[Evolution-hackers] Re: Translations not used in Evolution (was Re: Evolution and Evolution-Data-Server branched)

[Harish Krishnaswamy <kharish novell com>]

On Sat, 2005-09-17 at 23:40 +0430, Roozbeh Pournader wrote:
> Hi Harish,

> It seems that the HEAD branch was mistakenly used for releasing
> Evolution 2.4.0, at least regarding the translations. The affected
> translations (translated and checked in to the gnome-2-12 branch before
> release but not used in Evolution 2.4.0) are Welsh, Persian, Estonian,
> and possibly Bengali, Lithuanian, and Turkish (depending on when you
> actually released the package). Romanian and French may have also been
> affected, but if not, that's only because they have checked in their
> translations to both branches.
> 
Right and Sorry. I had used the HEAD for tagging evolution-2.4.0. I will
watch this in future.
I had made an assumption that the sources of HEAD and the branch were
identical [1] at that point - since 
1. The patch review mode was still on and I was monitoring all the
commits (excluding po/*) to ensure HEAD == evolution-2.4.0.
2. All commits to 'po' were being made both to the stable branch and the
HEAD. Any reason why this was not done in the case of Persian, Bengali
translations?

And I reckoned (wrongly, in hindsight) that what might be missed
b/w a release snapshot being tested and the tagging - would be a single
update - not an entire language translation. I appreciate the fact that
you (and other translators) are voluntary contributors (mostly, unpaid
too) and it would only be fair to accommodate any of your commits before
the release. 
   However, it would be helpful if you could make incremental updates
during the development cycle [2]  or have all your huge changes
committed at least a day prior to the deadline. This would limit the
damage caused by errors in packaging or missing commits.

I am handicapped by the fact that 

1. I do not know what languages I should watch out for - before carving
out a snapshot for pre-release testing.
2. The curtains are down not exactly when the tree is tagged but
precedes it by a build-test-fix-verify cycle that ranges from half a day
to one. So translators have no way of knowing for sure if their commits
on tarballs_due_date -1 would make into a release unless they put in
most of their work in advance or send me a personal alert.

I know the way things are done now sucks. I am open to suggestions on
how we can co-ordinate better in future.

> I and a few other translators from different teams would really
> appreciate if you could re-release a new version of Evolution, using the
> translations in the gnome-2-12 branch.
> 
Sure. I will release  2.4.0.1 (from *gnome-2-12* branch) with the
missing translations.

Harish

[1] The build verification by the Evolution QA team was done from rpms
obtained from HEAD not the stable branch at that point, due to lags in
updating the red-carpet channels.
[2] Most European languages already do this.