Re: Translations, GNOME and KDE

[Christian Rose <menthos gnome org>]

fre 2004-01-02 klockan 18.56 skrev Jeff Waugh:
> Some data on recent stable releases:
> 
>   http://lists.userlinux.com/pipermail/discuss/2004-January/002610.html

As for the stats presented, I don't know if they are correct or not
since we count "supported" differently, but they may very well be
correct. What's for sure is that KDE has, and has always had, a broader
language support in terms of number of languages supported, although
GNOME is to some extent slowly catching in in that area.

There are however, and this I think is *far* more important than just
the pure raw data on number of languages supported, notable "white
spots" on the GNOME language support map.
For example, the lack of support for some minority languages such as
Sami has already been mentioned, but what I think is even more
problematic is that GNOME, AFAIK, has not any support whatsoever for any
language solely spoken on the African continent. So there is a whole
huge continent that we are missing a lot of support for.

I've been in contact with translators for some South African languages
and been trying to invite them to improve the situation in GNOME, and
the response given has been positive, although the situation on the
support side has unfortunately not actually changed so far. So we must
remember that even if our support is up to par with KDE or even better
as with for example Indic languages, we have large areas where we're
missing out on support and need to improve. This is also an area where
everyone can help, even though you may not be a translator yourself --
if you know a language in your region or community that is not supported
at all or well enough by GNOME, then please let us know about it and if
you know some existing translators for it, so that we know about the
issue and can try to contact those translators and ask whether they are
interested in helping out the GNOME project aswell, and try to get them
started if the response is positive.

So, in my opinion the biggest problem for GNOME translation is, as also
other people have pointed out already, our marketing w.r.t.
internationalization and language support. We're simply not attracting
as many new translators for existing languages and translators for new
languages as we might perhaps need for the language support to improve
more rapidly. I think the GNOME release notes have improved dramatically
in this respect, so I think our major bottleneck right now is the
gnome.org web pages -- as a new translator and all other things being
equal, I know which of the http://i18n.kde.org/ and
http://i18n.gnome.org/ pages would attract me and give me the impression
of a more active community and my efforts being spent the best. I
believe this "feeling" might not always be true or correct, but
nevertheless I suspect the gnome.org translation pages are too static
and give too little help to attract the translation community en large.
We need help here to start making these pages incrementally better real
soon. You don't even need to be a translator!


As for the Skolelinux effort and Sami support in particular, I have been
trying to contact them and invite them to also help improve the support
in GNOME aswell, although without any response whatsoever so far. This
is not a single isolated case, but just one example and IMHO indicative
for another, perhaps even more depressing problem -- even though large
parts of our developers and hackers have learned that interdesktop
cooperation is in everyone's benefit, some translator groups on all
sides are still very much tied only to a particular project, with little
or no plans to cooperate with or support other projects or efforts.
Sometimes this may be for technical reasons such as toolkit rendering
support, but more often than not it seems it is purely for political
reasons. I think we all need to point out better that better and more
broad translation support for a language should be in everyone's
interest, and that this issue should really not be primarily related to
what applications or desktop environment the user happens to choose.


Last but not least, I think the general translation status in GNOME 2.4
was very much influenced by the decision to include so many new, huge,
and untested modules so late in the process. There were huge amounts of
new messages as a result of this, many of them not even proofread or
spellchecked as it seemed, and lots of other more or less i18n-related
problems coming from previously not enough tested software, and more
importantly too little time for fixing all of this.

Looking at this from a GNOME 2.5/2.6 perspective, and looking at the
"proposed" section of http://www.gnome.org/start/2.5/modules/, I think
there is unfortunately a danger of this piece of history repeating
itself, perhaps even more than before. There are modules like dasher
that are not even in GNOME cvs yet, and which GNOME translators have had
little chance to work on and little possibility to try to find the
problems in time.

Also, there is Evolution which is an *enormous* load of messages (over
6000 in the 1.4 branch!), and where translators have not even recieved a
single response from the maintainers yet about which branch should
really be worked on. As there seems to still be made releases from the
1.4 branch, this question is not as simple as it may seem. As a result,
translators are to this moment still primarily focusing on the
evolution-1-4-branch. And as the 2.6 release is not that far away from
now, and the amount of messages from Evolution alone is so immensely
enormous to say the least, I think there will be very little chance for
most translators to be able to fix that in time.
Since I'm worried not only about our language support but more primarily
the workload for translators, and since I'm also of the belief that
communication with translators is important and that we must be able to
communicate properly with maintainers, especially for modules in GNOME
core, I very much think that it's unfortunately too premature to include
Evolution already in GNOME 2.6. I'm sorry to say this, and I know there
have been tragic things happening to the Ximian developers, but looking
at it from a practical side of things from a translation point of view I
think we should better wait with Evolution for a later release.


Christian