[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Debian Debconf Translation proposal ( again )

On Wed, Jan 07, 2004 at 09:03:41PM +0100, Tim Dijkstra wrote:
> On Wed, 7 Jan 2004 16:14:53 +0000
> Colin Watson <cjwatson@debian.org> wrote:
> > That applies to people doing other work on packages too, not just
> > translations.
> I don't know what you mean here ... The point was, that if we could
> prevent the maintainer having to add each translation to his package,
> that would simplify things.

For you, perhaps, but as a maintainer I'd feel obligated to go hunting
around to check that these strange translations aren't too off-the-wall
(for the languages I speak or recognize), and I'd get weird bug reports
about them and be surprised because I didn't know the translations
existed or be able to look in my revision control repository for them,
etc., etc. Maintainers - competent ones, at least - are *not* uninvolved
with translations; it's always depressing when people assume they must

> Usually (when he/she doesn't speak the language) there's not a lot a
> maintainer can do with a translation other than merging it in.

Steve Langasek has already admirably refuted this, so I won't bother. I
take much the same approach as he does, with similar results.

> > And it would be less good; maintainers do pick up errors in
> > translations from time to time, which is valuable. One of the jobs of
> Some teams have a review system, which is undoubtedly much more
> efficient in finding errors than a maitainer (who doesn't even speak the
> language)

Again as Steve says, translator team reviews typically find different
problems from maintainer reviews.

> > a maintainer is to collate contributions to a package. Please stop
> > trying to subvert this or work around it! If maintainers are being
> > delinquent, they should be assisted or replaced. 
> I'm not trying to subvert anything, I'm just thinking about ways to make
> all our lives easier.

I'm afraid that this would make my life as a maintainer harder and
certain classes of problems reported to me significantly less obvious to
debug, which is one reason why I'm opposing it.


Colin Watson                                  [cjwatson@flatline.org.uk]

Reply to: