Re: planning for translation of help

[Susan Harkus <susanh -at- CARDSETC -dot- COM -dot- AU>]

Tom Brophy's mail summed up translation management issues really well.

The other issue is "writing for translation".

Translated technical information cannot be reviewed easily. In any software
project, different engineers review different sections of the
documentation. Even if you have one or two engineers with knowledge of the
target language, you will still fall short of being able to do satisfactory
technical reviews of translated text. I know a company that assumes the
translation won't change the content but I think that is a little naive
given that the translators are usually removed from the engineering site
and that in the normal course of writing, tech writers constantly confirm
their understanding with the engineers.

For technical accuracy reasons, in particular, I am anxious to use machine
translation, as well as for speed of localisation but translation software
is only as effective as the text is well-constructed. I'm trialling writing
consciously for machine translation. Have discovered that provided you
write with a keen eye for retaining structural signposts in the text,
machine translation works really well.  I am optimistic at the moment and
certainly learning a lot about the signposts that are required to enable
the machine to parse the syntax correctly.

Would really like to have a situation where the machine translated the
text, and the localisation phase did the edit and localisation of the
output.  Of course, some tools are also better than others, which I have
discovered by comparing output from different tools for the same document
sections.

My only concern is that I haven't had a translation software company put up
its hand and say its tool translates RTF files (Help files). Does anyone
know of a tool with that capability?

Susan Harkus

 From ??? -at- ??? Sun Jan 00 00:00:00 0000=