Reply To: Auto Generated Text for WebHelp

[Derek Read]

Note that the DITA OT code that outputs PDF does handle xml:lang at the map level (so that might confuse things).

If you do both types of outputs (or might ever need to) it would be best to come down to the lowest common denominator so both work (ie: code xml:lang inside topics). It makes more sense to me to do it there anyway. A specific topic contains the actual content, which must be in one specific language (possibly more than one, but at least one).

You can then use the same map with topics in different languages (English and Japanese for example). When working on the local file system the easiest way to do that would be to name all the topics the same for both languages, which is usually the case if you send something out for translation. When your Japanese comes back from the translator it would have the same filenames for each topic but the content is now in Japanese. You then put those into an identical file structure but different folder from English. Then the map doesn't need to change case (provided it uses relative paths) though you do need a duplicate copy (one placed relative to the English and another placed relative to the Japanese). Of course if the map contains a title and metadata that needs translating it will need translating too, but the rest of the actual structure of the map (all the elements) would not need altering using this strategy.

Reply