Indeed machine translation is really hard, but assistance into translation is a more reachable goal. Focus on the terms before translate whole stuffs, suggest meaning for terms, for example. Or suggest completion while translating stuffs. And as there is already datas to feed in in the Mediawiki ecosystem as translation of stuffs, take Wikidata translated pages for example. This is not a huge dataset and there is tons of untranslated things, so it may be something that eventually makes a difference, translating things make translation easier which could motivate people.

Le ven. 20 sept. 2019 à 16:42, Nicolas VIGNERON <vigneron.nicolas@gmail.com> a écrit :

Hi all,

As a reminder, for ideas of tool, there is https://www.wikidata.org/wiki/Wikidata:Lexicographical_data/Ideas_of_tools (feel free to add your own ideas or to take an idea to make a tool)

Machine translation is a very hard and complex subject, and anyway, it needs a large set of data. It's still a good goal for the future but right now, we should focus on tool increasing and improving the data.

Cheers, ~nicolas

Le ven. 20 sept. 2019 à 16:34, Thomas Douillard <thomas.douillard@gmail.com> a écrit :

Oh, one obvious usecase that comes to my mind, after reading a discussion like this one this reminds that we are a multilingual community that could hugely benefit from some help in multilingual discussions. So if we can eat our own dog food … it’s for the better. It could indeed help for the funding as if I recall there was a recent community call for need to help cross language barriers. Wikitools to translate pages in a wiki, or cross wikis, could probably benefit from sense annotation and translation suggestions thanks to Wikidata datas …
_______________________________________________
Wikidata mailing list
Wikidata@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata

_______________________________________________
Wikidata mailing list
Wikidata@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikidata