[mailto:omegat-development-bounces@... Behalf Of Marc
>Sent: Saturday, January 27, 2007 1:19 PM
>To: Mailing list for OmegaT developers.
>Subject: Re: [OmTdev] Transparent (automatic) Word > OOo conversion
>> I don't use it from OmegaT, but directly from the desktop. Right click on
>> suitable document allows me to convert to OpenOffice or back.
>> I have also a batch script to which I can drag a complete folder.
>If a project only has one file, I open it in OOo and convert it. If the
>project has several files, I use the batch conversion wizard (now within
That's what I did in the past, but it doesn't handle RTF files, and I have
quite a number of these.
>This amounts to the same thing as your procedure: a conversion step
On the principle, yes.
Practically, I find it more convenient.
When I demoed OmegaT, the unacceptable conversion seemed much more
acceptable done this way (righ-click/convert).
>The point is, though, that if the conversion routine were called from
>OmT, OmT would appear to support MS Word natively. For some people, it is
>totally unacceptable that OmT does not do this, but the same people find it
>acceptable that other CAT tools convert to RTF and back internally. My
>suggestion is more for marketing reasons.
I perfectly understand.
>> Configuration of OOo path in OmegaT.
>> How do we know where it is?
>It should be in the system path.
No OOo in the path under Windows.
>If not, the OmT installer should be able to
>search the system for it.
>> Calling OOo with -invisible is terribly inefficient. We cannot do that
>> time a project is loaded.
>Is this any more inefficient than opening OOo just to convert files
No, but you are not converting to .doc to .odt each time you load a project,
>I am sure that if you compared my suggestion for an
>automated procedure with doing the conversion manually, the former would be
>quicker. Another possibility would be for the files to be converted only
>once, and for OmT to load the .odt file if it finds one (as it would, if
>conversion had already been done).
That's what I wrote, a kind of cache.
Don't get me wrong.
I find the idea interesting, and have been thinking about it for quite some