^B00:00:13 >> Beacher Wiggins: I'm Beacher Wiggins. I'm Director for Acquisitions and Bibliographic Access here at the Library of Congress. Today is another one of Elsie's Digital Future and You series and we are talking to you again about bip frame. We wanted to keep bip frame front and center. When we were determining the title of the session, which I see has changed on me again, some of my colleagues thought saying bip frame goes international was a bit too forward-thinking if not pompous. And I said that we need to be proud of what we're doing and show that bip frame is indeed penetrating beyond the domestic borders of the US. So we should talk what we are doing and pump our chests out. So we're pumping our chests out this morning and I'll leave it up to you at the end of today's session whether you think our title is overly ambitious or not, even -- and I think we can say global and international law are synonymous. So either way, you decide at the end of the session. And I say that we with full acknowledge that most of the work if not all of the work is being done by the staff in the Network Development and Standards Office and in the Cooperative Instructional Program's division, they are making it happen and I simply point fingers from time to time to try to keep us on track. But it's been a good journey thus far and we wanted to share with you what we're learning about what's going on in the world. As I travel to conferences and programs, I am constantly bombarded with the interest in bip frame and in linked data, bip frame in particular because the Library of Congress is leading that and that is likely to be the way the library community moves into linked data. So you will hear about the international aspects from my colleagues today and we'll let each of us introduce ourselves as we present to you. I understand there will be some interesting points that you'll have to take away. So with that, I'll turn it over to our colleagues and I believe, Judy, you lead off? Sally leads off. No, they're pointing at each other. ^M00:02:36 [ Laughter & Background Discussions ] ^M00:02:42 Now we think it's Sally? >> It is Sally. >> Beacher Wiggins: Sally McCallum will lead off and she will tell you who she is in case you don't know. I doubt there are many of you who do not. Sally? >> Sally McCallum: OK. I'm Sally McCallum and I'm Chief of the Network Development and MARC Standards Office here at the Library. And so I'm going to talk about what's going on in Europe with respect to bip frame and some very, very interesting developments. When we first initiated the project back in 2012, we always had international participation. We had representatives. From Germany and the United Kingdom on a very small working group that we had to plan the project. In fact, it was that small working group that named the project bip frame. And so they have been there from the very beginning. We also had a liaison through Denmark to some other groups within Europe that we didn't know very much about the groups but they were very interested in having information coming to them. Then when Elsie started the bip frame updates at each ALA the next year, we've always had someone from Europe to give a talk, give a presentation, tell what they were doing, tell what was going on in Europe at each one of those. ^M00:04:09 So then in 2017, Leif Anderson who is at the Royal Danish Library organized a European workshop. He got much more ambitious with respect to having cooperation and collaboration within Europe. He wanted people to be able to share their implementation experience because they have their various meetings that are focused on different topics within Europe but they didn't have anything that was focused on bip frame and what I'm doing and what you're doing and what different ones were doing. It would bring together people who were working on the transition from MARC to linked data using the bip frame model and that's what he wanted to do. And I have several places here I have quotes and it's because I quoted someone from Europe who was saying those things. And this was a quote from Reinhardt Huvelman [phonetic] from the German National Library who made this summary of what the workshop was to do. So in September of 2017 we had the first workshop at the German -- it took place at the German National Library. There were 40 participants -- I must admit I was surprised that there were so many -- from 16 European countries. And one of the -- for them, one of the big outcomes, and a lot of people said, I can't build something new. I need to be able to buy it. I need to be able to get it from a vendor. And so one thing that the workshop decided to do was, OK, what are the questions that these libraries should be asking the vendors if they are trying to get pieces or parts of a bip frame implementation for their library. They -- the document which is online that one could use for doing a vendor solicitation is called Bip Frame Expectations for ILS Vendors. And it starts out with a very simple I just want to put some URIs in my MARC data so that it will be linkable. It goes on to I want to convert my database and then I want to not just convert it but I also want to input native bip frame into an editor of some sort. It's because you can do each one of these without the other to a certain extent. And then I want fully linked data. I want that data to be not just in the bip frame vocabulary but I want the links to be in there. And then explore and discovery. By then you could explore, discover, and share records because everybody was used -- in the huge MARC environment, they share records a lot and they needed to be able to do that again. And a decision was made that this was useful and let's have one the next year. So in September of 2018, we had the second European Bip Frame Workshop. It was held in Florence, Italy at the European International University campus, on their campus. And it was sponsored by Cassalini Ribri [phonetic] and the university. And you all, many of you will know, of course, Cassalini's name because they are one of our large book suppliers and cataloging suppliers. And they also have a very large bip frame project which I'll get to a little bit later. They are there -- for that meeting there were 70 participants from 17 countries plus some North, several, about ten I think people came from North America and someone came from Qatar. And the US initiatives that involve linked data and bip frame that were presented there, our own, the LD4 project from Stanford Chairs or is the leader four, and the PCC activities going on in the PCC with respect to linked data. We heard a lot of discussion. We had talks, tutorials, lightning talks, discussion groups and breakouts. And I want to talk a little bit or give you a little bit of a glimpse of a few people who are doing either production or are looking at things or are evaluating at this point, are experimenting still, or building and sharing. And so I will do that in some of the next slides. But first of all, this is a group and it was the buildings from the university that were used. It is an old Medici palace that they've turned into a conference center and it's a very well outfitted conference center and we had lots of very good facilities for the conference. And this is the group. ^M00:09:11 And this is the -- this, as you can see, good facility. You know, it had a huge screen. We could all sit and face each other. We weren't the classroom style. And it really worked very well in that conference center. We had also had very good facilities, I must say, in Frankfurt the year before. Excellent. So here's a glimpse of what we heard at this conference. I'm just going to talk about four or five of the papers or discussions that we had. First is Sweden. Sweden was in production. This was September of 2018. In June of 2018, Sweden converted its main union catalog into bip frame database. And they have -- in Sweden they have a union catalog that catalogers catalog into and then the records are sent back to the local systems where they can be given or displayed to the users however that university or that public library or whatever, however they think their users will respond to it. So it sort of separates. You can build it centrally but then you can have the -- you can display and serve your users as you, with who you know and not everybody would know and they're not all alike. They have the -- they call it the Libras XL and I frankly don't remember if they told us what XL meant. I don't know if it's 40 or extra large or -- ^M00:10:55 [ Laughter ] ^M00:10:58 Or a pop group. But any of them could be interesting. ^M00:11:02 [ Laughter ] ^M00:11:05 Or maybe something Swedish. I don't know. Anyway, they wanted to improve coordination. They wanted to improve reuse of information. And they wanted to enhance the user experience through having more links. Those were some of the motivations for them. As I said, they collaboratively edit in the union catalog and then they take the records to their local system for the user experience. But some of the expectations -- and some of the ones that we are also very interested in. They wanted to find out what in MARC they depend on. What is it that's in MARC that's essential that we bring forward into this environment? They wanted an easier maintenance. And that's part of what everyone expects to get from the link data aspect of bip frame, is easier maintenance. When something, a heading, is changed, it doesn't have to be changed in a lot of records. It is changed immediately in whatever when anybody retrieves a bibliographic record that contains that heading. They wanted to phase out redundant details and that is something we would all like to be able to do because MARC is very crusty at this point. It was developed under when cataloging rules said one thing and then new cataloging rules said something slightly different, so slightly different ways of saying the same thing were shoved into the format. So they wanted to bring all this sort of redundancy that's all over the place in the format together and get rid of it. And we have that same motivation. And they wanted to increase derivation. They wanted to be able to derive by using works and instances as in the bip frame model they wanted to be able to leave a work as it is and just attach a new instance, attach a new instance to it. ^M00:13:17 Some choice comments from the person who presented for them were, it's working versus perfect. They realized into their project when they had -- I think they were about six months late coming up which is not bad. They said, it's working. It's not perfect but let's go on with it. And then he also said not done, just real. And he -- we can all sympathize with that that it will never be quite finished but it is real and as of June 2018 it was real. ^M00:13:57 ^M00:14:01 In Hungary they have been looking and evaluating. The Hungarian National Library has -- they've carried out projects, experiments, in the past few years because they were getting ready to get a new ILS and they wanted to be able to go forward with the new ILS. But so when they'd been looking at various vocabularies -- BIBO, schema, various vocabularies that they might use for their bibliographic data. But when bip frame came out, they said, well, this one offers us an opportunity to have vocabulary that is really tuned to our data, to our bibliographic data. And so they were quite pleased to be able to turn their attention to looking and evaluating bip frame. Some of their questions, though, that -- and I'm taking