Following up on my last post, I made a transcript of my keynote at code4lib; here you go! In case you missed it:
- video (I start around 7 minutes in);
- my last post, which has links for various sites and ideas I mention throughout the talk.
There are numerous small audio glitches which I’ve filled in with my best guesses of what I was saying, where possible.
Architect for wanderlust: the web and open things
Nine years. Nine years ago we were in a much smaller building. Who was there? [pause for raised hands] Who wasn’t? [pause for many more raised hands] Who saw this coming? I knew there’d be one wiseass in the crowd. Hey, Mark. Nine years ago, except maybe for a token wiseass, no one knew we’d still be here for a tenth conference all this time later. That we’d have an IRC channel, a mailing list with three-thousand-some people, a journal — all these things we do. No one knew that it would evolve to be all this, but you built an open thing. You built a thing where lots of people can get write access and can build it all together.
Twenty-four years ago, someone else built another open thing. This may not be the first web page, but it is the oldest known one, and you can still see it today at info.cern.ch. The top post in my blog right now has all the links I’m going to be referencing during this talk, and I tweeted that out so you can find all this stuff for your clicky clicky pleasure. And this is the oldest web page we’ve got.
And why did Tim Berners-Lee make this? Well, physicists needed to share data. They had a bunch of different research stuff, but they didn’t have shared servers, they didn’t have shared presentation software, so they needed a thing. And he made this thing, and it seems to have gone well. So…why did it work?
Well. One big reason it worked was a determined agnosticism about formats. He didn’t care what format your data was in, what software you’d used to create it. The internet at the time had a collection of protocols. I remember gopher — I sort of have a soft spot for gopher. But it had a whole bunch of different protocols and formats and he built his protocol so you didn’t have to care. So that it was a generalized idea of information connection that could hold all the things, without being prescriptive as to their content or nature. And in fact he considered — I was reading the wikipedia page on the history of the world wide web, which is a great way to lose, like, three hours — he considered what should he name this thing, and went through ideas like “The Information Mine”. Go ahead and think about if he had named that “The Information Mine”, and we still had to call that today. “The Mine of Information”. But he settled on “the World Wide Web”, and what that says to me is that the important thing about what he was building out of all the experiments and hypertext he’d been doing in the past, the important thing wasn’t the information, it was the interconnection. The important thing wasn’t the information you put in it; it was the way it enabled people to connect to information and each other. So he didn’t tell them what to do with this architecture he’d created. But —
— he told them how to do it. If you read his original proposal to CERN for the money to support this thing, building the prototype and so forth, it says in there one of the conditions of the work is that he wants “to provide the software for the above free of charge to anyone”. If he hadn’t said that, we wouldn’t be here today. We would literally not be here today. But he wrote into this proposal — he used the tools of bureaumancy — to make sure this was a thing that anyone — who admittedly met a pretty high barrier for technical connectivity and knowledge — could use. And he wrote a ton of documentation. He told you exactly what you needed to do to download this and set up your own web server. And documentation is a brand of hospitality. And that made it possible for this thing he’d built to spread, and become a thing that everyone, and no one, owned, and everyone could build together, and as a result of that we today, twenty-four years later, have —
the Arab spring
our childhoods [delayed-reaction laughter as the audience reads the slide]
art and culture
and each other.
Let me tell you a bit about my origin story, and how the things that evolve don’t necessarily have anything to do with what we predict. This, also, is twenty-four years ago. I was at nerd camp, and a bunch of friends wanted to put on a scene for the talent show. They wanted to put on the scene from Monty Python where one is debating whether someone is a witch, and it becomes important to compare her to — a duck. And my then-boyfriend, now-husband, speaking of things you can’t predict twenty-four years ago, went to the mall and found this remarkably charismatic little duck. Which cost an outrageous amount, but it’s got this really cute houndstooth hat, and he bought it to be a prop in the talent show. And this talent show skit never happened. Because they weren’t just performing the scene from Monty Python; they were also performing the logician’s analysis of the scene which shows up in a BBC radio play, and which is so profane that no one is ever letting a bunch of fourteen-year-old boys do it on a stage. So the talent show act never happened, but we still had this duck. And I will get back to it.
