Translated documents

Question

I've done some searching in the forum, but haven't found what I need.  Sorry if this question is a repeat of a topic already covered and solved, but I'm a fairly new FrameMaker user.  Feel free to direct me to a thread topic if that'll answer my query, which is this:
We create a base Instruction manual (Book) in English that may then require translation in up to 16 different languages. We then have 16 separate books, one for each language.  The time-killer is when there are edits to the English version that then need translation and inserting into each of the foreign language versions.  To save money, we usually make a file for the translation service that contains just the edited text.  This text is normally dispersed at various locations within the document.  It's difficult and time-consuming to find the proper location to insert the text in each foreign language version, especially one with an unfamiliar alphabet (Russian).
Can Paragraph Tag types be broken down and/or expanded like Sub-Layers within Layers in Adobe Illustrator?  I'm wondering if then each Tag ("Layer") could then have infinite, sequential (Body1, Body2, Body3...) numbering "Sublayers" to the end of the document.  These would be similar to how Bible verse is numbered.  Then, each numbered Paragraph Tag (Sub-Layer) could then be searchable with Find/Change.  Would this be an efficient method of replacing these bits of translated text?
Also, rather than 16 separate documents, would it be better/easier to have one document (the bulk of the manual/Book) conditionalized to include all languages?

Answer 1

Steve,
The time-killer is when there are edits to the English version that then need translation and inserting into each of the foreign language versions.  To save money, we usually make a file for the translation service that contains just the edited text.  This text is normally dispersed at various locations within the document.  It's difficult and time-consuming to find the proper location to insert the text in each foreign language version, especially one with an unfamiliar alphabet (Russian).
Your best bet would be to talk with the translation agency how best to avoid this copy & paste operation. They are most probably using Translation Memory software and can easily handle so-called "100% matches". If you take into account your own time handling the preparation steps as well as the publishing steps after translation (×16!!) it might well be less efficient than moving the full book to the agency and receive a full set of files in return. There are some stumbling blocks:
Some (most?) companies don’t value the time of internal workforce as much as they should do. They only look at external costs... what a waste of resources!
Some source language documents are in a condition which requires major work (×16!) after every translation run. The best way to handle that is working with Structured FrameMaker, but I have seen documents from well-organized teams of a very high technical quality, well-suited for translation.
Some source language documents do not use terms and phrases in a way that allows precise segment matching with foreign languages.
The last two points follow the principle »garbage in – garbage out«, if the documents are not up to the task you won’t get desired results.
Also, rather than 16 separate documents, would it be better/easier to have one document (the bulk of the manual/Book) conditionalized to include all languages?
No translation tool I know of is prepared to handle multi-language sources.
IMO it is best to start with the basics before trying to find a technical solution. Of course, I don’t know anything about the technical quality of the documents you have to handle and a lot of conditions are very specific to your company environment. Try to get a seasoned consultant to optimize your processes.
- Michael