You are currently working on UAT 

  • State Suggested Answer
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 428 subscribers
  • Views 1141 views
  • Users 0 members are here
Options
Related

Trados Studio 2017/2019 bug | impossible to confirm certain segments when opening multiple files

Gaetan Chretiennot

Hello,

In a specific scenario, we cannot validate certain segments with CTRL+ENTER in Trados Studio 2017/2019.
This has been replicated on at least 4 PCs.

DOCX FILE of 30,000 words -> converted into SDLXLIFF at project creation.

Then we create 3 copies of the SDLXLIFF and rename them part1, part2, part3.
We add those 3 SDLXLIFF files to the project and copy to target languages.
In part1, we lock the last 2 thirds of the segments.
In part3, we lock the first 2 thirds of the segments.
In part 2, we look the first third and the last third.

This is because we want to receive 3 partial deliveries from the linguist and start reviewing them one by one before we inject everything back in the original DOCX/sdlxliff using PerfectMatch 3 times.

This method has been working well so far!

PROBLEM/BUG:
when the linguist opens simultaneously parts1-3, it happens for certain segments that CTRL+ENTER won't confirm the segment.

WORKAROUND:
CTRL+ENTER works fine when opening just 1 file instead of several.

Still, I thought this is a bug that shouldn't be there and thought it's worth sharing.

BR,

Gaetan

 

 

 
Rate translation
×
Suggest better translation
×
Moderator UI
×
  • Get AI Suggestion
Parents
  • 0
    Hello

    Thank you for letting us know about this.

    As for the error, indeed it seems like a possible bug, however, let me point out please that working on 3 such big files together use quite big parts of the computer's resources even if Studio runs alone on the computer, let alone if other programs also run in parallel (as most cases). This is similar to when one would like to process several huge documents in parallel in Word for example - the computer based on its computing capacity might start to push back.

    This aside, as a former translation project manager the described process itself seems to be more complicated than it should be.
    I understand that you would like the translator to deliver in batches to speed the project up, so you can process/review the already translated parts while they translate the next batch.

    What I would suggest, eliminating the need to work on 3 huge files together:
    1. Have 1 file only for translation, send it to the translator along with a project memory they would update with their translations as they go along in the file.
    2. Then the translator would send the file back with the first 1/3 translated, which the PM could send for review.
    3. Then the reviewer sends back the 1/3 reviewed file, and the PM locks the reviewed content in the file.
    4. Now the translator has translated the second part of the file, and instead of the file, they send the updated memory to the PM (with 2/3 of the translation).
    5. The PM pre-translates the file with the reviewed content already locked, and sends the 1/3 reviewed - 1/3 translated file again for review, while the translator works on the last third.
    6. The reviewed file with 2/3 of the content reviewed comes back, content gets locked.
    7. The translator now sends back the memory containing the last 1/3 of the translation, the file is pre-translated, and sent to the reviewer, who does the last third of the review and a consistency check on the full translation now.

    This might seem long in writing, but saves the long pre-processing and post-processing, with creating one file again out of the 3. And it also means less computing capacity and better performance since there is no need to process 3 huge files at the same time.

    Alternatively, it is also possible to split the file with the SDLXLIFF toolkit, if needed, and send those instead of the memory or the full file, and then the master file (with all content) could be pre-translated in the end with the memory updated with the batches. But again this takes more steps in the processing.

    I hope this can be a useful suggestion, something you might consider.

    Best wishes,
    Greta

    Greta Magyar - Customer Success Representative - SDL
Reply
  • 0
    Hello

    Thank you for letting us know about this.

    As for the error, indeed it seems like a possible bug, however, let me point out please that working on 3 such big files together use quite big parts of the computer's resources even if Studio runs alone on the computer, let alone if other programs also run in parallel (as most cases). This is similar to when one would like to process several huge documents in parallel in Word for example - the computer based on its computing capacity might start to push back.

    This aside, as a former translation project manager the described process itself seems to be more complicated than it should be.
    I understand that you would like the translator to deliver in batches to speed the project up, so you can process/review the already translated parts while they translate the next batch.

    What I would suggest, eliminating the need to work on 3 huge files together:
    1. Have 1 file only for translation, send it to the translator along with a project memory they would update with their translations as they go along in the file.
    2. Then the translator would send the file back with the first 1/3 translated, which the PM could send for review.
    3. Then the reviewer sends back the 1/3 reviewed file, and the PM locks the reviewed content in the file.
    4. Now the translator has translated the second part of the file, and instead of the file, they send the updated memory to the PM (with 2/3 of the translation).
    5. The PM pre-translates the file with the reviewed content already locked, and sends the 1/3 reviewed - 1/3 translated file again for review, while the translator works on the last third.
    6. The reviewed file with 2/3 of the content reviewed comes back, content gets locked.
    7. The translator now sends back the memory containing the last 1/3 of the translation, the file is pre-translated, and sent to the reviewer, who does the last third of the review and a consistency check on the full translation now.

    This might seem long in writing, but saves the long pre-processing and post-processing, with creating one file again out of the 3. And it also means less computing capacity and better performance since there is no need to process 3 huge files at the same time.

    Alternatively, it is also possible to split the file with the SDLXLIFF toolkit, if needed, and send those instead of the memory or the full file, and then the master file (with all content) could be pre-translated in the end with the memory updated with the batches. But again this takes more steps in the processing.

    I hope this can be a useful suggestion, something you might consider.

    Best wishes,
    Greta

    Greta Magyar - Customer Success Representative - SDL
Children
No Data