librarypunk - 144 - BIBFRAME must die

Episode Date: February 14, 2025

It’s in the title.  Media mentioned BIBFRAME Must Die,Jeff Edmunds https://scholarsphere.psu.edu/resources/fc19faee-70b9-44b3-9346-18e40a2cd990 BIBFRAME Must Die, Part II: the Official RDA Toolkit, JEFF EDMUNDS https://scholarsphere.psu.edu/resources/40be29ce-4342-475f-b380-d8ed065b3643 BIBFRAME Must Die, Part III: A Brief History of the Future of Cataloging https://scholarsphere.psu.edu/resources/6100de9a-5f6f-400f-90ea-fe517d16152d  “Missing the MARC: Utilization of MARC fields in the search process.” https://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=1314&context=lib_pubs  Transcript: https://pastecode.io/s/ssabs0oo  Join the Discord: https://discord.gg/zzEpV9QEAG

I'm Justin, I'm Skullcom Library, my pronouns are he and they? I'm Sadie. I work IT at a public library, and my pronouns are they them. And I'm Jay. I'm a cataloging librarian, and my pronouns are he him. No guests, just the crew. I also didn't get any library news for this one, but what happened with the Proton guy? I didn't really understand it from the Mastodon DMs or whatever, the Mastodon posts. Oh, just that he's like, J.D. Vance is good, actually, and the Republican are better about lobbying against big tech, TM, than the Democrats. Sounds like Coke.

Yeah. Sadie, you were muted. Oh, yeah, I just said that was pretty much my understanding. It's just chilling for J.D. Vance and gross. Yeah. It's kind of what I was worried about when you told me that Proton has like the storage stuff. And I was like, but what if the Proton people are like even worse weirdos? They do have like, you can do like, they have a crypto wall.

wallet. Oh, boy. Yeah. But, like, all the privacy people, weirdly are, like, into crypto because you can pay for VPN without leaving a paper trail or something. I don't know. Like, Mulvad accepts Bitcoin as a form of payment. But there is a paper trail.

It's literally a distributed ledger. It's everywhere. You can also literally mail them cash in an envelope, which is, like, literally why Mulvad, like, they're, like, the only, like, the true heads know, like, if you really don't want to do anything. You just mail $5 in an envelope to Mulvad for your account. Our sponsor this

week is Mulbad. Send them cash. I weirdly used to see Mulvad like bought ads for the for the T like in the MBTA stations. There would be like change up your like your location tracking like get

off at a different stop than you usually do Mulvad. Like it was like weirdly like little privacy tips and then it'll be like Mulvad VPN or like they have a new browser now that's basically Tor and I think they do it with Tor but instead of it being a Tor network it's a Mulvad VPN. Like if you want the version of Tor that's a VPN instead of an onion router, a Mulvad made one. That's apparently pretty good.

It's still based off of the same Mozilla under backing. The Mozilla VPN thing or the Mozilla Tor? Browser. Yeah. Yeah. Yeah, like Tor is based off of a Mozilla browser, right? Think so. Yeah.

Makes sense. Yeah, I've been thinking about switching up my browsers again because I feel like Brave, it just doesn't do anything, all that impressive. I just use Safari. It mainly, and this is going to sound so stupid, because on my computer, I can do Apple Pay with it. And also, if I have to get a verification code sent to me via text, it will auto load it into the field and make the mark the message is red. Like, it does have my phone.

that's good maybe I should use Safari on my phone more ecosystem I mean on your phone it'll do it no matter what you're in but on your computer like Safari you have to be using Safari for that to work but on your phone yeah oh is that Apple doing that I thought that was a bit warden and last pass were doing that no that's that's Apple doing that I only noticed it after I put last pass on my phone so maybe also was around the same time I got an update yeah and it's it's not like the multi-factor authentication where it's an authenticator sending you a code. It's like if you get a code texted to you. What Apple will do will be like, here's the code that just got texted to you, and then mark the message is red. So you don't have to go click the stupid message. I'm lazy. Yeah.

I think most people are. Speaking of notifications, I got notification that Guillermo del Toro was following us on Blue Sky, because I went through everyone Jay was following and just started following that. because I was kind of getting bored with my feed. So I was just like, all right, well, if Jay's already following the person, I'll just follow them. And I think Guillermo del Toro was on that list. And I think I accidentally hit followback when looking at his notifications and followed us for a second. Because as soon as I saw it, I was like, who called himself Guillermo del Toro?

And then I clicked on it. And I was like, there's no notification there. And I looked it up. It was like, oh, that was him. For a brief moment. Very funny. It was glorious. That or he didn't like what he saw.

There's been followed by a couple like, well-known people. So we're followed by a lot of authors and stuff, who I'm like, I don't know, your books. I don't know who you are. I think because we have David and Chuck following us. A lot of authors are like, oh, okay. They're librarians. We like those in theory. They must talk about books.

Yeah. Sometimes. So what are we going to talk about today? Not books. Yeah. We're going to talk about Bibframe, but actually we're going to talk about. these three, and there's probably going to be more, but so far,

as of yet, there have been three non-scholarly articles. It's a high old bib frame must die by a cataloger I believe you pinned. And yeah, I have read these, and then I've also in the Radcat list serve, which

if you're a cataloguing librarian and you're not on the Radcat list serve, you should get on the Radcat list serve. It's where all of the people who aren't aren't Libs go to talk about cataloging stuff. It's auto-cat, but good. And yeah, so then people have been talking about these. And I want to talk about them because, like, I agree with them, but I also have some critiques TM of them. And Jeff Edmonds, if you happen to listen to this ever, thank you for writing these and getting people talking about this. And I hope you take

our criticisms in good faith the way we... And if you're a catalogger out there, shout out, join the Discord. Let us know what you think about these. Because I'm honestly curious what other cataloging librarians think of this. Yeah, because like, I will come right out of the gate

and say that I'm a bit of an RDA apologist. Gasp, I know. And these articles should be retitled why RDA must die. Not why Bipbrae must die. but I, yeah, so I'm a bit of an RDA apologist, but I, the, the reasons I like RDA are all to do with the original RDA toolkit and not the official RDA toolkit. Anyway, I lost my train of thought there and I'm going out of order. So, Justin, do you want to do this or you want me to do this?

