Wikidata:WikidataCon 2017/Notes/Integrating a custom Wikibase Instance (Rhizome) and Wikidata via SPARQL

Title: Integrating a custom Wikibase Instance (Rhizome) and Wikidata via SPARQL

Speaker(s) edit

Name or username: Dragan Espenschied, Lyndsey Jane Moulds, Andra Waagmeester

Useful links: https://github.com/rhizomedotorg/Wikibase-SPARQL

Abstract edit

Rhizome is an internet art organization founded in 1996, with a collection of over 2000 artworks; websites, bots, software. The technical and cultural complexity of these works is very high, sufficiently describing them is outside the capabilities systems currently used in museums or libraries. Especially preservation data management is demands advanced capabilities, which we found out Wikibase was able to fulfill.

The project to provide digital preservation data in Wikibase, lead by Kat Thornton, can provide certain globally effectual data points, like what file format can be handled by what software.

Other types of information, like the technical composition of specific collection items or preservation data tied to a very specific use case—the arts—is best stored in a separate instance of Wikibase which Rhizome runs. It also serves as a sandbox for proposing data structures Wikidata might use to store preservation information. Rhizome has set up its own instance of Wikibase, created a tailored expert pipeline to RDF, and connected to Wikidata via a Blazegraph SPARQL endpoint; using established linked data practice, items in Rhizome’s Wikibase are matched with Wikidata items, providing numerous benefits. Rhizome is now seeking to adapt WDQS as a front-end to its Wikibase.

The plan is to present Rhizome’s journey towards using its own Wikibase instance, underscore the need and benefits for federated Wikibase instances and get community feedback on how to proceed with further goals.

Collaborative notes of the session edit

Gitlab/github/Gerrit confusion

Overview of the session edit

https://catalog.rhizome.org ← this should be made more presentable