How to avoid spending time on repetitions
Thread poster: Pascale Pluton

Pascale Pluton  Identity Verified
Netherlands
Local time: 06:41
Member (2005)
English to French
+ ...
Feb 5, 2008

Hello there,

I may soon have the opportunity to translate a large projects consisting of 7 files. The word total is 95.147, half of which are repetitions.

I am looking for a way of saving time spent on repetitions (the time when Trados runs throught the repetitions in the document and when you can do nothing else but watch, which is, in my opinion, a waste of time)

I was wondering whether it would be possible to somehow gather in one document only one occurence of all the segments (leaving the repetitions behind). Repetitions could be automaticly dealt with by Trados at a later stage (option "tools", "translate").

Would someone have an idea?


Direct link Reply with quote
 

Renée van Bijsterveld  Identity Verified
Netherlands
Local time: 06:41
Member (2007)
English to Dutch
+ ...
export frequent segments Feb 5, 2008

You could use the option Export frequent segments (in the Analysis window) to create a text file with all segments that have more occurences. Than you could translate this document, and pretranslate your other files using this TM with the frequent segments (if you wish with specific colours for full matches).
Then, when translating your docs, you should use Set/Close next no 100%.

Good luck,
Renée


Direct link Reply with quote
 

Angela Dickson  Identity Verified
United Kingdom
Local time: 05:41
French to English
+ ...
Faster computer Feb 5, 2008

I used to begrudge the time Trados spent when going through repetitions - but then I upgraded my computer, and it is much faster now...

Direct link Reply with quote
 

Vito Smolej
Germany
Local time: 06:41
Member (2004)
English to Slovenian
+ ...
"Export frequent segments" Feb 5, 2008

++

thats my way of doing it. Maybe in the following fashion:

start with N = 5 and repeat:

1. export Frequent segments > N
2. translate them and feed the TM
3. pretranslate
4. N = N-1

analyze and repeat 1 to 4 until you get down to N = 1

I found PlusToys (see my profile at http://www.proz.com/profile/91005) essential for 1 and 2: it allows you to create a word file out of frequent segments (i.e. out of the exported txt file) so you can just keep banging at untraslated segments without much interlude.

The relevant issue of course is, how much you get paid for all this. The suggestion above is only the suggestion how to boost your productivity.

[Edited at 2008-02-05 20:07]


Direct link Reply with quote
 

René Stranz-Nikitin  Identity Verified
Czech Republic
Local time: 06:41
Czech to German
+ ...
What about the segments the CAT isn't able to segment right? Feb 5, 2008

ReneevB wrote:

You could use the option Export frequent segments (in the Analysis window) to create a text file with all segments that have more occurences. Than you could translate this document, and pretranslate your other files using this TM with the frequent segments (if you wish with specific colours for full matches).
Then, when translating your docs, you should use Set/Close next no 100%.

Good luck,
Renée


Dear Renée,

all your ideas look great, but how do you handle all the segments, which the CAT is not able to segment right? At least in my language combinations there is no such thing as reliable segmentation rules and at most of the segments I have to correct the segmentation manually!

You do the analysis and the export of "frequent segments" with improper segmentation rules, than you translate a file received on the basis of unreliable segmentation rules, and after that you rely on the pretranslation of improper segmented documents on the basis of the translation of an improperly segmented export from the documents. At least in my language combinations such an approach would be quite useless, because the word order in every sentence can be assessed only on the basis of the whole sentence! You surely know, that our lovely CATs mostly cut sentences into pieces or consider 3 sentences as one. And segmentation rules were developed and work mostly only for the Western languages.

So how do you really want to save time on this?

Another René
www.uebersetzerrsn.web.wo.cz


Direct link Reply with quote
 

Renée van Bijsterveld  Identity Verified
Netherlands
Local time: 06:41
Member (2007)
English to Dutch
+ ...
Glad I'm not into non-western languages Feb 19, 2008

Hi René,
I don't know how to really solve this segmentation problem, for french and english I have few problems. Luckily in this case the translator herself has a choice if and how she uses the frequent phrases path.

The problems you describe lead to the conclusion you can't use CAT at all, is that correct (just curious)?


Direct link Reply with quote
 

René Stranz-Nikitin  Identity Verified
Czech Republic
Local time: 06:41
Czech to German
+ ...
Yes, but this forum is global and there are much more combinations than the western ones. Feb 19, 2008

Hi Renee,

ReneevB wrote:

Hi René,
I don't know how to really solve this segmentation problem, for french and english I have few problems. Luckily in this case the translator herself has a choice if and how she uses the frequent phrases path.

The problems you describe lead to the conclusion you can't use CAT at all, is that correct (just curious)?


I am glad, that you see the problem. Even if you would translate from English into German, it would appear. Just think about the German verbs, that divide so lovely, one part at the beginning of the sentence, one at the end. And it appears all the time in translation jobs from Slavic to western languages.

So my point of view is that in a globalized world there are relatively few language combinations, in which any kind of automated segmentation will do the trick. It doesn't! It "does" only in the non-relevant "dreams" of poorly educated PMs!

Of course, I use CAT-tools in my language pairs. They are useful, if nobody forces you to accept sales on the basis of CAT-analysis's, and to work with source files, that did not come originally from the end-client. You just have to use the tool honestly interactively, as it was intended in TWB with MS-Word. You should not be lazy to extend or shrink the segments as you see the sentences or real TUs in the original source file. This is a job that no machine can do for you. It is absolutely necessary to adjust every segment properly.

Greetings from Prague.

René Stranz-Nikitin
www.uebersetzerrsn.web.wo.cz

P.S.: English is not even my working language.


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

How to avoid spending time on repetitions

Advanced search







SDL Trados Studio 2017 Freelance
The leading translation software used by over 250,000 translators.

SDL Trados Studio 2017 helps translators increase translation productivity whilst ensuring quality. Combining translation memory, terminology management and machine translation in one simple and easy-to-use environment.

More info »
memoQ translator pro
Kilgray's memoQ is the world's fastest developing integrated localization & translation environment rendering you more productive and efficient.

With our advanced file filters, unlimited language and advanced file support, memoQ translator pro has been designed for translators and reviewers who work on their own, with other translators or in team-based translation projects.

More info »



All of ProZ.com
  • All of ProZ.com
  • Term search
  • Jobs