these from the slides that they presented -- were, is bip frame well known enough yet? They didn't want to put all their eggs in a basket that wasn't going to be, wasn't going to have longevity. Is there a critical quantity of data using bip frame to spawn tools? We're all -- we're very used to in the MARC environment where, which is huge, of many programmers and companies, various ones, have developed tools for us. And until there's a lot of data, you're not going to get that tool development. So they were concerned that, is there enough data that we don't have to go invent everything ourselves but we well get tools the way we have had in the MARC environment. ^M00:15:48 ^M00:15:51 Finland is experimenting with bip frame. They convert the MARC data to bip frame and then they convert it to schema.org which is the schema that's the schema child that was invented by Google, Yahoo, Microsoft, I don't know if Yendex was in there initially but they are one of the developers now. And it's for selling products but it's adaptable to bibliographic data and some people have done a lot of work at trying to adapt it to bibliographic data. But it doesn't accommodate all the data that libraries need and everybody recognized that. However, because it's known and used by Google and Yahoo and Microsoft, it makes a good channel for being able to get your data out there. So a quote from Osmon Sumanen [phonetic] from the Finnish National Library: It is useful to go by a bip frame, but for our current use cases and experiments, schema.org seems like a better fit. And so that's what they have been doing and we don't see anything wrong with that. That is, what we want is to see more and more experimentation going on in the environment rather than less, rather than everyone saying, we've got it now, we can stop. ^M00:17:23 And then there's the big Italian project. Now, we had a talk from the, from Tisiana and Mikali [phonetic] from the Cassalini operation. I think it was about a year ago and they shared again what they are doing. They are converting thousands of records into bip frame and building a huge database and then they give the records back to the library -- bip frame, a bip frame database of records, back to the library that sent them to them. And it's called share BDE is how we refer to it in the US. But it focuses on the creation of experimental linked data discovery environment which is what they have which is this huge environment as well as the enriched data that they can then send back to the libraries that have data that has links in them. It's, as I say, a very large database and it is a catalyst for the library when they get their database back to then go on experimenting and doing things, new things, with this data. So in summary, and this summary is in the words of one of the Europeans that attended the conference which rather than what I thought was came out of it. Reinholdt Huvelman [phonetic] from the German National Library, he first quoted Leif Anderson [phonetic]: nothing fits everything, which is apparent. But then he talked about the expansion of the community, that from the first meeting to the second meeting, there was an expanded community. There were many more people doing something and interested and that came out particularly in the little breakout sessions where people talked a lot rather than having someone talking at them. ^M00:19:29 People are at different stages and I think I -- I hope I've indicated that -- from production to looking at something to experimenting to building and so on. People are going from experiments to production now, which Sweden demonstrated. There are a variety of approaches, as the ones I picked out to say something about. Like, Finland has a very different approach from what Sweden is doing. There's -- he felt that a critical mass had been reached, that there was a lot of data beginning to be generated and out there and that people could safely go in, start doing their work, doing projects aimed at bip frame. And but that communication and cooperation were going to be essential if we were going to go forward in a manner that would make us able to share in the long run. We don't want everyone to go forward in a different direction which is a little bit of what we have right now as people experiment. Jodi Williamschen and I both went to the conference and we gave a number of presentations based on our work, at LC's work, and some of you who are in the cataloging directorate, your work. And but we were impressed. We were very impressed by what we saw and heard. Thank you. ^M00:21:04 [ Applause ] ^M00:21:11 >> Good morning, everyone. It's great to see so many people here interested in bip frame. I'm Judith Cannan, Chief of the Cooperative and Instructional Programs Division, and I'm going to be talking to you about the cohort. And a lot of people have asked me what is the cohort. And it's sometimes known as the PCC cohort. And that's the Program for Comparative Cataloging. But before I actually talk to you about the cohort, I need to give you some information that led to this. ^M00:21:46 In January, 2014, Paul Frank and I managed to get hold of the bip frame editor from the Network Development and MARC Standards Office. And we were extremely excited and Paul came rushing to me and said, I've got the editor. And I said, OK, to get the conference, I'll stand in for all the Program for Cooperative Cataloging meetings. You go back to the hotel and see if you can pull a PowerPoint together on this editor. And let me assure you, those of you in this room that have been around a long time and saw that initial editor, it was extremely primitive. But together, the two of us pulled together a PowerPoint presentation which we delivered twice at the LC booth at midwinter 2014 and it was a smash