I always mean to write reflections on conferences and rarely do, but here is my lightning quick take on Code4Lib 2020. First of all, the tools I learned a bit more about and may use soon: Elasticsearch, Docker, docker-compose, CollectionBuilder, Wax, Fail2ban, Hyku.
I always struggle to select particular talks because I hate the idea of leaving out other high-quality ones. But here are those that hit me hardest.
Alison Macrina’s keynote was inspiring and vigilant, as I expected it to be (humble brag: I was the one who nominated her to keynote on the Code4Lib wiki). Macrina has carved a special, unique space out of librarianship to focus on issues of privacy and technology. The Library Freedom Institute is a needed program. There were several delightful moments in this talk but when she said “I’m giving out stickers…if you see an [Amazon Ring] camera, I suggest you put the sticker somewhere the camera can see it” was my favorite.
Kate Dohe’s The Digital is Critical was such a powerful talk. I took almost no notes during it because of her steady and dense stream of ideas. I returned to rewatch the video and even copied the transcript from the closed captioning service. Dohe implores us to involve users, to code with (not for) libraries and our communities, to not mindlessly serve the hegemonic interests which can dominate our institutions.
Bess Sadler’s Design Sprints for Democratization made a compelling case that restructuring our decision-making can surface voices that would otherwise be lost and yield better project outcomes. When I hear people talk about the contrast between the “technical” and “theory” talks at Code4Lib—a topic I will return to—my mind goes immediately to this talk, which annihilates the distinction. Sure, it was not a technical talk in the same way that a coding live demo is, but it was deeply concerned with the praxis of design sprints, with how organizations actually operate, and used theoretical underpinnings related to democracy and marginalization in harmony with concrete courses of action.
Brian Foo’s Visualizing Moving Image Archives basically shocked the audience. Seriously, go back and look at the Code4Lib Slack during its time slot. Here’s the video of the talk because I can’t describe it. I’m not sure what I was expecting, but it certainly wasn’t a dizzying, psychedelic journey through public domain video. Easily the most aurally/visually striking Code4Lib talk I’ve ever seen.
Huma Zafar’s “A New Approach to Old Metadata” also defied my expectation, as I anticipated some screed about Bibframe or linked data or MARC’s inherent problems. Instead, Zafar delved into the history and structure of MARC as many have before but discovered something novel, how it resembles a programming language in places more so than a data carrier or metadata format or, per another innovative take, a markup language. Once more, I see this as a wildly compelling blend of the concrete and theoretical. Historical examples (MARC seems weird to us because it came before relational databases which structure so much of modern application design) and investigating particular tags (880 is central to Zafar’s argument, as explored on her blog) are used to weave a more abstract narrative about the nature of the standard. I’m not a computer scientist, nor do I find CS all that interesting, but this talk made me wish we had more CS minds interrogating GLAM concepts and technologies.
I heard, as I have before, grumbles that Code4Lib has “too many theoretical/social justice talks”. There were a fair amount of new attendees, perhaps higher than usual, and I suspect the perception of Code4Lib is slides full of entity relationship diagrams and ruby code. But we have, for years now (ever since I started attending? I don’t know), had numerous talks dealing with the social implications of technology and the lived concerns of library technologists. Our keynotes, almost invariably now, deal with social issues. Valerie Aurora, Alison Macrina, Tara Robertson, Sarah Roberts, Chris Bourg, Christina Harlow, Andromeda Yelton. Those are just the ones I can think of off the top of my head.
The stark distinction between theory and technical is misleading, a way to disempower us and subtly insert a value judgment about what the One True Purpose of professional discourse is (it goes both ways). Theory is neutered without application; implementing technics thoughtlessly serves no one. Theory frames and is framed by technical issues. Harlow employs Python code as a slide background while establishing her narrative and impels us to “get shit done thoughtfully”. Technical talks, of necessity, include some theory. Your explicitly stated design rationale is nothing if not theory, your technical choices (linked data, open source, analytics) carry implicit paradigms. Technics and theory are not the same, but they are related, intertwined, and inextricable. Code4Lib has a larger appeal when it acknowledges that, as the program committee has demonstrated excellently for years now. My opinion is of negligible importance, being one person among many, but I enjoy a diversity of topics and delivery styles. I was a dual English and Mathematics major; when I was bored with problem sets, I wrote a poem. Code4Lib embodies the same refreshing duality; it presents both the big ideas and the little lines of code that lead us to them. Neither on their own is nearly as powerful.