I don't know of any true "common practice" with this problem. At my
last job, we worked with a translation company, and I'm not sure how
they managed it. (From what I could deduce, I don't think they did
particularly well. I heard them saying something about importing the
XML into RTF -- including tags -- and then using basic text comparison
functionality.)
I haven't ever used it, but the Borges project
(http://sourceforge.net/projects/borges-dms/) is a content management
system aimed at solving exactly this sort of problem. If you search the
list archives, you should be able to find some relevant information on
Borges.
Jeff Beal
Jelmer Vernooij wrote:
>Hi,
>
>I'm one of the people working on the documentation for the Samba
>project (www.samba.org). All our documentation is in Docbook/XML and
>we're happily using xsltproc with various stylesheets to generate
>HTML, PDF, PS, nroff and text output.
>
>Recently some people have started to translate the documentation to
>German and French. I did a bit of searching on internationalization in
>docbook, but I haven't been able to find much. Should we just keep
>seperate trees for all available languages?
>
>Ideally, we would like to be able to generate a document with that uses
>(for example) French, where available, and English for all paragraphs
>where there is no translation yet. Also, it would be very useful for
>the translators to be able to see what parts of their translation are
>outdated.
>
>Is there a common practice for such situations? Anyone struggling with
>the same problem?
>
>Cheers,
>
>Jelmer Vernooij
>
>
>
