Posted by: patenttranslator | April 21, 2016

The Battle of Two Wolverines for Our Minds at BP16 in Prague

One evening an old Cherokee told his grandson about a battle that goes on inside people.

He said, “My son, the battle is between two “wolves” inside us all.

One is Evil. It is anger, envy, jealousy, sorrow, regret, greed, arrogance, self-pity, guilt, resentment, inferiority, lies, false pride, superiority, and ego.

The other is Good. It is joy, peace, love, hope, serenity, humility, kindness, benevolence, empathy, generosity, truth, compassion and faith.”

The grandson thought about it for a minute and then asked his grandfather: “Which wolf wins?”

 The old Cherokee simply replied, “The one you feed”.

An excerpt from: You Are Not Alone – by Frances Black

I am sorry I haven’t been posting much this month. But I have a good excuse – I just came back from BP (Business Practices) 16 Conference 2016 in Prague, Czech Republic.

Unlike during my trip to the last translators’ conference – IAPTI 3 Conference in Bordeaux IAPTI 3 (International Association of Professional Translators and Interpreters), a Conference held in Bordeaux, France in May of 2015, there were no major harrowing, nearly fatal experiences during this trip to a city that I used to know so well when I lived there for over a decade more than 30 years ago.

This time I only missed the connecting flight from Frankfurt to Prague when Lufthansa sent a message about a gate change to my iPhone, which prompted me to run for about 20 minutes from one end of the Frankfurt airport to the other, only to be told by a smiling German girl at the information desk that the gate had been changed again and that now I had to traipse on my tired feet back again to the same end of the airport where I had started the long journey of a thousand gates to my final destination.

Final destination for the day, anyway, not like in the films Final Destination 1, 2, 3, 4, and 5, the “Pentalogy Horror” where everybody dies in the end.

“Fortunately, you don’t have to hurry now, Mr. Vitek, your flight is leaving in an hour so you have plenty of time”, said the German girl when she saw how heavily the overweight old man in front of her was perspiring with a smile that seemed more than slightly sadistic to me.

Other than that, the trip went without a major glitch, especially considering that during my last trip to France a few months ago, my plane was diverted to another airport, which felt like a fiendish and terrifying hijacking. After that I had to spend the night in an overpriced hotel in hot, humid and dirty Atlanta, a young French guy who was sitting in the seat next to me then threw up on me after his second cognac (and instead of apologizing, he just sheepishly avoided looking at me during the rest of the flight), and for good measure I found that Air France lost my luggage when I finally made it to Bordeaux as you can read in this post.

But let’s get back to the topic of my post today, namely a few of my impressions from the Business Practices Conference BP 16 in Prague which ended only a few days ago. Because I tried to attend as many sessions at the conference as possible, sometime I would even visit half of two consecutive presentations if I could not make up my mind which one might be more interesting. Most were excellent and a few where kind of mediocre, I thought, although all were in my opinion worth listening to.

I will try to briefly compare, or contrast, two sessions in my silly post today. So as not to get sued, I will not name names, although it is possible that commenters will mention the names of the speakers later in the comment sections. Unlike the Democratic and Republican parties which see the need to prevent third parties from being able to participate in the political process as their main duty, I do believe in democracy and I do not prevent commenters on my blog from participating (unless they really piss me off for some reason).

One session was given by a young and relatively inexperienced translator who for the purposes of my silly post today will be representing one of the wolverines fighting for translators’ minds. Incidentally, I do not mean anything negative by the animal names inspired by the old Cherokee tale that is a favorite of motivational speakers and that inspired the title of my post today. Since I was also one of the speakers at the conference, I too was a grey wolf fighting for translators’ minds in my presentation about patent translation at BP16.

I believe that we, wise and experienced wolves and wolverines, can do important work … provided that we actually know what we are talking about and have something to say.

The main topic of the session of the first wolverine, a translator turned motivational coach, was how important it is for translators to understand and be accommodating to change, the only constant in our lives, as Heraclitus of Ephesus so eloquently put it 2500 years ago. The changes that she was talking about were mostly technological changes so cherished by “the translation industry”, such as fast computers, algorithms and machine translation, cloud computing, crowd computing, optimized management systems, the emergence of mega agencies, the use of computer assisted tools (CATs), technical tools enabling competition with translators in developing countries and other scary things like that.

