TM organisation: how do you do it? Thread poster: Eleni Makantani
|
|---|
Hello!
I use Trados and I have come up with a problem:
In the beginning I had several small TM's for different projects and when this became frustrating, I decided to unify my TM's for every language pair and for every field I work in (i.e. en-el technical, en-el legal...). The problem is that now these TM's have become too large (over 250.000 TU's) and they don't function anymore, so it seems that I am getting back to my initial small TM's.
How do you handle you... See more Hello!
I use Trados and I have come up with a problem:
In the beginning I had several small TM's for different projects and when this became frustrating, I decided to unify my TM's for every language pair and for every field I work in (i.e. en-el technical, en-el legal...). The problem is that now these TM's have become too large (over 250.000 TU's) and they don't function anymore, so it seems that I am getting back to my initial small TM's.
How do you handle your TM's? They are a real asset to every professional translator, so how do you deal with more and more projects and keep your TM's functional at the same time?
Thanks a lot! ▲ Collapse
| | | |
Jerzy Czopik 
Germany
Local time: 15:34
Member (2003)
Polish to German
+ ...
| No problem with big TMs | Jan 26, 2008 |
250,000 is not a very big TM.
It MUST work, so try to reorganise it. If this does not help, export the content, create a new one and import the content there.
As for me, I'm using one big main TM with now nearly 900,000 segmenst with no problems whatsoever. My Trados is SDL Trados 2007 SP2 (build 835).
Regards
Jerzy
| | | |
Eleni Makantani
Greece
Local time: 16:34
English to Greek
+ ...
TOPIC STARTER | So simple... | Jan 26, 2008 |
Jerzy, you were so right! I didn't know that the solution would be so simple! Thank you!
| | | |
Jerzy Czopik
Germany
Local time: 15:34
Member (2003)
Polish to German
+ ...
| Use attribute and text fields to mark you TUs | Jan 26, 2008 |
You can set up your TM so, that you will be able to mark translation units according to the project you are translating. This is the purpose of attribute and text fields.
J.
| | |
|
|
|
| ApSIC XBench and TMs | Jan 26, 2008 |
Hi Eleni,
For what it's worth, I began with a large TM and then decided to split according to different customers, just to be sure I used the correct terminology for each one (sorry Jerzy, never got to grips with attributes and things and generally forgot to change them with each project or something). I now regularly export my TMs and use them in ApSIC XBench, which is a great and incredibly smooth (free) tool for rapidly accessing TM content.
I decided to try it when... See more Hi Eleni,
For what it's worth, I began with a large TM and then decided to split according to different customers, just to be sure I used the correct terminology for each one (sorry Jerzy, never got to grips with attributes and things and generally forgot to change them with each project or something). I now regularly export my TMs and use them in ApSIC XBench, which is a great and incredibly smooth (free) tool for rapidly accessing TM content.
I decided to try it when I got fed up with Trados not finding terminology I knew was in the TM somewhere during a concordance search, and am extremely happy with it. I've saved all of my exported TMs as one project, and they're loaded in no time. An added bonus is that if I'm looking for frustrating and I type 'frustrat', the results will show all existing translations of frustrating, frustration, frustrate in all TMs, which isn't the case if you do a concordance search... OK, not very useful for this term (just given as an example), but fantastic when you're not sure how the French verb was conjugated (if I type 'frustr', I'll get frustré, frustrée, frustration, frustrant, frustrée, etc. etc. etc. you can, of course, also type the full word if you know what you're looking for.
For each translation unit found containing the term, XBench also shows from which TM it came, and you can juggle the importance of different TMs so that the results for the most pertinent TM are shown first.... and all with a simple Ctrl+Alt+Inser. Contents can also be easily transferred into the open segment.
I don't know what I'd do without it now...
(And, no I don't work for them!!)
Just thought I'd share that with you!
Katherine ▲ Collapse
| | | |
Eleni Makantani
Greece
Local time: 16:34
English to Greek
+ ...
TOPIC STARTER | Thanks for sharing | Jan 26, 2008 |
I'll try that for sure. Thank you, Κatherine!
[Edited at 2008-01-26 15:07]
| | | |
| Super TMs are not always a good idea | Jan 26, 2008 |
I still think the best thing is to keep separate TMs for each project, even if they are from the same client and have the same subject. That way, when you start working on a project, you can select the most relevant small TMs and make up a big one just for the current project. As you work on your current project, it is best to use text fields to label each new unit, or each unit that was in your existing TM but that you have modified during translation, with reference data so you can later disti... See more I still think the best thing is to keep separate TMs for each project, even if they are from the same client and have the same subject. That way, when you start working on a project, you can select the most relevant small TMs and make up a big one just for the current project. As you work on your current project, it is best to use text fields to label each new unit, or each unit that was in your existing TM but that you have modified during translation, with reference data so you can later distinguish between different TM sources when results come up in the concordance search or in XBench.
I have text labels set up for client, subject and project number. When I make a super TM for a project, I can always tell which segment came from which past project and can make better terminology choices much faster.
And please, do try XBench - Katherine resumed its main features very well, but there is a whole lot more to it. I find it extremely fast and efficient, much more so than Trados itself - and it IS free! How cool is that?
[Edited at 2008-01-26 23:39] ▲ Collapse
| | | |
| TM editing and search | Jan 27, 2008 |
I use SDLX Maintenance or Wordfast to search terms in my TM even the TM is created with Trados, Transit, Deja Vue etc.
I find SDLX Maintenance is easy to import, export, set atttibutes, edit, group, reorganize, create table, text file or translatable file. I use it with Microsoft glossary of about 1,000,000 segments without any trouble.
I use Wordfast for MS Word based projects since good functions of Word are quite helpful to edit the TM.
Regards,
Soonthon L.
| | |
|
|
|
Giles Watson
Italy
Local time: 15:34
Italian to English
In memoriam
Viktoria Gimbe wrote:
And please, do try XBench - Katherine resumed its main features very well, but there is a whole lot more to it. I find it extremely fast and efficient, much more so than Trados itself - and it IS free! How cool is that?
... Trados could incorporate XBench's goodies into its sadly limited Concordance feature!
Just a thought,
Giles
| | | |
Eleni Makantani
Greece
Local time: 16:34
English to Greek
+ ...
TOPIC STARTER
Indeed, Xbench is perfect... Fast, clear and user friendly!
| | | |
| I doubt they would ever do that | Jan 31, 2008 |
Giles Watson wrote:
... Trados could incorporate XBench's goodies into its sadly limited Concordance feature!
That would equal admitting defeat - to an absolutely free software, no less!
If only there was a way to have Trados interface with XBench... But I am already quite happy - I can't imagine working without it anymore. And if you have spare time (doesn't happen often in my case), you can align parallel texts (Canadian government, CAT.INIST, EU, etc.) and make XBench into a super dictionary, much more useful than online resources except for maybe the few useful dictionaries out there (GDT, IATE, etc.) because it would be customized to your particular needs. I did create a nice dictionary of environmental strings that way and I still can't believe how useful that is.
XBench is my Paddington Bear - you wouldn't have guessed, now, would you?
| | | |
Login to reply/comment 
Translation news related to SDL Trados
