quality dips are due to vandalism, quality jumps in the case of cell membrane. This is when a Wiki edu course decided to tackle cell membrane. The quality jump here is attributable to their effort. The clitoris article kind of crawled through the quality ranks early here as well as later. This actually achieved good article status by the end, which is quite an accomplishment in Wikipedia terms. Most of this growth in this phase was accomplished through the effort of a single editor. And finally reviewed at the end hereby by other collaborators. Early in its history also a different person, but a single editor kind of pulling this article up through the quality ranks. In administration by contrast, we found after the article was split away here in, it's in blue here, bottom left, after the article was split away from menstrual cycle, it makes somewhat steady although bumpy progress. As people sort of tried to understand how to convey information on this topic without violating sourcing rules and sort of reflecting people's lived experiences, dealing with misinformation and so on. The Philip Pullman article, the story of this person's biography is after the initial kind of growth and development phase, growth in this article then becomes very event driven. So the author publishes a new article or a new interview comes out, someone comes in and updates the article on Wikipedia about them. So very different kind of histories and life patterns for these articles, but we were able to extract a couple of themes that seemed to particularly characterize the growth of taboo articles. In particular, these articles required resilient leadership. The people who worked on these articles needed to not only persist through the typical challenges of writing an article in terms of the effort and the time and the sources and the synthesis that's required, but also face conflicts, attacks, both on Wiki and off Wiki and kind of persist in the development of the article. We also noticed identifiability challenges around people who did not have accounts trying to contribute and perhaps being blocked. Often the articles that are taboo were blocked from non-account holding editors being able to contribute at all. They also faced a great deal of vandalism, which was a challenge for their authors. We also observed in these taboo articles in particular a lot of what we call disjointed sense making. So what do I mean by this? For one thing, working on an article is sometimes somewhat disentrained where someone does something and someone else comes along, maybe they don't know what's come before them. They ask the same question, introduce the same misinformation and introduce an old conflict that then all of the people who've been here for a little bit longer have to go back in and kind of clean up and straighten out or perhaps an old conflict that was resolved rears up again with some new combatants and needs to kind of be struggled with all over again. And we saw this around both content decisions and kind of governance decisions around what belonged in the article and what didn't. We also observed kind of this emergent governance. So over the 20 year history of an article, Wikipedia's are inventing the platform at the same time they're inventing the community, they're trying to decide on policies, rules, processes, and they're making articles all at once. And so there's a lot of moving pieces all at once here as people try to understand how to resolve conflicts while they are indeed in conflict. Finally, we observed a lot of themes around audience and publicness when writing taboo articles. This is concerns about should an article be safe for work? Is Wikipedia censored or not? The decision being Wikipedia is not censored. Should Wikipedia be kind of include the very best information that people have access to at a given moment such as an image that some might find objectionable but that image be replaced in the future or should we have no images until an excellent image that's not objectionable comes along. And what's our impact on say school children or people at work if we choose different approaches toward this audience working in public at all times? In all, the process of writing a taboo article on Wikipedia seems exhausting and deeply tiring for those who are sort of contributing to that production and the heroism of these authors really comes through. One question that I have having watched this unfold kind of contribution by contribution as we walked through this 20 year history is to ask how my platforms better support cooperative work like that on taboo topics, work that is exhausting and vulnerability inducing, can we make it better, easier, less exhausting for these contributors? Thank you so much for your attention today. I'm really looking for a door discussion. Thank you.