Re: Translation status for "core" Gnome 2.6



fre 2003-10-24 klockan 23.20 skrev Danilo Segan:
> What does everyone think of putting even the proposed modules in  
> categories in translation status pages where they would belong if they  
> were accepted?

I'm not sure I think this is a good idea. I think we should very much
aim for *only* including modules on the desktop and developer-libs
sections that we are *absolutely* sure will be included, and try to
change the contents of those sections as little as possible otherwise.
That's FWIW also what we've done in the past.

I fear that people will otherwise spend lots of work on modules that may
not be included, or even avoid spending time with any modules on those
sections in the fear of that possibly happening, and I don't think it's
worth having that risk. In contrast, I think it's a great benefit to be
able to point at those sections and be able to say "these are modules
that are very likely to be included; some more will most likely get
added, but probably fery few or none removed", as we're currently able
to do.

So, in short, I think we should follow the "nothing is included until
it's decided to be so" motto of the release team for these sections.


> E.g. so far, rhythmbox, gnome-system-tools, gnome-network and gdesklets  
> (these reside in Gnome CVS) have been proposed for inclusion in Gnome  
> 2.6, meaning there's a chance they might get included.
> 
> I think that it's better that we add all the potential modules to  
> desktop category, so we would avoid the situation we had with 2.4 --  
> gnopernicus, gok, and other (not small at all) modules where added to  
> desktop category pretty late, and many translators didn't have  
> translations available for them.

The problems with 2.4 was IMHO that we realized too late that some of
the proposed modules even had translatable messages. In other words, we
added those modules to the translation status pages not until the last
minute, and of course noone had translated them before and had a look at
the messages and reported errors in them before that point, which later
proved to be a troublesome experience, as we catched all those problems
in the regular translation cycle.

I think what we can learn from that past experience for this release
cycle is this (and in the following "we" includes anyone that wants to
help the GTP with this):

* We need to monitor the desktop-devel-list for any proposed modules.
* We need to check that those applications support translations with
gettext and intltool if needed (and point that out on the
desktop-devel-list if they don't)
* We need to make sure those applications are hosted in GNOME CVS (and
point that out on the desktop-devel-list if they're not, and why we need
them in this CVS). Even if they're not yet hosted in GNOME CVS we also
need to take a look at the application's current messages to catch
possible serious problems in them early.
* We need to make sure the applications that fulfill the above
requirements are all added to the "extras" section in the translation
status pages as soon as possible, so that bleeding edge translators can
have a look at them early if they want to.
* We (as bleeding edge translators) need to translate those applications
early to catch potential problems and bug report them. If there are
serious problems (i.e. problems that are more than a few simple bug
reports) we need to point that out on the desktop-devel-list as soon as
possible, so that 1) they are known about, 2) can get fixed asap, and 3)
that the release team knows about it and can potentially consider it if
needed when deciding which modules are in and which ones are out.

I think that by doing this together, we can avoid most of the problems
we had with gnopernicus and gok and others in the 2.4 cycle from
happening again.


> Even this can be solved by adding a 'proposals' category, and putting  
> these modules there instead of 'desktop' category. I don't think we  
> should remove them from their current categories, because they might  
> still belong there too (eg. if it's in fifth-toe).

Having a "proposals" section sounds useful, but, as I said before, I'm
not at all convinced that we should even touch the desktop or
developer-libs sections before the 2.5 module list is final.


Christian




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