ROBIN FAY: Access is the main purpose of metadata. It allows users to access the information that they need and access that information on their own terms. It's everywhere around us and we're very familiar with it. Essentially, people have been using metadata their whole lives. Everyone kind of has to. CHARLIE BENNETT: So you're cool with people using metadata to, say, catalog cats? ROBIN FAY: I'm very cool with that. You got to catalog the world somehow. I think everyone needs someone out there to have to have thought about description and titles and these types of information. Especially now that so much is online, cats even aren't safe from metadata application, nor should they be. [MUSIC PLAYING] CHARLIE BENNETT: You are listening to WREK Atlanta. And this is Lost in the Stacks, direct from her heart to you, the research library rock and roll radio show. I am Charlie in the studio with Matthew, Ameet, Fred, Cody, and a guest to be named later. Each week on Lost in the Stacks, we pick a theme and then use it to create a mix of music and library talk. Whichever you're here for, we hope you dig it. AMEET DOSHI: That's right, Charlie. Our show today is called Metadata In A Material World. And while it's fun and somewhat accurate to define metadata as data about data, we want to get practical today and talk about how we describe things and relationships in the material world. FRED RASCOE: For instance, how do we describe resources in a library collection or not in the collection? AMEET DOSHI: Or not not in the collection. FRED RASCOE: Or cats. CHARLIE BENNETT: This is not fun. AMEET DOSHI: There are relationships outside our resources that we need to describe, and sometimes those descriptions are resources themselves. CHARLIE BENNETT: Ameet, is this like turtles all the way down or something? AMEET DOSHI: If P not Q. FRED RASCOE: It's cats all the way down. If you want to join the conversation, the hashtag for this show is LITS 430 for Lost in the Stacks episode 430. Feel free to tweet your thoughts, questions, or favorite metadata strategy with that hashtag. AMEET DOSHI: Our songs today are about creating a relationship, managing identity, and understanding things in the real world. And you're going to need to double check the metadata of these songs today, because they're all cover songs. The metadata world is linked to our material world. So let's start with a song about the material world. CHARLIE BENNETT: Ooh, I think I know what this is going to be. AMEET DOSHI: We all know what's about to happen. This is "Material Girl." CHARLIE BENNETT: Of course it is. AMEET DOSHI: By The Bloodsucking Zombies From Outer Space. CHARLIE BENNETT: What are you talking about? AMEET DOSHI: Right here on Lost in the Stacks. SPEAKER 2: And I'm the captain, we're just lining up for our departure. I'm pleased to say they've brought the time forward. AMEET DOSHI: Big finish. "Material Girl" by The Bloodsucking Zombies From Outer Space. CHARLIE BENNETT: I just don't think-- I don't think zombies suck blood. I think that's-- I don't understand anything anymore. AMEET DOSHI: Today's show is called Metadata in a Material World. And our guest is Robin Fay, a.k.a. georgiawebgurl, that's G-U-R-L, on the internet. She has done way too much for us to summarize her career, but let's say she's worked in libraries and the corporate world, in technical services and project management and database maintenance, and as a creative technologist. And through all of that, she does her best to help people use technology to make their lives better. Robin, welcome back to the show. ROBIN FAY: Thank you. Thanks for having me. CHARLIE BENNETT: So, Robin, when we were planning this show, one of the topics that interested you was, and I quote, "real world objects, things, linked data, kind of exciting library data stuff," end quote. And when I said that seemed like a good idea, your next message was about Prince. ROBIN FAY: Yes. CHARLIE BENNETT: The musician. The artist. Artist formerly and future known as Prince. So I was hooked. So let's define real world objects, things, and linked data for this conversation. Talk about why they're exciting library stuff and about Prince. ROBIN FAY: So real world objects is actually a metadata term or construct that's been around for a while. And the idea of it is we're just looking at everything that's out there in the world, both real and it could even be a topic or an idea. But anything out there in the world and then how can we describe those things. So that's been around for a while. And as the cataloging and metadata community started working on revising rules and guidelines to address the shift to all of these digital objects and also the rise of AI and all this other data that's being out there created by other people, then it started looking at all the other communities out there, like the semantic web community, the archives community, all those other communities, and then looking at what they do and how they describe things. And so we were expanding what we're doing in terms of how we describe things and