But first, I want to say some stories about the rest of you. This is where we come from. This is where the organizing committees for this code4lib come from. And it’s kind of a lot of places. And this is just the committees — this isn’t counting all the people I’ve met here. If I counted them there’d be a whole lot more Canada on that map; there’d be a bunch more states; I’d have to zoom this out to include Japan and New Zealand. We come from all over.
But we come from all over metaphorically, too, and that’s one of my favorite things about librarianship, is we are people of wanderlust who found a home, here. You talk to people and you ask, “what did you major in in college?”, and you hear English, and history, and math, and religion, and philosophy, and musical theater. You ask people what they did before librarianship, because just about everyone had a “before librarianship”, and there’s teachers and publishers and marketers and designers and computer scientists and there’s all kinds of disciplinary backgrounds and perspectives and toolkits that we bring to librarianship, and that we can use to inform our work. The wanderlust that brought us here is a thing that lets us all enrich one another with our different perspectives.
And that is important for everyone’s stories. This quote, “a disciplined empathy”, is from Sumana Harihareswara’s keynote last year, which I loved, which you should read or watch if you were not here. And one of the things that she said is that user experience needs to be a first-class responsibility. And how do you get there, in software, is you have that empathy that comes from being able to see things from a variety of perpectives, and also the discipline to make yourself actually do it. To observe people, to do the user studies, to talk to people, to do the ethnography or the reading or what-have-you. Everyone has so many stories, and part of what we do with library technology is try to find ways for them to interact with their own and others’ stories, for them to make their stories legible and to find stories that are legible to them.
So let’s get back to mine. This is my friend Allegra, the one transformed by joy, and my friend Sam. Allegra just graduated from the University of Chicago last year, Sam is a senior at MIT, and that piece of paper that Allegra is holding, I wrote in 1991, twenty-four years ago. And it had the Legend of the Duck, in the most bombastic way that teenagers can find to write a thing. I had the best handwriting, so that’s why it was me. And I told the story that I told you earlier, except with fancier language. But at the end I said that every year the Holder of the Duck would pass it on to a new Holder of the Duck, from now into forever. And I wrote a bunch of lines for people to sign their names. Not because I thought that would ever work — just because that’s how much paper I had. So Grant signed the first one, and he gave it to Meggin, and Meggin gave it to me, and I gave it to Frank, and so on and so forth, and Max gave it to Allegra, and Allegra gave it to Sam, and it came to be the most important thing at this nerd camp even today. And there’s a Wednesday in August when a hundred and fifty incredibly excited teenagers will gather in a room to see who gets this next. Because being the Holder of the Duck is the most important thing they can imagine. We had no idea it would do this, but we built an open thing that people could be part of, that people could inscribe their names upon, and make into their story, and facilitate their wanderlust.
Of course open things don’t always go so well. This Icelandic pony is not by Kathy Sierra, who takes these unbelievably ethereal photos of her ponies, and it’s not one of hers because even though they’re the best ones out there I’ve ever seen, she doesn’t have clear license terms on them, and I didn’t want to be just another random person from the internet in her inbox, because she’s had way too many of those. If you’re not familiar with Kathy Sierra’s story, she basically got chased off the internet years ago by someone who’s widely held to be a nerd hero in some circles. She was threatened and her children were threatened and it became altogether not worthwhile. She came back recently, pretty much got chased off again. This is not Kathy Sierra’s work. This is not Randi Harper’s work, or Zoe Quinn’s, or Brianna Wu’s, or Anita Sarkeesian’s. There are many people who are severely threatened by open and ungardened things. Because the things that grow in open places can be vicious indeed.
