Serious problems with missing translations!

Rostislav Raykov nostalgiafed at
Tue Mar 14 06:52:23 GMT 2006

On 3/14/06, Erdal Ronahi <erdal.ronahi at> wrote:
> Hi,
> > > What about changes made via the interface? Will they override strings
> > > that are published upstream?
> >
> > Yes, they have precedence, as they are another kind of "user uploads".
> Yes, but this IS a problem. Imagine this scenario:
> All Gnome packages are in GNOME, but a team does not use Rosetta as an
> official tool. The team frequently updates the upstream translations.
> These translations are not reflected in Rosetta.
> At the same time other people translate using Rosetta. They are
> newbies, playing around, have no contact with the official team, do
> not adhere to standards. The official team does not even know of them.

Actually this is what has happened with the Bulgarian translation of GNOME.
We have build a standartization system, have bug reporting system, an
infrastructure for the team as a whole. We have translated GNOME to 100%
(not one string missed) and we regularly upload to the i18n cvs repository
of GNOME. One day I take a look at launchpad to see if I will be able to get
some po files for ubuntu-specific translations and I see that some people
that haven't contacted the team are basically translating some files that
are in Rosetta at 50-60% but they're actually all translated. And this leads
to flaws in the translations. I'm not saying their translations are bad. I
don't know. I haven't had time to look at them, but they're for sure not in
touch with us and don't recognize the standarts for translation. Also when
the team translates, some things are discussed and put up for translation
before we make it a standard.
So I think the mainstream translations strings should have an advantage. The
Rosetta strings can probably be used when there is no translation for these

Best of all,
-------------- next part --------------
An HTML attachment was scrubbed...

More information about the ubuntu-translators mailing list