Translation Memory sharing (Dropbox) not working as expected

Hi,

I have been trying to share several TMs through Dropbox, as indicated here:

https://multifarious.filkin.com/2014/12/08/with-a-little-help/

Both Trados are 2017 SR1, and I am not using Project TMs. I am just trying to do a simple test with a small text file and the usual file based TMs.

I have named them TM-PC1 and TM-PC2. And I set them up like indicated in the above tutorial, this is: PC1 uses both TMs but only updates TM-PC1 and PC2 uses both TMs but only updates TM-PC2.

What I am encountering is this:

When I start a new project in both Trados, it works as expected, it reads from both TMs and when I open the file for translation there are several segments already translated, some of them coming from TM-PC1 and others coming from TM-PC2. Fine.

The problem comes later, when I translate and confirm a new segment in, for example, PC1. TM-PC1 is updated and Dropbox syncs the file in PC2, until here all normal. But then Trados will not show that concordance for the same exact segment in PC2 (the file I am translating is the same in both PC1 and PC2). This is, it does not work in real-time.

However, if I check "update" for both TMs in PC1 and PC2, then it does work in real-time. But this is not recommended because of potential write access conflicts.

Is this normal, or should it work in real-time also in the scenario where each PC updates only its correspondent TM? From the above article, I understand that it should. May I be doing something wrong?Thanks in advance.

Parents
  • Hello ,

    Sharing resources this way is a 'work around' and cannot be regarded as 100% real time or 100% reliable, as this there are many outside factors such as the processing speed of your computers, the speed and reliability of your internet connection and other factors.

    Limitations and restrictions of such setups is also explained here in more detail.

    https://gateway.sdl.com/apex/communityknowledge?articleName=000005098

    Its worth noting also that since Uplift was released with Studio 2017 (Paul's article was written a couple of years earlier) , this can slow down the access to the TM and thus the update. Fuzzy matches cannot be shared instantly.

    Lydia Simplicio | RWS Group

    _______
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

Reply
  • Hello ,

    Sharing resources this way is a 'work around' and cannot be regarded as 100% real time or 100% reliable, as this there are many outside factors such as the processing speed of your computers, the speed and reliability of your internet connection and other factors.

    Limitations and restrictions of such setups is also explained here in more detail.

    https://gateway.sdl.com/apex/communityknowledge?articleName=000005098

    Its worth noting also that since Uplift was released with Studio 2017 (Paul's article was written a couple of years earlier) , this can slow down the access to the TM and thus the update. Fuzzy matches cannot be shared instantly.

    Lydia Simplicio | RWS Group

    _______
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

Children
  • Hello, thanks for the response, I know that it is a work around, but it should have nothing to do with processing speed or reliability of connection, because as I mentioned, it works flawlessly in real time (of course, there are some seconds of lag until Dropbox syncs the files, but that's totally ok) when I tick "update" for both TMs, but it does not work at all in real time (not even after hours) when only one TM is being updated and the other one being only read. So it must be a Trados thing. Maybe it is the normal operation (which would be awkward, in my opinion), or it is a misconfiguration by my part.

    And it should have nothing to do neither with fuzzy matches, I am talking just about 100% matches.