Yeah, we can start off with what is BIP frame? So we did an episode 135 about linked open data and I think we talked about Bipframe and that. But Bibframe is a link data planned to be a replacement for Mark, but now it's discussed more as a successor to Mark, because Mark is old and therefore bad, and you got to use three number digit fields, like in the olden days of computers. And, I mean, Mark does have problems that are kind of unsolvable because it's so ubiquitous, so it's like, well, what are you going to do about it?

It's why a lot of legacy stuff lives on Fortran still, because it's like, well, unless there's a Y2K about to happen, we really don't have a reason to fix this, or spend all the money to fix it. Yeah, and I will say that, like, link data in libraries is not a totally out there idea. Because, like, I remember when I was in grad school, and just to let people know this was 2015 to 2017, that, like, the way that we were taught, like, linked data, because I took an ontology development course, like, I made a link data ontology in grad school. I've been in the shit, right? And I also took a metadata, like, in theory and praxis course, praxis, ha, practice. I just got back from an organizing meeting, so you know, it's on my head right now. And it was like, you know, in premise and all this, they use like URIs, right? They're like linked data ontologies, even if they're not, right? And I will say that like, especially in Europe, digital libraries.

So libraries have like scanned special collections and archives or like fine arts, especially in Europe, have been doing linked data for a long time. Like the Europiana, which is like their version of the DPLA, has been like a link data environment. I don't think it was ever a bib frame, but it was like using the concept of URIs and like semantic linking shit together.

And then when we learned about bib frame, especially like I've been at conferences where there have been like bib frame shills doing presentations. And their main selling point was if you have a patron from your library who is searching, for a book on Google.

In the little knowledge bar in the side, it will show up that your library has the book and they can get it. That basically using link data is a way to expose your collections to search indexing. And that was the main selling point

of Bibframe when I was in grad school. People didn't give a shit about connecting shit. It was expose your collections to Google so that people can find them on Google. And that was like it. That was what people cared about. So I'm curious how you two heard about Bibframe or how it was like sold to you ever at any point.

That's kind of pretty much the same thing that I always thought was that I was just like, oh, it'll pull it out and put it in places people are already looking. And like I feel like that's the purpose of a lot of discovery layers these days. It's to do something. Discovery layers don't even really do that. They'll expose it to Google Scholar. Huh. Well.

But Discovery layers mainly just bring in journal articles along with your books. Yeah. Well, yeah. And I think I've always kind of struggled to figure out what exactly the purpose of Bibframe is. I remember at one point in time trying to read about it launching off of something that I read on Tumblr and was like, I don't get this. I don't think I can wrap my head around this. like some parts of it, the open link data, like the previous episode we had, got that. But Bibframe's particular, I don't really understand.

I'm going to be honest here. What's RDA? We will get into that. Okay. Yeah. Yeah. Okay, because I don't remember what RDA is. Put a pin in it.

Justin. What's your Bibframe relationship? I don't know if we really ever talked about it in library school that much, except that Mark was old and Bibframe was supposed to be. be a replacement for it. And Bibframe could link more things together because a mark record is just a document and it has access points. So it acts like a physical document, whereas Bibframe could link things.

And really the way it was explained to us more was like the whole semantic web would be making the web understand human types, like what we talked about in the last episode where The real thing was allowing computers to actually reason through things because you could form triples and you could make sentences and data would be linked together and librarians and stuff would be the ones like cataloging books and publishers would catalog books this way and that would link all of publications together which was always like I always thought well that's a whole lot of labor and I don't think librarians could do all of that but it would be nice. and I always thought, you know, kind of the brute force way that we do it now, which is like through full text searching and like algorithmic matching and just cramming more and more data in there

was kind of like the stupid semantic web because like it doesn't really teach the connections between things. But I think at this point... It also is limited by just like, what do you do with a print collection if something's not scanned? Then full text searching does shit for that, right?

This is a preview into my main grievance with this whole conversation. Yeah. But when I, like when I mentioned that in the last episode as well, it's kind of like, yeah, but this, this type of linking through big data is just what's going to happen because it's actually cheaper and easier to do, even if it's messier and creates like it's just good enough linking. but kind of for most use cases, you only need good enough. So as we are learning with generative AI,

it doesn't matter if it's accurate 90% of the time, but if it's 80% of the time, that's good enough for people to use it and say like, oh, well, if it's only wrong one in five times, who cares? And it turns out, yeah, people don't care. Yeah.

And so these articles, some of their main arguments against Bibframe, are that they argue that it's already outdated because Bibframe was conceptualized before like algorithmic searching really came into fruition, right? Like back when you could actually still use Boolean on Google, you cannot use Boolean on Google more.

Like you can still use some of the operators, like the wildcards and shit, but you can't, the Google foo that we used to call it or whatever isn't really much of a thing anymore. But BitBframe imagine that Google would still sort of Like, yeah, Google was always an algorithm, but not the natural language processing kind of algorithm that it is now. And so Bitframe already didn't anticipate the way that people search for information, including in library catalogs for the most part. People don't use library catalogs the way we want them to.

People will copy-paste citations, people will ask questions and type sentences. Yeah, like people don't search the way that we librarians envisioned that they would. no matter how much we try to teach them otherwise. And so, Say, do you had a comment in the notes. Would you please explain your comment? I just, that one line made me laugh. It's seemingly enamored of the auricular pronouncements

of Tim Bernersley. Proponents of Bidframe have advanced an agenda divorced from reality. Like, that's particularly cutting to me, it seems like. Like, you dissed Tim Bernersley, you dissed everybody who believed in Bidframe. is just an excellent sentence all around.

Now imagine the world if instead of Tim Berners-Lee being the guy who shaved link data, it was our boy Ted Nelson. Imagine how good Bibframe could have been if it was inspired by our boy. You know, shout that to Ted. Yeah, because Tim Berners-Lee was all about like, like, the way I've described it is like reconstructing a like pre-tower babbled, like divine language, like everything. is these like URIs that the computer understands, like no matter what language, it's all just numbers, right?

And I'm like, no, that's not how this should be. Damn it. So yeah, that's a cool idea. Yeah, Bitframe and also Bibframe is of the link data forms. I think it's RDF, I believe is the format. Bibframe uses. I think the name of the editor that most, that I think like LOC uses is synopium.

or something like that. No, I don't know. You can do PIPframe and Mark Edit, too, by the way. If you use Mark Edit every day and notice, I've been open Mark Edit in like a couple years. Yeah. So, yeah, that's Bip Frame. I can go ahead and answer the question, what is RDA now as well, I think? So RDA stands for resource description and access.

And in cataloging, there are like standards and there are like schemas. and there are like schema and then there are vocabularies right? Mark 21 is a schema. Bipframe is a schema. Dublin Corps is a schema. This is the

structure, the house that the metadata lives in. And then standards are things like RDA or AACR2 or descriptive cataloging for rare materials or whatever it's called. How do you put

things in the house. That's what the standards do. And RDA replaced, kind of, again, kind of, a ACR2, which stand for Anglo-American cataloging rules or whatever. I never learned a ACR2 because RDA had already happened by the time I learned how to catalog. However, most catalogs in this world are mixed between RDA and ACR2 and the Library of Congress has not a fixed. adopted RDA yet. So if you are in connection and you find a library of Congress record, it will not have the 040 subfield E RDA. The two six instead of a 264 for the publication info, it will have the two, the old school 260, etc. They have not officially adopted RDA. And like back in the day, you just like, buy a fucking AACR2 book and be like here, here's how you catalog.

your standards for cataloging and Mark 21 have at you. AACR2 was functional but also had some stupid rules like abbreviating everything including like making abbreviations into Latin and people complain that RDA is too hard but then

they expect you to do all these Latin stupid abbreviations for AACR2. This is why I'm like most of the people who are mad in this conversation are also wrong about other things and I'm the only one who's right. RDA is a general improvement in a lot of ways over at ACR2,

despite its very significant flaws, because I agree with a lot of the criticisms of RDA, but I think people also aren't noticing where RDA is right sometimes. And so RDA comes along, and it's like, hey, what if, instead of just cataloging for Mark, we instead created these rules of description that can be applied to all kinds of schema, like Dublin Corps, or whatever they use in Europe,

or whatever, Like, here are like the ideas. Like, what's a title statement? You know, what's an imprint statement? Like, what's an addition statement? So instead of cataloging for mark tags, you get these, like, chunks of information, right? And the chunks of information follow, like, chunks of information about a work, chunks of information about an entity, chunks of information about a manifestation, and chunks of information about an item.

me. Everyone can understand what the I means. The item is this exact copy of the philosophy of social ecology by Marie Bookchin that I just grabbed off my bed, right? This is an item, right? The work is when you go to a catalog in a library and you search for the philosophy of social ecology, you kind of don't care if it's this exact item. You care about like the concept of this book that you think it, right? Then there are entities and manifestations and I never really understood the difference between them. Except manifestation is like all of the books that like are printed exactly like this, like this edition and everything, but I have this item.

But all of the ones that are just like this, like if I bought a second exact copy, Those would be of the same manifestation, but they're two different items. Stupid bullshit like that. That's not really helpful. Wumi's dumb, and that's not how people live. And notoriously, Wimmy is really bad for things like music. If you've got the Nats de Figaro, that is the work. But then manifestations in bullshit include, like, a cast recording.

It includes the score. It includes the libretto that doesn't have the score. it includes a videotape. It includes, like, all of that, is technically the same work Linocated at Figuero. And so in Primo, I would always have music librarians

yelling at me when I managed it being like, why is it so bad with music? And I'm like, that's not my fault. And I can either turn it on or off. Harvard turned theirs off because it's so bad with music. And that's the reason why, for example. But RDA thinks it with me.

You have, like, recording bullshit about the work, recording bullshit about the entity, recording bullshit about the manifestation and recording bullshit about the item. And that's like how you do RDA, right? RDA also thinks about like how things are connected to other resources,

right? Which now Mark does. There are fields in Mark that connect it to other resources that like can hyperlink in catalogs because of RDA, I think. Does that answer your question as to what RDA is? The main problem with RDA is that it costs money as a subscription.

Who owns it? Who do you subscribe? Who are you subscribing to? RDA. It's a company, like the RDA steering committee or something. So it's, yeah, it's proprietary.

And also, I learned original RDA, which is pretty functional, I would argue, and maps very clearly to mark. If you go on the RDA website now and look at their example records, they, these are pre-official RDA, These are original RDA.

They haven't been updated since like 2016. In 2017, so right when I'm finishing grad school, we start getting official RDA. Original RDA has not been updated since then. It is still available. It is the one I have been instructed to use as a cataloger. I didn't know official RDA was a thing because I haven't done cataloging since my grad school or had access to RDA toolkit, right? Official RDA is the most incomprehensible bullshit.

I have ever seen in my life because it's so hyper-focused on the link data aspect now instead of just like a different way of thinking about describing resources, right? Because like the thing I like about RDA is like the rescuers explains itself, right? You don't do abbreviations and bullshit. You don't abbreviate the title. You transcribe exactly as the resource is listed on the resource. There's a typo in that fucking title. You type that typo and then you provide an alternate title.

but like the point stands, you transcribe, the resource describes itself. Whereas at ACR2, you abbreviate and like put dot dot dots and all sorts of stupid shit. I like that. Official RDA, I can't even figure out how to use the website. I don't know where I'm supposed to start. I have tried. I'm like, what is going on here? So that's bad.

So official RDA, I think, is what these three articles is really mad about. Sorry to derail us, but yeah. So it's not even Bibframe, really? No, because no one fucking uses Bibframe except for Library of Congress in Europe. Probably University of Washington or some stupid shit. Like, because like in Primo, for example, in Alma Primo, they can like crosswalk mark to Bibframe for you. Right? So like, or like people are doing Mark XML instead of just Mark, right?

They're doing XML records. Who fancy. Mark XML is really complicated. It's stupid. No one actually uses BipFran. It's stupid. No one likes it. It's hard.

I think it's like Europe and Library of Congress uses BipFram. T, B, fucking H. But everyone's got their panties in a twist about RDA. And have since I've been in cataloging, like, since I was in grad school. Like, and I always, and like, no one ever argued for like, oh, ACR2 was better for patrons. They just complained about how they do. didn't like it. It was hard for them to understand. Because there's some like, RDA is more theoretical, right? There's some like theory that you need to understand kind of for RDA to make sense. And I was in grad school and so I was taught it. And so I was like, okay, this makes sense to me and this is the weird kind of conceptual bullshit I like, cool, I'm on board with this. But like, you know, no one pays for professional continuing development, right? No one pays for that shit, which is an argument I'll get into later. Anyway, I see the user tasks thing, which honestly, I learned about Ferber in grad school, and I, like, don't remember any of it.

I was like, Ferber, yeah, whatever, fucking whimmy. I don't remember what any of this, like, user tasks, if la bullshit is. So if Justin, do you want to illuminate for us? Yeah, well, Ferber is the entity relationship model. Right. Okay, yes, which I kind of agree with. So a user tasks of retrieval and access in online library catalogs and databases from users' perspective. And it's separate from cataloging standards.

It's a model that explains work expression manifestation item, which is whimmy. So it explains how work is an item is an exemplar of a manifestation, which is an embodiment of an expression, which is a realization of a work. Which is very Buddhist, I have to say. This is like Vajriana emanations out of the ground of being bullshit. Like, the Dalai Lama is an item. So a work is like a work, an expression might be additions or drafts, and a manifestation would be the physical embodiment. And then an item is a singular manifestation.

So one particular book. So copy one, copy two, copy three in your library. Yeah. So like the item record, basically. And who owns official RDA toolkit is owned by ALA, CLA, and CILIP. That's right. I don't know what's, I don't remember what Killip. I don't chartered Institute of Library and Information Professionals, but I don't know.

That's the UK. Okay. So I guess the three of them own it together. Yeah. And then Bibframe is Library of Congress, Stanford, UPenn, and U of Alberta have adopted, or have been pushing Bibframe and are involved in. And that's like it. Basically.

So the argument in the BipB Primis die papers is that the works, expressions, manifestation items is kind of absurd, probably always was. Users have record-based thinking, which is like they think in a Mark-like way. So they think kind of like a card catalog, which is they want a record with a single resource.

And Bipframe turns it into a cloud of entities and relationships between them. Right, which is why it's so annoying for music. Right. So people don't really think about it that way. They think more like, okay, this item, this record represents this item. And here's the information about that item. And it's all one thing. Right.

So most users don't have a need for those conceptual manifestations. Kind of the only reason that I ever understood was the purpose of it was to say like, well, this particular item is in this library and this item's in another. one. And if you wanted to know, like, where every single version of this edition of Hamlet was, then you could find all the connected libraries that have this exact edition. And you could find every single one that has how many copies of it. But, like, who needs that information? You just need to know what's at your library. Whereas, like, I would argue that, like, one of the use cases for this that would make sense, weirdly, is where holds are concerned. So, like, this is just,

came up at my library, like last week, where there was some hot new book that won a stupid award or something, and we had to buy a bunch of copies of it from a local bookstore because it was so popular that it was on back order through our suppliers, right? And they gave us a paperback instead of a hardcover. And we had already hard covers of it. And the paperback, like, sometimes you'll see an OCLC, like, a record, as long as it's a record. As long as it's a, like the same page numbers will, and as long as it's not different editions, will include the hardcover and the paperback as one record.

And we'll just list both ISBNs with, this is the hardback ISBN, this is the paperback ISBM, right? If they're not like different editions, they could share the same thing, as long as the pagination is not wildly different, which it shouldn't be. This one said like first,

it said in it both first paperback edition and first hardcover edition. It had both edition statements. in the thing, and we had found that another library had made a record for this paperback one. And we were like, if we did that, if we used that record and brought it in, this would not help our holds situation, because we couldn't then use these to fulfill the holds on the other one. And there's a lot of fucking holds on that other one. And we bought a lot of these.

So we're like, you know what? It's the same page. Like, fuck it. We're just putting it on our hardcover one because we can't differentiate between payverback and hardcover to fulfill holds, right? Like, unless someone, like, has a fucking reason. But for this, it's like, we don't give them a, we don't let them make that choice. So, like, if they were actually separate editions and separate records and everything,

then, like, we couldn't do that. But, yeah, holds are like ILL. I feel like this is where this makes sense. Say to you, we're going to say something. Yeah, I was going to ask, like, how do you treat large print editions different from that? Because that is one use case that I see people being like, no, I want this very specific, like. Well, that's a separate edition. It's just considered a separate edition.

Okay. I think so, yeah. I've seen very similar things happen in, like, that's a pretty, I think, perennial public library deal is like, which ones are going to satisfy? which holds so people don't get pissed off or when is it the last copy of something when you actually have like five hardbacks still left but you're out of paperbacks like so is it the last copy or is it the not last copy kind of kind of thing so but there are people who are like no I want the hard back and we say we don't care yeah and and you kind of have to just make yeah the judgment call on that like oops sorry you get what you get when it's super popular when it calms down you can probably have a

better chance at what the exact physical manifestation you want. But yeah, no, I just, large print is one of those things that also comes up in discussions like this. So, but yeah, as a separate addition, makes sense. But would academic librarians have thought of that? No, that's the other thing. They, yeah. And that is my main criticism with these pieces, is they are so firmly rooted in academic library thinking, of which I used to be guilty. because I did not yet work in public libraries. But now that I have been on the other, I have crossed the Rubicon or whatever the fuck.

I've noticed some shit now. I'm like, oh, like, these people don't think about anything. And to be fair, not a lot of public libraries do their own cataloging anymore. They either buy them from Ingram or Baker and Taylor already cataloged, or there's consortium, like cataloging consortiums that'll do it. But if you do have catalogers at your library,

Like, still, like, this is a use case, and, like, I feel like this entire discussion has been dominated by academic librarians. And that's one such instance where I'm like, ha ha, and more will come up. Not least the fact, and I will reveal my big Trump card at the end that I put at the end of the notes. But, yeah. The next thing, bullet point, though, is the labor costs. What do you think is the real actual issue here? in that, like, I mean, to quote, like, the guiding principle or whatever that Justin brought out, making things easier for computers at the expensive people.

But that's not really a labor cost thing. That's just we're changing the way that we frame what this is actually for. The real labor cost is like the IMLS grants or whatever for implementing RDA aren't existing anymore. RDA costs money to subscribe to and not every library can afford that. And also, this is, I think, is the other actual, actual, actual problem is we don't pay for professional development and continuing education. And catalogers don't leave their fucking jobs ever. And so the only people who are learning RDA are library school students who then can't get jobs in cataloging if they don't already have experience in cataloging and then how do they get experience in cataloging. but there are just like aren't jobs doing cataloging.

