InDEsign multiple languages from 1 document

Question

I have a longish document in English [also in Spanish Russsian adn Chinese] I need to  automate teh layout  for the language versions.
I have  successfully done a sample both as a data merge  [using CSV file and as XML. I have both InDesign  CS4 and 5.5 - windows XP.
Eitehr way - I must go through teh  document and define/tag each element whenever there is a change of style. For CSV - I'm copying it into a database and then  converting that to Excel and to CSV becaue I find it easier to enter data into MS ACCESS. for XML - I need to go through and tag each elemnt.
So far so good - but I  have a huge number of tags and I'm  getting lost in the  tag names - which could also be the table headings.  There is an option within INDesign to "show snippets" of structured/tagged elements - which shows the first few  words of a line - this is great and woudl help me immensely in naming/taggign my elements.
Is there any way of capturign those snippets - other than copy paste?
Does anyone have any  other suggestions on how best to proceed with this task?I can't believe it is unique - I'm sure there are  many other appplicatiosn for   language versions. Hint/Tips would be welcome
Ann

Answer 1

Ooookay. I think I can help you in a few ways - but first I have to tell you why what you are doing is not a good idea.
Especially since neither of us  can read Chinese.
Then you shouldn't be laying it out. You'll probably just make it ugly or hard to read, but there's a small-but-decent chance that you'll introduce critical errors. I often lay out languages I can't read - but that's after years of study of their widely varying typographical traditions, and also after years of having all of my layouts format-proofed (usually harshly) by the original translators. If you can't get that service from the translators, then you shouldn't be doing the job. Sorry to be blunt about it.
Source docuemtn is in English MS Word - it has been imported into InDesign - styles  applied - no overrrides, etc.
Need to create  same documetn in Spanish, Russian, Chinese and  a multitude of  other languages. For these the source files are also in Word.
Then why can't you just place the Word files, adjust styles to have fonts that support the scripts, and clear overrides? Why is automation necessary? It sounds like overcomplication to me. Such automation is helpful to someone processing lots of work quickly in a production environment, but to do so in circumstances where no one at all can read the target language is a recipe for disaster.
I don't know about the otehr things you mentioned TRados ,etc - they're probably way too expensive. And yes - I do feel as if I'm reinventign teh wheel but cannot find any alternative.
Basically, what such tools do is that they do automatically what you are trying to do manually with Data Merge or XML Import. In the case of Trados, what it does is harvest all text content from .inx or .idml export (are you familiar with idml?), and then flow the translated text back into the doc in the exact same place from which it came.  There are very good reasons why these tools usually cost thousands of dollars.
I have a gut feeling that you have all of the intellectual tools necessary to build a pretty good XML-based translation workflow yourself. However, my other gut feeling is that it would cost about 4x - 10x as much as jobbing it out to a professional. However, that's based on assumptions about your hourly rate (I am assuming here that you're charging the way a graphic designer would in Europe/North America/etc.) that are potentially flawed. So: yes, if you're going the XML route then you will need to manually tag everything in all of the translations. That will be a lot of manual work. You will make mistakes (as you can't read the target languages). Wouldn't it make more sense to use translators who use translation tools that respect those tags? Could you possibly go back to the translators who provided the text and pay 'em to apply their translation memory databases to the XML-tagged document? You could tag up the English manually, then provide that tagged English doc to the translators and (if they are professionals using the correct tools) return translations that already have all of the content tagged exactly as it was in the English.
That being said: if your translated content is table-heavy (as you imply) then there are good reasons to use Data Merge that have nothing to do with the fact that your content isn't in English. But if you can't read the language, then of course you're going to get lost! Your inclination to get the English snippets from XML import into your CSV file makes a lot more sense than it did before, but I have no idea how to do so without doing it manually by copying & pasting into Excel.