Re: Translation statistics for Gnome 2.4 desktop and developerplatform

[Christian Rose <menthos gnome org>]

mån 2003-08-18 klockan 04.37 skrev Luis Villa:
> I guess what I'm curious about is change over time- it seems like a lot
> fewer languages will be in the 90-95% range than there were for 2.2.
> Maybe I'm mis-remembering, though.
> 
> If I'm correct- does this indicate we need a longer string freeze in 2.6
> or what?

One reason is most likely that several not-so-tiny modules that didn't
use to have many translations have been added to 2.4, and judging from
looking at the stats for some not-complete-yet languages, gnopernicus
and gok seems to be the main culprits here.

But I'm not sure how to solve that, and I'm not convinced that a longer
string freeze would necessarily help that much by itself -- I guess much
of the difficulty here is coming from specialized accessibility
terminology that most translators are unfamiliar with, and also a
somewhat bad UI/terminology. In retrospect, I think that a general
UI/terminology review of these modules would have been needed well
before the string freeze, and perhaps also some documentation (or just
source code/po comments) that briefly explains the specialized
accessibility terminology to translators.

Also, as Ole Laursen points out, schemas messages are as always a big
load of work.
But I don't agree with the verbosity being the main culprit here --
rather, I think verbosity is often needed to properly describe these
keys, their values, and what they affect.
I rather think that there are perhaps too many keys. While it's great
that the descriptions for the keys can be localized, it seems, to me at
least, that some modules are starting to abuse the possibility of gconf
preferences -- while extreme and unnecessary configurability is banned
from the user interface these days due to usability demands, it seems in
some cases that this is only moved to the gconf level instead. While
that largely solves the usability issues, it still creates problems for
translators as all of these key descriptions still have to be
translated.
I can't cite specific examples right now though, this is more a general
feeling after having translated loads of schemas descriptions. Perhaps
just something to keep in mind for the future.


> Anyway, thanks for the general stats, Danilo- I'd definitely love to see
> them more often, and probably it wouldn't hurt to see them go to
> desktop-devel-list as well. More groups should do as the accessibility
> team has been doing and posting bugs/stats to the main list.

Yeah, sending them to gnome-i18n and desktop-devel is perhaps the best
option. I'd love to see weekly stats -- and perhaps also have included
how the percentages and level of support have changed for each language,
compared to the last report. That's the most interesting thing right now
to me at least -- to see in what direction languages are moving, and if
we perhaps need to contact some formerly-well-translated language teams
and remind them that GNOME 2.4 is actually going to be released soon,
and so that we can ask for their help in making their support complete.


Christian