Untitled
Anonymous
Plain Text
11/20/2026 2:24 PM
71.1 KB
6
No Index
Never
[Justin] [instrumental music plays] I'm Justin. I am some kind of academic librarian. My pronouns are he and they. [Sadie] I'm Sadie. I work IT at a public library, and my pronouns are they/them. [Jay] I'm Jay. I'm a cataloging librarian. My pronouns are he/him. [Justin] And we have a guest. Would you like to introduce yourself? [Thomas] I'm Thomas. I am an academic librarian, and my pronouns are he/him. [Soundboard] [cheering] [Justin] Welcome. [Thomas] Thanks so much for having me. [Justin] Yeah, it's been a long time in the making. You've been busy, and so were we. And I was getting married and moving, and you were out of the country, so yeah. [Thomas] Congratulations. [Sadie] [laughs] [Thomas] Yeah, I was also finishing up my master's. I'm freshly graduated, uh, for my master's- [Justin] Congratulations [Thomas] ... in library and information science. Thank you. [Thomas] And yeah, I also did an internship at Library Futures, and it's great to be here because Library Futures is actually... was one of the, the most valuable experiences I did during my master's, and I learned about it thanks to this podcast. Thanks when, uh- [Justin] Oh [Thomas] ... Jenny Rose Halperin, who's the ED, came on your show, I think around December 2024. [Justin] Probably, yeah. [Thomas] Um, yeah, and part of my origin story of, of interning with Library Futures is that that's... I, I vaguely knew the organization, but hearing that interview really, yeah, made me wanna to apply. So it's great to come here as a full circle moment. [Justin] Nice. [Jay] That's cool. [Sadie] Yeah, we love Jenny Rose. She's great. [Justin] Yeah. It's, it's always like the stuff that isn't the degree that's the most useful stuff in library school. I feel like my job was more important, like my graduate assistantship was more important than the classes. Me and Jay were talking about library school today, I think, about how the curriculum is not really standardized, and a lot of people go into archives. 'Cause I'm doing some archival research right now, and it's like how few jobs there are in archives. It's like, should we point people away from archives? 'Cause it's tough, and you need to make sure people are well-rounded. It's like, well, how do you make sure people are well-rounded in the library school curriculum? [Thomas] Yeah, I think, I think the internship came at a time where I was, I was just finishing up my first semester, and- [Justin] Mm-hmm [Thomas] ... I was kind of yearning for more. I was like, "I feel like there's more that I can do during this degree." So I think it's a great, like, if there's any library school students listening, I think internships like that are a great way to kind of find your people and get a chance to explore things outside of a class context that interest you. So yeah, I definitely really resonate with that idea of library school being like, you know, the degree you need, but then there's everything you do around it. I also, like, worked part-time. I was lucky enough to have a position part-time at the library where I now work at the same university where I did my degree, so that was really, a really valuable part of the degree as well. [Jay] Cool. [Justin] Oh, nice. I, I feel like a lot of them don't like hiring their own graduates. They were very like, "You will not get hired here after you finish your degree." Which makes sense. They don't want it too incestuous, but still. [Jay] U of I basically said, "You're not gonna get a job in the state of Illinois." [laughs] Like, "Don't even try looking at Chicago. There's too many of you." [laughs] [Thomas] Yeah, I think it helps that I'm in Quebec where there's- [Jay] Mm-hmm [Thomas] ... you know, only one French language library school, and so the French universities don't really have a choice to hire the graduates. [Jay] Mm. [Thomas] So yeah, I guess it helps that we're lucky also that most graduates don't need to move for jobs, which I understand in, like, the rest of English North America is a, is a reality. I mean, some do, obviously, but in that way, we're, we're lucky. [Justin] So it's a French language program? [Thomas] Yeah, so I went to the University of Montreal, which is the only ALA-accredited French, French language program- [Justin] That's cool [Thomas] ... in North America and anywhere in the world, 'cause the ALA programs are mostly in North America. I think there's a few in Europe that are trying to get accredited. But yeah, so it's kind of cool we're in this little niche. Yeah. [Justin] Yeah, I keep waiting for, like, there to be some integration between ALA and, like, Mexican librarianship because, like, it's right there, and it's such a huge country, and, like, there's all these, like, the UN, the, the, the autonomous university system is so huge, and again, right there, and it's like, you know, why, why is North American librarianship so disconnected from Mexico? [Thomas] Yeah, that would make sense. There's a trade- [Justin] Right [Thomas] ... there's a free trade agreement, so there should be a free library agreement. [Justin] Yeah, I really don't see why not other than the, you know, language barrier. But again, if you have a, a, a French language program, I feel like, well, that, that might open some doors. [Thomas] It can be done, yeah. And what's cool about Quebec is we get to kind of, like, mix also the influence of France, so it's, like, half influenced by the North American context and half influenced by France. [Justin] Mm. [Thomas] So it's an interesting, like, mixing of traditions and ideas. And that can be really- [Jay] French MARC records are such a vibe check. It's like, I don't know what they're doing. It's nothing good. [Thomas] Yeah, I mean, some of the advantages, for example, is, like, the, the translation, like the equivalent of the- [Jay] Mm-hmm [Thomas] ... Library of Congress subject headings, we manage that at one of the French universities. So we- [Jay] Oh, yeah. [Thomas] You know, we're, like, independent from the government or, like, a- [Jay] Yeah. [Thomas] Um, so we're, like, closer to the people developing those records, and I'm not a metadata expert, but my understanding, it's, like, easier, you know, to, like, actually make modifications and work on that. So there are some advantages. [Justin] Yeah, 'cause you have to run it through Congress, 'cause it is Congress's library. It's not the National Library. So you-- They can, they can change things unilaterally, like if some psycho is just like, "There's no queer people in the Library of Congress subject headings," and they can get it past a vote, that could technically happen. [Thomas] Yeah. [Sadie] Or, you know, it's called the fucking Gulf of America or whatever the fuck they wanted. [Jay] Yeah, that changed- [Sadie] Yeah [Jay] ... in subject headings. Yeah. [Thomas] Fuck that in particular. [Jay] And also Denali, they changed as well. [Justin] Which makes me curious about, like, if you have, like, unilateral control over the French translation, can you then change a little bit of the semantic meaning as well? [Jay] Mm. [Justin] And be like, you know, "We're going to use a more accurate queer term." You don't need, like, homosaurus. You can just, like, translate it better. [Thomas] Yeah. [Justin] Right? I wonder what happens. [Thomas] In fact, I probably shouldn't have said translation. I think they, they stand on their own. They just created equivalencies for consistency- [Jay] Yeah. [Thomas] You know? [Justin] Mm. [Thomas] But yeah, I think it's more than a, in a translation. They kind of stand on their own, and they're just a system of equivalencies with LCSH, with the, like, French from France system, which is, like, which was different. So yeah, I think it, it's really nice that it's kind of peer-managed by a few librarians in Quebec City who kind of review suggestions and are very close to the ground rather than this, this pretty, [Thomas] yeah, big organization controlled by the government. [Justin] Yeah. It has its advantages. It's, it, it is nice having kind of a national library system. [Thomas] For sure. [Justin] But who knows how relevant it all is after full-text searching. But anyway, we're going to talk about algorithms. So you did an internship with Library Futures, and were... I guess as part of it, you, the internship, you're doing a project the entire time. Is that how it works? [Thomas] Yeah. The internship is pretty project-based, where sometimes the project will come from Library Futures itself. Like, there's really cool stuff that they've done that existed, uh, like work that existed when interns came in. But most interns kind of pitch projects to work on that align with Library Futures' mission. And so I came in, and I had pitched, I had pitched this pretty long, convoluted ideas about algorithms and libraries and