[Ubuntu-l10n-eng] Leaving for upstream (possibly)

[Bruce Cowan]

Sorry about the lateness of this reply, but here goes.

On Mon, 2007-09-03 at 13:31 +1000, Sridhar Dhanapalan wrote:
> On Fri, 31 Aug 2007, Bruce Cowan <bruce.cowan at dsl.pipex.com> wrote:
> > I think I'll be leaving this team, not that I have done much work
> > anyway.
> >
> > My main reason is the fact that all of this work is for Ubuntu, and for
> > Ubuntu only. I don't like that situation, and saying as en_GB in GNOME
> > seems to be getting behind (95%), I will probably try to join them.
> 

I've changed my mind slightly, I'll stay in the Ubuntu team, but I'll
only work on Ubuntu-only things in Rosetta.

> I can't say that I can blame you, Bruce. For over a year, we have been trying 
> to pressure the Rosetta people to better integrate with upstream projects. I 
> know that this is on their TODO list, but for whatever reason 
> (priorities/resources/difficulty/complications/whatever) the improvements 
> aren't happening fast enough.

Fair enough, I was assured when I joined this team that work was being
done in this regard, but I have seen no evidence of this.

> Many translators are not very technically inclined, and Rosetta's gift to the 
> community is that it makes translating simple for those who can't be bothered 
> with technicalities like PO files. But if those translations can't be shared 
> with upstream, their usefulness is greatly diminished.

This is one major advantage of Rosetta, but I'm willing to learn the
"proper" way.

> The strength of a translations project at the distribution level is that we 
> can operate on a wide breadth of packages and projects at once. Working with 
> GNOME only allows modifications to GNOME packages. Working with Ubuntu allows 
> work to be done with GNOME, KDE, Xfce and so on. It also allows us to 
> identify and address consistency issues across the applications. It is not 
> uncommon to see the same element or concept named different things across 
> projects. A distribution-level translation team can see this and smooth out 
> the differences, making for a more consistent user experience.

This makes a lot of sense, but it seems slightly wrong that all this
work is only for one distribution. This is a very noble concept, but
without upstream participation, all the work is (IMHO) pointless.

> What we have been weak on is pushing those modifications upstream. As has been 
> already mentioned, Rosetta doesn't make this easy.
> 
> This is _not_ to say that we are ineffective as a team. What needs to be done 
> is better communications with and submissions to upstream, even if it must be 
> done manually at this point in time. Simultaneously, we need to keep up the 
> pressure on the Rosetta folks so that they improve their upstream 
> collaboration capabilities. If you're a coder, please pitch in and offer to 
> help them to make this into a reality.

I like the work the Brazilian Portugese team has done, see
http://www.ogmaciel.com/?p=397 for details of this.

I have signed up to the GNOME-i18n mailing list, I'll send an e-mail
asking for advice on what to do when http://l10n-status.gnome.org/
starts working again.
-- 
Bruce Cowan <https://launchpad.net/~bruce89/>