Poll: What is your biggest frustration with modern Translation Management Systems (TMS) and CAT tools?
Автор темы: ProZ.com Staff
ProZ.com Staff
ProZ.com Staff
ПЕРСОНАЛ САЙТА
Jul 29

This forum topic is for the discussion of the poll question "What is your biggest frustration with modern Translation Management Systems (TMS) and CAT tools?".

View the poll results »



 
Lieven Malaise
Lieven Malaise  Identity Verified
Бельгия
Local time: 23:44
Член ProZ.com c 2020
французский => голландский
+ ...
CAT tools Jul 29

My biggest frustration is the shift to subscription models of software in general. It apparently makes a good reason to shamelessly double, triple or even quadruple prices compared to eternal licenses.

WolfgangS
Sebastian Witte
Li-Hsiang Hsu
Luis M. Sosa
Renée van Bijsterveld
Philippe Etienne
Philip Lees
 
Sebastian Witte
Sebastian Witte  Identity Verified
Германия
Local time: 23:44
Член ProZ.com c 2004
английский => немецкий
+ ...
CAT tools Jul 29

1. Paying fees for minor incremental updates. (Or rather, what Lieven said).
2. Laggy platforms. (We do not use browser-based editors when working in our own CAT tool).
3. AI features forced into them. (However, these can easily be ignored.)

Even so, we're 100% faster than when working with no CAT tool.


Maria Laura Curzi
Luis M. Sosa
Zea_Mays
Andrus Lauringson
 
Li-Hsiang Hsu
Li-Hsiang Hsu  Identity Verified
Франция
Local time: 23:44
французский => китайский
+ ...
Complexity and inconsisitency - an example of XTM Jul 29

There are at least three to five ways to connect to XTM to manage tasks, translations, and invoicing; such complexity, cumbersomeness, and lack of consistency do not reflect high-performance engineering, but rather sheer incompetence.

And its TM and concordance research barely work well.


Andrus Lauringson
 
Luis M. Sosa
Luis M. Sosa  Identity Verified
Эквадор
Local time: 16:44
английский => испанский
+ ...
Speed Jul 29

Question is for both PMs and translators. But I guess only translators will show up here.

I have the impression that I already have said this in a previous related poll.

With respect to CAT tools, large projects (>15k with large TMs) turn them to virtual elephants. QA checks also become quite slow or nightmarish, and, believe me, it does not depend on the internet connection speed. Could give 3 names, but it might sound incorrect.


 
Zea_Mays
Zea_Mays  Identity Verified
Италия
Local time: 23:44
английский => немецкий
+ ...
ouch 07:04

Luis M. Sosa wrote:

With respect to CAT tools, large projects (>15k with large TMs) turn them to virtual elephants. QA checks also become quite slow or nightmarish, and, believe me, it does not depend on the internet connection speed. Could give 3 names, but it might sound incorrect.


Ah, this touches a sensitive point! I often have to dismiss hundreds of false postives, and if you don't know how to do this in bulk, it can really drive you mad and cost a lot of time.
In a recent project with various 10ks words I asked the client to disable the live QA as it was slowing the editor page to prehistoric speed.
A very weak point in current CAT tools is surely the use of AI for translation. So far, none of them could beat trained neural engines, meaning you'll have tons of inconsistencies, ignored glossaries and TMs etc. etc. (AI bots do not remember things for longer periods.) I just hope that AI subscriptions will become so costly in the near term that these gimmics will disappear very soon!



[Bearbeitet am 2026-07-30 07:06 GMT]


The Shredder
Christine Andersen
Maria Laura Curzi
 
Li-Hsiang Hsu
Li-Hsiang Hsu  Identity Verified
Франция
Local time: 23:44
французский => китайский
+ ...
Stupid AI QA checks 07:23

While I am quite OK with AI translation, which makes postediting rather like proofreading with somehow more mindful checks, I am really annoyed with so many stupid AI QA checks, which unnecessarily waste much time. In the end, I don't think AI saves us a lot of time.

[Edited at 2026-07-30 07:24 GMT]


The Shredder
Hideki Yoshida
Zea_Mays
Andrus Lauringson
Maria Laura Curzi
 
Dan Lucas
Dan Lucas  Identity Verified
Великобритания
Local time: 22:44
Член ProZ.com c 2014
японский => английский
I feel your pain 12:12

Zea_Mays wrote:
Ah, this touches a sensitive point! I often have to dismiss hundreds of false postives, and if you don't know how to do this in bulk, it can really drive you mad and cost a lot of time.

Now, imagine that same problem but exacerbated by the fact that you're dealing with a CJK source language, where things like unit numbers (units of ten thousand, hundred million, etc.) are different to those used in English. Neither Trados nor Phrase have sensible support for this kind of thing - I had to create my own tools to deal with these issues, and many others.

Dan


Zea_Mays
Maria Laura Curzi
 
Zea_Mays
Zea_Mays  Identity Verified
Италия
Local time: 23:44
английский => немецкий
+ ...
basics 14:45

Dan Lucas wrote:

Zea_Mays wrote:
Ah, this touches a sensitive point! I often have to dismiss hundreds of false postives, and if you don't know how to do this in bulk, it can really drive you mad and cost a lot of time.

Now, imagine that same problem but exacerbated by the fact that you're dealing with a CJK source language, where things like unit numbers (units of ten thousand, hundred million, etc.) are different to those used in English. Neither Trados nor Phrase have sensible support for this kind of thing - I had to create my own tools to deal with these issues, and many others.

Dan


You'd think that's basics any CAT tool should be able to handle. Curious about your tools... In Phrase, I ended up sorting all QA flags by category or comment, marking all identical ones like in excel (mark the first entry, hold the shift key and mark the last entry) and then selecting Ignore on one of them. Gone they are! Of course, this QA madness risks to let you overlook real issues, but that's the price (I do some random checks among the flags, though).

(Glad to see you in the forums again, I think you have been greatly missed by quite some people here! )


Maria Laura Curzi
Lieven Malaise
Michele Fauble
 
Dan Lucas
Dan Lucas  Identity Verified
Великобритания
Local time: 22:44
Член ProZ.com c 2014
японский => английский
Custom QA 20:03

Zea_Mays wrote:
You'd think that's basics any CAT tool should be able to handle.

You would, wouldn't you, and yet here we are.

Curious about your tools...

Basically xliff is text, so opening a file, extracting the content of segments, and looking at the results carefully is possible, although there are many pain points and irritating edge cases.

So, if you have something like 2億3千万円 in the source, you create a list of all the combinations that number could be when translated into English (e.g. 230 million yen, or 0.23 billion yen), then you check the target sentence for the presence of any of the numbers on that list. If the checking system doesn't see any of the numbers, it alerts the user, who then checks in the CAT tool.

Similar approaches can be used for all kinds of things, but there are many wrinkles and pitfalls for the unwary, as I know to my cost.

Dan


 


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

Модератор(ы) этого форума
Jared Tabor[Call to this topic]

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

Poll: What is your biggest frustration with modern Translation Management Systems (TMS) and CAT tools?






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! »
Anycount & Translation Office 3000
Translation Office 3000

Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.

More info »