Re: [Freecats-Dev] Underlying Unix base

[Marc Prior]

Henri,

It's crucial when considering the technical side of a new application not to 
forget the business side. And when considering the business side, not to 
forget the commercial side.

Wordfast *could* no doubt be ported to OpenOffice.org, i.e. to Star Basic, 
and certainly there are some people who would like to have it. I have even 
wondered whether Sun might be interested in including it in Star Office (i.e. 
as one of the benefits of buying the commercial product rather than its free 
sister, OpenOffice.org). I don't know how difficult it would be; I can only 
speculate. OOo, incidentally, has no hidden text function - a feature which 
Wordfast, like Trados, relies on. Perhaps that's easily surmountable - I 
don't honestly know. 

However, as Yves points out, although there is some interest, it is not 
enough to justify the work. There are other factors speaking against it: if 
you want an application which interacts with OOo/SO, Star Basic is not 
necessarily a wise choice. Sun has "inherited" Star Basic along with SO, but 
I would not count on them supporting it indefinitely. Java is clearly the 
programming language of choice - but that solution, of course, involves even 
more work. Then there is the question of whether an application which can 
only be used with one word processor is a good idea. If that word processor 
is Word, OK - the market is huge. But if it's OOo, that's putting all your 
eggs in a very small basket. There IS interest in OOo, but a lot of that 
interest comes from people who object to being tied to Word. Those people 
aren't necessarily keen to tie themselves to another word processor 
(admittedly, one which is open source and has an open file format).

Some time ago, I spoke to many of the TM application vendors, including Yves, 
about the possibility of porting their products to Linux or StarOffice. At 
the moment, though, I don't see any likelihood of it happening. Hermann 
Bruns, for instance, publicly said that this was planned for Metatexis, but 
if you follow the discussions of Metatexis on the mailing lists, you will see 
that he is busy enough already trying to get Metatexis to work with all the 
different combinations of Windows/Word. The situation is similar for all the 
TM vendors. The return on investment is much greater if they improve features 
for their existing user base.

Araya and an updated version of Frankenstein were both launched in the last 
two months. These are both commercial Java apps, i.e. they will also run on 
Linux.  I don't expect a great surge of interest in these products. There are 
twenty-odd TM applications already on the Windows market, and twenty-odd 
translators in the Linux market :-)) , so either way, they will have a hard 
time of it.

The reason OmegaT is different is that it is open source. It is a completely 
different philosophy. OmegaT's future does not have much to do with technical 
aspects. It is a question of whether mainstream professional translators can 
become enthusiastic about open source, and/or the open-source community can 
be persuaded that it is in its interests to reach out to the mainstream 
translation community. OmegaT is unique in that it is the only product, at 
least at the moment, which can act as a bridge between the two communities. 
Foreigndesk is also open-source, but can't fulfil this role because it is 
strictly Windows. Some enterprising (mainstream) translators have used 
KBabel, gtranslator, etc. for their work but these applications and the 
associated procedures are simply too technically challenging, by two orders 
of magnitude, for the average professional translator. So there's the big 
chance for OmegaT.

Marc