Wikimedia language codes

[Stephane Bortzmeyer]

On Sun, Nov 12, 2006 at 04:27:55PM -0500,
 Don Osborn <dzo at bisharat.net> wrote 
 a message of 235 lines which said:

> Take for example four related languages in western Uganda - Nyoro,
> Chiga, Nyankore, and Tooro.  Each has an ISO-639-3 code. In
> addition, Nyoro and Nyankore have ISO-639-2 codes which correspond
> to the -3 codes (neither of which have a -1 code). It might seem
> pretty straightforward that you have 4 categories. However they are
> so closely related and interintelligible to varying high degrees
> such that they might more appropriately be treated as a single
> unit. In fact, it turns out that since 1990 a standardized version
> for all 4 has been developed called Runyakitara.

For the specific case of Wikipedia, one can say that there is very
little demand to more unification of closely related languages or
dialects and a lot of demand for "I want my Wikipedia for my own
dialect/script/variant even if it is only ten pages" (there are five
or six Wikipedia for serbo-croatian, for instance).

There are a few very big Wikipedias (en, de, es, fr) and a lot of
"vanity Wikipedias" and most of the political fights take place on
these small projects with very little content.