culture. That was like the, the Venn diagram. And what ended up coming out of that was this zine called Track 10: A Librarian's Journey Beyond the Algorithm, focusing specifically on music recommendation algorithms of big commercial platforms, how they change our relationship to music, and how librarians can offer alternative ways of discovering music and relating to it. [Justin] Why did you choose a zine as a format? Was that something you decided or something someone suggested? [Thomas] It ultimately was something I decided, but I was actually reluctant to write a zine because I had just written a zine as part of a class I did, which was on a similar subject. And our kind of final paper, I-- my teammate and I suggested to our professor, "Can we just do a zine?" And she was like, "Yes, of course." So we had just done a zine, and I was worried that it'd be redundant. But one of the first lessons I learned is there's value in doing something twice. You know, I'm someone who tends to really value novelty, but actually this zine ended up being much different than the zine in the class, which was kind of an academic paper with, like, ALA-- APA, sorry, citations in the zine. And, you know, what I wanted to do with this zine was to make it still grounded in research, but a lot more personal. And yeah, I think it allowed me to talk about a complicated subject in a fun, creative, warm way. You know, I think peer-reviewed papers are great, but sometimes it's really nice to have something that's-- has pictures and photos and is-- you can hold in your hand. So yeah, I think it was, it was really a desire to [Thomas] have a sub-- like, have something grounded in research, but that's just more accessible and fun. [Justin] Yeah, and someone will actually read it, unlike- [Thomas] Exactly. [laughs] [Justin] ... peer-reviewed papers, which, you know, I-- many of mine have very few downloads and probably even fewer citations. So this is-- I think this podcast was also kind of like me giving up on publishing peer-reviewed stuff and being like, "This will actually get cited," which is true. It did. [Jay] It's so weird. [Thomas] I mean, you can-- I've listened to your podcast making dinners, you know, which I can't do reading if peer-reviewed paper, so. [Justin] I keep getting ads on Instagram for this thing aimed at clearly graduate students that's like, "Have it read your paper. Who can read 30 articles? Have it read your, your academic journal articles to you." I'm very curious if it actually works, because obviously most aren't formatted in HTML. They're, they're PDFs. So I wonder how good it actually is. It probably just runs the OCR through an LLM to make sure that it's in the right order. But I, I love all the accessibility possibilities of AI, but the, the way it does it is so bad. [laughs] It bums me out. It does such a bad job of it, and it does it in such a way that's so exploitative and undermines labor and undermines, like, the, the AI, you know, voice actors for audiobooks. And it's like, there's so many people who want voice acting jobs and would take $20 to do it. [laughs] [Thomas] Totally. [Jay] What is your history? You said you made a zine for, like, to fulfill another assignment. Had you made zines before that, or, like, had you read zines? Like, why a zine besides just it's more fun than a paper and more people will read it? [Thomas] Yeah, that's a good question. I hadn't made a zine before the assignment zine. [Jay] Mm-hmm. [Thomas] I had made in law school, because I actually went to law school before going to library school. And in some ways, I prefigured my library career by doing a zine about how copyright regulates the flow of information in universities. [Jay] Nice. [Thomas] And it was a pretty militant car-- it was these little cartoons with FAQs on the back, and it was pretty militant. Like, we angered a few people because, um, you know, it was arguing for open access, and this was in 2012. So this was, like, when institutional- [Jay] Oh, yeah. [Thomas] repositories were still pretty new, like, a pretty new thing, and, you know, we were explaining fair use. And again, the idea was to take this complicated legal concept and make it accessible to the public. And we, my team and I, went to the bookstore of the university and kind of put... We asked, "Can we, can we leave these at the cash?" And they were like, "Absolutely not." [Jay] [laughs] [Thomas] 'Cause the most popular one was about how expensive course packs were that were just, like, reprinted things where people ended up paying twice for the same content, that most of which was, was fair dealing, which is fair use in Canada. [Jay] Mm-hmm. [Thomas] Um, so they were like, "Absolutely not." So what we ended up doing is just putting them in the books, kind of guerrilla style. [Jay] Nice. [Thomas] So I think, I think that was kind of an in- an inspiration of, like, I wanna do something like this again. And of course, zines are also, like, militant, you know? And so this zine, you know, is probably not the most militant zine [laughs] that you can find, but I think it was also to, like, continue this tradition of, [Thomas] yeah, having a... Not just that it's fun and whimsical, but also that it has a message and that it- [Jay] Right [Thomas] ... aims to, like, shake the tree a bit. [Thomas] And one thing visually I really wanted is to make it, make it feel like the kind of collage aesthetic of the original zines, even though it's in color, so it's, like, a pretty premium zine, you know? It hasn't been Xeroxed 500 times. So I did... In the process of making the zine, I went to an in-person zine workshop at- [Jay] Mm [Thomas] ... you know, our, like, queer and trans bookstore in Montreal. And I wanted the physical experience of, like, cutting, doing the collage, and ended up making just a small zine on an unrelated subject with a bunch of New Yorker magazines. But that really helped influence the visual of the zine, even though I ended up building it in, in Canva because I had to collaborate with other people and, like, edit text easily. But I wanted that feeling of this was kind of cut and pasted [Thomas] to kind of, yeah, like, respect the, the tradition of, of zines. And in fact, the, the cover of the zine [Thomas] is a knitting pattern. So the- [Jay] Mm-hmm [Thomas] ... the X you see is... I, I don't know if people picked up on that, but, um- [Jay] Yeah, I did. [Thomas] It's a knitting pattern, which is a metaphor for an algorithm. And that's... I had that aha moment during that physical workshop 'cause there was, like, a knitting magazine from 2013, and I was cutting up the... You know, I was just cutting things up, and I was like, "Oh, this is like an algorithm." So there's, there's ways in which the, kind of the physical zine-making made it into the final product. [Jay] Cool. [Justin] Yeah, collage is my main style of doing zines as well, and so I bring it into a lot of stuff. I've, I've tried to avoid doing too many digital zines unless it's something that's, like, digital first. Like, I was doing, like, this manga fanzine, and it was just, you know, screenshotting, pasting, screenshotting, pasting, and then just writing my commentary around it. It was kind of the same thing. But at work, I've been doing zines and bringing stuff in. And so now, I, I kind of just leave all my zine-making stuff at work, and then we just do zines at work whenever we're bored, or we have, like, a, a long lunch or something. [Thomas] Nice. And, and part of the process also was, like, this is one of the themes in the zine is that you make different... You have a different relationship to things in the analog world. Um- [Jay] Mm [Thomas] ... and so it was important for me to, you know, that that come through also in the format of the zine, that I probably would've never had that aha moment if I had, like, stayed in Canva the whole time and never gone out into the physical world with scissors and cutting stuff. And, you know, part of it was representative of the kind of serendipity that I, I, I kind of argue for in the zine of, like, being in the physical world and discovering things without the help of, like, optimization and let's make this process more efficient, you know? Like, it's not efficient for me to take the metro and go for an afternoon to this, to this workshop, but there's value in that. [Jay] Yeah, it's like that... The whole time I was reading your zine, I kept thinking of that, I think it's that Kurt Vonnegut quote about him going to the store to buy a single envelope, even though he could very easily, like, order through the, a mail or magazine a bunch of envelopes and then never have to go to the store again. He's like, "No, I'm gonna go to the store and walk there and get a single envelope." [laughs] [Thomas] Yeah. [Jay] I'm like, "Yes, yes, go get the single envelope." You know, cut and paste things manually. Go to the, the bookstore, you know? [Sadie] Yeah. And I think that there, there is also, like, a certain amount of, like, creative challenge in having a set amount of physical things to work with. So it's like- [Jay] Mm-hmm [Sadie] ... you've got a stack of magazines to create a collage instead of being able to, you know, image search and just grab whatever you want. You have to, if you want to have a something in mind, you have to, you have to really work at it to get what you want across as opposed, you know, as opposed to just... Yeah. I really vibed with the idea of friction- [Jay] Mm-hmm [Sadie] ... that you- [Thomas] Yeah [Sadie] ... put in your, put in the zine 'cause that's... zine 'cause that's been something that's been rolling around in my brain for a long time now, so. [Thomas] Yeah, I'd love to talk about that. Uh, I think it's probably the theme that stayed with me the most also- [Jay] Mm-hmm [Thomas] ... since it came out. So it came out in, in December of last year. And I think, I think the zine is, you know, focused on music, and it's even on my personal relationship to music and, and libraries. But I think it tries to say something about the value of friction in general, especially in the age of AI. One of the challenges I had in the zine was to never mention AI, which I succeeded in. But you can probably see some parallels in that a lot of what, you know, music recommendation algorithms aim to do is to take the work out of discovering music. So I remember as a teenager, this is kind of the opening scene of the zine, I went to the Grande Bibliothèque, which is our kind of flagship national library in Quebec in downtown Montreal, and I would go, I'd have my little list of call numbers for CDs, and I would go through the, the shelves and, like, take out CDs and then bring them home and listen to them. [Thomas] And, you know, then, you know, in the 2010s, I would, like, discover new music by reading music blogs and, you know, as part of my... Like, it was a joy to kind of read music blogs and see what's new and listen to, to artists. And then when I made the switch to Spotify, kind of just before the pandemic, all of a sudden, I didn't have to do that. So in a way, it's like, great, we've optimized a process. I can spend that time on something else. But then I started having this uneasy feeling of, like, I actually don't relate to the stuff I'm listening to. Like, I don't know who these people are. I don't know their backstory as an artist. It's just sound. [Thomas] And one of the... In my research, I came across this story in Kyle Chayka's book, Filterworld, which was one of the sources I used. And he tells this great story of this band that one random day saw streams of a song on their album just skyrocket. And it wasn't a single. This album had come out years ago. And then they asked Spotify, and they're like, "What's going on?" And Spotify was like, "Oh, like..." It turned out that song was the most generic, so it created the least friction. Like, when it came on in the algorithm, people didn't skip it because it just didn't make an impression. So that became kind of their most... the song that did the best on Spotify because it was frictionless. And so I started, I started becoming, like, really obsessed with this idea of friction and turning off autoplay, because the reason this song, like, skyrocketed is, like, the day it skyrocketed was the day that Spotify turned on autoplay as a default feature. And so, you know, and that's the one thing that stuck since writing the zine is I never turned on autoplay again because I have this little... Now when, like, an album ends, I have this one, like, point of agency where I can decide what to listen to and instead of just, like, continuing this frictionless experience of background sound. [Justin] Yeah, I know that there was stuff with Spotify about how ambient music was making the most money for a long time because it was kind of geared towards the algorithm of these really long ambient tracks. [Jay] Yeah, my, my best friend does well on Spotify because he does a lot of, like, ambient music. [Justin] Mm-hmm. [Thomas] That's the career we should have chosen. [Justin] Mm-hmm. Yeah, I didn't realize how important it was when someone turns off the music, that that weighs more heavily than kind of anything else. I notice it a lot with... because I use YouTube Music because I feel like it was better than Spotify. Spotify would just get obsessed with playing the same songs again and again, and it wouldn't stop, and I couldn't make it stop. So I eventually just gave up on Spotify, and I went to YouTube Music because there's all this pirated YouTube music it pulls from. So it's kind of nicer because of all the pirated stuff, and that gets pulled in a lot more. But yeah, it's... I lost my train of thought. Completely left me. [Thomas] Yeah, but you're right that the... one of the biggest flag for these platforms is when you stop listening. [Justin] Mm-hmm. [Thomas] I think the industry term in one of the books I read was captivation metrics, you know, how cap- [Justin] Right. [Thomas] How captive you are of the platform. And I don't know if you guys know this kind of infamous quote from the CEO of Netflix that their biggest competitor was sleep. Um, and I feel if there was an equivalent quote for Spotify, it would be like their biggest competitor is silence, right? [Justin] Mm-hmm. [Thomas] Like, the most horrible thing you could do for the platform is do something without being accompanied by music because they want to maximize the listening time. [Justin] Yeah. Oh, that's what it was. The... when I'm, when I'm on YouTube Music, I feel like sometimes when I skip something just because I've heard it before, it will say, "Oh, we'll adjust your preferences." And it's like, "No, I just didn't want to hear that song right now." But, like, I feel like it's now made a decision about me in some way that, like, I can't really understand. And I'm like, "I don't dislike this song. It's just not what I'm looking for." I'm looking for something very specific right now, and I'll skip like 10, 12 songs in a row because I'm trying to pull something, either something I haven't heard before, which sometimes it gets in a rut, or I'm listening for something very specific, and it, it doesn't know what this weird niche thing I'm looking for is because why would it? [Jay] And I think something like, like one of the things in the, like, scenes in the zine that sort of stuck out to me was you talk about, like, going to the library to look at CDs, and that you note that your library has a whole section on, like, French language, like Francophone CDs. Because, like, even though, like, you know, libraries are curated, like, those were all chosen by either a person or a vendor, right? And then, you know, put out by the library. But the, that sense of, like, serendipity of like the, is the type of discovery. But, like, still someone was like, "Ah, yes, we live in an area where there's a community of people who speak this language, so we should probably at least have that represented," that then there can be, like, serendipitous discovery within that. And that you noted that, like, Spotify never suggested anything similar to you, which is baffling to me with all of the metrics that these sort of, like, algorithmic... Like, sometimes I'm surprised at the things that, like, Instagram will recommend to me. I'm like, "How does it even get the sense that I'm in-into that?" Or, like, it'll like, "Oh, you, you, uh, you looked for a reel about this thing. You want all of your Instagram to be this now," right? Like, how does Spotify not pick up that you might be, uh, like a French speaker, be interested in that, like, based on where you live? Um, and so it was just, like, this idea of, you know, like, the ability to sort of have serendipitous discovery or to sort of have your choices sort of pre-curated in any sort of way. Like, again, going into a library collection, it's not an unlimited selection that you have, right? But that the, there's some intention about not, like, selection based on what's gonna get you to spend money, like, in an ad way, but just, like, the sort of intentionality was... I don't know why it was so profound to me just to read that little bit, but just the, the way you... it just got, like, dark. The way you wrote that out was just really, like, duh, but it was just so surprising Spotify didn't recommend French music to you. [laughs] [Thomas] Yeah, I'm glad it resonated. And I think, you know, Spotify is not accountable to a community the way that- [Jay] Yeah. [Thomas] That a public library is. And, [Thomas] you know, I think that's also friction of me being confronted with the fact that, like, half the CDs are in French, and I'm like, "I never listen to French music," you know? And, like, why? You know, maybe I, I, I