By Trevor Owens
Senior Library Program Officer, IMLS
Following on the Linked Open Data for Libraries, Archives and Museums summit in 2011, interest in this topic has continued to grow among the digital cultural heritage professionals. For those unfamiliar, term “linked open data” can sound like a lot of different things. Conceptually, linked open data has been described as a cultural heritage researcher’s metadata paradise, but as it works its way through the hype cycle it seems to be finding a clear connection to the descriptive work that libraries, archives and museums are involved in.
In a pragmatic sense, a core idea in linked data relevant to libraries is to move toward using unique identifiers in metadata instead of simply using plain text. For example, instead of (or in addition to) having “Twain, Mark, 1835-1910” identified as the author of a work, one can use http://viaf.org/viaf/50566653/. That link, to the Virtual International Authority File, acts both as a unique identifier for Mark Twain and as a point of entry to a wealth of links to information about Mark Twain, from a range of places and in a multitude of languages. As the range of unique identifiers for people, places, and things continues to grow, there continues to be considerable promise for linked data approaches in the cultural heritage sector.
Work Continues on Linked Data for Libraries
I was thrilled to be able to join a set of library technologists at Stanford University for a two-day workshop to explore these issues, and I’m happy to share some of the activities and work participants discussed. As part of ongoing work at Stanford, Harvard, and Cornell funded by The Andrew W. Mellon Foundation, the event worked to inform the universities’ project and served to convene experts working on issues around linked data to share their work and chart future directions for the field.
Linked Open What?
For those interested in background on this area, you can read up on the issues and motivations behind the project. "Why Linked Data?" from the project proposal offers a useful introduction. For further context on the topic and the project, consider watching a video of a presentation Dean Krafft and Tom Cramer recently gave on the topic at the Coalition for Networked Information’s meeting in December.
Use Cases for Linked Data in Libraries
One of the central points of focus at the meeting was to work through a series of use cases in which librarians and library users might make use of linked data-based services to aid their use of library resources. These cases have guided the work of these teams on a range of particular implementation projects. Aside from that, they are quite useful in illustrating how linked data can help meet the needs of particular library user communities.
Linked Data Projects Moving Forward
Through a range of presentations and lightning talks, participants at the event shared a diverse set of examples of how different libraries are using linked data right now. Examples included:
- How Internet portal Europeana is using linked data to support multi-lingual use (for example, around a concept like “hourglass”)
- How UC Davis (supported by an IMLS National Leadership Grant) is working to transform its cataloging process to work via BIBFRAME
- How Cornell is working to describe a collection of Hip Hop flyers and connect them with the Musicbrainz data.
- How the Library of Congress plans to expand the id.loc.gov services which provide linked data access to various Library of Congress authorities
- How OCLC is approaching a range of linked data projects.
If you are interested in learning more about the results, presentation, and discussion at the event, you can review the Twitter stream for the event hashtag, #ld4l. Many of the slides and presentations from the event are also being posted up on the workshop page.