Posted by: patenttranslator | December 7, 2011

Genuine Enthusiasm About Machine Translation in the ATA Chronicle

It is never appropriate to submit unreviewed, unrevised machine translation output to a client as a finished translation.

(This is an actual highlighted quote from an article called “Questions and Answers about Machine Translation” published in the November/December 2011 issue of the ATA Chronicle).

There was an interesting article in today’s Washington Post by Paul Farhi about the fact that despite a law against it, “stealth commercials frequently masquerade as TV news”. The newspaper used the example of Alison Rhodes who “is one of a small army of hosts and reviewers” of products offering to Fox viewers advice on what to buy. According to the article, Rhodes sees no issue in accepting payment for her recommendation because “her enthusiasm is genuine”.

Presumably, the more they pay her, the more genuine the enthusiasm. There are laws precisely against this kind of peddling and huckstering of products masquerading as news, but these laws are not enforced. In many respects we are already living in the post-legal reality of the corporate America at the beginning of the 21st century. Laws are enforced or ignored depending on who is breaking them.

I don’t watch Fox News, but I read, or try to read,  the ATA (American Translators Association) Chronicle magazine. One of the three featured articles in the November/December issue of the ATA Chronicle is an article written by Laurie Gerber titled “Common Questions about Machine Translation”.

Although Laurie Gerber states at the end of the lengthy article “Please know I have no ties (other than professional friendship with former colleagues) to any MT developer/seller listed below”, as far as I can tell, instead of providing a useful analysis of the strengths and weaknesses of machine translation (MT), the article was written to promote further acceptance of MT by translators and “LSPs” (language service providers, which is the newspeak for translation agencies), see my post here, because all resistance to MT on the part of translators would be as futile as resistance to Star Wars-like or Terminator-like machines that were sent by a highly sophisticated and extremely malevolent culture of space aliens or intelligent machines to colonize the Earth.

As far as I can tell, the article is a long advertisement for a product, namely MT. It has perhaps two thousand words, with a handy list of vendors of MT products at the end of the article.

For example, this is how Laurie Gerber deals with one of the objections that shortsighted translators such as myself might have against incorporating MT in human translations:

Q: “Isn’t it easier to do a good translation from scratch rather than edit a bad translation?”

A: “This, plus MT’s inability to handle language nuances and culture, are condemnations of MT that most translators find quite satisfying. I offer three ideas for your consideration below.

1. Post-editing MT can be unpleasant, which makes it seem like it takes much longer.”

[Mad Patent Translator’s Note: These dumb translators must be thinking that it takes longer because they are mentally unstable. They are not even smart enough to realize that they have to start using MT to make more money].

“At the Translation Automation Users Society (TUAS) conference in October 2010, Mirko Plitt, of Autodesk, described Autodesk’s experience. He explained that while translators often feel that post-editing MT takes longer than translating from scratch, timing their work showed that post-editing MT actually yielded a 30% – 40% productivity increase.”

Etc., and so on, and so forth …. If you don’t use MT, you are a Luddite and a loser. But, the author cautions ernestly, “It is never appropriate to submit unreviewed, unrevised machine translation output to a client as a finished translation.”

How very true. It is also never appropriate to kill your husband and children in their sleep, although that may sound like an easy solution too.

Anybody can design an MT test that will prove anything about machine translation, or the exact opposite, depending on how you design the test. I use MT quite frequently because it saves me time as I don’t have to use dictionaries or look for words online. But I happen to know that it is absolutely not just my imagination that it is much faster to translate from scratch than to “post-edit MT”. I would be post-editing MT myself if it did work the way Laurie Gerber and Autodesk say it works. Machine translations of patents are now available for free to anybody on a number of websites (the website of Japan Patent Office, European Patent Office, World Intellectual Property Office, and other websites).

I will now try to demonstrate what I am saying above on my own test of MT of a Japanese patent. True, some short and simple paragraphs translated by MT almost make perfect sense, for example this one: “If you use an alkali metal fluoride as a catalyst, and the manufacturing method, etc. Specific examples of alkali metal fluoride is supported as described in the first embodiment.”

You could edit something like that without wasting much time. It might even not take much longer than translating the whole thing from scratch.

But most longer paragraphs are completely butchered beyond recognition, see for example how specialized MT designed for patents handled “language nuances and culture” in this paragraph:

Before the mind complex acidification matter wa, 2 or more metal elements wo constituent element su ru composite acidification things thou ri, the metal element , patients , Hikaru , Jewellery Neko , factors, Hikaru ni , Tatari , ji ru ni , ni , Getting Tatari , grades posted , ni Hikaru , ga , lead, su , lead, and other ga Ju ru. Acidification of the former compound material wa mind, one kind with wo shi Certified, and use of more than two wo shi Certified .”