still won't borrow them, but at least I'll have some kind of reflection. ... about, oh, maybe I should. Like, maybe this is something I want to do. And later in the zine, I go back to the Grande Bibliothèque, and I, like, go to the reference desk to ask for recommendations, and I purposefully, like, took the one French artist I really liked and asked for recommendations, you know, based on her. Because I was like, this is an opportunity to have a recommendation that is, like, grounded in my context rather than kind of, again, grounded in this, like, limitless availability of sound that's being, like, algorithmically fed to me. So yeah, I'm glad that, I'm glad that resonated. [Sadie] Yeah, that scene also, the, the going down to talk to the reference librarian for recommendations made me think, 'cause I mean, it's been a long time now since I've worked, like, a front desk sort of thing. I've been in the IT side for quite some time, but, you know, I, I remember being told how to do reference interviews and stuff and how to use the various databases we have to, you know, recommend books and stuff. And I was like, "I'm not sure there ever actually was anything like that for music." So like, you know, Novelist is the big one that has been in every single- [Thomas] Yeah [Sadie] ... one of the, my libraries, and I'm like, how would you start to go, like, how would you start to dig into, like, if somebody wanted a music recommendation? And that got me thinking, especially because I am very much, I'm a... Music is a big part of my life. So I'm like, "How-- What would I do if somebody came to the reference desk and asked?" And yeah, I wonder how, I wonder how much of that nowadays people would just be like, "Well, let me look on Spotify," like, already look at these already almerate- [Thomas] Mm-hmm [Sadie] ... al-algorithmic, like, services as opposed to, you know, something that's more curated or thought out through- [Thomas] Yeah, and that's, um- [Sadie] ... library principles kind of deal. [Thomas] Absolutely. And I think I, that's why I brought out the reader's advisory, 'cause I was like, you know, public librarians are used to recommending books, so what can we learn about comparing these two? And I think a big difference, to go back to something, Justin, you were saying earlier about, you know, sometimes I skip the song not because I don't like it, but because I just don't want it right now. I think a huge difference is that when a patron comes to the reference desk to do reader's advisory, they have agency in the process. They can say- [Justin] Mm-hmm [Thomas] ... like, "I'm going on vacation, and I'm looking for a book that is light and that is a mystery," et cetera, et cetera. That's a lot different than going to a system that says, "I'm gonna, you know, read all of your Libby history, notice how fast you turn the pages, when you stopped, when you put the bug- book down," basically this, like, hyper-granular surveillance of your habits, and, like, feed out three recommendations to you, which to me I find really alienating. But that's what we let do to Spotify, right? Like, we're gonna watch- [Justin] Mm-hmm [Thomas] ... your every move, your every behavior, and we think you'll like this, which is how I ended up getting, like, book Good Luck Babe by Chappell Roan recommended to me. Like, that was the song I was obsessed with all of 2024, which is a good song. Like, I love Chappell Roan. But I was like, "What about me screams Good Luck Babe?" You know? [Justin] Mm-hmm. [Thomas] And there's something kind of- [Justin] Yeah, it's like, "What?" [Thomas] Yeah. [Justin] "Why is it judging me like this?" Yeah, my Instagram, it's always like, "Why are you... What do you think about me? What do you think about me?" [laughs] [Jay] Just call me a slay already, you know? [laughs] [Justin] [laughs] [Thomas] Yeah, and I think, I think algorithms in general can be quite alienating. There was this quote that I f- I f- it was in a New Yorker article, but it was something like, "Algorithms..." [Thomas] I'm gonna get it wrong. [laughs] The quote was, "Algorithms don't show us what we want. They show us what kind of they think we want." I'm butchering the quote, but it's this thing of, like, it doesn't let us choose what we want. It kind of des- decides for us what we want, which I think also is, like, speaks to the sense of alienation because, yeah, when you go to the reference desk for reader's advisory, also there's stuff maybe you don't wanna share with the librarian, right? You decide how much you wanna share, which you don't really get that choice with these platforms 'cause they're just, you know, watching your behavior and, yeah, how long you lingered on this reel or whatever. So yeah, I think it, I, I think there's something there that speaks to the kind of the loss of agency when we just cede power to these platforms and want to maximize the time we spend on them. [Justin] Yeah, I mean, I was also thinking, how would I do a reader's advisory for music? But then when Sadie was just talking about it, I was like, "Oh, I know how I would do it. I would just direct the user to my special interest for the week," which is, like, what you would always, like... Like, I remember I had a student come up asking me about something for Michael Jackson, and I had just read and listened to, like, a video about his composition option, like, how, how he composed without being able to write music. And I was listening to these tracks of him, like, beatboxing. And so I was like, I found that track of, like, Thriller, of him beatboxing the whole thing, and I played it for her. And I was like, "Here." And she, so she wrote her paper about, like, beatboxing, uh, as a composition method. So I mean, there's, you know, the, the librarian is going to be reflected in the answer they give. Like, you know, the way I approach assignments is gonna be different from other librarians. Like, someone comes to me asking for three books, I say, "What's the assignment?" And they say, "I gotta write a paper." I say, "Well, you don't want three books. You want three sources that you can get through very quickly." And then I point them towards the easiest, quickest things to read that they can get their minimums in, and then they can worry about reading extra, 'cause I trust them to read extra if they're interested. So, but another librarian might not do that. They might say, "Okay, here's three books," or, "Here's three scholarly articles," whereas I would point them to reference articles if they're, you know, a freshman. I was working at a community college at the time, so that was when I kind of owned this approach of, you know, getting the, the minimum stuff out. So... [Jay] Yeah, like, I, I've worked in a couple of music libraries before, and it's not that that necessarily had, like, a reader's advisory service, but, like, it did sort of change the way that I... Like, like even bef- like my reader's advisory skills or even my like reference and like search skills, because if someone doesn't know exactly what they're looking for, there's like this entire other vocabulary for how you search for it. And so like, I think that's one, like music complicates every single aspect of librarianship, which is wh- I, I find music librarianship so fascinating. Like, I got into copyright because I worked in a, a music library in college and learned about music copyright and was like, "Oh, this is fucked." [laughs] Like, like, "This is fascinating. Whoa." And like, I, I'm reflecting now of like the way that like when I have friends who are in, who are like, "I don't know anything about opera, but you seem to really like it, and so I must be missing, like, why should I care about it? How do I get into this? If I wanted to, like, listen to or watch an opera, where do I even start?" And like the way that I, like, help narrow down for people, like, "Well, what kind of movie soundtracks do you like? Do you c- like, do you care about plot? Like, do you..." You know, what sort of all these like reader's advisory type of, type of questions. But like, I, I don't know why music is so much more complicated at that than like reading, because there's still emotion involved. It's still a very subjective experience that like no alg- like algorithms might think that they have clued into what it is that people like about certain music based on how many times you replayed a song or you played this song and then you played this song and then you went back to this song, or you always play this song when you go to the gym, which it knows because you have your location services on, right? Like, it thinks you kn- it knows why you like things, but there's such a emotional, un- like unquantifiable connection to why we like the art [laughs] that, that we do that has nothing to do with all of those other things. Like, why do you like that song every