And even for the people who don’t face that kind of peril, there’s a million quieter ways that open things — the openness of neglect — can be threatening or scary or overwhelming. If you have ever tried to teach yourself to code or take on the mantle of technologist, and you’re not a nineteen-year-old white man in a hoodie, you may have looked into technology and had a lot of trouble seeing yourself there. And that’s one of the things that comes up over and over when I teach people to code. It’s not just about, how do variables work or functions, although that’s challenging, but the more challenging questions are the questions of identity. When people look in the mirror, do they see someone who looks like themselves? When people look in the mirror and try to figure out how to piece together the disparate fragments of their identity and one of them is ‘technologist’, can they fit it together into a coherent whole with all the other things that they also are, and aren’t going to give up? The openness of neglect is a way of not noticing the barriers that don’t affect you personally, but that’s not the same as the barriers not being there.
That’s why this mattered to me so much. This is the commit history, part of it, from the CodeOfConduct4Lib. This is how you operationalize a disciplined empathy. I angsted for like a year about joining code4lib. I spent a solid year wondering if I was, like, cool enough, or smart enough, or technologically skilled enough to be part of code4lib. And I finally got over myself, and I joined the mailing list, and I started hanging out in the IRC channel, and it was fun, actually, and I liked the people I met, and I was having a good time. But I didn’t realize that I had been spending that entire time waiting for the other shoe to drop until Bess Sadler displayed the remarkable political courage to ask us to do this thing. And in the ensuing discussion — which, of course, had a variety of perspectives, which you would expect — but the thing that stood out to me is so many people in this community who have political capital, who have influence, who matter, were willing to put that status and capital on the line to be part of this thing, to draft it and to sign their names to it. And that was when I knew that it’s not just some weird freaky coincidence that nobody has yet been, like, a horrible misogynist to me. That that’s actually who you are. You’re nice to me because you’re nice. And I, I didn’t know that until people took this explicit step.
And it’s not just me, of course, right? There’s a lot of first-timers in this room, and I was a first-timer in this room two years ago, and now I’m on this stage, and none of us knew that would happen. And there are first-timers in this room who a year or two from now will be on this stage, or who will be writing the software that is the new hotness that all of you want to use, and you don’t know who they are. Hospitality matters.
It also matters because we don’t always do it as well as we did right here. This picture that I included on an earlier slide, I wanted something that was a story sculpture. I wanted something that showed people interacting in a really tangible way with a book that was larger than life. And this is what I found, and that was great, and I didn’t realize until much later that everyone in this picture is white. Everyone in this picture is young. Probably everyone in this picture is able-bodied. And I looked at the exif data that Flickr so graciously exposes and realized this is the Grounds for Sculpture in Trenton, New Jersey, which I know because my in-laws live just up the road and they always keep saying we should go there and we never get around to it. But because of that I know that everyone here had at least $10 and the free time to spend on this afternoon. Probably everyone here speaks English, probably natively. There’s a good chance they’re all really highly educated. My in-laws live up the road in Princeton. There’s a lot of really highly educated people who live in this area. Part of architecting for wanderlust is thinking about whose wanderlust you are architecting for. Whose stories are actually tellable in the systems that you create? Whose stories are recognized? Whose stories are writeable in the systems we write? It’s not just about not consciously erecting barriers. It’s about going out of your way to notice what barriers might have been erected and doing something to take them away.
Switch gears for a bit and ask, what is a library? This is a library. They don’t know it. My hometown, Somerville, has Artisan’s Asylum, which is one of the best-known makerspaces in the country, and that’s great, and honestly I pretty much never go there. I go here. This is Parts and Crafts. It is the kiddie, unschooling version of a makerspace about a mile down the street. That’s my kid. And pretty much every Saturday they have an open shop, and people come in and you can just kind of do whatever you want in a really self-directed way. There’s not really rules; the grownups don’t tell you what to do. But every so often, one of the Parts and Crafts staff will kind of wander over and say something like, “[dramatic voice] You know what? That thing you’re doing, you could do it better with a hot glue gun. Do you know how hot glue guns work? Do you want to?” And then they go find the hot glue gun, and suddenly it’s part of your eight-year-old’s repertoire, and next time she shows up she just goes over to the hot glue gun and starts gluing stuff to other stuff.