And if there are, they don't pay shit or they expect you to know their specific ILS or to know how to already do all of this shit because they will not pay to teach it to you or to train you or anything like that. And I think that's the real labor cost here is we aren't training people who are already in jobs how to learn this stuff. Because shit does evolve. Shit does change. And I think sometimes people are just being sticks in the fucking mud. But part of that is that they're not being taught, like, and paid to learn how this works or to be part of changing it. I think that's one of the actual labor costs here. Like, just teach people OG RDA and then also make it free or like a book you can buy,

which I think it can't be a book you could buy, but still. That by itself wouldn't get people into using Bibframe because the labor costs to transfer over to. Yeah, getting over to Bibframe has dubious benefits and it requires recatalogging pretty much everything. And there's already so much cataloging work that needs to be done to make things discoverable. So yeah, the thing I always bring up is it's a paper. It's called more product less process. And it's about the fact that people are over describing things in special collections. And then most special collections still have items that never get cataloged at all.

So no one knows that they exist. So it's better to just have a minimalist catalog record or a minimalist finding aid rather than nothing. Because if it's nothing, no one knows that it exists. this, probably not even the librarians. So, I mean, I remember one time we were looking for something and it was at the bottom of like a pile on a bookshelf that had like just a ton of like postcards and newspapers on top of it. And it was like really important donor material that we couldn't find. And my supervisor was the kind of guy who just like pile stuff on his desk like a mile high. So it ended up on the bottom of one of his piles.

Yeah, like one of the core texts, my sort of shaping ethos as a cataloger, is called Making Keywords Work, which makes the argument that like really rich catalog records are good or know that subject headings are still valuable in a keyword searching environment. Because we don't make artisanal records, the main thing that keyword searches are pulling from are just subject.

headings still. And so doing subject authority work is still important for that reason. You have to make the keywords work. Not that you have to make artisanal records, but if a keyword search is going to work and you don't have all this full text to be pulling from, it has to pull from somewhere. So like thinking about your access points is what's actually important for discovery. Or just relying on a shared vocabulary. So like the paper I wrote about keywords in archaeoanithology paper with a biology professor,

it was because basically this subgroup of scientists needed to come to an agreement on what they were calling things. And they needed their own vocabulary and to start tagging their papers with it so they could find each other's work and stick to it.

So they needed, they needed a common vocabulary and they needed to implement it in the keywords for their papers. when they submitted authors, supplied keywords. Yeah. So, yeah, that was another reason why I had to accept it linked up with data is not a good idea because unless we had state-sponsored jobs to clean metadata around the world all day, which would be good.

It would be a good make-work job if we needed full employment. But there's probably other things we could ask people to do like Great Forest debris and do others, you know, WPA kind of jobs rather than converting the world to bid frame. Yeah, I mean, and that's, and again, like, this is Bibframe must die, but actually it's argument against RDA. In a list survey, I'm in, and I won't say who's had a, one of the arguments against RDA that this person who I think is correct in all things said that, like, one of the main problems with RDA, or that there's, the main problem with RDA again is that it, like, costs money. And so this is a class issue. Like, again, not all libraries can afford it. But also, because then not all libraries can afford to use it, it causes disparities amongst records.

So not all libraries are cataloging to the same standard. And this is ignoring the fact that like rare books has its own bullshit standards that it follows. Right? Like you'll see like DCRM or DCRB, whatever, like records in WorldCat, basically. But that like my catalog has an ACR2 and RDA in it. And we have been instructed like for time, say, not to correct ACR2 records to be RDA comply it, unless we're already doing a lot of work to the record.

Like if we're already there, okay, make the changes. But if not, just like fucking leave it because it still works. But like because between ACR2 and RDA, there are some differences in how you approach describing things. This then causes inconsistencies and disparities between places, including like indexing issues, right? Like, if you correct a record in WorldCap to be RDA compliant, and this changes the title, then that causes a discrepancy between that record and then every library who uses that record, right? Which is like, could be an API, could be like an API issue.

It could be like there's all sorts of issues that this could cause or like upstream if it gets pushed to them and stuff, right? So that that was like one of the main arguments with RDA is that it causes messy, data because not everyone can afford or understand us. Including Library of Congress. They don't even officially, they haven't officially adopted RDA. They still see those 260s everywhere and they piss me off. I'm like, so one of the points this article makes is that discovery layers already do a lot of what we would want bib frame to do, which is to pull in data that is out in the open web, I guess, or in other places. The point that he makes is that marked data is far less importance in full text.

and other sources of indexable data. 80% of all records displayed in search results come from non-mark records, even though non-mark records are only 60% of the database. And I grabbed that study, and I threw that in the notes, but I didn't have time to read it. And I would say that's an academic library.

Discovery writ large happens elsewhere, either in discovery layers or in Google or on the open web. And also, I would say a lot of people, either, especially in academic libraries that use IP authentication, never even really go through the library because they're already IP authenticated on campus or on the VPN. So a lot of people don't really know how to use our discovery layer when they're off campus to get access to things or just the way that Tremo is supposed to work is it like tracks your cookies or something and keeps you logged in. I don't really understand it. But it's supposed to keep your browser logged in. Yeah.

supposed to keep your browser logged in so that you don't have to keep logging in, so you don't even notice when you're using your library subscription sometimes. And also the same thing where Google Scholar will say, your library has this because you set it to your library. And it'll take you straight to your library's copy of that article and authenticate you in, which is a problem I always had when I was working my first job, because all of our students were off campus, commuters, and did a lot of their work off campus.

And so I always had to make sure that the proxy was in the URLs that we shared out because that's what forces them to log in when they're off campus. And so I also had to start teaching people that again in 2020 when people were off campus for the first time and saying, okay, you really need to click the share button because that's going to inject our proxy into the link before you share it out to your students because they're going to ask you, why can't I read this? So when you're sharing something from the library, get the share button because that's going to throw the little special URL that's going to tell it. You need to log in. You should be doing that anyway because not all library authentication is IP-based. Yeah, I know you should, but I'm just saying these behavior. Yeah.

We also have Libkey, which is, I don't know how many people use it, but it works really well. And of course, on paywall, if you're logged in, it will shine green when you have access to it. So it'll take you straight to the PDF with one click. So even if you do work at an academic library and have access and don't need on paywall that much, it still is the easiest way to get a one-click PDF because it'll just show up as green and you can just click it and go straight to the PDF so you don't have to click around five times.

Yeah, which like, again, my main argument, my main arguments here are that that's still so academic focused because the majority, because even the public library where I work uses a discovery layer. It's a public library discovery layer called Bibliocommons. And there are other public libraries that use it, and it will bring in, hoopla stuff and Libby's stuff and other.

I don't think we have it set up so that article show up in it, but it also is like what our website backend is. So like reading lists will show up in your search results, for example. But that discovery layer is powered by the fact that like all of our like hoopla stuff is also Mark records in our ILS, right? So all of our digital. discovery layer is mark results, unlike in academic libraries where it's all just database results.

Book reviews. In public libraries, even if, yeah, in public libraries, even if, even if they have like ebooks and stuff, like a lot of it is still going to be physical materials or and or mark, even in a discovery layer environment. So I don't. And also with discovery layers, like bringing stuff from outside the web, someone has to set that up. It's not going to bring in Google search results.

And it's also not going to expose your stuff elsewhere. Stuff has to have an OAI feed or be part of some community content zone or whatever that you have to turn on. It's a pain of the fucking ass to, because I've had to do it in two separate discovery layers to set up like all of the external. shit and it's a pain in the ass. proprietary ones don't play nice. Primo doesn't play well with Epsco and vice versa. Epsco doesn't play well with ProQuest.

WMS, in theory, plays nice with all of them because it is provider neutral. In theory. But it doesn't have all the databases in there. So I don't even, I don't think Discovery Layers and Bibframe, like, I don't even think that's a correct comparison

to be making, to be honest. like, I don't, Discovery layers aren't about connection. Discovery layers are about just putting everything in one place, but not about this sort of like linked environment where things connect to other things. And like, yeah, I don't know. So I just like don't even agree with bringing that argument fundamentally. Thank you for coming to my TED talk.

Are we looking through the rest of the bullet notes? Another thing the articles bring up is the Duopoly problem. So the creation and maintenance of descriptive and authorities metadata in such an environment, a duopoly, is more challenging than it was. Ownership and control of collections have to ship it from libraries in the printage to multinational corporations. Does it make sense to overhaul the practice of cataloging in such a way that only minimally affects the system? And he's talking particularly about big e-book collections in which collections shift in and out of your discovery, and your collections that you own because they are big packages of e-books.

So I go to buy e-books sometimes and we already have the book, even if it just came out this year. So when I'm selecting for the sciences, it's actually very hard for me to spend my money because we usually have books pretty much automatically from the point in which they're published because we get them in big databases. So I have to go looking for like books about teaching science or critiques of the field. and stuff that isn't pushed out by Rutledge or Elsevier or any of the main publishers that we get big bundles from. So part of it is it's, I guess the idea of having,

I didn't copy these notes over very well, but it was, it's going to be very difficult to convince these companies to switch over to a link data model, which benefits everyone minimally. And Jay has a note that this is public focused, public library is still by a ton of materials in the paper, talks about how everything is basically just a strip mall, like everyone's got the same books. Collections aren't unique.

I still disagree with that. Yeah, that might be true. And it's a particular part about special collections, which are physical assets that are genuinely unique. And it's also the ones that Bibframe are least suited to describe because it was conceived to serve an interconnected online world rather than an analog one. So the data is too homogenous. I still disagree with that. like what academic libraries might all have the same collections because they all subscribe to the same databases,

but even that's not true because some universities can't afford to subscribe to all of the databases, and some might just have only J-Store, or might just have this other thing. So like, with books maybe, that makes sense for academic libraries. A lot of them do have exactly the same thing. Public libraries, even if there's a lot of overlap, there's still, at least for physical materials, like, largely curated still.

There are still, like, collection of element librarians. Like, and, like, when I was at the Gerber Heart, for example, like, all of those were donation-based, and that included a lot of, like, self-published authors, which academic or public libraries will collect a lot of self-published stuff, actually. And, you know, those get catalog records or, like, independent publishers. So like, I also just disagree with, I think this argument's still so academic focused, like the homogeneity of connection. Like, it's, does, then what's the point of world cat, if not to see how collections differ between libraries if we all have the same thing? Like, you know, like, I don't know.

Like, I didn't understand this argument. And also, a special collection, there are things in my library, which probably should be in special collections. but which aren't, because a curator hasn't gone that special and put it in the special collection. But they're very rare, quote, unquote, materials that aren't in special collections, so don't get cataloged as such. Special collections is a marketing term. Public libraries, even, and not even just mine, public libraries have, like, community materials, right? That could, they could be the only library in the world that has that thing. Special Collections is a marketing term.

Is a self-published romance novel that a page and printed out, like, got published, and now their public library has it? And maybe that's the only library. Is that not a rare book? Like, and that's catalogued according to RDA and not whatever rare book thing that rare book catalogers use, where they over-explained the shit out of everything. Like, I just, I don't like this argument, and I think it's very got blinders on to the realities of how things get described

and why things might be in special collections versus not. Also, our special collections, we have shit that other libraries have. Not everything in a special collections is unique. Yeah, well, I think the whole point is he's talking about distinct collections of the physical ones.

And so... Not even with special collections. Like, yeah, a lot of libraries have, like, overlap, but that doesn't mean every single library. If every library had all the same exact shit, we wouldn't need in our library. loan, even among academic libraries.

You know, if it was just special collections, that was the unique thing, which doesn't get in our library loan, you know? I just, like, didn't understand this argument, and I just thought it was, like, to me, this gets as, like, theory-brained as this article is trying to argue against. It's like, oh, RDA is conceptually about linking digital materials and unique whatever, but we're all two fucking homogenous now and the only distinct things are physical. I was just like, shut up.

Like, this doesn't, this doesn't making your point about why Bibframe is bad. This is a like, I don't know, like a retention policy problem. This is a collection development argument. This is a like, let us

own things or curate our own ebooks and articles argument. Like this is unrelated to a metadata. standard, I think. Yeah, I think the second article has a hard time connecting it back to Bidframe. I think it's focused on why RDA data doesn't work with all of the proprietary data that we get from our ebook collections. I think that's the connection there.