hit. It was really a sensation. So that really started things moving. And by 2014 annual conference, ALA annual conference, a lot of people were talking about bip frame. And one of them was a very enthusiastic supporter and that was Phil Skir [phonetic] from Stanford. And Phil made up his mind that he was going to get a Mellon grant to advance this. And by May 2016 he had this grant. But and there were only a few libraries that came in with him to do this. There was Columbia, Cornell, Harvard, the Library of Congress, Princeton, and Stanford. And they were working primarily on linked data or linked open data known as LOD also. And during this experiment, they focused primarily on ontologies. ^M00:23:51 So it was very limited scope. When the grant -- as the grant was coming to an end, Phil realized that he hadn't achieved nearly what he wanted to achieve, so he sought a second grant from Mellon. And Mellon gave him a second grant on June 2018. Now this is a two-year grant. And in order to pull this grant off, he worked with Cornell, Harvard, and somebody from the University of Iowa. But when he got the grant, Mellon had some requirements. And there were seven goals that Mellon had in mind. And I'm not going to go through all of them but I've given you a link to them. But what I do want to say is that Mellon recognized that if this was going to be a success, there had to be a much broader audience. And they told them that they wanted them to do this. Now, in order to get a broader audience, people have to have somewhere to work. And this is one of the things that they don't have at the moment. They have nowhere to really experiment with bip frame. We at the Library of Congress do. We've got the bip frame database. But the world out there didn't really have anything to play around in. The goals that Mellon enumerated are very important for the Library of Congress because we at the Library of Congress have not had an opportunity to build a bip frame database with multiple institutions, but a sandbox will make this possible because the amount of institutions will be able to add their bip frame data. Sharing bip frame data needs to be tested in order to advance bip frame. And I think that Sally made this very clear, that we have to experiment. And we have to try different things out. Second, our focus has been on developing the editor. And we've done very little about how this is going to impact discovery. We need to know how the end user is going to interact with this linked open data. And but we can't advance bip frame if we don't know what the discovery layer is going to look like. So we're now in linked data for production, LD4P2. ^M00:26:37 ^M00:26:42 Sorry. So the first thing that they're going to do is expand the involvement. And the need -- the grant has said that they're to build a sandbox which is what they wanted which is a place for people to experiment and to save their data, to look at other records. And so this is what LD4P2 will do. And invited PCC members to submit proposals where they could be selected to participate in the grant and work in the sandbox. And they had expected not many proposals but they got far more than they expected. So they had to go through those proposals and winnow them down to 17. And the 17 were selected based on the collections that they were going to put in the database and also on their willingness to adhere to the goals of the Mellon grant. ^M00:27:51 ^M00:27:54 These 17 institutions are going to be working in something called sinopia. Stanford had to build a database and they are building one right now. It's nearly complete and it's called Sinopia. And this is where these people that are part of the cohort where will be working. They'll be working in the Sinopia database system that's housed and managed by Stanford University. Now, Sinopia is using the bip frame ontology but their editor looks somewhat different from our editor. In fact, actually it looks quite different. But the ontology will be the bip frame ontology. And they have looked at our bip frame editor. And they've made modifications to our bip frame editor. They've also been working with us on the bip frame profile editor because as you know you have a variety of templates that you work in when you're doing your bip frame work. And that's what they've got to do is create these templates that the cohort members can use. ^M00:29:11 ^M00:29:15 So, the cohort. It's the 17 institutions that are participating in the grant and they're all Program for Cooperative Cataloging institutions. The Sinopia bip frame editor -- well, I should, we can call it that or the Sinopia Editor, whichever one you want -- focuses on cataloging and is being designed specifically for catalogers and for catalogers to use. They'll be submitting their data using the bip frame editor and the [inaudible] data that they submit will be saved into the Sinopia database and they'll be able to recall it and edit it. And as they go through this participation, they have to follow their proposals and they have to adhere to what they agreed to that relates to the grant that Mellon gave them. ^M00:30:24 ^M00:30:34 I don't know if this was a moment of weakness or a moment of strength, but LC agreed to provide the initial training for the 17 cohort members. And this task has fallen to Paul Frank, Les Hawkins, and Jodi Williamschen and you can see here on this slide what they've done. But I think the thing that's probably going to interest a lot of you in this room most of all is the fact that we have to create a bip frame manual. And that bip frame manual will be an LC bip frame manual. It will be the product that many of you in this room are using. ^M00:31:19 ^M00:31:22 And we've got two new senior instructors in the Cooperative and Instructional Programs