On the 1st of May 2017, the Linked Data in Research and Cultural Heritage Seminar (#DANSLOD) took place in Den Haag, the Netherlands. This seminar series was organized by DANS (Data Archiving and Networked Services) – Netherlands Institute for Permanent Access to Digital Research Resources. This seminar series was arranged by Herbert van de Sompel and its main focus was to present new advancements related to Linked Data with special emphasis on creating, keeping and using Linked Data in the context of a distributed and decentralized Web. According to the Schedule and plan of DANS Linked Data in Research and Cultural Heritage Seminar, the seminar was organized into two sessions, each followed by a panel discussion.
The first session of presentations considered different aspects of Linked Data in the context of a distributed and decentralized Web. Ruben Verborgh from Ghent University underlined the imperious need and extra care that proprietors of Linked Data should have when working with their data collections such that the isolated silos of Linked Data would be easily linked together in a knowledge graph. Afterwards, Tobias Kuhn from our group introduced nano-publications as a way of publishing data on the Web in a decentralized manner, a solution that would allow research results to be be both replicated and re-used. You can find the slides of this talk here:
Next, Sarven Capadisli from Bonn University argued how creating a Linked Data ecosystem would help connect contributors, libraries, institutions, publishers and researchers, etc. together and to also help in the social paradigm shift of Linked Data which lags considerably behind the technical aspects of this technology. Lastly, Michel Dumontier of Maastricht University introduced the FAIR (Findable, Accessible, Interoperable, Reusable) principles as a generic metric for evaluation of the quality of data repositories or collections.
The second sessions of presentations showed how Linked Data principles were applied by certain institutions when dealing with their data and how certain applications can help in the incorporation of these principles in the normal data workflow. Miel Vander Sande from Ghent University showed how problems like reproducibility and sustainability can encumber breaking the barriers of the data silos in the world of Linked Data. As solutions he proposed a combination of querying with Linked Data Fragments using the Triple Data Fragments interface and the Memento “Time Travel for the Web” protocol. Valentine Charles and Nuno Freire from Europeana presented their new approaches regarding future data acquisition. Among the technologies that their organization experimented with and that they want to include in the Europeana data acquisition workflow are IIIF (International Image Interoperability Framework), Sitemap and Schema.org. Next, Enno Meijers from KB (Koninklijke Bibliotheek) – The National Library of the Netherlands, introduced the roadmap, strategy and design for creating a distributed web of cultural heritage information within DEN (Digitaal Erfgoed Nederland) – Digital Heritage Netherlands. Lastly, Albert Meroño Peñuela of Vrije Universiteit Amsterdam demonstrated how Github can be used as a hub for (repeatable) SPARQL queries by using a tool called glrc.
The discussions that followed each session of presentations tried to address and tackle the current pressing issues in the world of Linked Data. Also, the presenters answered questions and raised themes for debate. Overall, this seminar was a good way to spend a day for a person interested in Linked Data. Both the presentations and the discussions that followed touched upon pressing and current issues in the field and left place for other open and unresolved issues. Linking data is not so simple after all!