Quoting sankarshan ([email protected]):

> Further along your response you mention Transifex. I don't really
> recall much about it, but till the time I did keep track there was a
> series of discussions between the Tx developers and Debian team(s)
> around various issues pertaining to the workflow.

"Debian team(s)" being actually mostly /me. Indeed, from time to time,
Dmitris Gleizos come back to me asking when we can indeed "sit" and
talk about a possible use of Tx....which I never found time for.

The main obstacle is mostly the very widely distributed nature of
Debian development (no single VCS/place, translatable material spread
all over single packages, etc.)

It's likely that it happens at some point (for instance, Debian
Installer with its quite centralized development model could fit quite
well)....as soon as we take time for this. At this moment, we're
mostly working to stabilize our infrastructure and make it less
hackish to be able to move it to one of the debian.org machines.

> 
> However, that is probably orthogonal to the discussion here.

Yep, slightly..:-)

> So, here's what I was thinking when I asked the question to Karunakar.
> In recent times, GNOME and LibreOffice (I assume that's what he meant
> by OOo) have put in some efforts and resources to firm up their
> translation infrastructure. While Damned Lies may or, may not be the
> best tool out there, LibOOo uses Pootle and receives regular input
> from the Pootle developers.
> 
> The part in your response that is highlighted (*with the various l10n
> projects agreement*) is also somewhat confusing. Most of the upstream

That is the point and it indeed fits the paragraph you wrote just
before. Any local multi-projects centralisation has to be done in
coordination with the upstream l10n infrastructures and should NOT be
parallel to it (The example of Ubuntu/Canonical's and inducing double
translation effort for GNOME comes to mind)


> projects are interested in whether the languages meet their inclusion
> criteria (another orthogonal point of discussion) by delivering
> translations (and, perhaps testing, although none of the projects seem
> to, sadly, mandate testing of translations) within the time-frame
> published. They aren't probably specifically interested in whether a
> language/locale/script decides to mirror their repository content and
> then push back submissions as a downstream-upstream relationship.

They're certainly interested to be sure that translations come back to
them...but we obviously agree about this..:-)

> 
> The reason I asked the question was simple - if an infrastructure is
> put up, it requires to be well tended. IndLinux, in recent times,
> hasn't seen that kind of care. Does it make it worthwhile to go down
> that path ?

That's probably one of the critical questions....which I have no
answer about, not being (slightly by definition) involved in
IndLinux....just a longstanding follower of the project.


------------------------------------------------------------------------------
Got Input?   Slashdot Needs You.
Take our quick survey online.  Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
IndLinux-group mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/indlinux-group

Reply via email to