aligning ourselves, I think, with that community a little bit more. So real world objects have existed as a topic for a while and as a metadata construct for a while, and we're starting to look at that as well. And so part of it would be how do we describe the resources we have in the library already? But then how can we link out to those other resources? And that's where Prince comes into it. CHARLIE BENNETT: Oh, excellent. So real world objects sound to me like everything that is. So what's the excluding quality? ROBIN FAY: Really, the only excluding quality for real world objects is, for example, if you have a web page, the information about that web page would fall into a different category. The web page itself would be a real world object, but that metadata, that information about the website itself would not. So that's really the separation. The administrative information about a record or that kind of information, that would be fall outside that exclusion. CHARLIE BENNETT: Gotcha. ROBIN FAY: Although some people would put everything into that category. CHARLIE BENNETT: Yeah, so I feel like this is one of those philosophical debates. ROBIN FAY: It is. CHARLIE BENNETT: Where you say, well, if you think of it in terms of my framework. But I also think that we've definitely separated the material oomph now from real world object. Things are things. Probably that's enough said about that. Can you summarize linked data? ROBIN FAY: Yes. so linked data is just a way of looking at existing resources that are out there. So it could be a Wikipedia article. It could be a author's authority record in the national authority file. Some kind of authoritative resource would generally be what we would look to. But we would just link out to that resource rather than copying or writing information or inputting information about that author. We would just link out to that record, or we would link out to that Wikipedia article rather than actually copying some of that information in. And so it serves the purpose that we wouldn't have to correct that information if it changes. So if something in the Wikipedia article changed, the name of a building changed, we wouldn't have to change the name in our data. CHARLIE BENNETT: It would change everywhere. ROBIN FAY: It would just change everywhere. Exactly. Linking to data. CHARLIE BENNETT: Love it. I hope. I hope I understand it. Because I like what I understand, but I don't know if I really understand it, much like Prince. So let's talk about Prince. Why does Prince come to mind when we start talking about real world objects and the world of things? ROBIN FAY: Well, I think about Prince in terms of all the different roles he has. So one of the other aspects of this is that relationships that real world objects have to each other. And so Prince, of course, has been in movies. He's been a producer. He's certainly a songwriter, a musician, an artist. He's been on television shows. He's had all different kinds of books and things like that. Plus he's had things written about him. So there's that relationship. Of all the books about Prince, the covers of Prince songs, all of those relationships back to Prince. And so that's why I think Prince is a really interesting example, because he does have all kinds of different relationships, and there's all kinds of different objects attached to him. Not only objects he's created, but objects that other people have created. All the fan art that's associated with Prince too. And then on top of that, you have the fact that he has changed his name over the years and how that then plays into his identity. CHARLIE BENNETT: So you say relationships, and obviously there's a kind of colloquial understanding of relationships. But when you say relationship, I think you're talking about as a creator or as a subject or as an object, how he relates to each of those instances, those real world objects. ROBIN FAY: So from a data standpoint, it is very much looking at that relationship between, say, Prince and fan art. So he would be the subject of fan art. CHARLIE BENNETT: Not the creator. ROBIN FAY: Not the creator in that case. Or it would be Prince as the creator of his music, or Prince as an actor on a television show. CHARLIE BENNETT: So we're almost done with this segment. I feel like we've laid out a lot of stuff. Why is it important that we can think about Prince as an analogy, but real world objects in these multiple relationships? What's the kicker here? ROBIN FAY: Well, I think the kicker is that traditionally, library metadata has been a little bit insular. So we describe the materials that we have, and that's really it. So we didn't capture the full sense of what a creator could do, or we didn't capture the full sense of a particular subject. So if you have a book about virtual reality, there's so much more virtual reality stuff out there that might be associated with that book in some way that we don't link to, we don't capture it. We don't create that relationship to it. So with Prince, because he's done so much, if you have, a new version of "Purple Rain" that comes out that you're working on. But then if you could link out to the movie and all the other different things that he creates, that would give you a more full view of all of the work that's out there, either about him or related to him. And as a user, we do this all the time. We're looking for recommendations. We're looking for things that are related, that serendipity of finding things that are similar that we discover on our own. But the idea behind linking to more real world objects is that we will help create and facilitate those relationships and discoverability. CHARLIE BENNETT: For so long, I've used Moby Dick as the center for my analogies and metaphors about virtual browsing, and I am pleased to say that I have a professional change today. I'm moving to Prince. This is Lost in the Stacks, and we'll be back with more from Robin Fay after a music set. MATTHEW: File this set under QA 76.9.D3I599. [MUSIC PLAYING] You just heard "Hanging On The Telephone" by Blondie. Before that was "Ain't That Peculiar" by Fanny. Those were songs about ways of creating connections and relationships. CHARLIE BENNETT: This is Lost in the Stacks, and our show today is called Metadata in a Material World. Our guest is Robin Fay, also known as georgiawebgurl. And we're talking about data, things, and libraries. So in the last segment, we finished by exploding our understanding of Prince's relationships to real world objects, both that he has created and that he is the subject of and that are just adjacent to him. You have a more maybe relevant professional kind of example of that kind of exploration of relationships from the long ago dark days of the early 2000s. Can you explain this project? ROBIN FAY: Yeah, so there was a project called Claim ID, and it was a project that anyone could participate in. And you would create your own little page within Claim ID, and it would assign a URL that you could use to share with people. But the more interesting part of it was that it wasn't just a biography page. It was a page where you actually listed all of the things that you had created. And you could also list things that you had participated in. So you could establish what your roles were for different things. So anything that was out there that was available either on the web or things that you would just describe that you had done, you could list. But the other part of it that was really interesting is that you could list things that you did not do. So for example, there are many Robin Fays out there on the internet. So I could say, OK, the fear of flying website, that's not me. The illustrations website that may be similar to me in terms of name, that's not me. So I could very clearly define what was not me. And I think what was interesting about that project is it was a way for either a professional, just average web user, it really could be anyone, to establish their identity and to create almost like a mini authority record, but pull together all of the resources that they had created across, well, anything. And then to also say these are not me. So it was very clear in terms of what you were doing. AMEET DOSHI: So on Wikipedia, is this the disambiguation concept? ROBIN FAY: It is. It is. And so that's the same concept that we use in terms of authority work too, where we say this person is not the same. We add dates or we might add the profession or something else to separate out the different Robin Fays or the different Charlie Bennetts to be able to separate those people out. CHARLIE BENNETT: There's a baseball catcher from the late 1800s and a screenwriter who worked with Alfred Hitchcock. Those are my Charlie Bennetts that I have to entangle with when I'm on the web. So who ran Claim ID? ROBIN FAY: I'm not really sure. CHARLIE BENNETT: OK, so that is the thing that I'm interested in. We talk about authority records. And obviously there are these civic institutions or professional organizations that kind of claim a certain amount of authority and responsibility. But then Claim ID was just a project? I mean, did a grad student try and throw something together? ROBIN FAY: Yeah, it just seemed like it was originally meant to be, I think, a bigger project and to allow people to also use it as a way to authenticate, if they were signing up for a particular resource. But what was interesting to me is that it just completely sort of faded away. I mean, Open ID still exists where people can use that to authenticate, but it's not the same as creating this little page where you can list things. Now, the challenge with that, of course, is that since anyone can create a page and there really was no vetting. I could create Charlie Bennett and I could put all kinds of stuff in it that had nothing to do with any of the Charlie Bennetts. And there was really no vetting of that process. So because it was that user contributed, user created process, there really wasn't that vetting or any kind of control over it. FRED RASCOE: Do you think it had anything to do with the fact that Claim ID was about making that, I