Hi all,
I just found this today, from New Scientist: "learn a language, translate
the web"
http://www.newscientist.com/article/mg21328476.200-learn-a-language-transla…
It's an article about a startup (from the same fellow who did ReCaptcha)
that provides language lessons by asking the students to translate
sentences from websites - Duolingo
http://duolingo.com/ The examples used
in their own video and also the New Scientist article are all about
translating the English Wikipedia into Spanish. Has anyone had any contact
with them before?
Whilst this project provides language lessons at no-cost I do NOT expect
this system to be "free" in the FOSS sense. Nevertheless, if the
translations are valuable, and the project proves to be popular (generating
lots of translations), do we think it would be worthwhile to contact the
organisation to try and feed their "best" wikipedia translations back into
the Wikipedias as suggestions? Perhaps a bot could place it on the talkpage
of existing articles by under the heading of "suggested content from en.wp
by crowdsourced translations"? Though, I don't know how it would work for
articles that don't exist in that language yet...
From a legal standpoint I believe translations are
derivative works and
therefore, because of the Share-Alike principle, the
translations are
already legally compatible to be re-imported.
Just a thought, no idea if it can work in practice though. In any case,
Duolingo seems to be an interesting project and time will tell whether it
actually is a useful method for people to learn a language (or not)!
-Liam
Peace, love & metadata