LD4LR: Linked Data for Linguistic Research

On February 6 and 7 CLARIAH WP3 organized a workshop to discuss the application of Linked Data for linguistic research. The workshop that went under the appropriate acronym of LD4LR, invited presentations from a number of foreign experts as also a number of representatives from CLARIN centers that had made some experience using Linked Data in their projects. The workshop concentrated on the perspective of the linguistic researcher that is increasingly confronted with all kinds of information about Linked Data and that needs to know what Linked Data can bring to her research. A number of prominent Dutch linguists were invited to present their current research topics where subsequently our experts could make suggestions on how to apply Linked Data paradigms to the researcher’s benefit. The invited experts that next to being Linked Data experts also are active linguists presented their efforts with Linked Data in the fields of Lexica, Phonetics and Treebanks.

Overall there was sufficient time for good discussions, where the experts tried to avoid too specific terminology and concentrate on user needs. In the round-up summaries from day one and two Sjef Barbiers and Jan Odijk concluded that although interesting things happen with Linked Data in linguistics, it seems not immediately useable for the end-user researcher unless they themselves are very familiar with Linked Data as the invited experts are. To make the potential of Linked Data use benefit a broader group of linguists, we need a better bridge between technologists and researchers. Dedicated pilots in WP3 should stimulate investigation of the usefulness of Linked Data application to different types of linguistic research, esp. Lexical resources (DUELME). From a data provisioning perspective, the benefits of Linked Data for interoperability purposes are clear.