One of the wonderful things about DITA is that you only have to manage small chunks of information. However, this adds a particular challenge for translators who – from a logic and ease-of-understanding perspective – should translate the objects in a publication starting with the first page and ending with the last page. Currently, all objects that are sent for translation are spit out into one large folder that contains the hundreds (or thousands) of objects that require translation, sorted alphabetically or by time stamp or some other non-helpful way. If a translator wants to translate the objects in a logical way, they spend a lot of time hunting for the next file to translate on the file system.

It would be nice if SDL could build in an *option* to have all objects associated with a chapter or appendix combined into a single translatable “chapter” file, which contains the text for the objects that need to be translated - in the order in which  they appear in the map. When the translated “chapter” file gets reimported to the system, the contents are automatically reinserted into the correct objects.

Many years ago, Bryan Schnabel presented his open source tool that does this at an SDL conference. Perhaps this could be a good starting place for SDL to build on: https://sourceforge.net/projects/ditaxliff/files/