Grants talk:IdeaLab/Translation platform: Minority Translate

From Meta, a Wikimedia project coordination wiki

Similar initiative already in progress[edit]

See https://www.mediawiki.org/wiki/Content_translation . How is your idea different? Calliopejen1 (talk) 02:35, 13 December 2014 (UTC)[reply]

I am fully aware on this other thing. For instance Minority Translate:
  • is more focused on fast creation of articles and building a framework -- something rather important for smaller language versions;
  • is usable when your're offline;
  • supports all languages.
So far it has seen a rather minimal development specially compared to Content translation tool, but I think it has a lot more potential and with full functionality it would easily outweigh the CT. Some remarks based on my experience with Estonian Wikipedia:
  • Bigger versions don't actually need translations that much: they are fully capable of building their own content (full article translations are extremely rare in et wiki; we've even had translation competitions and tried to get people into making translations, but the interest is very low in this). With more manpower it is still good to use some parts of the existing stuff, but you can build upon it and new quality emerges (something you can't have with exact translation). And well... making translations just isn't that fascinating for most of the people. You really need to add something new to it.
  • Small wikis (like võro for instance) need to get their article count up. To help them, you need to focus on speed. How to increase the content in wide variety of topics as fast as possible to reach some threshold, where wiki really becomes usable and starts to attract new content creators (in et.wikipedia it happened in ca 2006). + Very important function could be recommendations (i.e. what articles to create), as this is often one limiting factor (when you don't know what to do, then the easy solution out is not to do anything / it takes time to find the "right" articles to start working with / with better coverage on more important topics you'll have more reach).
Maybe it could be put like that: Content translation is more like man helping a machine and with Minority Translate it is a machine helping man. First one is indeed more effective at making a lot of long translations, but last one is more interesting. Not to mention the limitations. When considering the needs of small wikis (that could benefit the most from translations) and the human factor involved at creating Wikipedia (creative work as it is), then I would place may bets on the MT.
Kruusamägi (talk) 23:55, 13 December 2014 (UTC)[reply]

Other translation project[edit]

The biggest Wikimedia translation project is en:Wikipedia:WikiProject_Medicine/Translation_Task_Force. It would be nice for these two projects to partner somehow. Blue Rasberry (talk) 12:13, 17 December 2014 (UTC)[reply]

This tool can download articles in any list, that makes it faster for working with specific set of articles. It should also make the translation process faster, even thou it is still waiting for some additions, that would add a lot of functionality. So if someone is interested of testing it with medical translations, then I'm interested of hearing the feedback (and using it as a input for future development of the tool). Kruusamägi (talk) 15:53, 19 December 2014 (UTC)[reply]
Retrieved from "https://meta.wikimedia.org/w/index.php?title=Grants_talk:IdeaLab/Translation_platform:_Minority_Translate&oldid=10789275"