[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: Adept 3 translation
From:       Gudmund Areskoug <gudmundpublic () gmail ! com>
Date:       2008-10-24 15:40:21
Message-ID: 4901EC65.6090106 () gmail ! com
[Download RAW message or body]

Sébastien Renard skrev:
> Le vendredi 24 octobre 2008, Vít Pelčák a écrit :
>> I do believe, that web interface would help. Not that it would be only
>> option, but additional.
>>
>> Why? It would help with collaboration. Button "translate this app"
>> could be put into any KDE app and would point to KDE web interface
>> regardless distribution. Anyone could translate and then admins would
>> approve or repair.
> 
> I did not though it could launch such discussion when I asked how to translate 
> adept...
> 
> Well, for this specific point, I don't think web translation is a good thing. 
> At least from a quality and consistence point of view. After 7 years of 
> active KDE translation, I think our translation workflow can be enhanced in 
> many ways but does not need a translation web interface. Not at all.

I can see three points where a web interface might help:
1. For existing, established team members on the go, to enable access to
QA and other translation tools that might exist on the server
2. For posting queries to the general (interested, initiated and logged
in) public to help resolve translation knots
3. For languages that are in such dire need of help that they'd really
be better off with web interface translations by the public at large
than none at all.

In at least the cases 2. and 3., upstream, real team translation should
*always* take precedence, by default.

My experience from contributing via Launchpad/Rosetta just to test if it
would be a viable way of fly-by contribution was not particularly good.
It seems to be too stilted and unflexible where it should be geared to
the needs of the upstream translation teams.

Resolving a problem with my (spam-free) login e-mail address being
autoinserted and published without warning proved to be extremely
difficult, since there didn't seem to exist any workable communication
between  Launchpad/Rosetta and the project that was open for translation
(in this case Schooltool).

Unless Launchpad/Rosetta improves on these counts, and/or another web
interface open to the public takes such things into consideration, it's
hard to see how it can be of any true benefit to KDE or any other project.

My 2 öre, BR,
Gudmund
[prev in list] [next in list] [prev in thread] [next in thread]