Re: [Freecats-Dev] Open Office & Wordfast integration - asking for a vote

[Stanislav Visnovsky]

Hi

I'm the maintainer of KBabel (http://i18n.kde.org/tools/kbabel) and I'm
following this list to find out if KBabel could use your TM server.

In case you think KBabel UI could be adapted for your project, I would be
glad to do that (if time permits - it's a free software project after all
:-).

Stanislav

P.S. Sorry for broken threading of this message :-(((

---------- Forwarded message ----------
Date: Sun, 16 Feb 2003 18:17:45 +0100
From: Henri Chorand <address@hidden>
To: Free CATS Dev list <address@hidden>
Subject: [Freecats-Dev] Open Office & Wordfast integration - asking for a
    vote

Hi all,

I believe we might establish a contact with Open Office team. There is
another reason for this apart from what I stated about their conversion
filters and spell checking module.

Ideally, and as explained in the specifications document, we translators
would like to work within a standalone translation-oriented editor, as
we know it's a good thing to keep document authoring & document
translation environments separate.

The present state of our project and the present lack of enough
experienced  developers available make me think we might begin
otherwise, even if we do remain convinced a standalone translation
editor would be a good thing and want to keep it as a medium-term goal -
I spoke about it with Julien and Marilyn and they seem to agree.

I often heard that the user interface part of a software often is the
one that requires the biggest effort, and I rather agree with it. This
may not be as much true now that several powerful GUI builders are
available, but anyway...

So the idea is, would it not be quicker to begin by implementing
translation client features from within Open Office, in a way similar to
what Trados and Wordfast do from within MS Word?


I see the following advantages to this approach:

- OO's file conversion filters
As explained before, we badly need OO's conversion filters in order to
allow translators to be able to work on various file formats. If we work
from within OO, we don't have to adapt these filters any more, as we can
use them straight away, from within OO. We "only" need to define our
working bilingual format differently, by designing custom tags (like
with Trados, or Wordfast that replicated them) which will be used from
within OO's Writer format.

- Reduced GUI work (to be confirmed)
If we do so, some work remains to be done. We still need to implement
the following functions at the client level, this time from within OO:
- establishing connection to a Free CATS server and one of its database
- sending calls to Free CATS server and handling returned data
- parsing source document and inserting custom TU-delimiter tags within it

In order to assess this option, we need to know how easy/difficult it is
to do the following, in a way compatible with OO API:
- write plug-ins (language, constraints, etc.)
- perform basic parsing/editing functions...
Any experienced MS Word knows that, to do this from within Word, one can
use a (rightly despised) macro language, and it seems there is yet no
such all-purpose macro language within OO.
Still, as OO is open source, a developer can have an in-depth knowledge
of OO's functions and add customized functions, in a way not possible
with MS Word (far from it).

To me, this looks quite feasible, but I personally lack any knowledge of
OO internals.

Anyway, if we contact OO team and ask for help, they might like the
whole idea - in fact, they might find it more attractive than the other
option, because we would put OO's Writer module at the center of our
solution.

Please, everybody, especially project team members & developers, let me
know what you think about it!


Note:
I'm sending this message to all members of Dev. list subscribers,
including Yves Champollion, Wordfast's developer, which I've recently
contacted in order to propose him to collaborate on our project.

My original suggestion was to make Wordfast able to work with a Free
CATS TM server, in order to be able to make use of our server with an
existing client. I also clearly stated that this collaboration would not
in any way alter our freedom, as a standalone free software project.

 From Free CATS project's point of view, as long as there are not any
more developers to help, with the actual team, we can reasonably hope to
have our TM server up and running within next year, thanks to the help
which the ENSTB engineering school may provide, but I'm not sure we can
go much further within this time frame.

Yves seems to have liked this idea and quickly saw that his software
could benefit from this option, but he has not yet provided any clear
feedback about whether he was willing to contribute to our project. So,
Yves, fee free to make up your mind, whatever your decision, we'll
respect it.

As Wordfast is an ethically-priced CAT software, and as it's built by a
single developer, its position is somewhat different from that of the
other proprietary CAT tool editors which made us launch Free CATS, and
this is why I contacted him in the first place. I only want to state the
following here:
My above suggestion is NOT contradictory to what Yves may, or may not,
decide to do. Anyway, we do not intend to make a Free CATS plugin for MS
Word, and if somebody is wondering about the a clear border between our
project and his software (cooperation versus competition), I think this
makes it clear enough.

And as with anybody else here, Yves is free to decide whether he wants
to contribute to Free CATS.

Henri



_______________________________________________
Freecats-dev mailing list
address@hidden
http://mail.nongnu.org/mailman/listinfo/freecats-dev