What the hell happened there ↑? I have no idea. And in all probability, neither do the MT programmers who created this freak of nature, I mean of silicon.

Under most circumstance, with most texts for which anybody would pay a translator any money to translate, post-editing of MT takes longer than translating from scratch, often much longer.

But according to what Laurie Gerber says in her article in the ATA Chronicle, translators simply need to get used to editing somewhat imperfect translations like these because  “MT post-editing is a cultivated skill. Embarking on any new activity where we lack “fitness” is frustrating, embarrassing and feels very uncomfortable.”

I think that machine translation could be in fact a very useful tool not only to translators like me who don’t have to open up dictionaries as much anymore when they translate if they are armed with a free MT product, since MT already is or soon will be also very useful to so called LSPs (translation agencies).

I can think of at least two reasons why some agencies, especially the larger ones who believe in the sanctity of the corporate model, could have warm and fuzzy feelings about MT.

1. MT will put the Fear of God into the hearts of translators. It is easier to negotiate rates down for human translation if you carry the big stick of MT. All you have to do next time if a translator dares to ask for a higher rate is to remind him of the big stick hanging over his head. Given that he is competing with MT, he should be glad he’s paid anything at all.

2. It should be possible to design a model in which instead of paying translators a rate based on the word count, you pay them a lower hourly rate because they are no longer translators. They are mere MT post-editors, and most people who have some familiarity with a foreign language should be able to edit MT product. If you can reclassify translators as MT post-editors, you should be able to pay them for instance 30 dollars an hour and then sell the post-edited MT product for more than twice as much.

MT is in fact a very useful tool in the hands of translators who can use it as an additional resource for human translators. But if the same translators are not smart enough to realize that the same tool is not really very useful to people who can’t translate, it is also a tool that can be used by smart business people to keep dumb translators in their place.

Advertisements

Responses

  1. Gives opting out a new luster, doesn’t it?

    P.S. This blog makes me even happier that I opted out of the ATA.

    Like

  2. I have been an ATA member since 1987.

    I don’t plan on quitting it now.

    I will probably even go to the ATA Conference in 2012, but mostly because it will be held in San Diego and my son just moved there 3 months ago.

    Like

  3. I might see you in San Diego, Steve. Long overdue for a family visit to the area, and the fishing down there is good. And if there are speakers propagating this MT nonsense at the conference, it’ll be fun to skewer them and have a barbecue in the sun.

    Like

  4. I am pretty sure I will go this time. I haven’t been to an ATA conference since 1998 and to San Diego since 1984.

    Hope to see you there!

    Like

  5. I was considering going to San Diego too, but given Kevin’s barbecue plans am having second thoughts.
    On a more serious note, payment is of course at the heart of the issue. Personally, I would have thought that paying by the hour is potentially a much fairer model than by the word. But the question is maybe less about the payment unit than about the ownership of the means of production (MoP, for the acronymophiles), if you forgive me that kind of ideological reference on your blog. It is of course problematic when the benefits from the productivity increase resulting from technological progress are not controlled by those who actually do the work. That’s very different from, say, architects who invest in CAD software and may then use part of the benefit from their increased productivity to offer more competitive prices, which is how the technological progress eventually also benefits the customer.
    No-one who builds a house would force an architect to use the latest AutoCAD version to negotiate their brick rate down (or whatever the word equivalent would be in construction). Instead, the architect will quote and at the end anyway invoice for at least 30% more — because for reasons beyond their control it took longer than planned. Same analogy for garages but also consultants etc.
    I wonder why in the translation industry we’ve ended up with a setup that makes it so cumbersome to achieve technological progress.

    Like

  6. I think that translators who are able to work mostly for direct clients are not forced to do anything other than deliver good work, because otherwise they will lose their clients.

    Translators who work mostly for agencies, on the other hand, are forced to do a lot of things that they may not like, such as to sign demeaning contracts in which among other things they have to agree to pay “reasonable attorney fees” should the agency decide to sue them, or agree to use certain CAT software such as Trados (almost every day somebody who typed “I hate Trados ends up on my blog), which is touted as a tool for increased productivity, but which will in fact ultimately reduce your income based on “fuzzy matches” if you work for agencies.

    The attitude of the ATA article that I used for my post, namely that all translators should use MT because “all resistance is futile”, is just the latest expression of how many, although not all, agencies view translators: as temporary hired help that is really only doing fairly simple data entry on a computer.