Division, and one of their first assignments -- a little trepidation here because one has been here two weeks, one arrived yesterday [laughter] -- is going to be helping to pull this manual together. So they're taking a crash course and they're in here today to learn a little bit more about bip frame. So anyway, this manual will be used by the Library of Congress and we have let those PCC institutions that are part of the cohort know that we are going to provide this manual and it will be out there for global use but that it will focus on LC bip frame, not the sinopia. However, in a spirit of generosity and also out of curiosity too, we are going to provide the initial training to those 17 institutions on the sinopia editor because we feel that it's very important that we know what they're using and what kind of templates they're creating. So Paul Frank and Les Hawkins have agreed that they can undertake this. ^M00:32:42 ^M00:32:47 Here I do want to say something, that I think we often forget as we're moving forward, that bip frame when it was initially designed was not designed just for catalogers. When the Library of Congress undertook this, it had a much broader scope in mind. But lately we've been focusing more on catalogers. But the way that it was designed was that it could take all kinds of information and so I've got something here that I ran by people that I'd like to read to you. Sinopia focuses on cataloging whereas LC designed bip frame with the express purpose of providing a foundation for the future of bibliographic description, both on the web and in the broader network world that is grounded in linked data techniques. So beyond the cohort, what I do want to tell you is that the manual that we're preparing will be used to explode the pilot participants from 48 to 100 by September 2019. But once Paul and Les have given the training to the PCC cohort on the sinopia editor, we're bowing out of any kind of training for the cohort. The cohort will be on its own and they will be expected to support the rest of the community that can come in and work in that sandbox which is so critical to the development of bip frame. So it's now over to Paul and Jodi to explain to you the shared VDE. ^M00:34:37 ^M00:34:41 [ Applause ] ^M00:34:49 ^M00:34:53 >> Jodi Williamschen: So I'm Jodi Williamschen from the Network Development and MARC Standards Office. >> Paul Frank: And I'm Paul Frank from the Cooperative and Instructional Programs Division. And our portion of the presentation is about Share VDE so we're going to share the presentation [laughter]. But I want to start by saying I want to thank my predecessors who just spoke because Sally McCallum mentioned the European bip frame workshop where they talked about needs and particularly things we're going to ask ILS venders to help them with and two of those needs towards the bottom of the list were discovery or discovery and shareability. And then Judith also mentioned that same thing. So we're very excited to present this part of the bip frame presentation because Cassalini has actually been developing that discovery layer. Most bip frame work up to this point has focused on being able to take MARC data and convert it to bip frame, to take bip frame data convert it to MARC, to actually input bibliographic descriptions that are sharable as linked data. What hasn't been really explored too much is how users will deal with what we've done. I'm speaking to you as a group of catalogers. How will the work that we do as catalogers be used by the public in a shared environment? And that's what Cassalini has done a great job in starting that move. ^M00:36:34 >> Jodi Williamschen: So, the Share VDE project started in late 2016, early 2017. It's a group, it's Cassalini Libri which is the Florence-based Italian book supplier and cataloger. They have a large contract, I believe, with the library. And Atcalt [phonetic] is an Italian-based data processing company. They also have a small ILS system. And it was the library-driven, vendor-supported initiative to really bring linked data forward. The interesting thing about it is that it really started out with data conversion and then moved very quickly into discovery. They have had three different phases of their development and the first two were really about data conversion and now the most recent one is focusing on the discovery. And to guide these developments they have started a user council which has about 21 member libraries and community experts. There is a lot of overlap between the institutions that are involved in shared VDE and the institutions in the LD4P cohort. The transformation council was started fairly recently to address some of the data conversion issues that have come up as the libraries have been evaluating the data that Cassalini and Atcalt converted. And they've made a lot of recommendations for testing and changes. And this work continues. There are members from the library that are part of these councils so that our work is more aligned with what Cassalini is doing. ^M00:38:12 ^M00:38:14 [Whispering] Is this yours or mine? [Giggling] >> Paul Frank: I'm going to talk a little bit now about this shared environment. By the way, VDE and Share VDE is virtual discovery environment. But as LC staff members, imagine that you're working in the cataloging module. Maybe you're working in the OPAC, the LC OPAC. You're dealing with resources from one institution. It's pretty straightforward. It's one of the largest institutions in the world but still you're dealing with data from one institution. But now I want you to think about a situation where you're working in a shared environment with data from multiple institutions. This is the future but this is something that needs to be dealt with now and is being dealt with by