Library Linked Data

Why do libraries have catalogs? If you work in libraries the answer seems obvious: we have catalogs because our patrons need to search our collections.

Catalog OPAC image
A patron searches the catalog at the public library [CC0 public domain image]
But each of those components—our patrons, how they search, and our collections—have undergone significant changes in the last 20+ years, while our catalogs and their underlying bibliographic data have remained fundamentally the same. They were optimized for a time when our collections were comprised of physical materials, and when our patrons used the library in person, not from mobile devices.

It makes sense that we should meet our patrons on the web, and yet, as Rachel Fewell from Denver Public Library describes it, our library data remains locked in a closed system—a building with only one entryway. What we need is to open all the doors and windows, go outside, and meet our patrons on the sidewalk, not force them to climb a flight of stairs and enter a security code. In other words, we need to move our data to the open web and embrace Linked Data.

What is Linked Data?

If you search for “linked data” online, you will get a list of rather technical definitions, including:

  • “a method of publishing structured data so that it can be interlinked and become more useful through semantic queries” (Wikipedia);
  • a “collection of interrelated datasets on the Web” (W3C); and
  • “a method of exposing and connecting data on the Web from different sources” (Webopedia).

Reading about Linked Data might get some people excited, but sorting through esoteric models and metadata standards with acronyms like RDF, OWL, and SPARQL leave most of us scratching our heads.

Alphabet soup aside, Linked Data is a model for how data from all different sources can interact meaningfully on the web. Instead of closed data silos, we will have connected datasets on the open web. Instead of human-readable data, we will have data with explicit, semantic relationships that computers can understand.

Linked Data has profound implications for searching, browsing, and discovering the inter relatedness of things. It’s going to be much bigger than your catalog.

Current Library Linked Data Projects

At a recent meeting of the Resource Sharing Task Force (RSTF), members pointed out that Colorado libraries have resource sharing opportunities in the area of Linked Data. One such opportunity is the Libhub Initiative, a Linked Data project where libraries from around the world publish their bibliographic data to the web so that it can be linked together and searched via web browsers. Libhub institutional partners from Colorado include Denver Public Library, Anythink Libraries, Arapahoe Libraries, and Douglas County Libraries. The work of these partners will eventually pave the way for smaller libraries to join.

Other Linked Data projects currently underway include:

LC Linked Data Service. The Library of Congress has published datasets of its authority records and vocabularies. Other libraries can link to these records using their unique identifiers.

OCLC Linked Data Pilot. In September 2015, OCLC began piloting Linked Data with seven libraries. The results will help plan future products and services.

Virtual International Authority File (VIAF). This dataset links together name authority files from around the world.

Barriers to Implementing Library Linked Data

The real impact of Linked Data won’t be felt until more libraries and institutions of all types publish their datasets to the web. So what’s holding us back?

Awareness and education. Right now, Linked Data is a hot topic for a small percentage of libraries, but not necessarily for the rest of the library community. What’s missing is a strategic plan for getting the word out at the grassroots level, using accessible language and concepts.

Library technology. We are limited by our MARC-based systems. As Karen Coyle points out in Semantic Web and Linked Data, we’re still waiting for a user-friendly interface to make Linked Data happen. But most library technology vendors aren’t going to move forward with Linked Data until they are compelled by the library community. We can help drive demand through awareness and education initiatives.

It’s change. Our age-old resistance to change plays just as much a part here as with any major shift in our profession. And this one is going to be big—publishing our data on the open web demands different technologies, expertise, and workflows.

What can you do now?

Educate yourself and your staff. Start reading, discussing with your colleagues, and attending trainings. Headed to PLA 2016? Check out the session The Visible Library: Exposing Collections Through Linked Data.

Talk to your vendors. Ask them about their timeline for addressing Linked Data. While you’re at it, ask them what their plans are for BIBFRAME, the format that will replace MARC.

Share what you are doing. Is your library publishing datasets? Are you still in the discovery and discussion phase? Wherever you are, share your journey on your library’s site, at conferences, through your professional networks, on social media, or through the CSL Networking and Resource Sharing team.

The Colorado State Library is on this journey with you. As we learn more about Linked Data, we will look for ways to lower the barrier of entry for all Colorado libraries. After all, the more libraries that participate in Linked Data, the more valuable it becomes for everyone. Everything’s better when you share!

Amy Hitchner
Latest posts by Amy Hitchner (see all)