Midwinter 2014 Report from the Linked Library Data Interest Group
The ALCTS/LITA Linked Library Data Interest Group hosted 2 sessions at Midwinter 2014. (1) A managed discussion, (2) its biannual business meeting. Both sessions are described below.
(1) The managed discussion took place on Sunday, January 26, 10:30-11:30. Somewhere between 40-50 people attended. Considering we were scheduled at the same time as two other linked data programs – the Library of Congress BIBFRAME Update Forum, and the CaMMS Cataloging and Classification Research IG session focused on the use of Linked Data in libraries and cataloging in RDA environments – this seemed like a good turnout!
We have been scheduled at the same time as the LC BIBFRAME update for the last 3 ALA conferences! We are pleased to report that, finally, in Las Vegas (at 2014 Annual), we will NOT be scheduled at the same time as BIBFRAME. Though this required excessive whining and groaning, the IG officers have since regained their dignity and are organizing for the Annual sessions where, they are pleased to announce, the next managed discussion (in Las Vegas) will be 90 rather 60 minutes long.
As for the managed discussion in Philadelphia, we opted to host a "Presentation Facilitated Discussion" (as described at http://connect.ala.org/node/197451). In November we were very pleased to recruit a truly distinguished speaker from the semantic web community (private sector) in New York who, in December, fell out of touch! We had to find new speakers, so IG officers started recruited from “people interested in attending” our session in the conference scheduler. There were some great people planning to attend, and the following three agreed to speak at our event at our invitation:
(1) Erik Mitchell, Associate University Librarian for Digital Initiatives and Collaborative Services at the University of California, Berkeley.
(2) Jessica Hayden, Technical Services Manager at University of Northern Colorado.
(3) James Morris, Informatics Scientist at AstraZeneca.
Each speaker spoke approximately ten minutes (their slides are available at http://connect.ala.org/node/217014). Erik Mitchell began with a talk called “Linked Data in core LIS curricula” and raised a number of questions about linked data instruction in the LIS curriculum. Jessica Hayden followed with a talk called “A New View of Old Data: BIBFRAME Practice” where she largely reviewed and analyzed tools on the bibframe.org site. James Morris finished with a talk describing and analyzing a SKOS and vocabulary management implementation at Astrazeneca (a global research-based biopharmaceutical company).
Approximately 25 minutes of discussion followed, with the speakers serving as table facilitators. The fourth table was facilitated by one of the interest group’s co-chairs, Theo Gerontakos. The other Co-Chair, Sarah Quimby, floated table to table to ensure discussions did not lapse.
Discussions not only did not lapse, facilitators were not required. Discussion was very lively. This is a time of tremendous interest in linked data (LD) in our community. Our interest group attracts all levels of interest in linked data, from novices to experienced semantic web developers. The three speakers/table facilitators complimented each other well, as it accommodated this variety well, from Hayden’s introduction to some relatively easy-to-use LD tools, to Mitchell’s broad considerations on how to generally represent LD practice to potential practitioners, to Morris’s somewhat detailed description of an implementation.
In the final 10 minutes of the session tables reported on their discussions as follows:
Table 1 (Erik Mitchell’s table): they raised the question, should we teach linked data? They discussed the roles of LIS education, how that education relates to employment, and the types of students found there. They agreed yes, we should teach linked data (LD), so that the main question can become how we teach LD and to what groups of students, keeping the coursework focused on practical applications, enabling students to demonstrate some sort of practical familiarity with LD.
Table 2 (Jessica Hayden’s table): they started with Jessica’s talk about BIBFRAME, but most at the table were not familiar with BIBFRAME. They then discussed why catalogers need to know LD. Some had seen the LITA linked data webinar series, but they felt the webinars were not terribly helpful for non-tech folks. They discussed the idea that catalogers don’t need to explore LD right now – and maybe not until it’s in-their-face – but heads need to know and help plan in using LD in search and discovery environments.
Table 3 (James Morris’s table): they discussed vocabularies, mapping, conceptual mappings, SKOS. They discussed the complexities of mapping terms in multiple vocabulary environments. They remarked on the importance of getting terms in RDF, using them and extending them, adding more relationships. Jon Phipps reported for this table, and added an observation that although the release this week of BIBFRAME vocabularies were important, RDA LD vocabularies were also available and were just as important to libraries. Library LD will most likely be RDA LD before it is shared as BIBFRAME LD.
Table 4 (Theo Gerontakos’s table): as Jennifer Marill (Head of Tech Services at the National Library of Medicine) was at our table, talk started with a discussion about LD at NLM. Some expressed the need for MeSH headings ad LD. NLM is thinking about offering MeSH headings and a subset of their bibliographic data as LD. Even Elsevier has been asking NLM for MeSH as LD, so that may go somewhere. Others at the table were at various stages of LD projects, a couple at the very beginning. Some expressed problems around practical issues at their organization: for example, lack of money to carry out LD projects, lack of knowledgeable staff, lack of even a place in the library for LD workers. Then some perplexity around the publishing of linked data was expressed. Richard Wallis (OCLC) was at the table and he suggested we were working with outdated concepts: with LD we will just leave the data where it is and let the mergers happen.
After the discussion Richard Wallis addressed the group and announced OCLC would soon be releasing (in 3 weeks, around Presidents’ Day in the USA perhaps) a very large quantity of bibliographic data as LD using schema.org schemas.
(2) The business meeting took place the previous day, on Saturday January 25, 2014. Minutes were taken by Sarah Quimby as follows:
Linked Library Data Interest Group Business Meeting, ALA Midwinter Meeting, January 25, 2014
In attendance: Theo Gerontakos (co-chair) and Sarah Quimby (co-chair) convened the meeting. At various times during the meeting, Suzanne L. Thomas, Thomas Dukleth, Jim Morris, Annie Glerum were also in attendance.
1. Welcome and introductions.
2. Reports and announcements:
2.a. Reports from past webinars:
(2.a.i) December 5: Linked Data Primer, Jackie Hsieh.
(2.a.ii) January 14 Coding experiments to transform MARC to linked data, Jeremy Nelson.
Theo Gerontakos attended both webinars. They each had 72 registered attendees. The December 5 webinar provided a broad overview of RDF, and the second was a very advanced session on transforming MARC with Python.
2.b. Upcoming events:
(2.b.i) January 26: LLD IG managed discussion January 26 at Midwinter.
The two co-chairs spent some time discussing how best to manage the table discussions during the LLD IG meeting on Sunday, January 26.
(2.b.ii) February 11: webinar on SKOS, SPARQL, and vocabulary management, Bob DuCharme.
There was some discussion on how best to advertise this webinar more broadly. It would be nice to see more than 72 attendees for these webinars. Perhaps the IG could take the initiative by posting the information on ALA Connect and other listservs.
(2.b.iii) End of June: Pre-conference at ALA annual: (Title, date TBD; presenters: Galen Charlton, Jodi Schneider, Dan Scott, Richard Urban.
Action items for the June 27 pre-conference were discussed, among them the possibility of getting a sponsor to pay for photocopying costs. Also discussed was the audience for the pre-conference, which is likely to be a mixture of beginners and advanced programmers. The deadline for a final program description is February 24.
(2.b.iv) Co-sponsoring with the ALCTS International Relations Committee a session at ALA annual: International Developments in Library Linked Data: Think Globally, Act Globally.
Theo Gerontakos will check into some of the remaining questions about this session, such as a proposed virtual preconference and who the speaker might be.
(2.b.v) LLD IG session (managed discussion?) at ALA Annual, Las Vegas. Do we want to continue with this format?
The two co-chairs contrasted the merits of the managed discussion session format versus the 90-minute uni-directional presentation. Uni-directional presentations lend themselves better to recording and making freely available on the web; whereas managed discussions are easier to put together and involve the audience more. It was noted that it is difficult to capture the entire content of the table discussions; perhaps at a future discussion table facilitators could be recruited beforehand.
3. Future event ideas: preconferences, webinars, co-sponsorships -- something else?
No-one proposed any future events at this time.
4. Volunteers needed for co-convener after Annual.
5. Open discussion
A question: what is the audience for the LLD IG? Those interested in the IG include those who have been working with linked data since 2003 and brand-new beginners. Gerontakos and Quimby discussed ways to reach both audiences while at the same time staying true to the intended audience for the IG, which is the linked data community of practice: linked data practitioners and programmers. One possible way would be to work with other sections to get more cross-sponsored sessions; another suggestion was to have two separate tracks for LLD IG programming. The possibility of co-ordinating a program with an organization or event such as LibHack was also discussed.
The meeting wrapped up with a discussion about how to get people more involved in the IG. Is there a way to ask for volunteers for web coordinator or other IG officers? Is it possible to start talking to others and delegating tasks?
The meeting was adjourned at 11:39am.
[END OF REPORT]