GnuCash Localizations

[Christian Stimming]

Am Samstag, 16. Februar 2008 11:25 schrieb Thilo Pfennig:
> GnuCash is one of the
> applications where localization is more than just translating the
> interface - or potentially it could.

Yes. On the other hand I think this has been handled in the gnucash core 
distribution just fine. Language teams that wanted to submit more than "just" 
a translation have very much the possibility to do so: In addition to the 
normal translations, the gnucash core repository accepts account templates, 
the documentation translation, even locale-dependent features (like 
the --enable-locale-specific-tax, as broken as it might be, but at least it 
is possible). I think from a technical point of view enough options are 
offered.

> So my idea or question is if it wouldnt be a good thing to split up the
> repository. So for example there is one core GnuCash module - and that
> this is what people get as a gnucash.tar.gz, while they would have to
> install other packages to get what makes sense for them. For example
> one could have something like gnucash-germany which could have
> dependecy to (certainly) GnuCash core, but also to some accounting
> methods or features. 

I'm sorry, but I disagree here. I don't think it is a good idea to split up 
the repository.

As I said above, IMHO the infrastructure already exists for many different 
levels of localizations. IMHO it is merely an issue of country / localization 
contributors - we don't have more localization because no further 
contributors are there. Encouraging another form of "splitting up" for 
third-party "downstream modules" IMHO would give you almost no benefit here, 
whereas you immediately get a whole bunch of drawbacks. To mention a few: API 
synchronization issues, requirements of handling more than one translation 
domain (which will be a pain in the backside in libglade), different levels 
of code auditing and proof-reading, much less cross-platform testing for the 
downstream modules, etc. etc.

Nevertheless, this is GPL software and we already have a plugin system up and 
running, hence nobody is hindering anyone to offer a "gnucash plus my 
preferred localization extensions" to the public. But speaking as a gnucash 
core developer, I can't see how a repository organization different from our 
current one would give anyone any noticable gain in terms of localization 
features.

Regards,

Christian