Which, like, with RDA, like, if you look at an AACR2 mark record and an RDA mark record, do you want to know what the difference The main big difference, it's the 260 field. It's the main one. I swear to God, that's the main thing. For publishing information, ACR2 is a 260 and lumps it all together into one thing including copyright info. And the 264 has all these stupid fucking second indicators that indicate whether it's publication or manufacture or copyright

or I forget the fourth one. I don't care because you only ever use one of four, unless you're a nerd. Four is just for copyright. literally all it's four. And that's like the main difference, that and the abbreviations. And like the article makes this point

that most libraries who say they are doing RDA, what that means is that they have put the 330, 6, 7, and 8 in their records, which is like automatic and OCLC now. And like, also that they do a 264 feel instead of a 260 feel. And that's like it. That is the big difference between RDA and ACR2.

Or you put like, a subfield e author in your 100 field and a subfield e editor in your 700 field or whatever. Like you say how the person, like what relationship the person has to the thing they described. Did this person write it? Did they illustrate it? I think that's nice. I like that. I think we should keep that.

That's good, actually. So fucking, I'm so annoyed at people who get mad at this, even though I agree with them. but they're fucking stupid nerds. There's no different. It's all mixed in. It's all mixed together. Literally, this was going to be one of the points.

And one day, maybe I'll write this. The Erotics of Metadata article I want to write. We talk about the main reason why I don't like doubling cores because I think it's boring and doesn't look good. Even though I actually also have structural issues with it, I just, I don't like that it doesn't nest.

And I like nested XML records. and like aesthetically I don't like Dublin Core but I made the argument about like the pleasure of cataloging and that most people, their gripe with RDA is an aesthetic one because they don't like it because it's not fun for them

or they don't understand it or it makes them feel stupid like outside of the labor thing and the money thing most catalogers probably aren't thinking of that most are just like I don't understand what a noman is or I don't understand what like an entity is I don't know what all the theory is. And like, why would they? Right?

And especially in official, the official RDA is just like incomprehensible. It's stupid. It's dumb. It's bad. But that like, when I tried looking up, like actual tests of like RDA versus ACR2 for like user studies, there weren't any, right, which should tell you something. But like, also these people weren't like making that argument. They weren't thinking about like, well, is this better?

for the way that patrons search or not. It was just, I don't like it. I think this is bad, which like there is something to be said for authority, right? And expertise. But like, I was like, the main reason people don't like RDA is because it's not fun for them personally. If you just like take your blinders off and you go, what if I just change this 2602 and 264 and then change the second indicator. Congratulations, it's RDA now, basically. I know that's grossly oversimplifying it. But that is in most libraries, that is the way that people interact with RDA is a 260 versus a 264 and then some like relationship designators. I'm not even making this out.

Damn it. I kind of love that you're on the public side, the public library side of things now, Jay. I'm so spicy now. I like these fucking bourgeois. You're that guy in, um, in party girl. Oh, God, I am. When they're sitting there talking about, do you go public or academic? Howard doesn't approve of academia.

He thinks it's for wimps. It is. Yeah. No, I was talking with my roommate about that because he loves that movie, and he showed it to one of his boyfriends. And I was like, oh, yeah, the part where they're at the table, like, bitching about public libraries versus academic libraries. And I'm like, I understand the public people now.

Like, they're right. Academic librarians are up their own asses. I was part of the problem. mean, it is true that most of the original cataloging at scale that gets done is academic libraries, especially at like, you know, fucking Princeton or Yale or UPenn or University of Washington, like even outside of special collections. Like, that's where you're going to get a lot of like really artisanal shit. But, and not every public, again, some public libraries don't even have catalogers.

It's done at a consortial level. or they just buy everything pre-catalogged. Like, it's true that, you know, most of the cataloging labor that happens in this country is probably happening at academic libraries, at least for people who might be doing like original cataloging, right? Or, like, more advanced copy cataloging. That's going to be happening in academic libraries.

But still, people use libraries at public libraries, damn it. me and Sadie, you're going to take over the podcast from you, Justin. Well, it's, I mean, when I worked at my first university, we had no physical acquisitions because we didn't have any money for them. So the library was so starved that it was really just evidence-based acquisitions, and it was all just knowledge-based data that flowed in and flowed out. So, you know, I would talk to people at bigger universities and say, like, you know, you could run a library without catalogers because you just run it on a shoestress.

and everything just comes to you from the vendors. And they, like, didn't believe me. And I was like, I don't know. And then I know a lot of big universities that, like, don't have catalogers. And so it's like a, you know, I was trying to explain, like, the precarity of their position, but they just got mad at me. Like, I was saying their job didn't matter. I was like, no, it does.

It's just that, like, you, like, you can get by without it if you are, like, small enough. If you're big, you have to have a cataloger. But if you're, like, a small community college, like they might not have one. And I don't know, I think that was kind of the point they was making in the second article about we don't own our data. And so like why make it linked open data? Because I think it was saying like, why fix Mark? Because like why do we need our data to be linked and open in this ecosystem where it's dominated by two companies?

So I think that was the connection it was trying to make. Because I wanted to like get that point out because I didn't want to like be unfair to this article. It was just I was trying to like, I think it's not quite clear. what the connection is in that one. I think the one's kind of the weaker part of his argument where it just kind of... Right, and I agree with that point. Like, especially like, again,

I think the argument is beyond, it's not a bib frame. It's like, the problem isn't necessarily bib frame. The problem isn't, oh, because some of the problem is RDA, but the problem is more so like, because like, I feel like, you know, bib frame isn't, like, there are other link data on, there are other like linked data ontologies like used in libraries or like schemas and stuff like it's not like completely unheard of it's just that most of it's not describing stuff you would use mark records for most of it is in digital library spaces again so like digitized special collections and digitized archives but not like tom clancy novels bibframe is for tom clancy novels you know like bibframe is for tom clancy novels you know like bibframe is

meant to replace mark. Bipframe is meant to replace cataloging and not metadata, that distinction, kind of. And then they make everything Bipframe. But yeah, like, I agree with, like, I think Bibframe is a, I used to be like, like, rooting and tootin for Bibframe.

