- From: Yves Savourel <ysavourel@translate.com>
- Date: Tue, 19 May 2009 08:34:52 -0600
- To: 'Camille Bégnis' <camille@neodoc.biz>
- CC: <public-i18n-its-ig@w3.org>
> I guess I must add some context myself about how we intend to manage this: > - Before the original file is translated, automatically add a unique xml:id > to every element that is translatable (according to ITS) > - Make a first translation > - Update the original > - Add xml:id where missing if necessary > - Increment a revision number associated to every xml:id of elements that > were modified > This way we can know exactly what elements need translation update. Does > this make sense to you? Yes. When you can control the modification of the original document like that things are much easier. Note that things will be more difficult if you have attribute values to translate, since assigning ID to them will be more complicated. So you do want to avoid that if possible. It is one of the XML i18n best practice anyway [1]. > What would contain the translation package in this case? You could use a format like XLIFF to store all the parts to translate or edit, and use the various state values for the <target> element [2] to indicate what is to be done. If you can access the previous source/translation of the modified entries, you could provide those pairs as in <lat-trans> elements, or create TMX files that can be used with translation tools. "How to package the translatable parts" goes somewhat beyond ITS, and you may find more idea and inputs for this in the XLIFF TC [3] (where you'll find a few familiar names :) [1] http://www.w3.org/TR/xml-i18n-bp/#DevAttributes [2] http://docs.oasis-open.org/xliff/v1.2/os/xliff-core.html#state [3] http://www.oasis-open.org/committees/tc_home.php?wg_abbrev=xliff Cheers, -yves
Received on Tuesday, 19 May 2009 14:35:26 UTC