When GLAMs upload media to Wikimedia Commons and volunteers add and correct the metadata, the information is rarely read back to the originating institution. Now that Structured Data for Commons is making editing and reading metadata much easier, it can be expected that there will be many more apps and tools to play around with and enrich the openly available media. The institutions will be able to take advantage of the cumulated knowledge, with new tools to read that data back to the intitutions’ own collection systems alongside their own data. This is what we call data roundtripping.
The Wikimedia Commons Data Roundtripping project run by the Swedish National Heritage Board is working with GLAMs to identify the needs and desires regarding user contributions in their content, as well as barriers to making use of them. A survey among 38 respondent GLAM representatives revealed that user contributed data is still modestly adopted by the organisations. There are no established technological means to tackle the challenge. User contributions that are adopted, are usually gathered in the most traditional ways, by individual feedback by phone or mail. Trust and verifiability of the contributions were mentioned most often as barriers to incorporating the input into the institution’s own data.
The research also looked at how institutions are taking advantage of authority data as a way of linking across collections, and whether Wikimedia projects could leverage that capacity.
Adding translations of descriptive texts into other languages
Identifying people, names and aliases, locations and subject matter
Linking to authority data and using it to retrieve third party contributions from other memory organisations
Data Roundtripping pilots
The Wikimedia Commons Data Roundtripping project will prototype a tool that addresses these three issues. Each functionality of the tool is developed in collaboration with a museum.
Swedish Performing Arts Agency's pilot is created around an upload of 1200 glassplate photographs. Earlier they have translated all descriptions of images into English before uploading them to Wikimedia Commons. In the pilot the images will be uploaded with Swedish descriptions only. They are translated in a translation campaign and the translations are fed back to Swedish Performing Arts Agency's collections management system.
The purpose of the pilot with Nationalmuseum is to take advantage of authority IDs that have been added to the institution's contributions in Wikidata by the users. The institution is interested in collecting the Wikidata IDs themselves and ULAN IDs found in the Wikidata entries. This will aid Nationalmuseum to get additional information about artists in their collections.
The topic of The Nordic Museum’s pilot is to record metadata of the imagery from their exhibition about British fashion. The audience is invited to describe the exhibition pieces in a description campaign. This enriched information is ingested into The Nordic Museum’s collections management system as multilingual Wikidata tags.