the Share VDE project. As Judith mentioned and I think Sally alluded to it as well, the Share VDE is collecting MARC data from multiple institutions and returning that data to the users in bip frame. But they are also taking that data, reconciling it, deduping [phonetic] it, providing an extra discovery layer, and doing all sorts of data manipulations on a super big data level that even the Library of Congress does not have to deal with yet. So they've taken a step forward in this shared environment and the test case for this or the cohorts that Judith mentioned, the institutions participating in the LD4P2 cohort group. So those institutions are giving their data to Cassalini to Share VDE. Share VDE is generating consistent identifiers for entities and works across all the collections. So we'll look at an example of Moby Dick in a second. We maybe have multiple copies and editions of Moby Dick. But probably Stanford and other institutions have additional ones that we might not have and have some of the ones that we have. So those either need to be deduped or identified as unique. And then there will be an ongoing exchange between the LD4P2 participants and the cohorts as well with Cassalini and Atcalt on the data transformation and the creation of work identifiers. You notice that work shows up twice on this slide. This is a real critical part of the process identifying works. ^M00:40:43 >> Jodi Williamschen: But this was the first step of the data conversion which was taking the MARC data and first assigning the identifiers and the URIs that you see in subfield zero. And then it was converted to bip frame. And so a lot of additional data is added in. Just getting the URI in means that the library can take the MARC data back, put it in their own local systems, and they have the URI. And having the URIs really does facilitate the conversion to bip frame. ^M00:41:19 In addition to all of the work they're doing just to put the URIs in, the Share VDE project is doing a lot of enrichment outside just our id.loc.gov database. They are doing a lot of analysis to reconcile different entities to figure out that Herman Melville with the birth and death dates is probably the same as Herman Melville who wrote Moby Dick but the author as heading doesn't have the birth and death dates. And they have a lot of proprietary software that they have developed to reconcile entities and store them and reuse them as they analyze different libraries' data. And the biggest thing they're working on is what Tiziana [phonetic] calls the clusterization process, so that there's like a constellation of data revolving around set works and set authors and publisher entities, I believe, also. And they're also, as they evaluate and convert more data, then they're looking, do they need to create additional clusters? Does this incoming data match an existing cluster? And so on. And this is an example of their discovery interface for Antonio Vivaldi. On the left side you can see that they have links out to multiple sources, including our id.loc.gov and wiki data and the Bibloteque Nacional de France in Viof. They've also pulled in a bit of a wikipedia description. On the right they have variance, I think. It's hard to read [laughing]. And other links. And then there's the line going down the middle to works and then under that is where the works are clustered. Isn't that right, Paul? >> Paul Frank: Yes, yes, they are. >> Jodi Williamschen: Right. >> Paul Frank: Right. >> Jodi Williamschen: And now here's our Moby Dick example. >> Paul Frank: Right, and the previous page. So I have to go back [laughing]. >> Jodi Williamschen: No, go on. >> Paul Frank: I just want -- I mean, is this the OPAC of the future? I mean, this doesn't look like any OPACs that I know right now but this looks more to me like a Google search on something where you get the pane on the righthand side with the blurb of data that generally comes from wiki data, right? But you can -- you get that constellation effect in a Google search or in an internet search. Well, the Share VDE discovery layer is sort of mimicking that discovery that you might find through Google, right. So this is Vivaldi with reconciled authority data about Vivaldi links to additional information. That's part of the link data, right? Linking out to other parts. But then as Jodi went on to say that that line down the center goes to what catalogers probably would be most interested in and that's the list of works here. So I don't think it's going to take any of you very long to identify, hey maybe there's some on this list that should be deduped, right? This is part of the process. I saw some duplicated data here. But that's the task that Cassalini and Share VDE has taken on, to reconcile all of this data that's coming from multiple sources. ^M00:44:35 >> Jodi Williamschen: And so then here we have a search for Moby Dick and there's a title that's highlighted about halfway down. And then this is the display just for Moby Dick. So it's the works with additional links out to instances, variance on the right, subject headings below that, and then this is how the individual record displays. And at the very bottom is the permalink back to our catalog. So this is our data that Cassalini and Atcalt converted last year. [Inaudible whispering] >> Paul Frank: Yeah. So well, that's actually the last slide for us and our presentation. But I do want to say, looking to the future, so the Cassalini firm has been looking at a future discovery layer. This is the one they have now that's used but they're looking to the future and there's been some very good discussions of this and one of the -- I'm just going to leave you with one thought that I find very interesting. And that's that in the future, maybe there won't be that distinction