[Foundation-l] A simple question on languages.

Fri Jan 25 01:29:54 UTC 2008

On 25/01/2008, Gregory Maxwell <gmaxwell at gmail.com> wrote:
> On Jan 23, 2008 9:42 PM, Brianna Laugher <brianna.laugher at gmail.com> wrote:
> > What resources are you speaking of?
>
> Any kind of resources.  The fundraising video would have us expecting
> WMF to be somehow spending money on the native languages of the
> developing world in the future.  How should that money best be spent?
> Part of the answer depends on knowing the impact of languages.

Part (and *only* part), but since we don't have six squabillion
dollars, we will never get to the end of the list that enables 99.999%
of the literate world to contribute. Since we have the equivalent of
spare change found down the back of the couch we can pick a much less
contentious number like 70% and still not go close to reaching it.

> > Volunteer resources?
> > If so, I find the question fairly moot, as we can't really "require" anything.
>
> Common misconception. You know better, think about it for a bit. We
> can exert pressure in all sorts of ways even in our existing volunteer
> systems. Consider commons, "To be featured an image's description must
> be translated into at least three languages, including at least two
> from this list".  That would have a clear impact on common's
> accessibility to people of many languages.

Limited accessibility, yes. I don't see how we can dangle a carrot to
get people to translate policy or help pages though.

> Even without stuffing it in as a requirement having clear information
> about the impact of supporting other languages will allow us to make a
> better argument to the volunteers, and providing that will hopefully
> shift their priorities a little.
>
> Going beyond that, other heavily volunteer organizations are quite
> able to set clear goals and achieve them.  That we are somewhat broken
> in that regard doesn't mean we will never get better, or shouldn't get
> better.

I set a goal for POTY to have complete translations and committee
language contacts for all the natural languages of the top 15
Wikipedias (which at the time were also all the Wikipedias with >
100,000 articles, since then Romanian and Catalan have snuck into this
list).
We did pretty well, although there was a lot of cajoling from me as
the end of the year drew closer. ‪Norsk (bokmål) (no:) was the only
one that didn't really get there in terms of translation completion.
That was 13 languages other than English, and that was quite
difficult. I would like to see us have consistent, solid results for
that set (or a similar one) before tackling any other dozens of
languages.

It's also useful to note the difference between material intended for
within the Wikimedia community (WMF info, election info, policies) and
that of external (fundraising, maybe some basic website info).
For material intended for the Wikimedia community it makes sense to
prioritise the language list according to projects listed by number of
users. [If someone has a better metric like active users in the last X
months then maybe use that instead.]

<http://meta.wikimedia.org/wiki/Table_of_Wikimedia_Projects_by_Size>
sorted by users:

1      en.wikipedia 	
2	es.wikipedia 	
3	de.wikipedia 	
4	zh.wikipedia 	
5	fr.wikipedia 	
6	pt.wikipedia 	
	commons.wikimedia 	
7	it.wikipedia 	
8	ja.wikipedia 	
9	pl.wikipedia 	
10	nl.wikipedia 	
11	tr.wikipedia 	
12	ar.wikipedia 	
13	ru.wikipedia 	
	en.wikibooks 	
14	fa.wikipedia
15	fi.wikipedia 	
16	no.wikipedia 	
17	vi.wikipedia
18	id.wikipedia 	
19 	sv.wikipedia 	
        en.wiktionary 	 	
20	he.wikipedia 	

Now that's a list that is pretty different to the languages of the top
20 Wikipedias.

Anyway, my point is that it will be decades before we need the answer
to your original question, since we are still struggling on this basic
kind of measure, which makes me wonder what the point of the
discussion is besides a flamefest.

cheers
Brianna

-- 
They've just been waiting in a mountain for the right moment:
http://modernthings.org/