Wikidata as the future of cultural heritage data − for Wiki Loves Monuments and beyond

Jean-Frédéric Berthelot

Playlists: 'wikidatacon2017' videos starting here / audio / related events

More than 1.4M monument have been gathered from more than 70 countries across the world since the inception of Wiki Loves Monuments in 2010. Up until now, these monuments and the meta data associated with them were stored in the Monuments database. Built by harvesting lists from Wikipedia projects, this is to the best of our knowledge the largest collection of data on built heritage in the world.The database and the API on top of it allowed tool developers to build visualizations (maps), maintenance tools (categorisation bots, consistency checkers), and beyond. However, the database imposed some important limitations to the storage and accessibility of the monuments data. For example, multilingualism was not well supported by the database, tracking the origin of data was difficult, and data sanity was too complicated to be achieved. In this talk, we present an overview of the migration of monuments data from Monuments database to Wikidata: we will show an overview of the technical work that has been done to use Wikidata as the backend for surfacing monuments data, the opportunities and challenges we have faced in this transition both in the social and technological sense, and the state of the current transition. We also will present some of the technologies that have been powered, thanks to this transition, to help Wiki Loves Monuments achieve its mission more effectively.