time you go to the gym? Like, what is it about it? You know, what, how do you connect to it? That just like an algorithm can never get at. But like, weirdly, librarians are very good at asking the kind of like, "Oh, what, tell me about that thing. What is it about that that you're interested in? And I can help you find more, more things." Like, we, we're like, I feel like reader's advisory is one of the only areas where we throw neutrality out the window and are very honest about that. And we just really get into the personal, "Tell me why you like this thing, and I will help you, and we're gonna get into it." Like, I, reader's advisory is so cool. Shout out to reader's advisory people out there. [Sadie] Well, and, and I think so much of that does, like it comes back to the friction because you're- [Jay] Yeah [Sadie] ... two people with two very different perspectives- [Jay] Mm-hmm [Sadie] ... trying to find a harmony so you can- [Jay] Yeah [Sadie] ... you know, recommend something that this patron will end up liking despite your personal tastes and all of that. And it's, there's no conversation like that with, like, I, you know, this is going back to what you were talking about earlier, Thomas, with like there's no conversation. There's, you're, you're not talking to Spotify telling it what you like. You don't have to think about it and how to articulate it. So it is just very like, yeah, there's, there's, there's no friction. And friction is good actually has become my go-to phrase lately. [laughs] [Thomas] Yeah, I'd love to talk about friction. And I think part of the friction is that, you know, maybe the recommendation won't be the "best recommendation," you know, that will be like mathematically computed. But that's rich because sure, maybe, you know, I like this genre of music, but maybe I'm ready to try something different. And I think algorithms can also lock us into, you know, this box of like all of our past, you know, like our past behavior, at least that's been tracked by the platform. And maybe it's really rich that I'm like meeting a librarian who's different than me and actually is gonna make me discover something different. And maybe I'll hate it, right? But like Kyle Chayka makes this point too, like that's taste requires effort, you know, like developing our taste. And part of that effort is, yeah, like being exposed to stuff we hate, which you might get on the radio, but you wouldn't if Spotify is just optimizing for stuff that you'll enjoy and hopefully not, you know, again, in this kind of background music way. I think, you know, it doesn't give us the opportunity to define, even just define our taste, even let alone change it or expand it. [Thomas] And that was, yeah, I, I think it's interesting that people are finding ways to reintroduce that into their life. Like a lot of the reactions I got from the zine from my friends is like, you know, say a friend like asked for Christmas for a radio that only is a radio, you know, like a, just tune, tune to the local channels. And if you listen to radio, you get exposed to a bunch of stuff you don't like. And maybe it's less pleasant than Spotify, but at least it tells you something about what you like. And, you know, it's also grounded again in your local context. So I think there's definitely value in, yeah, the, the recommendation not being the best recommendation. Maybe that's what you need. [Justin] Yeah, I do worry about the amount that librarians rely on algorithms. You know, if I was just to do reader's advisory based on, "Well, let me put that into an algorithm for you," it seems like a very bad approach, and that's probably one of the things that should be addressed when talking about like AI reference, is that this is going to have a smoothing effect on what gets recommended, which means, you know, why is the library suddenly recommending this one particular book at a 300% rate or 10,000% rate? Because something, it tickled something in the algorithm, and that's really the only reason, and that doesn't tell us anything about its quality or its... But then that infects its own user statistics if people actually go and check it out or click on it, especially if the, if it's generating links to e-books and someone clicks on it and then you get a click and then it becomes a cycle and you don't really at any point go, "Did anyone look at this?" I'm curious, because you, you went to law school, and you did this copyright approach. What got you thinking about algorithms so much in the first place? [Thomas] I think it was just this personal experience of, like, alienation from my own music taste, and specifically Spotify Wrapped, because I think Spotify Wrapped is the kind of, kind of grotesque demon of the, the algorithm personified, because it's like, "Here's what we think you like." And so that's-- I talk about that in, in the first track. The zine is organized into 10 tracks, 10 chapters. And yeah, during the pandemic, like, my Spotify Wrapped was also like, "You're the biggest Taylor Swift fan." And like, I like to describe myself as, like, a critical Taylor Swift fan, at least, you know, like- [Justin] Mm-hmm. [Thomas] I can appreciate the songwriting and, and her early work and everything, and I, I did like the pandemic albums. Anyways, but I was like, "This is not-- Like, Taylor Swift is not my favorite artist," and I was a bit offended. And again, it, it kind of [laughs] because, you know, it just tracks also how artists have also, like, reacted to what the algorithm wants, right? So the, the endless albums on Spotify or in general, like the-- with the deluxe version and, you know, the, like, two-hour albums. Again, that's because if someone plays the album, like, they're gonna get more streams if there's 20 tracks in the album than if there's 10. So, and I think someone like Taylor Swift has, like, mastered the, the art of, like, optimizing for the Spotify algorithm. [Thomas] But I think, I think, yeah, the-- what got me interested in algorithms was that experience of alienation from my music taste, which is really personal, right? I think more personal than our taste in music, uh, sorry, than our taste in movies, maybe even in books. Like, it-- in, in some ways, our music taste really, like, goes to the core of our identity for a lot of us. So to kind of watch that taste be distorted, especially in the, in, in the, in the form of Spotify Wrapped, was kind of the, the thing that really made me want to explore this. And I didn't think I'd get to explore it in library school, which is why I was so glad that I got to do this, because in my mind, it was just this personal thing of like, "Why don't I relate to music the same way?" And a happy coincidence is that the zine came out the week that Spotify Wrapped was released last year. [Thomas] And so-- and I feel last year's Spotify Wrapped was especially egregious with the, the age thing. I don't know if, if you guys remember how it would tell you your music age. [Thomas] But, like, these kind of gamified reflections of, of your data turned into, like, your identity. And also, like, I remember I had kind of, like, this, this horrible day last year, random day, like, that was just kind of, like, a bad day, and I listened to this one song on repeat, and then the Spotify Wrapped was like, "You loved this song on August 7th, 2025." And I was like, "I forgot about this day, but now you're putting it in my face because you think I love this song." But it's just like, this is the song that got me through a shit day, you know? And so, yeah, I think Spotify Wrapped is, like, the, the epitome of, you know, this kind of, you know, you think this is my taste, but actually this is not my taste because there's other stuff that's not quantifiable, right? [Thomas] And I read about these people who even, like, get-- will, will get, I think-- I guess there's apps or AI agents or whatever there is these days that can, like, play music for you, and you're not listening just so it, it, like, affects what your, what your Spotify Wrapped is, right? [laughs] [Justin] I kind of like, I kind of like that idea. It's like I'm using my, my bot to trick his bot. [Jay] This is some Zizek teledildonics- [Thomas] It's like- [Jay] ... that sex toys are fucking, and so you don't have to. Like... [Sadie] Junk data, you know. [Jay] Oh my God. [Sadie] Data poisoning or whatever, but it's like- [Justin] Synthetic data. [Sadie] Oh my God, that's- [Justin] That's what it is. It's literally synthetic data. [Thomas] Yeah, but it's the logical absurd of-- the logical extreme of, like, the machine talking to the machine, right? Like, who-- you can just take yourself out of the loop. They can just talk to each other. [Justin] Now I can go-- Yeah, now I can go enjoy my day. [Sadie] Can you even call that music taste then? Like- [Jay] Right. Like, I remember-- So in 2018, uh, 2018 was the year my mom died, but it's also the year that I started transitioning, which meant, like, it was the year my sexuality changed as well. So suddenly I had a dead mom who I had a complicated relationship and was also sad and gay about it. And so naturally, I listened to a bunch of Sufjan Stevens that year- [Thomas] Yes. [Jay] ... particularly the Carrie & Lowell album, right? But I had bought that on Apple Music, like, so I didn't have to listen to it on Spotify, but I listened to it a lot because I was like, "Oh, this is how I'm feeling right now." I've al- me too, Sufjan, like twins. And so when I looked at my-- because I was still using Spotify at the time, when I looked at my Spotify Wrapped, I was like, "This doesn't reflect the music I listened to at all this year," because I was over here listening to Rufus Wainwright and Sufjan Stevens on repeat, and I wasn't listening to those on Spotify. I would use Spotify to, like, listen to Bach while I was at work because I can't listen to things with lyrics while I work. And so it was like, "You really like Bach." I'm like, "I know I do." But, like, it, it would just get so, like, "You're the top listener of, you know, the, the Brandenburg Concerto," or whatever. And it was just like, it was so funny how, like, the areas where I listened to music affected, like, I know that's not what I listened to all year or what I even cared about because I was over here listening to it, but it doesn't know that. It was just like, it was so fascinating that year to see, "Oh, you listened to this this many times," and it just not feel reflective of the year that I had with music at all because I didn't spend it all on Spotify. [Thomas] Which is a way to hook you to the platform, right? Because, like, if I listen on CD, if I-- what if I want to switch to YouTube Music instead of Spotify? But what if I change my mind six months later, and all of a sudden my Spotify Wrapped isn't representative, right? [Jay] Right. [Thomas] So it's-- Yeah, that stuff is, is, you know, is like a retention strategy. [Jay] Yeah. [Thomas] And again, there's nothing wrong per se with, like, keeping- ... track of what you're listening to. Like, I remember in my iTunes library, I would make these smart playlists- [Jay] Mm-hmm [Thomas] ... that were, like, based on the number of song listens, right? But again, I controlled that. Like, I designed the self, the, the, the playlist, that data about what I was listening to was stored locally. It wasn't like, Apple wasn't like, you know, computing stuff with it. And so I think there, it goes back to that piece of agency, I think, also of like, I can track what I want to listen to if I want. There's also last, last.fm, I don't know if you remember this, this service that would track your listening across multiple apps. So again, it was like an idea of like interoperability of not we're locking you into this one commercial platform. We'll give you a way to track your number of plays across multiple platforms, which is also like harkens back to a different kind of internet that was more independent from these commercial providers and interoperable, which is also thing, something we've lost with these like huge commercial players where there's a cost between switching between Apple Music, Spotify, YouTube Music, or all these others. [Justin] Yeah. I, I do remember... Well, actually, let's go to alternatives to commercial streaming, because I wasn't familiar with this Edmonton, what was it, like Capital Records program. Because now I'm thinking like, why doesn't Boston do that? Do you know much about how these local programs work where they get local artists to license with the public library and get their stuff played? Like, how does, how does the player work? Does it go through like a Last.fm sort of thing, or do they run like a radio station? [Thomas] Yeah. So these, this model was pioneered by the Iowa City Public Library, and it's been kind of replicated across North America. I chose Edmonton because it's in Canada. It's a city that I've been to, that I had a personal connection with. But the basic idea is that the library will build its own streaming platform, and they'll license directly with the artists who get paid so little by Spotify anyways that, you know, the library can just kind of cut out the middleman and become, [Thomas] yeah, a publisher. Not in, in similar way that like academic libraries are trying to cut out like the, the big publishers. You know, the public library will just license the music directly, and it's local music, and typically they kind of build a committee that's made up of librarians and representatives from the local music community. And yeah, there's different models, but at, at Edmonton, like if you're a patron, you can download the music offline, but anyone can stream it. And they, they'll kind of once a year, they'll choose which artists to add. And what I like about Edmonton also that I talk in the zine is that they also built kind of a whole oral history project around it, and just history in general and context. So they talk about the music venues and the history of music in Edmonton. I believe, I don't know if Edmonton does this or another library that follows this model, but like sometimes they might organize events. So it's not just this disembodied digital experience, but it's grounded in a community, it's grounded in a place. I think philosophically also they see it as kind of a digital public space. Like, what would it mean to create a public space digitally? [Thomas] So I think that model is kind of points to a possible future, especially if we can replicate it in more places. I think, yeah, every, every library, [Thomas] you know, would benefit from elevating their local artists who typically don't get a lot of visibility when, again, they're, they're lost in this endless stream of music [Thomas] on these streaming platforms. So yeah, so as part of the zine, I kind of was listening to these artists based out of Edmonton. And again, I was like, some of them I didn't like, but some of them I did, and I got a chance to listen to music that reminded me of a certain, you know, place. Like the, the, that gave meaning to the music beyond just the sounds and the lyrics of like, oh, this came from a place, and here are the experiences in my life that I had in this specific place. So I think, I think that's a really interesting model because [Thomas] it's, you know, it's, it's all good and fine to be nostalgia about the past, [laughs] which part of the zine is that, like the time of CDs and mixtapes and physical media. And there are some advantages to that, but also it's like, how can we take the good things about those things and bring them into the digital age? [Jay] I know, uh, University of Virginia was trying to do something similar where they were working with, I think this person had been like a janitor at UVA, um, but was also like a local, like rap artist, and they bought the streaming rights, like for his music. So like if you, like, can we put this on our streaming thing as like a local collaboration with, with you? I'm trying to find that, but because I was at a music library conference where they, they talked about it. But yeah, some academic libraries are starting to get into that as well. [Thomas] Yeah, it's a model in academic libraries that we do with, you know, let's promote the research or as scholars and make it open access. And of course, the economics of that are a bit different, but I think it's, I think it's definitely, you know, an interesting model. [Jay] Yeah. [Justin] Yeah, and definitely with like a closed access system. I mean, I do a lot of, not a whole lot, some licensing of films that can't be gotten through a vendor. So we do independent licensing, and then we put it on our platform, and then we, you know, provide access to authorized users. And so I feel like there's a way that, uh, the infrastructure is kind of there, although it's not cheap, and it's definitely not getting cheaper because of all the AI stuff picking up hard drives and server space and things like that. But it also shouldn't be too prohibitively expensive. So I am curious how other places could do it, because I just went on Capital City Records, and it seems like it's kind of open for anyone to stream, not just cardholders. [Thomas] That's right. It's anyone can stream, and cardholders can download. Um, and I believe also it's a DRM-free download, which, you know, is a whole other thing we could talk about, but they just-- it's this tr-- like, trust system. Don't quote me on that, 'cause I haven't tried it. But if you're listening and you're a part-- patron of Edmonton Public Library, let us know. [Thomas] But yeah, that's, you know, that's what I think is so interesting about the model, because the li-- these libraries can also, you know, with a certain budget, license the