Hi Phil,

I'll work on the basis that you still want to see these segments as the context may be useful.  So in this case, if it was me I'd do this:

• First I'd run the analyze batch task and "Export frequent segments"
   
• This will create a folder in your Project Folder called "Exports" and in there you'll find an sdlxliff that contains the segments that are repeated throughout your project.
• Add this file to your Project and translate it first.  Then run a pretranslate across your entire project... maybe lock 100% matches automatically too and then when you come to translate the files these segments will be skipped as you work so you don't have to worry about them.

Hey Paul,

Thanks for your answer.

In this case, the export file will only contain the "most frequently" repeated segments, so in the 'second step' (processing all the other files) there can still remain many many words to translate that only occur once and therefore were not integrated in the export file (> so in this case there would be two translation steps + 1 revision step).

In the ideal case, there should be a workflow by which it is possible to export all the 'no match' text (to translate) to an external file (+ creating a separate TM) and once this is done pretranslating the 80.000 words with this separate TM.

In the final step (proofreading phase) it should then be possible that once a (frequently repeated) proofread segment is completely ok, it should be applied like this in all the remaining files + + possibly blocked and highlighted in another colour. Especially this last step is crucial in my eyes for the efficiency. Is this possible in Studio?

Forgive me my English if there are mistakes in expressing myself.

Thanks for any suggestions,

Phil

Hey Paul,

Thanks for your answer.

In this case, the export file will only contain the "most frequently" repeated segments, so in the 'second step' (processing all the other files) there can still remain many many words to translate that only occur once and therefore were not integrated in the export file (> so in this case there would be two translation steps + 1 revision step).

In the ideal case, there should be a workflow by which it is possible to export all the 'no match' text (to translate) to an external file (+ creating a separate TM) and once this is done pretranslating the 80.000 words with this separate TM.

In the final step (proofreading phase) it should then be possible that once a (frequently repeated) proofread segment is completely ok, it should be applied like this in all the remaining files + + possibly blocked and highlighted in another colour. Especially this last step is crucial in my eyes for the efficiency. Is this possible in Studio?

Forgive me my English if there are mistakes in expressing myself.

Thanks for any suggestions,

Phil

Will it be translated by one single person? Or do you want to split the task to several translators?

Hi Jerzy,

Thanks. Yes, to translate by 1 person :-)

Hi Philip

So the situation is very similar to what I do. In such case I would not try to go a way Paul describes, mainly because your repetitions need special treatment.

TBH I would open all the files together and try to work me along with the AutoPropagation solution set to ask me before propagating and to confirm all AP segments afterwards. Maybe using the display filter for the repetitions may be of a little help, but I am afraid exporting the no matches and translating them separately or the repetitions to do the same will not make the process faster or smoother.

Usually I open files of such size alltogether and work in the described way.

You can gladly contact me off-forum, if you wish - my phone number is +49 231 801645, provided we could speak German, English or Polish.

Kind regards, Jerzy

Thank you for your kind Answer Jerzy.

I was also thinking in that direction: opening the files together and then playing with filters and options, but I don't know which filters/options/etc. during the translation phase and the proofreading phase? It would be kind if I can call you therefore in the next days?

Have a very nice day,

Phil

Feel free to drop me a line via jerzy at czopik dot com or to call - talking about it is much faster than writing.

The display filter has several options, like showing repetitions only, only the first occurrences and so on. But you can also filter by words or more with regex.

Have a nice day too and best regards - Jerzy

Hi both... post back in here what you finally found to be the fastest method.  It would be interesting for me and others I think.