First group message

[laseray]

Hi,

This is the Bitext2tmx groups first message.

I am the lead developer on the project. Unfortunately, no work has
been done on the program in a long time. I hope to spend some time
adding features in the future, but right now I am working more on
OmegaT+ (omegatplus.sf.net) translation processor.

One group member asked:

"Bitext2tmx is a great alternative to the proprietary ABBYY
Aligner!!!! Just wondering whether there are plans to add the BATCH
alignment feature to Bitext2tmx."

Batch processing is an idea, but the aligning itself can definitely
not be near 100%. You must be aware of this. The main purpose of B2T
is to allow you to produce the alignments in the form of a TMX by
editing the "pseudo" alignments (let's call them). There were some
thoughts to develop some better way of producing alignments. Unless
extreme changes are made in the design, no matter what is done there
will always be a need for post editing by someone fluent in the
languages involved (e.g., when a sentence in one language corresponds
to two sentences in the other). The program just does not have the
intelligence to work out the semantics to determine the proper
"mapping" in every situation.

Batch processing is not ruled out, and it would be fairly easy to
implement. The results would just be some realtive multiple of the
errors that you see presently when aligning a single document against
another.

Anyway, if there is more interest that will encourage me to spend time
on improving B2T sooner. If you have questions just ask. File any
feature requests or bugs on the SourceForge project page so that they
can be tracked and eventually taken care of.

Thanks.

Raymond