Well, scary to us, human translators, but so exciting for “the translation industry”!!!

We have to understand these changes and learn how to adopt them in our own line of work if we want to survive, the first wolverine was saying in her presentation. She was using slides with graphs that were demonstrating on colorful rising and falling curves the progress of machine translation that seemed to be pretty exponential, at least in her graphs. We have to anticipate that within about 10 years, there is likely to be a breakthrough in machine translation that will result in something that is almost as good as human translation, she said.

When I raised my hand to point out that “the translation industry” has been claiming that machine translation that will be “just as good” or “almost as good” as human translation would be here in about five years is something that “the translation industry” has been saying for the last 30 years, she just nodded her head but did not pursue the thought further. (I felt bad for disrupting her train of thought.)

The conclusion that most translators would probably reach from her presentation would be that we translators have basically two choices: either we stop foolishly resisting technological changes and bravely adopt “translation technology” or “language technology” as human translators using technological tools for the purposes of “the translation industry”, for instance by becoming human post-processors of machine translations, or some of us may choose to become highly valued specialized translators in certain coveted fields who can more or less afford to ignore “translation technology”, (because they are so special, by which I mean both the fields and the translators).

I could probably more or less agree with the last conclusion, although I would tend to disagree with most of the other things that this translator coach was saying. But if I had the transcript of the session in front of me, it would be in fact my pleasure to tear the entire presentation to pieces because most of the so-called facts and conclusions were in my opinion about as wrong as the anticipated technological breakthrough in “language technology” that will result in a machine translation that is almost indistinguishable from the way you or I translate.

But it would be a very long post, and possibly not a very interesting one. Fortunately, I will not be writing it.

The main topic of the second wolverine, an older and much more experienced translator, was about something else, although her session was about a related topic – namely about the elusive definition of what quality in translation means and whether there is a correlation between quality and price.

Full disclosure – I’m definitely biased here because although I only met the second wolverine in person at the conference a few days ago, I have been talking to the second wolverine online in various translation discussion groups for about 25 years now, and she sometimes also leaves a comment on my silly blog, usually when she is moved to do so by yet another insanity that I post online.

So don’t expect me to be impartial here. If I were a judge, I would have to recuse myself. But I am not a judge, so I can pretty much say whatever the hell I want on my own blog.

As I was saying, the topic of the second wolverine was about whether there is a correlation between price and quality. She had the same text, about one page, translated by about five translation agencies. At least one of these agencies used a machine translation that was then post-processed by humans (my hearts goes out to them, as does yours, I hope). The other suppliers included a cheap agency that uses cheap humans and a more expensive agency. Unfortunately, as I came late to her presentation, I did not have the color-coded sheets of translations corresponding to the more or less expensive LSPs, which I am told stands for “Lame Service Provider.” But the two young and very pretty translators sitting behind me, one was from Russia and one from Italy, helpfully shared them with me (thank you so much again if you are reading this).

Although I had less time than most people because my translation sheets were shared, the experiment has shown even to me that it was pretty obvious that there is a correlation between price and quality because machine pseudo-translation post-processed by a pitiful, cheap human, which was the cheapest solution, was so gratingly unnatural that I would hesitate to even call it a translation.

It did make sense, but after the spirit of the original text, which was a clever piece originally written by an educated French writer, was murdered by the machine translation, there was not that much that a human post-processor could have done with it. And anyway, given how little these human almost-but-not-quite translators are paid, why should they bother to do more than just remove the most glaring mistakes even if they had more time to do something with the cadaver on their hands, which they don’t?

The additional translations provided by other translation agencies sounded more natural, and some even sounded good or pretty good to some people. But the interesting thing was that when we got to vote about which translations were the best and which we thought were the most expensive ones, there was no agreement between the translators and the other translators who voted differently.

Beauty is in the eye of the beholder, and the beauty or ugliness of a translation is in the mind of a translation critic.

We cannot measure objectively translation quality, although some translation agencies like to pretend that we can and that we do exactly that by using neutral and reliable quality metrics and standards that are internationally approved and agreed upon, called for example ISO (insert a number, updated for greater impact every few years), or EN15038 (this number is also updated for greater impact every few years).

Although the quality standards that some LSPs are advertising on their websites as an assurance of measurable quality are nothing but a big lie as they have nothing to do with the quality of the translators or translations and only relate to the way in which the papers are shuffled around a desk by various mendacious Lame Service Providers, they are very useful for advertising purposes, which is the only thing that matters in “the translation industry”.

But just like most of us can tell the difference between art and pornography, most of us can also tell the difference between a good translation and a really horrible one, which will always be the inevitable result of a machine translation, even a machine translation that has been subsequently post-processed by an underpaid human, as the spirit of the original text will always be killed by a non-thinking machine.

That, among other things, was what I took away from the presentation of the second wolverine.

The way I see it, the problem with the first wolverine was that she was looking at things only from the perspective of “the translation industry”, and also that she did not realize that the translation industry is not the translation occupation. She was sincerely trying to give good advice to translators, but her advice failed to take into account the fact that there is a big difference between these two concepts and that the “translation industry” is not the world. In fact, “the translation industry” represents only one segment of the translation market.

“The translation industry” needs us to work for them because otherwise it can’t make any money. But do we really want to work for “the translation industry” under the current conditions? And if not, what are the other options that translators have?

That question that was not examined in the first wolverine’s presentation at all. And yet, I believe that this is the most important issue facing translators at this point in time.

Let’s hope that she will eventually be able to examine the issue also from the viewpoint of translators, instead of just painting a picture of immutable reality to which we translators must surrender if we want to survive.

She probably will be able to do that at some point, I think she is very talented.

The second wolverine made translators look at the problem that “the translation industry” has with translation quality, which made her presentation so fascinating to me. She was thus also making us examine the sorry state in “the translation industry” while implicitly telling us that they way forward for the translation profession is to concentrate on quality, which is something “the translation industry” is often unable to even ascertain, let alone evaluate, as it is an industry of brokers who don’t really understand much about translation.

To me, the voice of the second wolverine at the BP16 Conference in Prague was much more interesting, much more important  and therefore worth to be listened to and fed in my heart.

 


Responses

  1. The “first wolverine”, who may well be paid by or operate a translation company that relies on CAT tools exclusively and thinks they are the answer to everything, makes her appearance at every translation conference in one form or another. A talk of this kind was delivered some years ago by Gavin Wheeldon, the monolingual owner of the disastrous agency that was awarded all the interpreting services in the UK because they were backed by Capita, a huge organisation that contributes massively to the Conservative Party. Wheeldon amply demonstrated his total misunderstanding of interpreting when he connived with the UK government to reduce court interpreters’ fees to below minimum wage and made a massive loss instead of the much vaunted alleged saving of £18 million (a figure plucked from the air by the Justice Minister). I also recently attended a translation conference (in Warsaw) and the only talk I found useful was one that told translators about posture and how to sit for long periods at a desk without getting back ache, sciatica, etc. The rest were of the “look at me aren’t I clever” variety.

    Liked by 1 person

    • I wonder what you would have said to my presentation about patent translation at BP16 in Prague, Josephine.

      But I have to say, it never occurred to me that the first wolverine might be actually paid by “the translation industry” to more or less propagandize on behalf of it.

      Liked by 1 person

  2. I have no problem with the first wolverine. I have met her by e-mail in several incarnations. However, the essence of my reply to her has always been the same. “Thanks, but no thanks”.

    Liked by 1 person

  3. Actually, one of the few useful presentations I have heard was from a translation company that specialised in patents. They gave me some useful phrases such as “prior art” that I have found very useful when translating patents. The other thing I wanted to say is that I find it amusing that most of the translators and interpreters who think they know it all and talk down or write books to their fellow translators and interpreters mostly work in Spanish to English or English to Spanish. This means they are totally ignorant of the kind of problems encountered by those of us who work from or into languages that do not use Latin characters and those that have a very different “world view” such as Chinese that has no plurals or Russian (and other Slav languages) with no definite or indefinite article. Then there are the problems of the diacritical signs. I had to translate an article about a Czech priest and had no end of problems with the hacek. Yes, I know a little Czech, my name may be Bacon, but it is really Bakonova, my ex-husband is from Moravska Ostrava.

    Liked by 1 person

    • “The other thing I wanted to say is that I find it amusing that most of the translators and interpreters who think they know it all and talk down or write books to their fellow translators and interpreters mostly work in Spanish to English or English to Spanish. This means they are totally ignorant of the kind of problems encountered by those of us who work from or into languages that do not use Latin characters and those that have a very different “world view” such as Chinese that has no plurals or Russian (and other Slav languages) with no definite or indefinite article.”

      Ha, ha, ha.

      You just said something that I never dared to say aloud myself for fear of appearing arrogant. But it is absolutely true that people who translate languages like Spanish or French to English have no idea of the kind of problems that those of us who translate languages like Japanese have to deal with.

      Liked by 3 people

  4. There is a big difference between the concept of a change and the advice given. It seems the advice is focusing more on driving and mediating very specific types of changes rather than exploring the concept of a change and how to deal with it.

    At its core dealing with changes has a lot to do with independence and confidence (i.e. taking control over one’s faith instead of passively going through the motions), while the advice given to translators these days about adapting to changes seems to be about taking away independence (mostly of thought), increasing dependency (and on brokers no less) and generally surrendering and submitting to others as if translators are inherently powerless and ignorant about the work they are doing.

    The translation prophets usually either have a stake in the game, or are information mediators who take their information and cues (a process they often refer to as “learning the subject”) directly from those with a clear, and often cynical, commercial interest in driving these types of changes, and presenting this as their own insights and analysis.

    Liked by 1 person

  5. “There is a big difference between the concept of a change and the advice given. It seems the advice is focusing more on driving and mediating very specific types of changes rather than exploring the concept of a change and how to deal with it.”

    And the question we need to ask ourselves is the eternal one:”Qui bono”?

    Liked by 2 people

  6. In a moment of whimsy, I tried to imagine how the ‘medical profession’ could benefit from the LSP model and be improved to deal with the escalating costs associated with basic medical services (medical profession: you know, doctors, who look at you, take your blood pressure to demonstrate that they know what they are doing, then take a wild guess at what your problem might be and consult the pharmaceutical company’s database for a solution – “if it does not clear up in a couple of days, come back and we’ll try another drug”). Wow, who needs them! 🙂

    I imagine the patient, using a mobile phone from his or her sickbed, calls a ‘Medial Services Provider or MSP’ and speaks to a ‘Case Manager’ who writes down the symptoms, summarises them as best as he/she knows how (not being a medical practitioner) and runs them through a medical database.

    The database provides the best possible diagnosis based on a secret (commercial and confidential), but, trust me, a very complicated but reliable algorithm. The case manager then sends the results to the client patient’s phone in the form of an SMS, together with a script notice and instructions for taking the medication. The script is filled by the MSP’s parent company (a very large pharmaceutical company) to avoid double handling by a pharmacist with unreliable, human skills and knowledge, and sent to the client patient by courier.

    An invoice ‘for services rendered’ is automatically originated and sent to the patient’s insurer (also owned by the aforementioned pharmaceutical company) or Medicaid/Medicare (now also owned by the aforementioned, pharmaceutical company after both were privatised to make them more cost efficient).

    If the database is unable to provide the answer with a high enough confidence interval, set by the MSP as part of their QA protocols, the case manager sends the summary he/she made of the symptoms to a short list of selected ‘specialist’ and accredited (by the MSP)’ doctors/medical enthusiasts for an answer. He/she then feeds all of the answers into a QA database and allows the computer’s algorithm to pick the most suitable diagnosis, eventually paying the originating (winning) doctor/medical enthusiast of same a bonus of (say) 25 dollars over and above the 10 dollar fee paid for responding. The computer will also grade the originators on the basis of the ranking of their answers, which will be reflected in their future pay rates; ‘pour encourager les autres.
    The case manager then sends the selected diagnosis to the client patient (together with script advice and instructions for taking the meds).