    The architect in your example presumably does not work for a middleman unless he is really dumb. That is why he can use AutoCAD software any way he wants, or not use it at all. Most translators probably do work for agencies. Work from agencies is low hanging fruit, you can obtain it relatively easily from an agency as long as your rates are low. So most translators go for the low hanging fruit.

    Translators own the means of production – namely their brains, but only if they find their own customers.

    When they work through a broker, the broker owns their brains and gets to dictate to them just about anything, including what software to use.

    The latest twist is the perceived threat represented by MT. There is a push on the way to to use MT in order to create a new type of business that would reclassify translators as MT post-editors. These post-editors would then be paid less, whether by the word or by the hour, because according to this logic, most of the work is done by the software and the post-editors simply put together the final file.

    I don’t think that this is a viable model, but many people would disagree with me, especially those who don’t know much about translation, i.e. business people who themselves are not translators.

    Like

  7. Hi, Steve and all,
    If I were to do post-MT editing, I would charge by the hour, not the word, and I wouldn’t charge a lower hourly rate than my standard. It would be up to me to figure out how much of a particular output (excretion?) I could edit in an hour; then I could provide a reasonably accurate quote for a job. But I don’t even like to edit post-translator output unless I know who the translator is or can review the translation before quoting. So I don’t figure this applies to me.

    It’s interesting to compare translation automation with monolingual English editing automation. In the editing field, all efficiencies of automation (e.g., macros and special software) accrue to the editor. Not to say that there aren’t editing agencies that operate like translation mills, but they don’t deduct for repetitions! It should be emphasized that in some styles of writing, and in some languages generally, frequent word and phrase repetitions would be unacceptable in a final product.

    The author of the ATA Chronicle article you discuss does say that MT is not viable for all types of translation and subject areas. I’m rarely asked to use CAT tools and even less frequently asked to base my translation on MT output.

    I think translators have to look for clients outside of the translation industry, or at least get an idea of what customers are willing to spend on their publications and documentation. The insular nature of our profession is one reason translators do not understand the value of translation and editing skills.

    Like

  8. “The author of the ATA Chronicle article you discuss does say that MT is not viable for all types of translation and subject areas.”

    That’s true. She says that briefly. But the article is not a serious analysis of the advantages and disadvantages of MT. It’s a puff piece designed to sell MT, which is usually garbage, especially from languages like Japanese. So I could not resist to add my two cents.

    I did not not know that you did monolingual editing too.

    I think that it would be fair to say that the rates have been frozen or lowered by customers and in particular agencies in the translation industry during the last 10 years or so. Do you see the same trend also in the monolingual editing field?

    Like

  9. I have only recently gotten serious about pursuing editing and proofreading work outside the translation arena (since this time last year), so I can’t answer about trends in editing rates from my own experience, but if I run across any discussions about rate trends, I’ll let you know. I can say that the rate I charge per hour for translation review has been acceptable to my oblivious-to-translation-trends editing clients.

    Here is a rate schedule (culled from surveys, not proscriptive) posted at the Editorial Freelancers Association site:
    I usually charge a standard rate and the price increase comes in taking longer to do more complex tasks (pages per hour decrease as complexity increases).

    I completely agree with you that translation rates are stagnant. The pressure is definitely downward. I’ve about doubled my rates since 2002 (when I set up shop in the US), but I still get inquiries at my 2002 rates, and am surprised when my current rate list doesn’t elicit some kind of comment from an agency — or an outright, “your prices are out of our range.” I occasionally offer lump-sum quotes when I calculate that I can earn my hourly rate on a job and avoid word-counts altogether. Sometimes that’s attractive even to low-paying clients. They slice and dice the numbers however they want and I earn a decent rate. (Also, I agree with your assessment of the Chronicle piece, and your two cents is worth way more.)

    There is something not quite right about the manner of discussion of rates within ATA. I don’t mean the fact that we can’t recommend rates (that’s probably a good thing in an organization composed of buyers and sellers both), but I feel like there’s a kind of beaten down atmosphere. (The few outspoken high-charging independent translators have aquired a kind of rock-star status, which in my opinion just reinforces the wretchedness* of the rest.) Maybe it’s inevitable in an organization without any barriers to entry — but I wouldn’t change that part. If anything, perhaps it’s too insular — we’re all looking inward and selling to and buying from each other, whereas we should all be facing out, looking outward.

    Well, you definitely struck a chord with me.

    *The word I want is perfectly captured by “bijeda, bijednost” in Croatian, you probably have something similar in Czech.

