You are currently working on UAT 

Search in potential duplicates over multiple TM

Former Member
Former Member

Since TM size is growing, we have to split them (I generally try to keep them under 1 Gb), it would be useful to have a tool for serching dupilcate segments across multiple TM. Does anybody know if such tool is already available as SDL app?

Parents Reply Children
  • Former Member
    Former Member in reply to Alison Field
    Yes, I installed it a couple of months ago but it required the use of a .NET version which resulted not compatible with SDL trados 2019, I then deleted the .NET and Olifant. I had to move to Win 10 because mty SDL Trados was not working anymore on Win 7.
    Is not alwyas easy to find a working configuration, then is better not to touch anything when you find the right setup.
  • Hi again 

    The only SDL AppStore tool I know of that handles TM duplicates is the following:

    SDL Translation Memory Management Utility

    With that, you can eliminate all duplicates but (as far as I know) only within one TM at a time and I don't think you have any control over which one remains (if the target translations are different).

    My preferred method of eliminating duplicates (from when I used to be in charge of TM maintenance for a previous employer) was to just export the TM to TMX then to reimport it into a new TM based on the original one (to keep any customised settings). The duplicates automatically overwrite themselves through that process until there is only one that remains. The reason for this was, as far as I recall, it is the one with the most recent 'last used on' date that remains so hopefully the best translation...

    Otherwise, I know of no way to help you work on duplicates in several TMs at once in SDL Trados Studio 2019.

    Sorry!

    All the best,

    Ali

     

  • Former Member
    Former Member in reply to Alison Field
    Thank you. The reason for a tool handling multiple TMs would be useful indeed. As an example, every year I have 400.000+ units from the EU Court (TM are about 1 to 2 Gb), I have to split every year or even less, I had one one TM with 8 Gb(!), which cannot be handled and is responsible for crashes and instability. Many segments are repeated in almost all TMs, resulting in a slowing down of SDL. To be sure a term is included, I have to take into account 14 years, since 2004. As you can imagine, is a heavy (and inefficient) task for SDL to cycle 14+ TMs, totalling 10+ Gb, and finding out that many segments are repeated.