Hi for the WIKI -1 because : I tested a translation software software to translate the documentation: OmegaT (java / gpl) I have taken over quite quickly and it allowed me to make a beginning translation of our user guide. See here http://starsseed.free.fr/doc/userguide the quality of the result (I spent about 5 hours to install, learn,configure and translate 1% of our user guide) This software allows you to translate an HTML or XML document, generating pairs of text to create a "translation memory". The translation memory is stored in an XML file ( .tmx ) concise, easily manipulated with SVN, with no-need to store the translated target files ( you work offline and then send a small SVN diff for many files translated :-))) ) At the end of treatment, every target files are compiled into a target directory without touching the source documents. what do you think about this way to work ? Friendly Starsseed