guess, identifier, for lack of a better word, for a person rather than a thing? Do you think it's a lot easier to identify a thing? And I'm going to look at that from the example, in my own library world, scholarly publishing and things like that, a scholarly article has a DOI. And that is really widely established. Lots of publishers use that. As far as identifying persons, there's this thing called ORCID, the ORCID ID, which is trying to pick up steam a little bit, but it seems like doing it for people is a lot harder than for a thing. ROBIN FAY: And I think part of the reason why doing it for people is a little bit harder is because we have this national authority file that's been used in libraries so much. So there's already this big practice, very well established practice for libraries, but there hasn't been in the scholarly communications and publications community, the corporate community. We just haven't had that same sort of thing. And I think ORCID is an attempt to get to that. But the challenge with ORCID is since it's very researcher focused, if you're a person like me who does things outside of that arena, there's not a place to put some of that information. CHARLIE BENNETT: I feel a vaguely paranoid resistance to doing what would make it much easier for us to understand relationships between people. Claim ID is self-motivated. You go and you'd say, I'm me. This is all the stuff. And I think one of the articles that we saw about Claim ID was save yourself from cyber spite. It was a reclaim your domain style. This is my internet identity. But then also internet identity is something that maybe we've been trying to get away from. You and I were speaking off air about Facebook and how I have ejected myself from Facebook. I wonder if there is a easy fix for that. Can you imagine? What would be appropriate for everyone to take part in to try and claim these relationships or identify them? ROBIN FAY: Well, and I think that's one of the challenges with identity management. Because as metadata practitioners, web editors, anyone who's describing content or writing about content, we do want to be able to identify people and point people to the resources that are relevant to that particular author. From an author standpoint, an author may not want to be identified that way. So the identity management aspect and the privacy part of it, I think, are really things to-- as a large metadata community, we're still kind of working out. Because we've recorded information over the years about people and in authority records, it's been pretty limited in terms of what we can record. Under the new guidelines, we can record gender. We can record a person's address. Those can really bump up into privacy. CHARLIE BENNETT: My blood pressure just went up a few points thinking about that. ROBIN FAY: And so there's been a lot of discussion about whether or not this is a good thing. How do you approach collecting this information about people? Or do you ask authors for this information? There's a lot of discussion about how to handle that. So I think as librarians and metadata practitioners in the library community, we're just now really starting to look at identity management and then that privacy aspect too. So my thinking is it has to be some opt out or opt in. Whatever it is that we're going to do, we need to allow people to have that option. And then of course, that creates a challenge. If you're describing, you're trying to catalog or describe something, and the authors opted out of this, how do you handle that? And I don't have an easy answer for that one. CHARLIE BENNETT: Oh, there's never an easy answer. FRED RASCOE: And to bring it back to Prince, I definitely don't see him signing up for any authority control websites. ROBIN FAY: No, no. Well, and he's interesting in the fact that his name has changed so much. And there was a lot of challenges in how to describe the symbol. FRED RASCOE: He definitely was interested in controlling his identity. We'll be back with Robin Fay for more about the real world of things on the left side of the hour. MICHELLE CASTO: This is Michelle Casto from the DC Punk Archive at DC Public Libraries. You're listening to Lost in the Stacks on WREK Atlanta. [MUSIC PLAYING] (SINGING) Dream team It's just a taste And it might not come this way again CHARLIE BENNETT: Today's Lost in the Stacks is called Metadata in the Material World. And I'd like to read from a work called The Life and Death of Data by our guest last week, Yanni Loukissas. "Institutions of collecting produce, manage, draw together, and disseminate knowledge through diverse practices with data. To understand contemporary collections, we must ask, what do data know? This might sound like a question exclusively for networked age collections. However, cultural and scientific institutions have long pondered how to organize and expose their artifacts and knowledge thereof through data. It has long been true, but only recently clear that data are the first order material traces of our most sacred ontologies." File this set under ZA4240.F39. [MUSIC