Re: open translations database

[Stefan Rieken <StefanRieken SoftHome net>]

Hello,

Well, I am sorry to say that my reply is going to be short: Evolution crashes like a nightmare all of a sudden. After editing 4 replies and getting crashes at the end, one gets the idea and makes his replies shorter.

> Dear Stephen & All

It's Stefan! ;-)


> My name is Aoife Dunne and I am the project manager responsible 
> for the GNOME Localisation at Sun.  

This is very nice. Now let me race through your topics before fate strikes again (it must have something to do with today, I broke everything already today).

> Terming Tool 
> ------------

Does it produce relevant terms? I mean, we're not waiting for a database containing words such as "electron" and "hairdryer". How many relevant terms are there anyway, and do we need a program to retrieve them? I'm mainly thinking of terms such as "scrollbar" and "listbox".

These terms may vary between different GUIs. For instance, the Amiga had - I believe they were called - a "dropdownbox", but not a "dropdownlistbox". Collecting these terms and agreeing upon them might be a little bit harder than this. I'm not sure. Anyone?

Even if we had relevant terms, people would have to look at these terms. This is where the TM comes in mind, see below.

> Translation Memory 
> ------------------

The TM is the idea I had in mind, but then come true. It would just be great to have this running on e.g. linuxi18n.org.

You wonder about licensing terms. Maybe the community wouldn't even need an open source license, if they can just use it online.

But there's a whole other thing. The TM is not feature complete, and it probably never will be if it doesn't allow for community contributions. You might call me mad on this one, but you'll have to admit that a lot of "Business Scheme" features in my proposal aren't in the tool that you describe. Most other ideas that came in reply to my mail, aren't there as well. There are language-specific needs for features as well.

Anyway, what I'm trying to say, this would be a heck of an open source project, with aid from developers of all languages, each knowing what is important for their language, and everyone working on their (good) ideas.

BTW, the Terming Tool gets really useful if you can couple it to the sentences that the TM has to work out, so that translators get the translations for the terms right away and only have to construct the sentences.

> Style Guides 
> ------------

Sounds swell. Do you think we can use these guides in such terms that the GNOME Documentation Project would be likely to adopt them? If these guides get their place at developer.gnome.org, they would get wide attention.

> * possible act a the host for the translation memory database, 
> populating newly translated products.

If this is going to work out, I think linuxi18n.org is a good starting point. We should contact them if this gets serious.

> * provide linguistic quality assurance feedback and implement 
> linguistic changes if necessary checking for grammar, spelling, 
> inconsistencies etc.

"Linguistic quality assurance feedback" is a good word for Scrabble :-)

However, I don't exactly get what you mean here: that Sun would hire translators to check for grammar? Or is this also a software solution?

Now, I am going to press "Send" (OK, "Versturen" :-)...

Greets,

Stefan