I was like, yeah, this sounds like cool. This sounds like a cool idea. This sounds good. Like, sure, why not? Fuck it. Let's go. I was for it. But, like, it was being developed by, like, L.C. or something and then got,

it's like a private company now or something. And so just like the development of it got put into private hands, I think, or corporate hands or something. And I just think it's a failed project that took too long and didn't evolve at the speed of information retrieval. Because again, like I said, in Europe, they were using link data for fucking ever. And it was fine. Like that's what I was always taught, like European digital libraries were ahead of the game in a way that we just never did over here. I don't, I think what the actual problems are in these articles, if we can, like, summarize them, is, like, the cost, which is around, like, RDA, like, if we're going to use RDA, it should not be a closed standard. It should not be a proprietary standard. It should be open. So that all of the libraries can use it. And it shouldn't be incomprehensible. It's incomprehensible right now. It's pretty fucking down. And there is the open rules for cataloging out there.

People have tried to make. Yes. But I feel like it's going to be hard for a library to make an argument to actually use it. Like we would need to like all as a profession. Be like we're using this now. It is how do you code for it, right? Unless it's basically replicating RDA. And then I feel like the other problem is like bib frame and,

RDA and like all of these standards, not just that they left out public librarians and stuff, like in their conception, but they just like left out most librarians when they were being developed, like, input from them and everything. Like the majority of the profession was not part of creating these standards or saying, we're going with this now. And so like, of course, it's not going to be in touch with the way that people actually search because people who work with patrons didn't develop it.

That's why it's so theory and conceptual without being tied to anything. It's not grounded at all. To me it feels like when people like you have to add a fancy computer thing in order for people to care about it anymore.

We saw this with the digital humanities, which I like but still, and then now we're seeing it with AI, you have to add in the new computer shiny to get money, I feel like that's what's just this whole endeavor was. Like I don't, I feel like these things like Bipframe and RDA and all this, like they aren't necessarily the problem, their symptoms of a problem. I feel like these articles are trying to get at. But like I make in my fucking notes, why aren't we also mad at OCLC?

I know some of us are. But they're also cakekeeping. You got to pay them to use their mark records, right? you got to pay them to be in World Cap. That gate keeps like 90% of interlibrary loan, no matter which sort of method of interlibrary loan you're doing, if you're not a fucking OCLC member, too fucking bad, I guess.

Like at least you can access like Mark and then the OCLC sort of explanation of Mark like for free. And I actually think their explanation is better than the Library of Congresses. Like I prefer the OCLC Bib Formats website over the Mark 21 website, TPH. but like you still got to pay OCLC if you want

because like there are web browser extensions for like Amazon and shit or even like what do you think Google or whatever or any of these things is pulling from for the little widgets that are like oh you're trying to buy this book in Amazon well like your library has it or if you're in goodreads it's like find this in the library or like any of that shit it's just like WorldCat data that's all that is that's just searching WorldCat.

That's what, that's all we want for Bibframe is just to replicate what WorldC already does in a widget, right? But we got to pay OCLC for that to even work. So why aren't we also mad at OCLC? Why aren't we saying OCLC must die? You know, you know. I mean, there are, there are definitely people who are saying OCLC must die. Like, but not as many as they're saying RDA must die.

I believe you on that front, but, I think, I, I see OCLLC and I'm like, oh, God, what now? But, you know. Yeah. I wasn't really taught AACR2 or RDA because they knew RDA was like literally about to come out the year, my second year of library school. So they're like, here's AACR2. Don't get used to it. But RDA is not out yet.

So we can't teach it to you. So I never had any real interest in cataloging because they were like, well, I guess I'll figure it out eventually. And then I remember they made all the librarians like on the. the same day do like an eight-hour RDA training. And our cataloger was hissed. She had to be in that meeting all day on her computer. How else do you do professional development and continuing education? Well, she just didn't want to be on an eight-hour meeting.

Going to school stocks, I don't know what to say. Like, wow, you got to learn shit somehow. God damn it. Well, she'd rather be cataloging books and getting them cataloged for the first time because it was a dark archive. Yeah, well, like even non-RDA, like standards should be living and be able to change and you got to learn when those changes happen.

I don't know. People get so like, I learned the thing once. I know it forever. It's like, no, you don't. It sounds like a labor issue, honestly. I'd rather be doing this thing. It sounds like you need another cataloger. Yeah, so that you can go take your professional development.

Anyway, it's got me riled up more than I was expecting. I also want to say I did a control F in all three articles, and at least referring to public libraries, the word public shows up a grand total of zero times across all three articles, whereas large academic library shows up like six times in one of them and then shows up a couple times in the others. Just saying. All right. God damn it. I also went through the open rules for cataloging page. And like a lot of them are blank, but it looks like they're still active. But the idea is they're filling in gaps for areas of cataloging that are not freely available.

So if there's already a freely available version of the rules or description type, we're going to skip it for now and focus on areas that are not really available. Yeah, like I think the rare books. description stuff is freely available or like the archives one is freely available so that probably wouldn't be in there yeah but this looks pretty active like provide feedback by january 31st 2025 they've got new committee members as of november 24 so yeah because i know one of the leading people who did that is no longer a librarian because i'm friends with them i was going to say they're friend of the pod i don't think they've been on ever before but i've i've done things with them i know them

Friend of the you Friend of the me Okay Well maybe we'll do an open rules For cataloging episode And dive into what that is more And see if that's something people should learn about And if you disagree with my takes

Join the Discord And we'll talk Because I can be convinced actually When I'm wrong I weirdly can have my opinion changed If I learn more I just sound cranky right now Don't forget to ring the bell

Sure On YouTube where we get cross-posted. Mm-hmm. Are our comments open on our YouTube videos? Yeah, we get comments sometimes. Nothing interesting. Oh, shit.

Oh. Usually just like, that's cool or peanut sorts in bio. I don't know. It's usually like spam or just like nothing interesting. Like, great episode. Cool. So yeah, comment. Great episode.

Yeah. Learn about VPN. London fog ice cream bar now. People kept tagging me in Discord while we were recording. so I thought maybe breaking news is happening in library world, but no, it was just I've been in Discord all day and people were following up with shit I've been saying. All right, good night.