    Liked by 1 person

  10. “Bida” means something like destitution or wretchedness in Czech. It could be the same word, but it could also mean something else, you never know which “same words” in similar languages are false friends until you get to know your friends real well.

    “If anything, perhaps it’s too insular — we’re all looking inward and selling to and buying from each other, whereas we should all be facing out, looking outward.”

    I would go one step further – there are so many different market segments in the translation field. Most people concentrate only on one or a few segments of customers, for instance agencies, even though they may pay really low rates because they are what I call low hanging fruit.

    Although translators are not the only people who feel downward pressure on rates. Look for instance at the construction industry. Even here in Virginia, cheap Mexican labor has put experienced carpenters out of work, they are only given work that requires fluency in English, or so I was told.

    I think that looking forward, as you put it, should also mean identifying different segments of customers in the immense translation market that you are compatible with you and that may be paying higher rates. This is not easy but it can be done. I will try to write a post or two about it at some point.

    But instead of looking forward and trying to figure out how to beat the system that is stacked against them, so many translators are deathly afraid of MT. Have been for the last 10 or 20 years, I think.

    Like

  11. “But instead of looking forward and trying to figure out how to beat the system that is stacked against them, so many translators are deathly afraid of MT.”
    Exactly. The current technological changes, including MT, actually represent a huge opportunity for translators to change things to their advantage, but they have to actively seize it and not leave it again to some middlemen. From a technological standpoint, mere translation brokerage is not needed anymore today–just look at how crowdsourcing functions–, so there is much more room for direct client-translator relationships, which is in the interest of both the client and the translator, for many reasons, not only financial.
    The main challenge is probably what has been referred to here as the insular nature of the profession. So logically, it would be up to translator associations to step up and propose a fair business model, i.e. not just payment recommendations and best practice advice and client outreach kits but actually create a platform that hooks up translators and clients, using the full possibilities that today’s technology offers.

    Like

  12. I don’t think that translators’ associations will be very useful when it comes to finding new direct clients.

    I have been listed in the ATA database for more than 2 decades, but the only direct clients who ever find me there are individuals who need a birth certificate or the like. OTOH, I do get quite a few requests from agencies from this listing. So it is still a useful listing.

    My main customers are patent law firms, but they don’t even seem to be aware that this association exists, at least none of them ever asked me whether I am a member of this or any other association. All they care about is how much I charge and what is it that I can do for them.

    Translators need to do the hard work involved by themselves, for example by creating a well functioning website that will be found by search engines, networking (which I describe as “schmoozing” in one of my posts), etc.

    Translators’ associations are useful for a number of things, but they mostly promote dependence of translators on so called “LSPs”, which is the newspeak for agencies.

    Like

  13. Why not take the control back into your own hands and build your own MT engines, you have the perfect data to build an engine suited to your own needs. Have a look at http://www.smartmate.co which is designed to be used by translators to improve efficiency and contro, the MoP

    Like

  14. But there is free MT software already customized for patent translation available from Google Translate, Japan Patent Office, etc.

    I am not spending money for something that is available for free.

    Like

  15. […] Will Help Us All The Different Levels of Politeness in Different Cultures and Languages Genuine Enthusiasm About Machine Translation in the ATA Chronicle Essential Office Equipment: External Hard Drives & Regular Backups Lingo Woman’s […]

    Like

  16. […] some MT enthusiasts are promoting now for early adoption by translators as I wrote for example in this post about an article in the ATA (American Translators Association) Chronicle, translators either need to integrate machine translation in their work, or risk becoming modern […]

    Like

  17. […] problem is, many people who know nothing about translation, (and some who are selling translation for a living), are often unable (or unwilling) to make a distinction between machine translation and real […]

    Like

  18. […] the way how these issues are presented in the ATA Chronicle (when they are mentioned there at all):“MT and CATs gooood, pointing out that MT is crude and doesn’t work or resisting demands…“, talking disparagingly about the wisdom of processing of the MT detritus baaaad – all […]

    Like

  19. […] spite of what various snake oil salesmen and saleswomen have been trying to make us believe for several decades now, machine translation (MT) does not make […]

    Like

  20. […] ATA (American Translators Association) Chronicle has been publishing articles celebrating the progress of post processing of machine translation for many years now. Not once has it published an article that would question the premise behind the […]

    Like

  21. […] which happens to be an opinion that most translators share. As I commented on my silly blog here, here and here, all the other articles published in the ATA Chronicle were written by cheerleaders […]

    Like

  22. […] because articles of this type are unfortunately so often found in the ATA Chronicle. For example in this post which is now already almost five years old, I compared the propagandistic nature of these articles in the ATA Chronicle to the predictable […]

    Like


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: