Cats “Я” Us, but We Like Dogs Too!

Language Translation Software - CAT tools

The stuff below, I am sure, has all the ingredients for a cool and bizarre performance, an avant-garde sort of thing. The curtain falls, and a sombre, masked, Greek-type chorus appears. You feel some eerie gravity on stage, the chorus' collective voice is silence. Until a spotlight falls and a solitary question sounds. The chorus comes to life. One by one, voices are rising.

For an outsider, mysterious creatures that anonymous personages call "cats" make no sense. But even in their absence, they are omnipresent. Can we do without them? A whiff of conflict is noticeable in the air.

Whether an existentialist drama or rather a Monty Python farce thereof, it doesn’t matter. The script for the play is a verbatim reproduction of a discussion in a popular Facebook group for translators ("Watercooler: for translators…"). I didn’t turn "CAT-tools", as they are generally referred to, into mysterious "cats", and took liberty only to anonymize/acronymize the names of the group members. I left myself visible though and also highlighted a few points in the others' comments. (Translator's note for an outsider: "CAT tools" stand for Computer Aided Translation tools, a competitive term is TEnT – Translation Environment Tools, i.e. software.)

So why put it here? At first, I wanted to save a few comments to Evernote for my own reference. However educational and entertaining such discussions in Facebook groups are, they have a very short lifetime and disappear in the abysses of Facebook's history very soon. Deathless (ahem) drama, poetry and prose aside, this stuff doesn’t lack argumentative ingredients that are well worth remembering or, at least, being made undead.

There are many more translators who are concerned with the subject of language technology or who simply think about using (or not using) CAT tools than professional translators who use Facebook. Perhaps this script/discussion can serve as a refreshing CAT detox for all of us. The ingredients are there, by all means.


DK: I would like to know if there are still translators working without CAT-tools…

KS: I'm a Japanese-English translator who tried using CAT tools and found them to be more trouble than they're worth. Most of the things I translate have very little repetition.

DK: Same with me, K., that´s why I asked. Thanks a lot for your reaction!!!

IT: Same here. I just use them when I am required to.

SK: I know a few of them!

FF: I work with academic texts and aside from a few key words, there is practically no repetition whatsoever. I've tried CAT tools but I find them to have very little value for my area. I think they're wonderful but ultimately it's about what works for you

IHM: Definitely, I only use it when I see beforehand that the text has repetitions and that I could gain time using it. Most of the time I find it more time-consuming because I always rework on my text in Word.

JS: I know many translators which never ever use a CAT-tool – because it is simply of no use to them (marketing, litterary etc. – that is to say, all those cases where they are juggling with inspiration and language

ES: I think there are some types of translation that are still best done without, but that most translators who work on manuals, legal contracts, and such do use them. If the texts are creative, literature, etc, it is more common to not use them. Although I must say that if I translated a book I would want to know which adjective I used last time and it is easier to search in the memory….. But this is probably because I have been using CAT tools for 16 years or so….

JF: I know plenty of translators who don't use CAT tools, mainly because they aren't interested in the idea and prefer to work in Word.

ES: CAT tools gave us many options for years, but the current 'packages' are also making it less interesting for smart translators to use the tools….
Not being able to change the source text is one of the worst ideas ever implemented in CAT tools.

SK: That's true, especially since I often find typos in the source text!

CM: Sometimes they're handy even in well-written texts to help you ensure consistency over many pages. But then the job becomes more like editing than translating. A lot of my colleagues use their Trados more for checking than actual translation though, of course, that and as a quickfire dic.

TF: E., although I'm the last person to be an expert in CAT tools, I easily managed to change some terribly written source text in Hebrew in memoQ. So if I can change the source text, anyone can

ER: Interesting E., that's a kind of cross-pollination of use of CAT in various texts that can only come about from years of experience with them. I only use CAT once every two years or so on large texts with repetition and also to keep terminology consistent within a text or for a client, etc. Otherwise all non-CAT work. (Dutch-English)

ES: TF, it depends on the CAT tool and the translator. I can too….. but it has become a very complex matter n the NEW CAT tools, as the agencies try to make sure you do not translate the wrong file or make any other mistake. It is now fool-proof which in my opinion makes it only useful to fools!

TF: In memoQ, E. (just edited my original post)

DK: Thanks a lot for your answers!!


EW: Me! But I translate novels…

ES: TF, MemoQ is not walking the fool-proof package route….. Kilgray is smart….

PW: Use them for jobs where there are lots of repetitions or the content is very technical, also when requested by customers. But for a lot of marketing/PR/magazine texts etc. there is no point and dividing the text into segments actually hampers the creative process. So worth using for certain kinds of jobs but not for everything.

Me: I increasingly find CAT tools to be of little use when translating PowerPoint or InDesign documents. Typically, PowerPoint presentations consist of keywords and cues. The usual lack of context is aggravated by the "wrong" segmentation, if you just open such a document with a CAT tool. In fact, you need both to pre-process and post-process presentations when using CAT tools and can save much effort and time (and often achieve better quality results), if translating the original document directly.


WWW: Yep. Me!

MM: I'm one of them too, so far.

JF: V., my experience is the opposite – I find Trados Studio extremely helpful for translating PPT files, and find it much easier in Studio than in PPT. For example, I've never found a quick way to select all the text boxes in the PPT document and change the language to the source language. And I find it much easier to just type in Studio rather than placing the cursor in each PPT text box to overtype. Have you seen Kevin's recent video on translating PPT in Studio and memoQ?

YZ: I only use it when required.

ES: Even when working on 'creative' texts I find my CAT tools useful (mainly memoQ). The material is segmented in easy chops and processing them becomes more streamlined. No window hassle with source and target. But maybe that's just my need for structure . I also agree that consistent use of terminology and exploring your terminology archives from 15 years (not inventing the wheel again) is much easier with the terminology tools at hand (at screen!) when translating in a CAT tool.


JD: Me

GBM: I now have Wordfast and will use on projects where it is useful (and for agencies wanting CAT-tool output), but I refused to use them until this year after trying several and uniformly detesting the experience. I honestly think they harm quality. YMMV.

ES: GBM, If you translate 100 thousand words and then do an addendum of only one thousand using the CAT tool you are happy they exist!
Of course if they do not mess up the DTP you could use Tracked changes/Revision control to find the differences, but this does not work for tables and such!

GBM: E., definitely—there are absolutely times when a CAT tool is a plus.

MDM: I use Trados

CD: I don't use them! I did at first, but when I bought a new computer a couple of years ago I realised I couldn't remember the last time I'd used the CAT tool I had, so I didn't install one on my new computer. I haven't used any CAT tools since.

MDM: Many agencies work with cat tools. Here if you don't use them, you don't work.

WWW: Where's here, M.?

TF: [E., Why do you always use a person’s full name when answering their posts? It seems so artificial and unnaturall to me, living in a country where even 3-year-olds address elderly people by their first names Aren’t we all on a first name basis here too? ]

MDM: Italy but I often work with extra EU countries…

ES: T., if I use the link FB provides the full name in FB is included. When I just use the name, i.e. without the link I just use the first name!
I hate being addressed formally, so it would not be my first choice either!

ES: AND there are few Ts in our groups, but more than one Es, who are very very active!

WWW: I quite like to be addressed as 'Mister WWW'.

MDM: E., you could just do that, like me!

TF: Thanks for the laugh, W.

ES: Mea culpa, I am lazy, I use what FB dishes out


IHM: At the 2009 BDÜ-Konferenz, there was a panel discussion with three or four top notch translators (Chris Durban and three guys whose names I have forgotten) about translation quality and they asked for a show of hands who did or did not use CAT tools. The majority in this room did not.

ES: Interesting, I.!

IHM: Interesting point that emerged from that talk: one of the panel members was the head of IKEA's translation department. He explained that they had been investigating into the notable shift in translation quality. The texts for the catalogues had been slightly "off" and not only for one language but all. And the result of their investigation was that their translators had all acquired a CAT tool and were translating in a more segmented and less creative way. Stuff to think about, innit?

WWW: Now that is reassuring

ER: Yup

PG: All this is music to my ears – I find TM useless for most texts, and I hate the way agencies are trying to force it on us. There was an ITI survey about a year ago that found 40% of people didn't use it.

ES: I would expect less than 4% of people not to use CAT tools, but maybe you are right that 40% of professional translators do not

MDM: I always use them, not on my own initiative…

DC: Do any literary translators use them?

LRK:Have not yet found a way to make sense of CAT use in my work. I do encounter repetition, but the language I translate from routinely includes hugely diverse collections of meanings (even antonyms) for a single word, such that the assumption of automated plug-in for almost any term is not supportable. Passages also rarely repeat in the discursive materials I translate. I want to learn more about CAT, but I feel sure that my translations would suffer if I got too enamored of it. That said, hand-crafting very deliberate and sometimes necessarily intuitive translations is hugely time-consuming, and not yet a viable way to support myself. Given market trends, I'm not sure that financial viability will ever be possible in my language pair, but CAT will not be the way to break out of that limitation.

NF: For me, the main point of CAT tools does not lie in repetitions, but as was already said, in being able to search through TMs or corpora. Terminology management is also wonderful and allows consistency. I use MemoQ for all my projects, whereas I encounter repetitions in less than 10-15% of my work.

AK: I use CAT tools almost from the first day of work and while I do work with manuals and user guides, so there are repetitions, the feature I value most is how a CAT arranges my work in one, clear and convenient window divided to boxes, each with a certain information provided. I use CATs even in less repetitive texts, because they're just more convenient than text processors, which would require me to jump between windows and two walls of text (source and translation). There's no way I could skip or omit anything in the source text thanks to how a CAT helps me arrange the source text to translate.

TG: That's one feature of memoQ, A., that I love (from the little I've used it). I can't skip words or sentences, something I've normally a strong tendency to do.

PG: AK and TK: that seems like using a sledgehammer to crack a nut. If you're worried about leaving out text, why not just write the translation underneath each paragraph?

JDD: Because then I would have to delete everything, PG, after I am done. And I could not for the life of me always remember how (or that) I have already translated a sentence one way or another (and done the research). In the tool, I can click on "Clean" and voila! I am still hanging on to an older version of WB (not Studio!) because I like working in Word or TagEditor. I despise the table format. Nothing less intuitive than that. My next investment will be MemoQ but I could not imagine working without a CAT tool. Yes, there is a tendency to "segment" but I can expand segments and shrink them as I would do without an underlying tool. No compromise there.

TF: It's not the same, P. [Caveat: Not that I’m a great fan of CAT tools. Just mentioning a saving grace ]

KS: I decided not to use CAT tools when I saw that there were clients demanding their use for utterly inappropriate jobs (translation of open-ended consumer questionnaires, for example) and discounts for matches.

JDD: You do not have to grant discounts for matches. I am sorry but I purchased the tool (and it was not cheap!) and I have to purchase the updates and pay for training. That is enough moolah out of my pocket. I cannot grant discounts for the use of a tool that I purchased. Makes no sense to me. Now, if the client/agency provides the TM and I can skip 100% matches (at their risk), I won't bill for them. For a context check, I will charge my editing fee. Works not for everybody but it is a way to separate the wheat from the chaff… (agency-wise). And, yes, they will tell you that ALL translators give discounts. But that's not the case…and we (who do not grant discounts for fuzzy matches) survive, too…. quite well, I might add.


AK: I do not grant any discounts either. My CAT tool is for my benefit, not any agency's. After all, *I* paid for it.

Me: Finally, I feel like I am not alone Really enjoying the discussion. Feel like copying/archiving/preserving all the comments (arguments for future use) or perhaps making them public and easier to find, accessible not for the members of this group only. Anonimising the comments and posting a compilation on a blog? What do you think?

KS: As other people have said, the odd segmentation is often a problem, especially in Japanese>English translation, because the word order is so different. In addition, in a program like Wordfast Pro, with its placeables, sometimes I don't know where to put them because they're next to an element that has no direct English translation.

FK: Guilty as charged.

ES: Valerij, seeing your blogs are usually interesting and well-written you have my blessing!

Me: Thank you, E! I am really playing with the idea to put it all together, strike out the names (or leave only initials) and post on a page where it can always be found and commented if need be. The power of centralization instead of centralization of power. I have no idea about the percentage of those who use CATs and those who don't, but I think it's worthwhile to offset the idea that you're not a translator if you don't use CATs. And – so many valuable arguments pro & contra, would be a pity if all this disappears in the depths of Facebook history.

IHM: Well, there is a whole bunch of us still alive and kicking from the time Crados didn't even exist. They sure as hell changed the course of history but not in a way they can be proud of. Nor should we be proud for swallowing the deal hook, line and sinker. It would be interesting to include in the thread the number of people who actually got a CAT tool and keep it in a drawer, so to speak, because at some point they were tired of thinking and talking in segments and using ideas a machine suggested instead of their brain. I actually have two and I hardly use one of them. And I will be obliged to get a third, because that's the way it is now. I don't have to like it.

AK: My CAT doesn't suggest any ideas I hadn't put into it first and I use my brain, thank you very much. (CATs are not a machine translation!) I'm not a worse translator just because I use a tool I consider convenient. There surely are translators who started working before computers were in common use. I wonder what they think of all those who use computers today – too lazy to use ol' good pen and paper? Not everything has to be for everyone and or fit everyone's style of work or habits, but I don't see a reason to offend those who chose a different style of work.

PS: People who don't use CAT tools are those who still can't appreciate smartphones over b/w Nokia phones from 10 yrs ago. And you can edit the source also in WF Pro.

Me: Actually, this sounds wise:

Seth Godin - Now It's Ruined

IHM: A., I don't know for how long you've been doing this, but believe me that the suggestions from a tool are coming from a machine, even if you put them there in the first place and the way the machine suggests possible translations to you is totally dependent on the respective CAT tool, the context, the job details that you configured and the time of day. And yes, it's not machine translation

AK: And I have brain to accept the suggestion or to reject it and translate the text in the new context accordingly, or improve my translation (as with time and experience happens; I already dumped my database twice as outdated and/or useless). Using a CAT doesn't mean I blindly accept any suggestion. My language pair, English to Polish, often means that English identical sentences are not 100% matches in Polish due to complicated Polish grammar. I just don't appreciate the claim I don't think when working. But if there is something that makes my work smoother (and as I wrote a bit earlier, it's not even the repetitions or the terms database), I don't see a reason not to use it. If someone doesn't want to use it, I don't tell them they're worse translators and I'd rather they didn't attack my choice of working style or my intelligence.

Me: AK, I really don't think anyone said translators who use CATs are worse translators (it's rather the opposite BTW, least professional at least, that is what the CAT-less hear). I don't think anyone attacked here someone. No offense meant!

AK: Maybe no offence was meant, but I took the comment about not using brain quite personally. It was hardly a compliment.


DK:
No, I just wanted to know if I'm one of the lot of the few translators who still NOT usins a CAT-tool

IHK: In every discussion someone will invariably feel offended. That' bound to happen in a multicultural and multilingual context. I still think that a brain works differently when it has to come up with a solution from scratch or get one presented on a silver salver. Not saying that there aren't colleagues here who think twice before accepting that solution. We all should. But there are people who don't.

JD: There is some evidence that the same person will translate differently with and without a CAT tool and even some discussion as to whether what was supposed to help our profession has actually had a negative effect. No, I don't think it is a question of people being techno-phobic or not (as per the Nokia comment). It's a question of working habits, text types and what we value.

Me: Great words, J.!

ES: P., I have embraced computers, CAT tools and more for longer than most. I was a COBOL programmer at 16 (with punched cards) and LOVE computers. I am also a person who only uses the phone to be called in case of an emergency. I have a very old mobile, and replace it with a new model basic phone when required through its demise. On the other hand I translated the newest options for mobiles for years. I just do not see the point of being able to be at someone's beck and call when out of the house and/or office. My mobile/cell phone is with me to warn me of a major emergency, all else can wait.

My children have smart phones and use them I do not need it. BUT I do need fast computers and fast tools.

Even if in most cases your assessment is true, there are exceptions!

PS: I really had the feeling that someone claimed as "brainless" the use of CAT tools the sort of tools for trained monkey (Computer+Ape Translating) VS. fine, intellectual and sophisticated linguists who are real artists, while the others are mere labourers. CATs help me not to skip segments, to maintain consistency and to translate faster. I can really use my own memory as a complement to the automation provided by CATs. I can't really see how a technical translator could live without CATs.

GG: IHM, talking about IKEA… I know that they use a CSM system for some of their online stuff… that's segmented and there are strict limitations for text and you can't use a CAT…

HF: I use CAT tools for my technical stuff, but quite often, with marketing, press and other media texts it doesn't work out, as consistency is counterproductive and a stilistic nightmare in these texts quite frequently.

GG: Also, most (good) agencies are aware that using a CAT for some texts is not ideal (see the IKEA story), nevertheless, their clients insist on them because they save money (read pay less for repetitions). So, they are prepared to put up with lower quality in order to save money…

JD: Some light reading: http://usuaris.tinet.cat/apym/on-line/translation/2010_technology.pdf

Translation technology

GG: Well, as usual, the problem is not the technology itself, but how, when and why you use it…

Me: In other words, G., the problem is – always was – humans (just translating your words)

GG: Sometimes it's the technology too…

IHM: Maybe someone will compile on a blog the countless shouts for help in FB and other groups, related to CAT tools not working properly. That seems to be the downside of increased productivity.

ES: In the old days we ran out of paper…..

IHM: Oh, as to that, I usually run out of ink when I need to print something urgently, which is rare. And I know where to find paper, wheras it seems to be complicated to get support for some CAT tools ('nuff said)

ES: If and when I have a problem with a specific CAT I just check the Internet. Other people have invariably had the same issues and I find what I need. I must say it happens once in a blue moon. I have more often than not resolved other translator's issues with their CATs. I am a CAT lover and know how to stroke them so they will not scratch me!

JD: Someone needs to invent translation software with DOG as the acronym. Goodness knows, our industry is short of things to debate.

Me: JD, I already have something with DOG for the opening (no acronym, but both funny and memorable I hope

ES: Translators never work like a DOG, and are never DOG tired as they only work with or without a CAT tool

 

  1. Kevin Lossner (@GermanENTrans)’s avatar

    You know, Valerij, if I ever wanted to wish confusion on those whose professional ambitions as translators I hope to see frustrated, I would encourage them to plunge head-first into such discussions and internalize their contradictory messages and misinformation. The same would be true of discussions which look far more favorably on such tools. Pro and con, I often see the worst side of the profession in such discussions. 

    Reply

    1. ValerijTomarenko’s avatar

      Well, Kevin, this discussion, one of many or part of an ongoing one, coincided with with the publication of the results of a study on the impact of CAT tools on creativity and consistency (PDF, p.17, in German). I don’t see anything wrong with this survey (TU Chemnitz) or a discussion that the study was based on.

      On the other hand, this also coincides with the publication of the German contribution to the the development of the new DIN EN ISO 17100 norm to replace DIN EN 15308. The draft perpetuates the notorious “Four-Eyes-Principle” and effectively denies individual translators the right to operate on their own: “Der ÜDL (translation services provider) muss gewährleisten, dass die Übersetzung redigiert wird. Der Redigierende, der eine andere Person als der Übersetzende sein muss…“.

      These two events have nothing to do with each other, but if I ask who profits from the so called (1) “language technology” and (2) the old/new ISO norm, it is the so called “industry” or, in your own words, LSPs as in “language sausage providers”.

      If I try to sum up what CAT tools did for our profession (after some 15 years of using them myself), I would say that CAT tools helped the establishment of the sausage industry, were detrimental to the economic interests of genuine translation providers, led to and predated the development of MT systems.

      Having started with segmentation of texts, CAT tools are running full steam ahead to fragmentation and pulverization of translation projects. In the long run, I am afraid, they are going to eventually converge with machine translation systems.

      Perhaps I misunderstand your words about the discussion as revealing “the worst side of the profession”, but I certainly think there are worse things for us nowadays.

      Reply

      1. Kevin Lossner (@GermanENTrans)’s avatar

        Well, TM was actually spawned by old work in MT, but essentially you are right, Valerij. I think it's telling that one of the first things a company does when it buys a translation server is to engage in a great and largely useless effort to align a massive body of old materials, many of which were not well translated in the first place. I would probably be more inclined to start building corpora and studying the terminology more carefully.

        I've been wondering for ages if those idiots who keep talking about the four-eyes principle will ever start using the usual term for that in English: "dual review". "Four eyes" is a schoolyard put-down in our language and those not exposed to the suggestions of Central European standardizers would probably never think to say such a thing without a gun to their heads. But that seems to be part of what is left in the Imprint of the panzer tracks rolling over quality and good sense.

        I'll be the first to agree that CAT tools have brought many damaging effects with them while at the same time offering many opportunities for improvement. To get the full benefits of these technologies, we will have a lot of work ahead to overcome many misconceptions about the tools, in particular among their most ardent advocates.

        Reply

      2. Diana Coada’s avatar

        I've tried to use one and God knows I couldn't get used to working in that environment. I don't like CATs but I sure love my Dragon 🙂

        Reply

        1. Kevin Lossner (@GermanENTrans)’s avatar

          Did you see Kevin Hendzel's piece on voice recognition software in translation? http://www.kevinhendzel.com/professional-quality-translation-at-light-speed-why-voice-recognition-may-well-be-the-most-disruptive-translation-technology-youve-never-heard-of/

          I know a few translators who simply dictate their work, then align it in a CAT tool quickly if some client insists on having a TM. From there it's simple enough to apply any software-based QA for terms, etc. which you feel you need. There are many ways to mix and match processes to get good results. What I do not find helpful are ignorant statements about technical features based on rumor or a limited familiarity with one tool. These are then usually affirmed quickly by other ignorant and fearful persons. I see the same cycle of stupidity in arguments against using tools like Dragon – people who inform you that if only you would learn to type properly you wouldn't need such a thing. The best way to win arguments like that is to tell them they are right and should stick to their guns while you continue to do what works 🙂

          Reply

        2. Shai Nave’s avatar

          Personally, I tend not the blame the technology itself. I can see the benefit in using a Translation Environment Tool – a name that I much prefer over CAT tool because I think the latter sending the wrong message. What I tend to blame is the technology abuse, which is the work of humans, and the translators themselves are partly to blame.

           

          The obsession with translation technology is very unprofessional and is plaguing the profession. People are obsessed with workflows, ROIs (on their stupid investment that most likely was made out of the wrong reasons after believing to false claims about inhuman productivity gain and cost reduction, the latter applies only to the translators of course, don’t be silly!), compounded efficiency metrics and other empty buzzwords.  It is just one manifestation of the "industry", an amorphous, artificial, and opportunistic being that was spawned by those without a single "translation bone" in their body, nor any respect to the profession and the importance of the work, who just saw the increase in demand as a money making opportunity(ies), that when the well will eventually dry up,  will move on to leech off some other “industry” that they profit from.

           

          Technology is a tool. Professionals use tools to do their job better. The tool is there to assist the professional, it is not the professional who is there to assist the tool, and this is where so many get confused. The historic passiveness demonstrated by the translators and their penchant for being their own worst enemies is partly to blame, the MLVs and their hordes of numbskull disciples and opportunists of all sorts are partly to blame, but one party that seems to avoid most of the criticism in this debate is the tool developers themselves. Not only that it is clear that some of the tools are developed first and foremost according to the needs of large corporate end-clients (who use a CMS system) and the MLVs, who in reality usually conflict with those of the translators and more importantly don’t really account for the professional aspects of the translation work, they add insult to injury with the messages they spread about what these tools are all about and the false claims accompanying them.

           

          Translation Environment Tools have reached a plateau, most of the innovation is done by the marketing department who find creative ways to recycle old concepts and functionality as new and groundbreaking. There is so much that can be done to improve those tools – really improve them from a professional perspective, and turn them in to a real production environment (production in the sense of usefulness, not in the sense of a production line). But hey, I guess that this conflicts with the self-serving interests of the MT lobby and other snake oil vendors.

          Reply

Reply

Your email address will not be published. Required fields are marked *

53 Shares
Share28
Tweet25
Share