Pages in topic: < [1 2 3] | How are you supposed to fix these kinds of CAT tool segment problems? Thread poster: Jeff Whittaker
|
Chris S wrote: What's wrong with the second option? It's not up to you to sort the TM, especially if they haven't sorted the source text. Probably it is a matter of preference. What takes preference at translation time? TM or target document? To me, it is always the TM. It just seems logical. I never get prepared projects or XLIFFs from clients. But, yes, it is an extra task to go through the entire source document to catch broken paragraphs, join lines and then, once you are done with the translation, check the target document to see whether you need to break those paragraphs again, for whatever reason, be it layout, etc. ================== EDITED TO ADD: I am sorry. I think I misunderstod what you meant by "second option". Now I am thinking that you probably meant the tool mentioned by Michael Beijer, the unbreaker. Sounds like a good tool to use, but I am not familiar with it.
[Edited at 2015-07-09 15:21 GMT] | | | I'll definitely try this program. | Jul 8, 2015 |
Thank you. Michael Beijer wrote: You can either use a better PDF converter, tweak the settings in your current PDF converter, or use a special tool to fix them in the Word document after conversion. Currently, the best tool to fix these things is TransTools, which has a special module called "Unbreaker" exactly for this kind of thing: http://www.translatortools.net/word-unbreaker.html[Edited at 2015-07-08 16:08 GMT] Directly editing the source document is easier for me than playing with the xliff file. | | | Selcuk Akyuz Türkiye Local time: 09:02 English to Turkish + ... subtitle files | Jul 9, 2015 |
Very good suggestions here, thanks to all. Wrong segmentation rules or pdf files sometimes result in bad segmentation. But it is always the case when translating subtitle files with character limitation. 50-55 character limitation in each line, sentences divided into 2-3 rows in excel files, make it almost impossible to use CAT tools. And the agency still asks for consistency in terms translated, and sometimes provides a term list. | | | Merging segments in Studio | Jul 9, 2015 |
Alas, Studio has very poor merging functionality. Other tools can merge any segments you want, Studio can only merge certain segments (not even sure what the criteria are). | |
|
|
Emma Goldsmith Spain Local time: 07:02 Member (2004) Spanish to English Studio workarounds and other tools | Jul 9, 2015 |
Olly Pekelharing wrote: Studio can only merge certain segments (not even sure what the criteria are). In Studio you can only merge segments that do not have a hard carriage return at the end (a paragraph mark in Word). If they have a hard return, then you need to go back to your original source file to fix the problem there. Another workaround, in addition to the one you added to your original post, Jeff, (and thanks for the mention, Miguel!) is to translate segments in the natural target language order and lock them so they don't contaminate your TM. In your example, you'd lock the segment where you translate "Información" as "Systems". I also agree with Olly. All other CAT tools handle this situation much better than Studio. | | |
Emma Goldsmith wrote: Another workaround, in addition to the one you added to your original post, Jeff, (and thanks for the mention, Miguel!) is to translate segments in the natural target language order and lock them so they don't contaminate your TM. In your example, you'd lock the segment where you translate "Información" as "Systems". Emma, Can you please clarify? Do you mean locking the segment or marking it as "rejected"? When I am forced to translate A for B and B for A in broken paragraphs, I mark the segments in question as "rejected" in order to avoid contaminating the TM with the wrong translations. | | | Emma Goldsmith Spain Local time: 07:02 Member (2004) Spanish to English locking/rejecting segments | Jul 9, 2015 |
Miguel Carmona wrote: Do you mean locking the segment or marking it as "rejected"? When I am forced to translate A for B and B for A in broken paragraphs, I mark the segments in question as "rejected" in order to avoid contaminating the TM with the wrong translations. I translate the segment and lock it. That way it doesn't get sent to my TM. I'm not sure how changing the status to "rejected" would work. If you've already confirmed it, it'll be in your TM, and if you haven't then your method works as long as the segment isn't changed by accident later on, e.g., by autopropagation. I lock it to prevent any possible accident. I'm not suggesting that locking segment is a best practice. It's just a workaround, like your "rejection" status approach. | | | locking/rejecting segments | Jul 9, 2015 |
The way I do it is I mark the segment as rejected and, without confirming it, I move to the next segment. This way the rejected segment does not get incorporated into the TM, but it will be part of the translated document. If I confirm the segment, even if I had previously marked it as "rejected", it will be incorporated into the TM, which is exactly what I wanted to avoid by rejecting the translation. ================= EDITED TO ADD: Yo... See more The way I do it is I mark the segment as rejected and, without confirming it, I move to the next segment. This way the rejected segment does not get incorporated into the TM, but it will be part of the translated document. If I confirm the segment, even if I had previously marked it as "rejected", it will be incorporated into the TM, which is exactly what I wanted to avoid by rejecting the translation. ================= EDITED TO ADD: You are absolutely right Emma. It is locking what prevents the segment to be sent to the TM. Rejecting the segment works too, but it is not as safe as locking it, because, if after rejecting it its state is changed, to the TM it goes, as you said! I learnt something useful today.
[Edited at 2015-07-09 16:38 GMT]
[Edited at 2015-07-09 17:00 GMT] ▲ Collapse | |
|
|
Jeff Whittaker United States Local time: 02:02 Member (2002) Spanish to English + ... TOPIC STARTER I'm beginning to wonder why we go through all this effort.. | Jul 9, 2015 |
.. for absolutely no purpose. | | | Thomas Pfann United Kingdom Local time: 06:02 Member (2006) English to German + ... Fixing the problem is easier than using a workaround | Jul 9, 2015 |
Don't forget that these are all just workarounds to a problem which usually is not too complicated to fix. I actually encountered the same problem only today: The source files were PDFs, my client (a translation agency) created the sdxliff files via the InDesign (idml) files, but my xliffs contained all the hard returns from the original PDF. This meant that some sentences were split over three or four segments. This slows down the translation, waters down any leverage from the TM a... See more Don't forget that these are all just workarounds to a problem which usually is not too complicated to fix. I actually encountered the same problem only today: The source files were PDFs, my client (a translation agency) created the sdxliff files via the InDesign (idml) files, but my xliffs contained all the hard returns from the original PDF. This meant that some sentences were split over three or four segments. This slows down the translation, waters down any leverage from the TM and contaminates the TM with lots of useless new entries. Solution: I explained the problem to the project manager, she apologized and asked her localization engineer to prepare the files again (this time properly) and a couple of hours later I received new xliffs. Problem solved. With less complicated source formats (eg. Word or Powerpoint) it is easy enough to fix the source files yourself (taking out all unwanted hard returns). But if that's not possible (like in my example from today) just get back to whoever created the project because it is them who skipped a step when preparing the files thus causing the problem in the first place. ▲ Collapse | | | Pages in topic: < [1 2 3] | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » How are you supposed to fix these kinds of CAT tool segment problems? 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 » |
| Wordfast Pro | Translation Memory Software for Any Platform
Exclusive discount for ProZ.com users!
Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value
Buy now! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |