Re: 2.4 still relevant?

[Christian Rose <menthos gnome org>]

mån 2003-12-29 klockan 17.46 skrev Tomas Kuliavas:
> > > I've backported a few of my translations for HEAD to gnome-2-4. But
> > is
> > > there really any point to keep doing so? According to the release
> > > schedule, there's not going to be any more GNOME 2.4 releases, so it
> > > looks like any effort directed at gnome-2-4 is wasted.
> >
> > Yeah, I think 2.4 is more or less obsolete now, and that almost all
> > translation efforts should go into 2.5/2.6 now. I've ignored 2.4 myself
> > for the last round of updates that I've committed for other translators.
> 
> It is easier to translate 2.4 first and then merge 2.4 strings with 2.6.
> 
> Instead of translating bigger .po, translator translates smaller one first
> and then translates rest of the strings in 2.6. And you get two branches
> translated instead of one.

This would be true if there weren't many messages in 2.5/2.6 that had
changed since 2.4, and where translating those 2.4 messages instead of
their 2.5/2.6 counterparts brings nothing at all. Instead of translating
once you have to do it twice, and one of those times with a large
potential of it being wasted work to begin with (see below).

Furthermore, there will most likely be many new proposed modules in
2.5/2.6 which weren't included in 2.4, so you'll miss out on all those
modules too if you are still focusing on 2.4 at this point.


> I think when debian sarge reaches stable, it will contain gnome 2.4 and
> not gnome-2.6. 2.4 translations will be used there. I think, even now
> Sarge/testing contains more gnome translations than Gargnome 0.27.1 build.

Believe it or not, everything is not Debian. There are many more popular
distributions out there, and most of them are way more uptodate than
good old Debian. So using Debian releases instead of the actual
gnome.org development work and timelines as the basis for the
prioritization of the work seems way backwards. We cannot control what
versions distributions decide to put in their releases, but we can
decide what we put in our own. So using some specific distro version
decisions as the primary decision factor for our own work is way
backwards, IMHO.

Also, there's no guarantee that there will be any more 2.4 releases. So
even if people translate 2.4 now and put their stuff into cvs, there's
no guarantee it will show up in Debian Sarge at all.

In addition, you're comparing apples with oranges. The reason there are
more translations in 2.4 is that it has been string frozen for many,
many months (almost half a year I think). GNOME 2.5/2.6 is not anywhere
near string frozen yet, so it is naturally bound to have many times
fewer complete translations. These things will change (as they have also
always done in the past) as soon as we get into string freeze and
translators can catch up.

So, to sum this up, I don't see why you're proposing translating 2.4 at
this time at all. It seems it's based on a lot of misunderstandings
about our actual development procedure. People should work on 2.5/2.6
translations now, and backport translations to 2.4 if they want to. But
2.5/2.6 should be the primary source for development now.


Christian