Advice on translations
Posted: Fri Mar 29, 2019 12:42 am
I author all my content in English. For my translations, I use a translation agency that can handle DITA files. The process is pretty simple - I send them the ditamap and dita files, and they send me new files with all the content translated. Then I open the new ditamap and run my transformations.
My company is global and we have bilingual (English and local language) speakers in many countries. For this reason, our managers are considering dropping the translation agency and using our internal employees to do translations.
If this happens, I want to minimize the amount of manual (ie, non-automated) work that I have to do. I do not trust the reviewers to work on a structured file format such as DITA XML. I also do not want to teach the reviewers how to use Oxygen. So at this time, my only idea is to do this:
1. Copy each string from every step, substep, note, etc into a spreadsheet.
2. Send the spreadsheet to a translator.
3. Copy my current English DITA files into a new folder.
4. Paste all the translated strings into the copied files, resulting in new-language DITA files.
This process sounds awful! Is there a better way? I imagine that I could use XQuery to automate step 1 in the above workflow. But can I automate step 4? IE, is there any way to automatically import all the strings back to where they came from? I can't think of any way to do that.
I am wondering if Oxygen WebAuthor would solve this problem. I could create a new set of DITA files with the English content, then the reviewer could translate the content directly into the WebAuthor. This would (hopefully) eliminate the possibility of a reviewer making an error in the XML structure.
Does anyone have any other suggestions on this problem? A friend of mine mentioned XLIFF, has anyone used that successfully?
Any thoughts on solving this problem would be helpful.
My company is global and we have bilingual (English and local language) speakers in many countries. For this reason, our managers are considering dropping the translation agency and using our internal employees to do translations.
If this happens, I want to minimize the amount of manual (ie, non-automated) work that I have to do. I do not trust the reviewers to work on a structured file format such as DITA XML. I also do not want to teach the reviewers how to use Oxygen. So at this time, my only idea is to do this:
1. Copy each string from every step, substep, note, etc into a spreadsheet.
2. Send the spreadsheet to a translator.
3. Copy my current English DITA files into a new folder.
4. Paste all the translated strings into the copied files, resulting in new-language DITA files.
This process sounds awful! Is there a better way? I imagine that I could use XQuery to automate step 1 in the above workflow. But can I automate step 4? IE, is there any way to automatically import all the strings back to where they came from? I can't think of any way to do that.
I am wondering if Oxygen WebAuthor would solve this problem. I could create a new set of DITA files with the English content, then the reviewer could translate the content directly into the WebAuthor. This would (hopefully) eliminate the possibility of a reviewer making an error in the XML structure.
Does anyone have any other suggestions on this problem? A friend of mine mentioned XLIFF, has anyone used that successfully?
Any thoughts on solving this problem would be helpful.