    Apart from not having to leave home to attend a medical practice (where, as we know, you can catch all sorts of terrible diseases), the great benefit of this system, in addition to its efficiency derived from the elimination various unreliable and expensive human/professional intermediaries, is that the database keeps on expanding automatically, becoming spectacularly reliable and accurate. This will eventually remove the need for doctors/medical enthusiasts altogether, making the system incredibly efficient and thus reducing costs (provided the MSP and its owner pharmaceutical company/insurer do not to take advantage and charge bigger fees/premiums/drug prices. This will not happen, of course, in a free market in which prices are determined by ‘unrestricted and fair’ competition between privately owned businesses. – trust me, I have an MBA :-).

    Liked by 1 person

    • Your description actually reminds me of the business model of private insurance companies in United States. But unlike translation agencies, they do not pretend that they provide the services, they admit openly that they are just brokers, and also they share the profit quite generously with doctors.

      Their model created a very expensive medical insurance system, much more expensive than in any other country, so incredibly expensive that tens of millions of people still have no access to healthcare, even those who pay high monthly insurance premiums because copayments and deductibles are in the stratosphere.

      Obama’s so called “Affordable Patient Protection Act” is a big lie. The only word that it true in its name is the word “Act”. But it works very well for the insurance companies that wrote it, and that’s all that matters.

      Liked by 1 person

    • Unfortunately, you’re describing a situation that appears to be occurring on the ground. A good friend of mine is a doctor, and the kind of complaints she has are rather similar to the ones I do.

      My favorite of her complaints about how the medical profession is progressing is that a software company has set up a questionnaire for her and other doctors to follow – which takes the entire allotted time of the appointment to fill out. “But what if I want to ask something that’s not on it? What if the patient has a complaint that can’t be given a rank between 1-10?” she’s asked, plaintively. Between that piece of software and the rule that patients can only bring ONE complaint to each appointment, and must make another appointment for any second issue, she ended up leaving the hospital practice she had been involved with and opening her own clinic – only to encounter the newly-instated system of billing codes “ICD 10” and the unbelievable demands of U.S. health insurers.

      Which implies that the medical profession is also running into this scourge of rational-sounding profiteers.

      I believe that neither medicine nor translation (nor, in fact, most other fields of human endeavor) can be done well in this semi-automated manner. I wonder how cumulatively expensive it will be to persuade the world/economy of this.

      Like

      • The service is likely to be inferior in such a medical system, and it usually is, but that does not matter. The system is set up to provide superior profits for the middleman and for the doctors, not superior service. And it absolutely does provides superior profits by providing shoddy services and preventing access to medical care to tens of millions of people.

        Liked by 1 person

  7. I think that the major problem with “post editing” is that agencies expect to pay less for it. However, every single translation job, except for a few pesky uneditable PDF files or rare language pairs, is potentially a post editing job because any translator can create their own machine translated text for pennies. Therefore, the value of the machine translation provided by the agency = zero. There is no difference between a standard translation assignment for a standard rate and a “post editing” project for a post editing rate. The agency is not adding any value by providing a machine translated document. The only reason for a translator to accept a lower rate for doing post editing work is the tacit agreement for the agency (not the client) to accept a lower level of quality – at least until the customer complains – then it’s all the translator’s fault. It basically boils down to a trick to get the translator to provide a decent translation for 1/3 of the cost. And what’s worse, no one at the ATA even thought about this.

    Liked by 3 people

  8. I think that post-editing of machine translation by humans could be best described as an attempt at wage theft, which can work only if gullible, unsophisticated customers think that they are getting a good deal in this manner. Since machine translation is for the most part free, and the post-editing can only catch a few mistakes but by no means all of them, both the customers and the translators are being exploited in this scheme.

    Liked by 1 person

  9. An “attempt at wage theft”? How? A translator accepts a specific job for a specific word rate.

    By the way, are we still 200 years away from a post MT editing world? Maybe only 50 years away? 5 years?

    Like

  10. If you don’t get it, you don’t get, there’s nothing I can do about it.

    Liked by 2 people

  11. Interesting post. May the better wolverine win!

    Like


Leave a comment

Categories