LIBER’s Linked Open Data Working Group aims to paint a picture of the current state of Linked Open Data (LOD) among European research libraries and to provide insights which help research libraries to develop their LOD activities.
The group recently completed a review of practices which research libraries follow in making data linked and open. The review was based on a survey, which looked at processes for making data semantically interoperable.
Challenges and possibilities, from both a technical and librarian perspective, were also covered. The LOD Working Group has now published the findings of its survey.
From the survey, it is clear that many libraries already use LOD in their processes. The following points were highlighted:
- Linked data projects are diverse in their character and scope. At the same time, there are certainly situations where it would have been possible to use less divergent approaches.
- The most notable expense related to publishing linked data is human labour. Providing guidance in the form of training and how-to guides is therefore of paramount importance.
- There is no one-size-fits-all tool. A great variety of tools are used — commercial, open source and specialized — alongside locally developed routines.
- The most commonly used vocabularies are GeoNames, VIAF, ISNI, Wikidata and Dublin Core. Wikidata stood out as the most common external resource that the projects were linking to.
- Data schemas used are often LOD-related: primarily SKOS and Schema.org, with mentions of FOAF and Dublin Core as well. A sizable minority opt for librarydomain specific schemas like the Europeana Data Model or BIBFRAME.
- Libraries are keen to cooperate and exchange ideas. This fits the character of LOD which intrinsically demands acting and thinking globally even when doing things at a local scale, however the networks enabling this are still somewhat thin on the ground.
The survey results will be compared and combined with results from other studies. From this, the Working Group will assemble best-practices and produce a basic workflow to guide institutions through the various steps of a LOD project. The group will also aim to include case studies highlighting how individual projects dealt with LOD and how that data was further used. Semantic interoperability is another topic which the group will aim to discuss in more depth.
The group aims to complete this second phase of its work by summer 2020 . If you would like to be involved, to be featured in a case study or to contribute in any way, please contact Matias Frosterus, Chair of the Working Group.