PLAYING] AMEET DOSHI: "12XU" by Minor Threat. CHARLIE BENNETT: Whoa. AMEET DOSHI: Before that, F-R-I-C-T-I-O-N, Friction by Echo and the Bunnymen. Nice find, Fred. Those were songs about understanding real things in the real world. Our show today is called Metadata in the Material World. Our guest is Robin Fay, who is georgiawebgurl on the internet. And we're talking about how we link things to people and define relationships between things. CHARLIE BENNETT: So this is the library segment. I think that's where we've landed. Because we've been circling the work that we do. We reference authority records or authority control. But all of this discussion about relationships and how we describe them comes down to now, how does this help us provide resources, access to information, to users? And how does it hurt us? ROBIN FAY: Well, I think in terms of how it helps us, for one thing, if we do use more linked data, that would mean less kind of cleanup when things change. It would mean that our information would not get outdated, because it would just link out to that information. It would also mean if there were other resources that were linked to that thing we were linking to, then our users might be able to get a little more context or a little more information. CHARLIE BENNETT: I'd like to interrupt you. ROBIN FAY: Yes. CHARLIE BENNETT: Where is the data that we're linking to? Is there an easy answer to say where that stuff is that we link to, to avoid duplication and the need to hand switch or hand revise things everywhere? ROBIN FAY: So for libraries right now, that's going to be to what we would consider authoritative resources. So these are established resources that have been out there for a long time, like the Library of Congress subject headings. They're online. The Library of Congress and the national authority file. So that list of the authors' names and their identities there. But it does give us the ability to link to things that might be a little more transient, like an author's web page or Wikipedia, a Wikipedia page, things like that. And so that, I think, is the question for the community of practice is how far do you want to go in terms of linking to these outside resources? And do you want to just stick to established resources that you know that you'll be able to link to, that the address will never change to that resource? The link will never change to that resource, so you won't have to ever worry about that. Or do we want to link to things that may change, move, things like that? And that's, I think, one of the things to discuss as we go down this road of looking at linking to these outside resources. CHARLIE BENNETT: Because individual librarians working with their collections, they're the ones choosing what to link to even if there's the established, yeah, link to the Library of Congress, link to whatever the oldest university in America is. I can't pull that out of my head right away. But these things that we imagine are as permanent as we can expect and also authoritative. But then an individual librarian could decide to link to a Geocities page in their catalog. ROBIN FAY: That's true. And I think that is one of the things that really will be a-- really will come about in terms of a community of practice as we go down this road. I think we're going to see more best practices established. And there's already sort of some best practices being established in terms of the examples that are used in the documentation. It's linking to authoritative resources versus an author's website, although that's certainly within the ability to do. CHARLIE BENNETT: So if you're introducing this concept to a practitioner, what are the things that-- what are the high level things that you want them to grasp first before they enter into the process? ROBIN FAY: The most important thing is just understanding that the work that they're doing today will save them work down the road. So they're not going to have to go back and keep updating that record. And that they're sort of building data for the future. And I know some people are like, well, this is my work today. Why am I building data for the future? But it will help in terms of their daily work down the road. And I think also for the people who are contributing to the national authority file, the librarians who are doing that work, they are going to be doing more work, maybe up front, but down the road it will save work once we get to the point where most of the records already have these links. CHARLIE BENNETT: Yeah. We talked earlier about KPrince and we talked about things that someone might not care to link to, like fan fiction, for instance. Maybe even adult fan fiction. But then there's going to be communities where that is what someone wants to link to. Is there a layer of more metadata about all of the various links and not linked data that's out there? ROBIN FAY: Definitely. That would fall into the data about data. What is not linked to in a particular record, as we think of records now. CHARLIE BENNETT: In a particular place too, right? ROBIN