between a traditional library OPAC and a traditional library cataloging module. We work in a cataloging module. That data goes to an OPAC. The prototype that Cassalini is looking at is maybe having that one layer be both for cataloging and for discovery, so all work would be done in a layer like that and I find that very exciting. We'll see how it pans out but it would be a big change from the way we do things now but it's in line with all the other changes in our environment. So thank you very much. ^M00:46:24 [ Applause ] ^M00:46:32 ^M00:46:36 >> Les Hawkins: Hello. I'm Les Hawkins. I work in the Cooperative and Instructional Programs Division with my colleagues Paul Frank and Judith. I'm a concert coordinator. I'm going to talk a little bit about some activities and developments having to do with link data, semantic web, and bip frame in China, Australia, and New Zealand. And as Sally pointed out when she was talking about the European Bip Frame Workshop, it's true here that these universities and government agencies in this part of the world are in different stages, from experimentation to just preparing data, adding URIs to their data, developing vocabularies that are link data friendly. So that's kind of a common theme, I think. The other thing is that they're also working with vendors and their systems to be able to provide some of the experimentation with bip frame and some of the data conversion. So some examples from China -- recently in the beginning of March this year there was a seminar held called Link Data and the Semantic Web at Hong Kong Baptist University. And it was a conference that attracted people from the US -- Phil Skirr [phonetic] from Stanford attended to talk about the LD4P2 project and to give, talk about moving towards link data and kind of the challenges and, you know, the rewards for looking at that and the risks involved with that and the payoffs. Some other libraries sending representatives were Shanghai Library System which was talking about its efforts to publish library data as open link data. And the last example that I have here I thought was very interesting -- from Hong Kong University of Science and Technology, the person who presented at this conference and in several other conferences in China was Kee Tat Lam [phonetic] talking about their bibliographic link data learning platform. And I'm going to show you a few slides from that. But the link up here is to an open access presentation. It was given at a different concert -- [tisks] ah, concert. [Laughter] A different conference, at Beijing University. But it's freely available there and it describes their project. And let's look at some of what they've created. They worked with their vendor to convert a portion of their library catalog to bip frame 2.0. They've set up this platform from this page. There's the url. You can play with it. It's freely available. But this is intended to let you compare link data formats. You can put in a comparison of bip frame, RDA, and RDF format and the MARC XML format, kind of compare the formats. This page also gives you an access to doing a sparkle input query for the bip frame data that's been converted from their library. And then they've also experimented with searching their database to pull up data using knowledge cards, the wiki data knowledge cards. And I'll show you some slides from that. This is just the apogee where you can compare the different formats, do a search for a resource and then compare it looking at the bip frame format, RDA, and RDF, Jason LD, and you can look at the MARC's XML format. This is a learning platform so it is intended for librarians to kind of get familiar with these formats and see some differences. ^M00:50:38 This is the Sparkle query form and this will search against their bip frame triple store. Then this is searching their database with knowledge cards. And so the knowledge cards are that wiki data that you'll see if you do a Google search. You'll see on the righthand side. You'll see a little box there with some information. That's what this data is. It's pulled from wiki data and it matches data in the library's bip frame triple score. And so you get a display and you can kind of navigate your search through these knowledge cards. ^M00:51:17 ^M00:51:21 And looking at Australia and New Zealand, relied on some contacts that Judith Cannan and Paul Frank had in these countries. And I would say both countries here are kind of in a stage where they're monitoring developments and in fact there are people from universities and government agencies in both of these countries that have been following bip frame for quite a long time, from its beginnings. And they've, you know, participated in regional and international and vendor conferences and they're part of the world for a number of years. But I think it's fair to say that their representatives from both the universities and government agencies are kind of in a monitoring and watching and a data preparation phase. So we have an example here from the Australian government. They formed a link data working group which is kind of a clearinghouse of information for laborers that are interested in publishing their data as link data. They also provide some learning resources there. The National Library of Australia, as the cataloging agency for their country, has been monitoring bip frame 2.0 particularly in terms of how it may be, you know, how their MARC data may be converted to bip frame 2.0. And these are only a couple of university libraries from Australia but these are the ones that we had some contacts with and they are very much in a monitoring kind of mode and working with their data, adding URIs to their authority data. They're kind of in a preparation mode and monitoring. ^M00:53:00 ^M00:53:03 I mentioned