Clean Sheet?

[Peter Damoc]

On Wed, 12 Jan 2005 12:07:46 -0800, Jonathon Blake <jonathon.blake at gmail.com> wrote:

>> Translations via Rosetta should be encouraged by software maintainers
>
> What is in it for the software maintainers, to encourage Rosetta based
> translations?  Especially when they deem their existing L10N tools
> "adequate".

Better exposure which might bring better accuracy. Think about a centralised translation system where you get suggestions for a string based on previous projects, now think that one of this suggestions is wrong... maybe is mistranslated... it could be made simple for a translator to track that wrong suggestion and fix the translation of a string in program X even if he/she never heard of that program before...

> The biggest enemy of "excellent", is "good enough".
>
>> however if let's say GNOME's maintainers say something like "we don't
> want Rosetta"
>
> For software projects that have not yet established a formal L10N
> policy/procedure, then a web based approach _probably_ will be
> favorable.
>
> For software projects that have large, pre-existing L10N projects, and
> active support for them, convincing them to switch to any web based
> translation tool is going to be problematic --- especially when they
> can incorporate the web based translation toolkit into their existing
> L10N support.  [Assuming that the web based translation tools are
> distributed under a FLOSS licence, and is also available.]

A centralised translation system might mean "more" information for a translator... this should be the selling point... you join the translation community and you might even get an entire program automatically translated in a bunch of languages due to "common" strings. As I said before you might even get translators to accidentaly help you with fixing the translation of certain strings.

> If Rosetta, or Poodle or any of the other web based translation tools
> gets directly incorporated into a projects website,  translations can
> be added into the cvs more or less on a real time basis.  [Spend an
> hour translating strings for the project, save it, and viola, it is in
> the cvs, and the next build incorporates the translated strings. (Yes,
> I am talking about a hypothetical ideal, where the user does
> everything whilst connected to the web.)]

I don't think that such a system needs to be incorporated in the website to accomplish CVS synchronisation, you could as well make a connection between the project's CVS and a central translation system.

>> is a BIG chance that GNOME's translators will never touch Rosetta and
> it will be rather hard to find translators willing to take on such a
> beast.
>
> Rosetta, Poodle, and similar web based tools do offer
> maintainers/developers of projects that currently have no existing
> L10N program some tremendous advantages. Those are the projects that
> web based translation tools can target.
>
>  >However if GNOME's maintainers say something like "all translations
> must be ported to Rosetta by next release" you've got the exact
> opposite situation...
>
> Why should Gnome, or KDE, or any other similar project insist that all
> translations may only come from Rosetta?  Especially if they can add a
> web based translation tool similar to Rosetta to their existing web
> based support package for L10N projects?

again with the exposure... not only would the translations already in GNOME or KDE will bring a lot to the community but they in turn could benefit from other translation projects...
A web-based translation tool on its own might not bring all that much to a project especially if the project is small BUT a web of translations could provide a lot of automagic.... even for big projects like GNOME or KDE.

A centralised translation tool could provide, at least in my view, another benefit.... celebrity of the translation project...  think about world wide exposure of this centralised system with teachers in certain countries assigning translation homework to their students.

well.... maybe I dream to much :) but hey... is better than not dreaming at all. :)

> xan
>
> jonathon


-- 
Peter Damoc
jack of all trades, master of none
http://www.sigmacore.net/