music for anyone to stream, not just a patron. So the music can reach beyond the, the city, but the library is kind of curating it and anchoring it in its local context. [Justin] Makes sense. Yeah, I, uh, I do wonder how much can be done in a digital space, like, especially as someone who works with, you know, licensing and digital resources and has done open access and open scholarship work. There's a certain amount of built-in limitations because of copyright and because of the way that copyright laws tend to get stricter. And we have, you know, that loss with the Internet Archive case, uh, which put, I think, a-- killed a lot of momentum that was happening that was accelerated by COVID for, you know, controlled digital lending, which could have been done for anything. It wasn't just books. Controlled digital lending could have worked for films that were in copyright. It could have worked for all kinds of things. [Thomas] Really good for sheet music. [Justin] Yeah, sheet music, all sorts of things that, you know, it's kind of closing a door if we aren't able to work this out, because otherwise your options are to work it out through licensing, which I don't think is a good idea, although people do do it. They're like, "Oh, we're doing controlled digital lending by licensing full book interlibrary loan." And it's like, well, that's not really controlled digital lending. That's just good licensing terms. Like, it's, it's just still licensing. The whole point is I shouldn't need a license to do this. It kind of annoys me that people call that controlled digital lending too, although it is technically, meets all the criteria, but you have to have a license allowing you to do it, which isn't really the point. The point was to utilize fair use in a, in an interesting way and combine it with for sale through an interesting little legal maneuver, which also tickles the super lawyer part of my brain that's like, "Let's, you know, let's, let's outwit the system." [Thomas] Yeah, I think I talk about in my personal journey in the zine, one thing I did is also I was trying to find ways to play these CDs that I borrowed at the library, so I went to all these thrift stores 'cause I had nothing to play a CD with. And I also realized these thrift store were selling these CDs, mostly of '90s divas, so I ended up buying a lot of Mariah Carey. And, you know, it was, again, a reminder, it was kind of an-- I don't talk explicitly about controlled digital lending, but it was a nod to that. I talk about the book The End of Ownership, um, that, that talks about how, yeah, by moving stuff in the digital realm, we're kind of losing the ability to, like, you know, resell, uh, or donate things that we bought. And I think the challenge is, what would it look like to take those principles and apply it for libraries in a digital space? And we haven't, like, figured that out yet, but I think you're right that, that the principles of controlled digital lending, or whatever we want to call it, would go beyond books. [Justin] Yeah. And it would just kind of open up a lot of things that otherwise would have to be done through legislation, which, you know, because of the Berne Convention and this international approach to copyright regulation, is just really unlikely. And so I-- there is a certain part of me that's like, we have to introduce friction through physical stuff, even though the users want more and more frictionless approaches. And, you know, sometimes I feel like just telling faculty members, "You know how hard this used to be to get you an article?" They're like, "I can't wait two days for interlibrary loan to get it to me." It's like, "You know, you, you would have to go to the stacks to get this before, right? You know that, right? Like, this would have taken you, like, you had to physically go there." You remember the 99. Like, he had to go to the library to get articles to do research. You, you should maybe spend a minute wondering what you want to look at. But it's-- I feel like it's only going to get worse with the expectations of AI-powered research. They want to go faster and faster, and they're going to want, you know, MCP servers that just feed the data directly into their personal AI, and they're going to have to license those servers, and is that going to cost more money? Almost certainly. So it's very frustrating that, um, you know, I, I, I wonder at this certain point when we have to push back and say, "You have to put up with friction," because otherwise-- but I-- that'll have to be a societal thing. But we'll have to be able to say, like, "You have to take longer to do things. You have to slow down." And I feel like this, you know, the whole, like, slow librarianship thing that people used to talk about, or slow research, things like that, like, that's actually going to have more purchase now. Like, it was prescient in the sense that people weren't ready for it, and I think now people are ready for it, um, where they're saying like, "Yeah, we do have to do things slower," which I think overall is good because that could also be good for a lot of societal problems that take too much, go too fast, thinking particularly like climate change. [Thomas] Yeah, I think a few things about that. There's a really interesting talk that I listened to today by Dave White, I think his name is, called The Inconvenient Library: Friction and Dissonance as Central to Developing Critical Literacy. I took note of the title. And I found it was really-- it was a talk given at a library conference recently, and it was interesting how it, uh, and it was in the context of AI and especially teaching critical information literacy. But he was talking about designed friction, so like the idea that, yeah, the, the friction, it's, it's kind of related to what we're saying, but this idea that the, the friction is-- brings especially learning, especially for the students, right? That one of the big problems with generative AI is it removes the learning experience, and just like to make a link with the zine, like taste is created through effort and friction, like learning also, right? Learning skills is- ... created with effort, 'cause he talks about his, his frustration with universities saying to students, "You can use AI, but you have to think critically." But by using AI, they're losing the skills to think critically, right? [Justin] Mm-hmm. [Thomas] And so the idea of friction is not being something is not working well, but we've designed this intentionally so that there's value in the fact that, yeah, maybe it's slower, maybe it's more complicated. So he makes a distinction between, yeah, like, just a system being bad and not working, which is not what we want as librarians, right? We wanna make people's lives easier and all that, but, like, adding friction with inten- intentionality, which I found, like, an interesting distinction. [Justin] Mm-hmm. [Sadie] I'm gonna watch that later. I found it on YouTube. That's fascinating. [Justin] Yep, I've got it in the notes. [Sadie] Yeah. [Jay] Yeah, like, I remember, you know, back in the early days of the pandemic, and even before then, but, like, especially during the pandemic, like, one of my worst qualities is that I'm really into the sort of, like, productivity, personal knowledge management, like, note-taking sphere of stuff. I got really into that during the pandemic. And I was taking, like, yes, buddy, a tutorial with Bob Doto, who does, like, Zettelkasten stuff, especially for, like, if you are a writer. And he, in his, like, setup, he was talking about how, like, so much of this space is about, like, reducing the amount of friction between steps, between recording info and between retrieving info and between putting it all together, when, like, it's actually good for this if you actually introduce friction back into it, because you have to be able to actually consciously think about something and actually actively, like... You don't wanna take all of your agency and all of the, like, out of the process of, like, recording information and bringing it back in and synthesizing it. Like, there is an element where you need to be slowed down [laughs] and to think about things, even if there are other things that can make it easier, especially to, like, collect it somehow. [Thomas] Yeah, 100%. And I think one thing we haven't talked about in the zine, but that's probably my favorite part, is I went back to the history of- [Justin] Mm-hmm [Thomas] ... music recommendation algorithms and to- [Jay] Yeah. [Thomas] Which, yeah, if you've read the zine, nothing I say will be new, but if people haven't read it yet, [Thomas] there's this... The first music recommendation system was called Ringo for the Beatles. [Justin] Mm-hmm. [Thomas] And it was a, a project of a master's student at MIT in the early '90s. And what I found really fascinating about his thesis, which you can go and read online, is you can see all the math, and they explain how, yeah, like, they found people with neighboring tastes. But what I found so uncanny is that in many ways it, you know, it, it kind of announced, uh, things to come, because one of the problems they ran into with the system was that it was an email system, so people would, you know, communicate with it by email, and they would assume it was a real person. They would talk to it in natural language. [Justin] Mm-hmm. [Thomas] Uh, so they actually had to change the boilerplate language to make it really, like, impersonal, so people would stop talking to it, like, in natural language, thinking it was a real person answering them and giving them music recommendations. And there's this quote at the end [Thomas] that [laughs] says, "As long as people believe that an agent is a reflection of themselves, they will enjoy using the agent. Who doesn't like looking in a mirror occasionally?" [Justin] Mm. [Thomas] "Here is a system which operates 24 hour, 24 hours a day, grows, and in a way, quote, 'learns.'" [Thomas] And I was like, "Ah." [laughs] [Thomas] And this was 1994. [Thomas] And so, yeah, even in the first kind of, you know, we had Ringo, now we have Claude, you know, like, let's name these agents, uh, with human names. And this sense that, you know, the more... One thing with Ringo also is the more data you fed into it, the better it was, right? That's why it says it learns, because the more users interacted with it, the more data algorithmically it had to place you in the ma- matrix to show the, the neighbors in your kind of music taste neighborhood. And so, yeah, it's always interesting. These, these things we think are new sometimes have actually been going on for a while. [Justin] Mm-hmm. [Justin] I think that covered everything I wanted to do. Was there anything anyone else had a question for Thomas about before we wrap up? Parker? [Jay] [laughs] [Justin] He's been tapping me the whole time. He does this when I'm working from home, is his new thing, is to just tap me, and I finally had to kick him out of the office the, the other day, 'cause he would not stop. Tap, tap, tapping. [Jay] He's chirping. He's got a lot of things to say. [Justin] Yeah. He's not usually got the zoomies when we're recording. Thomas, was there anything else you wanted to mention about the zine, or something you're working on that you want people to know about and go check out? [Thomas] I think the final tension for me in the zine is the question that I tried to answer and have it answered, but I would pose to everyone is, what does it look like to create a technology-positive future for libraries? [Justin] Mm. [Thomas] I think the tech can feel so dystopian, but what, you know, what does it mean to use it for good? So that's my unresolved question that I was kind of sitting through. [Thomas] And in terms... And maybe on that theme, something else that came out recently from Library Futures is the second project of my internship, which is called As Long as the Sun Shines: Indigenous Digital Libraries in North America. [Justin] Mm. [Thomas] Um, so you, you can find it on Library Futures' website. And it's a digital story map built with ArcGIS. So it's... If you've ever kind of had those kind of immersive scroll-through narrative journalism pieces, it's kind of in that style. And it takes, yeah, it takes you through a trip across North America to read, to, to learn some of the things, exciting things that are happening with digital libraries helping preserve and strengthen indigenous cultural sovereignty, to kind of reverse the legacy of glam institutions stealing from indigenous people. So yeah, there's some interesting stories there about kind of pioneers, indigenous librarians who paved the way. If you're a metadata nerd, there's a lot of stuff around- I know. [laughs] [Thomas] I know my audience. Uh, there- [Jay] [laughs] [Thomas] ... there's a lot of stuff about how do we make metadata kind of reflective of indigenous worldviews and not just from a Western, [Thomas] you know, view. We're hearkening back to our discussions about the, the subject headings. Yeah, some things around digital repatriation, which is the idea that, you know, if the artifacts were stolen and were, are now in a colonial institution, why not let the colonial institution do the expensive work of preserving it, but through digitization? Then we can build our own, as, you know, indigenous nations have been building their own digital archives and libraries with the digitized version, with their own metadata, with their own protocols around access, possession. If you're also a copyright nerd, there's a lot of stuff around there about how copyright fails and what we can do instead. So that's perhaps to, to stay on the theme of a technology-positive future, that's one way that, yeah, technology can, can help us do things better than we've done in the past. And otherwise, if you read French and you're interested in neutrality and controversial events, I just published an article in French, which is peer-reviewed, so [laughs] will not be read as much as a zine, but maybe it will, about... It traces a history of controversial or contested events in public libraries in Quebec from 1960 until today- [Jay] Mm-hmm [Thomas] ... to have a discussion about what do we do when intellectual freedom and inclusion clash. Uh, it's kind of based on the equivalent book that was written by Whitney Kemble, which is a librarian at the University of Toronto, which I want to give a shout-out to. She wrote a book called Contested Spaces that traces this history in Canada, so I kind of wanted to do the same thing for Quebec in our little distinct Francophone context. So yeah, those are some other things that interest me and that have been published recently. [Jay] Nice. [Justin] All right. Well, thank you so much. I will add anything you want to the notes so people can find it, and if you want to put your socials in there. If not, I understand. It's a whole other conversation we can have about making our algorithmic friends. Luckily, when you have a podcast, you just meet other creative people, so you don't have to just find them by clicks and likes. Truly useful. Highly recommend it. Get a project, go outside, fall in love. [Thomas] Yeah. Yeah, I actually don't have, like, professional socials except my LinkedIn, but I want to put that there because, like, I don't know, it's just too much pressure to, like... I know I... When my indigenous libraries piece came out, people were, like, screenshotting, like, you know, people on Blue Sky who liked it, and it was so nice. And I was like, "Oh, I should be on Blue Sky, and I should..." But part of me is like, then everything I post, I have to, like, assume will be screenshotted forever, and I don't know if I want that pressure in my life. So for now, no socials. [laughs] But- [Justin] Yeah. I... Same. I... All of mine are pseudonymous. I don't- [Thomas] Right. [Justin] Just, just my first name, and I have, like, no personal socials. Like, my coworker's like, "Are you on Instagram?" I'm like, "Yeah, but not for you." [Thomas] Yeah. [Justin] Like, not, not to talk to you. Like, I talk to you at work. Like, we don't, we don't gotta bring Instagram into this relationship. [Thomas] Yeah. [Justin] That's for looking at, you know, animal pics and web comics and not people. Well, thanks so much, Thomas. I'm really glad you came on. [Thomas] Yeah, thank you for having me. I hope... I feel like I got... I feel like I was, like, really stiff at the beginning, like I was interviewed on, like, national television, but then I hopefully, like, loosened up. I tend... I got, I got media training when I was doing climate activism, and it, it was my- [Justin] Oh. [Thomas] Really, like, your 30-second pitch and eight-year message and everything, and, like, so I have to deprogram that brain and be like, "We're having a conversation." [Justin] Mm-hmm. It's tr- it's tricky sometimes. A lot of people are used to speaking professionally, and so they, you know, only know how to present or not. And, you know, it's, it's... That's the fun part of, like, having library people on, is you get the wide range of, like, people who talk and don't and have various levels of ability to, you know, converse or present an idea. So that's, that's why we have the notes document, and that's why we have, you know, the, the pre-show, the pre- the pre-show idea of like, "Eh, don't worry about it. It'll be fine." [Sadie] My favorite is when somebody goes, "Wait, can I swear on here?" And we're like, "Yes, please." [Jay] Girl, I, I said "fuck" so much, Justin had to put an explicit tag on it. [laughs] [Justin] They will get mad at you if you don't tag it explicit. [Jay] They will get mad at you. [laughs] [Justin] And good night.
Editor is loading...
Leave a Comment