Hoi,
Several new developments have happened since the last update..
Members of the WMF localisation team have been to several conferences and
we discussed the notion of collaborating with several organisations on
collaborating on the shared needs for quality information about languages.
One of these organisations develops LibreOffice and, the person I spoke to
mentioned a project somewhere where people could enter data about their
language. The beauty of this project is that the data entered was used to
enable a language for OpenOffice and also creates the XML needed in the
format used by the CLDR.
On the subject of the CLDR, it is really painful when you have to append or
amend information. One of our developers found several errors in the names
of languages in his mother tongue and found himself unable to change them.
This seriously begs the question if the people of the CLDR are really
interested in providing a usable interface and get information on the 6000+
languages they still have no data on.
We have also reached out to all the Wikipedia communities to join us and be
part of a language support team for their language. Many languages are
supported in MediaWiki, some have a Wikipedia others are still in the
incubator but at this time we already 77 language support teams. This is
however only a fraction of the number of languages we do support. (there is
for instance nobody who is willing and able to support English .. :) ).
As it is so vital for a language to be available for use in software like
word processors and as we do need the information in MediaWiki as well, it
could be added as a requirement in the process of making a language
eligible for a project in the Wikimedia Foundation. Typically a first
project is a Wikipedia. Given that it is so hard, almost impossible to add
data to the CLDR directly, we need a plan. Maybe the project mentioned
above can help us out. One fringe benefit is that the people who are so
active in proposing new languages, languages they do not know according to
their Babel info, will have something that does show they are serious about
what they propose.
Technically in MediaWiki, more languages are getting support with input
methods and web fonts. The official language of Bhutan, Dzongkha, has
support in WebFont, this was one of the tangible results of the Pune
hackathon. It still needs review and it became available for review in
translatewiki.net. When we visited the offices of RedHat in Pune, we worked
on the input methods for languages like Punjabi and its documentation. The
benefit of working together is not just that things move faster but also
that the sharing of code becomes practical. With a shared code base, by
sharing tools, the results of the definition of input methods or web fonts
become available to many more people. This is where collaboration, this is
where Open Source really shines.
Thanks,
Gerard
http://translatewiki.net/wiki/Language_support_team