a couple of conferences. International conferences, regional conferences, and vendor conferences are important for most of these countries, people and practitioners in both of these countries to communicate with one another, give updates on their developments, and to learn, here's a couple of conferences. The Australian Library and Information Association has an annual conference and throughout the past several years I have found there have been bip frame presentations. There have been link data presentations. It's an ongoing theme of that conference. The international conference -- the Association for Information Science and Technology located here in Silver Spring, Maryland, their headquarters, actually. But they're going to be holding their conference in Melbourne later in this year, October 2019, and we understand that there will be some invited papers involving link data and bip frame. ^M00:54:02 ^M00:54:06 So I mentioned -- let's go up, moving on to New Zealand. I mentioned that the vendor conferences are very important and one of those for both New Zealand and Australia is the Australia New Zealand Regional Exlibras Group, ANZREG. And this has been a forum for several years for people to communicate about link data. In 2020 it will be held, their conference will be held in New Zealand and we understand that some of the invited papers will be covering bip frame. And the last example bullet that I have here, our National Library of New Zealand. They've told us that they are in the mode of monitoring bip frame developments. They're preparing their data, adding URIs to their authority files. They've also recently transformed to link data format, Morari Subject Headings Thesaurus. And I'll -- there's their page that leads to that subject heading thesaurus. So that's all I have. ^M00:55:11 [ Applause ] ^M00:55:18 >> Beacher Wiggins: So we have tested what the impact will be. We have sent in the data and the analysis. I think it was about four to five pages based on the work that you did testing. There were a set of questions that the participants used. We did this in the last two weeks of February. I sent it off to the ALA Publishing, to the chair of the RDA board, and to the chair of the RDA steering committee so that we can have some influence on what the outcome will be. We made suggestions in terms based on what the testing showed in terms of improving both the language, the impact of the 3R process, the LRMPs. We have no control over the LRMPs because the reference model is now the underpinning of RDA. So what we'll have to do is to manage that process. Judith Cannan and her staff in the [inaudible] operation will be focusing on the training that will be necessary for our staff to be able to use the 3R project, the outcome of the two left of that. We have been told that once the beta version of the toolkit is solidified then there will be a year before there will be the final implementation of the new toolkit. So we will be working on the policy statements, the training that are necessary, and determining how we want to make this successful. And we figure that if we figure this out for the laborers, since we have the largest catalog in operation, we will help the cataloging community to move forward with this. So that's all we can tell you at the moment but we are focusing on that quite intently. We haven't -- the other part of your question was how that interfaces with bip frame and the work that we're doing and we will see that too. Sally McCallum and her staff and Judith Cannan and her staff meet weekly now to make sure they're on the same path. So we will be coordinating whatever efforts are needed to -- how should we say? Help bip frame as well as to help us with the training preparation. Does anybody want to elaborate further on that? Nobody wants to speak on that? OK. [Laughter] Other questions? Regina? >> Regina: Yeah, of course it's about serials. >> Beacher Wiggins: Yes. >> Regina: Related to what I think one of the goals of bip frame was to be able to realize fully the RDA focus on elements as opposed to records. And I keep watching and waiting for examples of deconstructing records and reshuffling them so that we can on the fly present the whole history of a serial or just selected elements to be clustered in different ways as a result of deconstructing right now which is a very monolithic MARC record. Has anybody that's seen examples from these various experiments seen this capability in action or even good provisions for this capability? >> Beacher Wiggins: I have not. Sally and colleagues, do you have any prime examples of that? And Sally, will the conversion specifications that we're working on now for converting from MARC to bip frame and bip frame to MARC, would that be useful at all in this process? Will anything come out of that? That's the only area that strikes me as perhaps having some impact. >> Well, I don't know if it would be useful but converting from bip frame to MARC -- >> Beacher Wiggins: Oh, yeah. >> Converting from bip frame to MARC makes it very clear that the data models are very different and that you can do things in bip frame that you couldn't have done in MARC. But also you can't get into MARC very easily either. And so you know, it hasn't all played out yet. But some of the things that you're talking about, we're all interested to see how they will, what will happen. >> I'll stick around! ^M00:59:48 [ Laughter ] ^M00:59:50 ^M00:59:59 >> Beacher Wiggins: OK. There does appear to be no more questions. So with that we thank today's participants for helping to show how bip frame is penetrating the global stage and stay tuned. We will do much more work here locally as well as exerting influence more broadly. Thank you for attending today's session. ^M01:00:25 [ Applause ]