They’ve got a lot of stuff besides hot glue guns as well. Here’s a close-up of one of the shelves, which have this wonderful mishmash of this, and yarn, and tongue depressors, and motors, and this-that-and-the-other. All kinds of stuff. And what makes this a library to me is that it is about self-directed exploration. It’s about transforming yourself through access to information in ways that matter to you. And it’s supported by this remarkable collection, and a staff who don’t tell you what to do with it, but who are intensely knowledgeable in what they have, and able to recognize when their collection is relevant to your interests. So to me, this is a library, and it’s one of the best libraries I know.
What is library software? This is not library software. [applause] This is a screenshot from my local public library’s OPAC, and this is what happens when you do a keyword search that returns no hits. It dumps you on this gigantic page of search delimiters that probably not even a librarian could love, and let me tell you, if I didn’t get any original hits from my search, limiting it to large print Albanian will not help. [laughter] And this page outright angers me, because there are so many options it could have that would allow you to continue wandering. This could have an Ask-a-Librarian feature. This could have something that told me that ILL existed, which I didn’t know until library school, and as you recall I already had another master’s degree. This could be something that did a WorldCat search of other libraries for this thing. This could be any number of things that let me take another step that had a chance of success, but instead, it gives me a baffling array of ways to get large print Albanian hits for a book that doesn’t exist.
Other things that are not library software include API keys you can’t get, documentation and examples behind paywalls, twelve-billion-step ebook checkout processes. These are not library software. These do not facilitate wanderlust. These do not let people transform themselves through access to information and one another.
This is library software. This lets you — not just wander among hyperlinks — but write your own things. Write your own things in text, write your own things in code. This lets you build and generate.
This is library software. Partly maintained by Misty De Meo, who’s sitting right there. [applause] For those of you who are not familiar with Homebrew, it’s a package manager, and if you’ve ever tried to install things by, like, going to the web site, and trying to find the thing, and then realizing you don’t have the dependency, and finding another thing, and then, like, your whole house is full of yak hair, this does that for you. You type ‘brew install the thing’, and it goes and finds the thing, and all the things that the thing needs, and it just does it for you. And this is library software because it lets use make the things and learn the things we wanted to make and learn with fewer impediments, and totally nonjudgmentally.
This is library software. This is something I wrote during a Harvard Libraries hackathon a little while back, intersectional librarycloud. The Harvard Libraries has an API that returns the kind of collection data you would expect, but it also returns this thing Stackscore, which is a weighted 0-100 average of various popularity measures. So what intersectional librarycloud does — it’s one of the things you can link to, you can try it out — it lets you search for subject terms, and it brings back the most popular things in the Harvard collection that match that subject term. And it also examines their subject metadata to see if they have any terms consistent with women’s studies, or African-American studies, or LGBT studies. And I wrote this because I wanted to see, when students and scholars are forming their mental models, their understanding of how the world works, at one of the most eminent universities on earth, are these perspectives included by default? Because the consequences of them not being there is you get out into the world and you have to have all these stupid arguments about misogyny or what-have-you because people think it’s not a thing because hey, they didn’t study it in school, right? It’s not a part of the mental model they built when learning about history. This is a search for ‘history’, and as you note, looking at those grey and grey and grey blocks in the columns, when Harvard thinks ‘history’ it doesn’t think women’s history, or African-American history, or LGBT history.
And I also wanted to see, you know, if I search for something that really is, like, ‘women’s studies’, do I get hits in any of these other columns, too? Are the perspectives we see intersectional or is it all just kind of separated? Is it like, oh, well, if you’re studying women, you’re clearly not studying gay people, or whatever. That’s bunk. Unfortunately, that’s how it works, as far as the Harvard library usage data are concerned. Chris Bourg talked about this in a great keynote she gave at the Ontario Library Assocation just a couple weeks ago, mentioning — She was talking about how our cataloging systems can reinscribe prejudices and hide things from us. And she mentioned the book ‘Conduct Unbecoming’, which is really the foremost history of gays in the military. And, if you were looking for it at her library, you would find it shelved between gay porn. Not that any of us have a problem with gay porn, but you don’t find it shelved in the military history section. So if you were looking, if you were browsing the shelves, if you’re looking at a subject search, right there, for the history of the military, the history of gay people in the military is not a thing. So I wanted to interrogate whose wanderlust we’re supporting.
