Tips for building a better TM Thread poster: pennylane33
|
|---|
Hi all
Is there any one would like to share some tips for building
up a better translation memory?
or is there anything i need to know in order to build up a
TM with high reusability?
( not the ways to build up a TM but some points that I need
to bear in mind when trying to build a TM)
Thanks a lot!
| | | | Heinrich Pesch 
Finland
Local time: 15:09
Member (2003)
Finnish to German
+ ...
| Keep translating | Nov 9, 2011 |
You don't build a TM but translate using a Translation Environment Tool and the TM will grow. The quality of this TM depends on the quality of your work. Reusable is a TM only for material from the same author and subject you have translated before. Otherwise you can use the TM for terminology search of words or chunks of text you have translated once and forgotten since.
| | | | | Edit your source text and use the maintenance feature in your CAT | Nov 9, 2011 |
I hope all this makes sense, but I will try and explain what I do with Trados...
First of all:
Make sure the text is formatted or punctuated to divide logically into the kinds of segments that recur. There will be many in some kinds of texts and very few in others, but the TM may provide good concordance references, even where there are no segment matches.
Decide on a policy for segments like mail addresses and telephone numbers, or placeables in general... See more I hope all this makes sense, but I will try and explain what I do with Trados...
First of all:
Make sure the text is formatted or punctuated to divide logically into the kinds of segments that recur. There will be many in some kinds of texts and very few in others, but the TM may provide good concordance references, even where there are no segment matches.
Decide on a policy for segments like mail addresses and telephone numbers, or placeables in general. Then keep to it, and format your texts accordingly.
Do the same for standard phrases and terminology - make sure they are in a form that the CAT will recognise, and be consistent.
Learn where Trados - or whichever CAT you use - divides segments, and make sure your source text is edited to divide logically.
Make sure you know how to merge or divide segments logically as you translate.
First of all, it is a good investment of time to check your source text before feeding it into the CAT.
I forgot this today, and my 320-word file showed no matches! I was a little surprised, as I guessed it should have matched 70% - 80% with a job I did last week - the client had simply added a page.
But all the lines ended with hard line breaks, and practically none of the segments matched apart from the heading. 
Check for things like:
Hard line breaks in the middle of sentences or logical segments.
Remove them, or necessary, use the line break with Shift+Enter instead. Trados allows you to print it as a line break, but does not force it as a segment break.
No division where you want to divide a segment into two.
Studio 2009 allows you to divide segments wherever you want to, but it is not so easy with the earlier versions of Trados. You have to insert something that Trados recognises as a segment break - a full stop, comma or the like.
________________________
Proofread an uncleaned version of the text, and update the TM by cleaning the finished, error-free version of the text.
After you have translated, you CAN use the maintenance feature in your CAT to remove any segments you will never use again - names, addresses, obsolete terminology.
It is quite time consuming, and may not be worth the effort, but if you exchange TMs with colleagues, you should make a habit of removing SENSITIVE material or making sure you know how to filter it out when you make a TM export.
(That also calls for planning in advance, to set the filters before you start translating!)
Think along those lines, and you will be able to find your own ways of optimising your TMs to suit the kind of work you do.
Best of luck! ▲ Collapse
| | | | Phil Hand
China
Local time: 20:09
Chinese to English
| Chinese specific point | Nov 10, 2011 |
Trados and other tools often don't get the Asian comma, so you end up with a segment with 一、 sitting at the front of it. If you can remove those during translation, do.
At the same time as building up a TM, I find it useful to build up a termbase (separate program, but you can add to the termbase from within Trados). A termbase gives you individual technical terms, rather than whole segments. Next time you meet those terms, Trados will remind you to use the standard translation ... See more Trados and other tools often don't get the Asian comma, so you end up with a segment with 一、 sitting at the front of it. If you can remove those during translation, do.
At the same time as building up a TM, I find it useful to build up a termbase (separate program, but you can add to the termbase from within Trados). A termbase gives you individual technical terms, rather than whole segments. Next time you meet those terms, Trados will remind you to use the standard translation in your termbase. ▲ Collapse
| | |
|
|
|
Richard Foulkes (X)
United Kingdom
Local time: 13:09
German to English
+ ...
| Maintenance worthwhile? | Nov 10, 2011 |
Just on Christine's point of whether maintenance is worthwhile...
I'm currently running a TM of over 700,000 segments in Trados 2007. It certainly seems pointless to remove a few 'useless' segments here and there, but is there anything you can do to really 'maintain' a TM of this size and is it likely to fail when it reaches a certain size?
Any thoughts appreciated.
| | | | | Some time ago... | Nov 10, 2011 |
I used to work with TMs containing up to a 1M segments. It took Trados 2007 about 20 minutes to reorganize it. In order to optimize them, I sometimes ran an export-import cycle. I exported the TM to TMX and then imported it back to Trados. This way, if there were similar segments with different dates, only the latest versions were saved.
Richard Foulkes wrote: I'm currently running a TM of over 700,000 segments in Trados 2007.
BTW, if you open the TMX exported from Trados in Aprsic Xbench and export it again into TMX (using a different file name), you will get a file about half the size of the original TMX. Trados inserts much junk...
| | | | | thanks a lot ! | Nov 10, 2011 |
Thanks for the answers!!
I do get some insights from your guys!
Thanks~
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Tips for building a better TM | CafeTran Espresso | You've never met a CAT tool this clever!
Translate faster & easier, using a sophisticated CAT tool built by a translator / developer.
Accept jobs from clients who use Trados, MemoQ, Wordfast & major CAT tools.
Download and start using CafeTran Espresso -- for free
Buy now! » |
| | Trados Studio 2022 Freelance | The leading translation software used by over 270,000 translators.
Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop
and cloud solution, empowering you to work in the most efficient and cost-effective way.
More info » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
Translation news related to CAT tools
