Topic on User talk:Frimelle

I just watched the talk https://media.ccc.de/v/wikidatacon2019-1060-new_usages_of_wikidata_to_support_underserved_language_communities#t=204 and it seems to me you start of talking about Wikidata in a way that ignores the ground reality. The uptick on nl-language labels in descriptions isn't because we have an active community of dutch speakers. It's because we have one bot that fills out nl labels by copying the label from other languages which in most cases would otherwise be shown as a language fallback.

Thanks for reaching out! I agree with you, and it was a very shallow interpretation of the numbers of editing and edits, because I wanted to put the focus more on reuse of the data in other projects. There is a lot more in-depth to say about how this language bias was created. However, there is also a relatively large community active on Wikidata that speaks Dutch, so I wouldn't disregard their influence (in a very positive way!) on the creation of content in Dutch. https://www.opensym.org/wp-content/uploads/2018/07/OpenSym2018_paper_18.pdf

I always thought this is very promising, as we can see that even a relatively small language community can influence the coverage of their language content- for example by bots. Could you point me to the bot that copies the labels from other languages?