This is library software. The New York Public Library does these amazing things with the vast pile of cultural heritage data they’re sitting on top of, and their remarkable software resources. And they use that place as a cultural institution to create software that connects people to the world around them and to their own cultural heritage in a way that is creative and inspiring and moving. And so this, for instance, there’s so many things you could look at. They just got a Knight grant to do this Space/Time thing, which is basically like Google Maps mashed up with a historical slider, so you can look at stuff at different points in history  address change, and you can look up addresses that no longer exist, and it’s not really built yet, but they got money to do it, so it’ll be awesome. But this is their menus page, which you can look at right now. Adn they digitized a whole bunch of menus from different times in the city’s history, and so you can see what were people eating, what sort of things did people aspire to eat, what counted as high-class or low-class in people’s brains at the time during all of these different decades. This is the 1920s and it’s mostly things you wouldn’t see in restaurants today. But it’s fun! It’s neat. And it also — one of the great things that they do is they have APIs. So Chad Nelson, who may be somewhere in this room unless he had an early flight, he wrote this adorable little Twitter bot called @_badtaste_ that mashes up dishes from this with truly distressing words to create…vomitous menus, actually. It’s pretty funny. Don’t browse it right before lunchtime. But it’s funny. And so the fact that they had an API makes it even more library software, right, because it made it possible for Chad to have fun with this, and to build his own thing that let him explore this cultural heritage data in his own way, and connect to it in his own way.
This is library software. Probably a lot of you have seen Ed Summers’ @congressedits bot. It checks out anonymous wikipedia entries that are edited from Congress and posts them to Twitter, and has gotten actually kind of a lot of media coverage. But the thing about that that really took it to the next level was he put the code on github and he documented it, and that made it possible for people to fork it and made their own bots that checked up on South Africa, or Israel, or Switzerland, or whatever parliament or what-have-you they wanted to look at, so this became a worldwide phenomenon. And this is library software not just because it’s creative and playful that way, but because it’s something that lets us be more engaged with the world around us, more connected civically. It’s code that lets us find and tell stories that matter to people. And it’s a platform. It lets people tell more and more.
This is library software. This is something we created all together. It’s got little bits of many, many, many of us, and it’s changing every day, and it’s our little thing. Zoia was kind of transformative for me, also, in becoming part of code4lib, because the first time I wrote a plugin for zoia, which was a thing that totally intimidated the heck out of me — but then I did! And people helped me deploy it, and then I saw people using it. And within, like, thirty seconds, I saw people mashing it up with the outputs of other zoia plugins, and making it their own thing. And then I wrote documentation of how I did it, and subsequently I saw people use that documentation and make even more of their own things. Again, documentation is a form of hospitality. It’s a way that we bring more people into the community and let them write code that matters to them in their own ways. So this is library software.
Year ten. We’ve come an hour and a half up the road, and nine years. We’ve spent the last nine years inventing code4lib. I totally had a fencepost error in the first version of this talk, by the way, but this is the tenth conference, ten minus nine is one — anyway. We’ve spent the last nine years inventing code4lib. And I want to think about what we spend the next nine years inventing, and how we spend the next nine years inventing code that is deeply informed by library values. That is library code.
I want us to spend the next nine years inventing, building, library software. Building systems that question our own assumptions. That intentionally remove barriers and make space for all kinds of people, from all kinds of backgrounds, to tell their own stories, to build their own technology, to use in their own ways, that transform themselves in ways that matter to them. I want us to decenter ourselves so the systems we build aren’t things we own but things we give, and can then evolve in ways that we can’t predict. I want us to build library software. Architect for wanderlust.