FAY: In a particular place. So that information. And I do think that we're just at the beginning of thinking about this and the beginning of thinking about data in a more semantic, web oriented way. So there's still a lot of things to consider. Because as librarians, as metadata practitioners, as catalogers, as people, we're describing digital objects, we do have that role of curation during the process, where we're deciding what information to record. Semantic web community, they're coming at it from much more of a machine generated data approach. So there's probably going to be a little bit of both for us, where we have more machine generated data available, more machine generated data that can happen from our materials. But there's, I think, always going to be that curation process, where we're making that decision. And that's really that value judgment, too, in terms of what's important for our community. In terms of how different communities may represent a particular resource, I think that's something we're going to have to talk about more in the future. CHARLIE BENNETT: So we only have a minute left. What is, for you, you personally, what's the next step? What's the first tool or concept or thing you're working on in this moment? ROBIN FAY: Well, for me, the thing that I'm looking at is really looking at how all of the resources that are out there, regardless of what platform they're using, regardless of what the record structure is, and where can we find that commonality between data so that data is transferable and there's that interoperability between resources? And how can linked data make that process happen better? So regardless of what the metadata schema may be, how can the resources that we're describing fit together? And then looking at that larger semantic web community too, so that we're building data or modeling data that the machines can then use to do amazing things. CHARLIE BENNETT: This is Lost in the Stacks. Our guest is Robin Fay, also known as georgiawebgurl on the internet. She is a media practitioner, a creative technologist, media practitioner, a metadata practitioner, a creative technologist, and a librarian. Robin, thanks for coming back. ROBIN FAY: Thank you. CHARLIE BENNETT: Hope to see you again. ROBIN FAY: Yep. MATTHEW: File this set under NK1089.T446. [MUSIC PLAYING] You just heard "Everybody Is A Star" as covered by Fishbone. Before that was "Cut Your Hair" by Superorganism. Those were songs about manipulating how you were seen and how you're managing your identity. CHARLIE BENNETT: Our show today is called Metadata in a Material World. And Ameet, to sum this all up, you got caught in a little thought there, didn't you? AMEET DOSHI: Yeah, I want to bring this back home. And certainly the patron saint of today's episode is Prince. But maybe, it might be, he may be the saint of the show. So David Byrne may have to move, be relegated to a lesser saint. CHARLIE BENNETT: I think we should switch the pathology around every once in a while in the show. AMEET DOSHI: So what I need is a subject heading for the rain that fell on Prince during his solo of "Purple Rain" in Miami during the Super Bowl 2007. CHARLIE BENNETT: Oh wow, that was such an amazing show. AMEET DOSHI: As it turned from liquid to steam. I need a subject heading for that. CHARLIE BENNETT: Transcendence. AMEET DOSHI: I like it. Transcendence. CHARLIE BENNETT: Wow. Hey, Matt, just roll those credits, man. [MUSIC PLAYING] MATTHEW: Lost in the Stacks is a collaboration between WREK Atlanta and the Georgia Tech Library, produced by Charlie Bennett, Ameet Doshi, Wendy Hagenmaier, and Fred Rascoe. FRED RASCOE: Matthew was our engineer today, and the show is brought to you in part by The Library Collective, which is currently taking session proposals for next year's conference. You can find out more at thelibrarycollective.org. CHARLIE BENNETT: Legal counsel and a pristine copy of "Purple Rain" on vinyl were provided by the Burrus Intellectual Property Law Group in Atlanta, Georgia. AMEET DOSHI: Special thanks to Robin for being on the show, to everyone who grapples with linked data, and thanks, as always, to each and every one of you for listening. CHARLIE BENNETT: You can find us online lostinthestacks.org. And you can subscribe to our podcast pretty much anywhere you get your audio fix. FRED RASCOE: Next week on Lost in the Stacks, a show about walking around and observing. CHARLIE BENNETT: What? FRED RASCOE: It'll make library sense. Promise. AMEET DOSHI: It's time for our last song today. Metadata or metadata describes the people, the places, the things of our informational world and thereby connects us to the people, places, and things in the real world. So let's close with our last cover song of the day, a song invoking stark imagery of relations between objects, thoughts, and words. CHARLIE BENNETT: Let me guess. "Darling Nikki." AMEET DOSHI: This is Prince's classic "When Doves Cry," as performed by Patti Smith. CHARLIE BENNETT: Hey. AMEET DOSHI: Right here on Lost in the Stacks. Have a great weekend, everybody.