[Freecats-Dev] Re: Free CATS and gettext - some homework for us ; -)

[Henri Chorand]

Hi,

Nice to hear from you, Tim.

> (...) I have just been pointed to the website of a piece of free (GPL)
> software which certainly appears to do everything I was trying to do
> myself for FreeCATS, using most of the same components, i.e. it's
> a server-based database (MySQL/PHP) with a web interface allowing
> users to collaborate on a translation project held on the server. Each
> translated segment is updated on the server, and all other
> collaborators get the benefit of the centrally-held translation memory.

It looks interesting.


> It's based on the Free Software Foundation's "gettext" format (see
> http://www.gnu.org/software/gettext/gettext.html), which was
> designed more for software localization than for "proper" translation.

This is typically an area where more investigation work should have been
made by the early promoters of our project - myself included ;-)

Projects like gettext should have studied earlier. Why was this not done?
Because we did not really know about it, or nobody who knew about it told
us.

Duplication of efforts is also to be avoided, if possible - Free CATS would
become a practical joke if it was only to try to reinvent gettext while
ignoring what it does!

Not that they solved all our problems, but it certainly would be an error to
ignore it.


So, may I ask each person subscribed to this list (myself included) and
still interested in this project to:

1) Read gettext manual, at
http://www.gnu.org/software/gettext/manual/html_mono/gettext.html
(It's not too difficult to read, at least the beginning)


2) Keeping in mind Free CATS spec. documents, try to answer the following
questions:

- Among the gettext set of tools:

2.1    What seems most important/relevant for Free CATS ?

2.2    What seems contradictory with Free CATS ?
(I'm not saying there is anything like that, but we have to ask this to
ourselves)

Obviously, once we succeed in this:
- We should see how to integrate items found in 2.1 within our own project
- We may try to solve issues found in 2.2, or decide we want something else.


For instance, there will be pros and cons about their PO file format, which
seems to be a nice solution for translating C language source files
containing strings, but maybe this is not what we consider ideal, and maybe
we should favour something more along the lines of XML.


And once we know a little more about gettext, we should be able to contact
them in order to let them know about what we will be trying to do. I believe
they should like the idea of an all-purpose (read: non limited to C language
style resource files) TM server.


(Yes, I know we talked about this .po file format before on this list, but
it's too bad nobody mentioned it was associated to the gettext project)



Well, I really hope I can read each of you soon,

Henri