Thursday, March 29, 2012

Quality is Dead, Except in My Backyard


"Could it be possible?! This old saint in the forest 
has not yet heard of it, that God is dead!"

A young mommy is reading a book to her little daughter. Children’s books are filled with illustrations and words in big, big fonts. They also use simple, short words. Typos stand out that much more. She comes to a page in which the book depicts a sheep and has the word “sheep” written in big tall letters. Except the word is misspelled: SHEIP. She goes online to a social media site to post a picture of the offending book and, implicitly, comment on how standards have fallen and “what is the world coming to?” etc.

Okay, I’m a forty-year-old man who has no children. At this age, you notice that the lack of children introduces a major wedge between your world view and that of other people who do have children. They have this wary, self-righteous attitude about the world, as if it is full of dangers and bad, bad stuff that is lurking behind every bush to turn their precious little babies’ brains into mush. Moreover, being a first-time mommy for some women becomes such an integral part of their identity that they have to insert the fact in the prologue of every sentence. (By the second or third child, I’ve also noticed, the babies start to begin to be shunted to a spot farther and farther back in their sentences.)

Anyway, the cri de coeur about the mangled sheep led to a discussion about the meaning of typos in general. I said something along the lines of not understanding the fetish about proper spelling and grammar. My view is that, beyond a certain point at which we have to follow its rules, it has a lot to do with a slightly pathetic petit-bourgeois flaunting of cultural capital (I did not actually say it this way, I only implied it tactfully). To me, it sounds like a cheap way of saying: “I went to college, but not to study anything useful like biology or building bridges, but to do angry, post-feminist deconstructions of The Scarlet Letter (conclusion: it is patriarchal) and To the Lighthouse (also patriarchal, albeit in a more indirect way). And after that, I got married and had a gaggle of babies.”

We live in such a relativistic world that there is no way to get a rise out of anyone by saying anything offensive anymore. You can make jokes about the Pope, the Dalai Lama, the Sai Baba, or any other religious leader with a funny hat, but not a peep ensues. You can satirize anyone’s political ideology, whether on the left or the right, and no one challenges you.

But beware of the passions you can unleash if you have the temerity--nay, the uncouthness--to express the opinion that proper grammar and spelling are perhaps not the ideal indicators of intelligence!

Snap! Oh, no he didn’t!

This mommy adopted a very “mommy” tone with me (although I am probably ten years older than she is). The tone subliminally meant: “I am a mommy! My job is to make the world safe for my daughter. Irresponsible, childless miscreants like you do not understand this sacred duty.” She suggested that following these rules can be compared to “hygiene” and that they are “the reflection of a beautiful mind.” To which I responded that beauty is more than following rules and that, as far as hygiene goes, there is a difference between washing your hands before supper and compulsively scrubbing your hands 78 times a day every time you think some random crazy thought.

Except I know this woman is also an l10n entrepreneur who is very much in tune with the faddish notions pouring out of Silicon Valley about how Low Quality is the future. Which created a certain cognitive dissonance for me. So I tried to gently steer the conversation in that direction: “I’m curious. What do you think of the theory that quality doesn’t matter and that the ‘TEP model’ in translation is obsolete?”

You could literally see the light bulb going on inside her head: “Ohhhh! Right!” (Facepalm!)

Immediately, her tone changed. In a flash, she went from mommy-ish to very corporate-y and jargony. It was like watching a metamorphosis, from Mommie Dearest ranting about wire hangers to the lady from the friendly folks at Omnitouch. She wrote: “I like it. It is a model conceived to make many expensive copies from an initial model. When that is the case, it is the best.”

Which means… uhhhh? (My brain hurts!) I must have missed that l10n seminar. I was probably dawdling with the croissants and free coffee, or reading a printed book, or doing some other unproductive idiocy.

The thing is that, by this point, I was already mentally writing this blog post. I was all in.

She continued: “However, if you give priority to agility (software, tweets) over other things, you make different conclusions. I’m not dogmatic.”

Boom! Gotcha. So I went in for the kill. I turned the conversation around 360 degrees back to the exact point where it began, with the bloody, mangled sheip-sheep: “But maybe the editor of your daughter’s book isn’t dogmatic either and just has different criteria about which texts require quality control.”

Check and mate.

Again: Boom! Which I’m sort of proud of, because, like George Costanza, the perfect witty riposte usually only comes to me hours after I need it.

I wish I could relate the rest of the conversation, but here it abruptly ended. My interlocutor declined to continue this fascinating philosophical exchange.

So her silence opens the door for me to complete the dialogue in my head. In there, the silence says: “Quality is dead, oh future post-editor, but if you dare show my child a misspelled sheep or a machine-translated children’s book, I will tear out your eyeballs like a lioness defending her cub!”

Which sounds to me a lot like some Tea Party nutcase saying: “Take your government hands off my Medicare! Not in my backyard! NIMBY! NIMBY!”

Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Monday, March 26, 2012

The Single Woman Hotness Phone App: Jaron Lanier and Super Sad True Love Story


I have a long review of Jaron Lanier’s You Are not a Gadget in the works. For those not in the know, he is a pioneer of virtual reality who now works as a consultant for Microsoft. The piece should provide a lot of background to some of my crazy rants. In the meantime, Lanier continually makes public appearances that always yield a good quote or two for the tech-savvy Luddite. This is from a talk last week at the SXSW festival, the same event, incidentally, in which homeless people were hired as wireless hotspots. The following quote provides a glimpse into how revolutionary the second wave of social media will be:

One of the problems is that if you say advertising is the only official business plan for open information, you’re inviting everything to turn into bullshit. You know, over time. It’s just a fact. What I’ve noticed with Silicon Valley start-ups is that there is this ideal time after they start when they have good information. At first, they’re just too small. They don’t have enough good information. And then they get flooded with bullshit and then their value goes down. Like, three or four years after something starts, then there’s good information. A year or more later, it’s all crap. That’s just the way it is. I’ve seen that in Google search results. Since I work with Microsoft, I shouldn’t… but in a way, it’s true.

They get gamed. It’s what happens. The thing is, uhm. Uhhh, God! It’s weird with network effects! The validity of the information doesn’t matter so much as lock-in. I’ll give you an example of this. I’m on this committee at UC Berkeley where we evaluate the business plans of graduate students in engineering. Because we want them to be entrepreneurs and show up with their start-up at "South By," right? A couple of years ago, we had these guys show up and they said: “We have this great idea for a start-up! And here’s how it works. We’re going to have a mobile app where a guy is in a bar. He sees all these attractive, unattached women and he’s going to enter all this information for his friends, so that they can come in and find these girls.” And I’m like: “Is there even the slightest chance, EVER, that this thing will have any good data on it? Will anyone ever do this?” And they looked at me and said: “No, of course not! This whole thing is based on hope. We’re just going to sell liquor ads. We don’t expect there to be good data ever. It’s all based on fantasy.” They got an A-plus.

So the thing is these business plans don’t depend on real information. And that’s a problem for society. 
One of the advantages of making information be real and grown-up and consequential, which means there’s money behind it and people are making a living and that someone is responsible for it, is that it will become less junky. 
The fascinating thing is that the mobile app where you sit and get information about the degree of hotness of the women sitting at the bar evokes Gary Shteyngart's Super Sad True Love Story. In the proximate future imagined by Shteyngart's comic novel, people's mobile phones immediately rank everyone else on the basis of physical attraction, credit rating, and money in savings account. So immediately when you walk into a bar, for example, your smartphone tells you that you are the fifth hottest male with the thirteenth best credit rating and fourteenth insofar as money in the bank is concerned.

Now notice that I am not saying that Lanier's example means "Oh my God! Shteyngart's dystopian future is coming true." Don't be stupid. You're better than that and that's why you read this blog. The reality is worse than the bad future where we are all reduced to vulgar hotness and credit ratings. In Shteyngart's future (circa 2020), the data on which these ratings are based is of a high quality. In our real near future (i.e., the one suggested by Lanier's example), you will have the information that at O'Neill's Pub in central Madrid there is a group of three unaccompanied female sevens and two eights, but this information will have been uploaded by the bar owner, a liquor company or who knows who the hell else. It will be junk information, like the junk we currently get on Facebook and Twitter.

At least in Shteyngart's nightmare, the data we use to discriminate against each other is accurate.

Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Thursday, March 22, 2012

ALS’s Gavin Wheeldon: A Case Study in Cheap Translation


Bialystock: Step 1: We find the worst play ever written.
Step 2: We hire the worst director in town.
Step 3: We raise two million dollars. ... One for me, one for you.
There's a lot of little old ladies out there!
Step 4: We hire the worst actors in
New York and open on Broadway and before you can say
Step 5: We close on Broadway, take our two million, and go to Rio.
The Producers (2005)

The tribulations of Applied Language Solutions (ALS) are everywhere in the blogosphere these days. It remains to be seen whether the company will hold on to its monopoly of court interpreting services in England and Wales despite its dismal performance to date (the catastrophe is discussed here at length). 

However, allow me to take a step back, because ALS CEO Gavin Wheeldon’s overexposure to the British media provides a cornucopia of material to conduct a brief case study of cheap translation in flagrante delicto, as it were. Watching Wheeldon’s TV performances on Dragons' Den and The Secret Millionaire, you literally see the yucky pig lips being inserted into the paste processed by the sausage machines. Wheeldon as a businessman (and a moral specimen) is truly riveting, albeit in sort of the same way that Hannibal Lecter is fascinating as a gourmet.

1.- There is no Revenue from Free Translation (!)

The following exchange is from his appearance in the BBC’s Dragons' Den. The format is as follows: five independent and separate private equity investors hear pitches from entrepreneurs and decide in front of the camera whether they will invest in the businesses. The ten-minute clip is a fascinating snapshot of the cheap translation sector in the early twenty-first century. Among many other issues, it is indicative of how machine translation currently has a grip on investors’ minds, as well as the way in which savvy shysters exploit it to the hilt for hype value (bubble, bubble, toil and trouble), despite the absence of any real substance.





Watch how the Wizard of Oz instantly turns into the shabby man behind the curtain (in three, two, one...):

Dragon 1 (Richard Farleigh): You do two things. One on the Web and the other one is actually live translation…
Wheeldon (interrupting): It’s human. The one on the Web, it’s machine translation. It does it instantly. It does it on the fly. The one on the Web, it’s about 70% accurate. It’s just a gimmick. It’s a good tool. People use it. It attracts visitors. But, obviously, that’s just a driver…
Farleigh: Okay. How is your revenue split between these two activities?
Wheeldon: There is no revenue from free translation. It’s all professional translation.
Farleigh: So all your 3.2 [million]…
Wheeldon: It’s all human translation. Professional translation.

Did you catch that? It is so pristine and simple that you almost expect it to come from the mouth of a Zen master. “There is no revenue from free translation.” What!? Listen to that, world: NO REVENUE FROM FREE TRANSLATION!

So there it is: Free translation does not provide revenue. The simplicity of this tautology is so beautiful, so absolutely beautiful, it brings the slightest little tear to the eye, like a twelve-year-old watching the closing scenes of E.T.

Can you imagine that? It’s hard to make money from free stuff! We should frame this phrase and hang it above the desk of every Cheap Translation CEO across the world.

I can imagine Henry Ford going: “You know, we’ve noticed that giving away Model-Ts tends to hurt our bottom line.” Or Warren Buffett writing to Berkshire Hathaway shareholders to say: “Well, it turns out that Project Free Hamburger was not the money machine we hoped.”

2.- There is No Profit from Cheap Translation

Wheeldon cites £3.2 million in revenue. When prompted for a profit figure, he vaguely estimates £400,000 for the year. He adds that his request of £250,000 for 4% of the shares is based on a P/E (price/earnings) ratio of 15. However, the second Dragon, Theo Paphitis, finds some clash between this figure and his own due diligence. He calls Wheeldon on it. The budding l10n entrepreneur replies that the 15 P/E for a £6.25 million valuation is based on “net profit.” Now, I am not an accountant, but net profit could (and usually does) mean almost anything. And, in this case, it means even less than that. After some prompting, it turns out that the slippery Mr. Wheeldon means “profit including [i.e., before] tax,” which, if anything, should actually be called “gross profit.” Listen to the investor schooling the weasely presenter like a stern schoolmaster:

Paphitis (Dragon 2): P/Es are calculated after the deduction of tax.
Wheeldon: Okay, well, I’ve learned something…

You have to hand it to Wheeldon. Caught red-handed in a transparent bit of obfuscation, he doesn’t even flinch. At most, he only seems slightly deflated. Paphitis, smelling weakness, presses on: the real P/E for a 4% equity stake costing £250,000 based on profits of £300,000 is not 15, but a whopping 21 to 22 times earnings (which in technical financial terms is “super-duper high” for a company outside the tech sector). Wheeldon accepts the analysis meekly, but offers an explanation for his creative accounting (viz. ignorance):

Paphitis: Do you expect me to feel a little bit uncomfortable?
Wheeldon (smiling): Once again, I’m learning here, Theo. In terms of a…
Paphitis (irritated): This not for learning! This is not a lesson!
Wheeldon (sheepishly): It certainly seems that way…
Paphitis: You come and ask me to invest £250,000?! And you ask me to teach you at the same time?!

The investor’s rebuke is harsh enough that it wipes the smile from Wheeldon’s face. The narrator sums it up: The valuation cited by Wheeldon is based on projected earnings (i.e., not actually in the bag yet) and it included taxes (i.e., money investors will never see).

Now, none of this is outrageous, but note that Wheeldon has been caught in the space of three minutes in several worrisome fabrications. You might argue that the professional investors pictured on Dragons' Den do this for a living, but is it at all possible that Mr. Wheeldon’s creativity with the truth played a role in securing the contract from the Ministry of Justice? I have my own opinion. I leave it to you to draw your own.

To lay this out as simply as possible, imagine that you have £250,000 in the bank. You can either place it in UK government gilts at 5% interest or, alternatively, you can invest it in the budding business of McTranslations Inc. in exchange for a 4% equity stake.

If you invest the quarter-million pounds in the British government bonds (the risk-free rate), within one year, you will have earned 12,500 pounds as interest.

If however, you take that chunk of change and sink it into the H.M.S. McLocalization Titanic Applied Language Solutions, at the end of the year you have a claim on 4% of the profits, which the chief executive officer “estimates” at £300,000.

Do the math. That is £12,000. That is 500 pounds less than the laziest, safest, most unimaginative thing you can do with your money, aside from leaving it to rot in a savings account at a negative real interest rate. The British Government has never defaulted on a loan since it started asking for money centuries ago. How does that compare to a company founded nine years ago in some dude’s bedroom in Manchester? Is it more or less dependable as an investment? Once again, I leave it to you to arrive at the answer.

The thing is that private equity guys such as the Dragons only become interested in businesses that make sexier returns—on the order of at least 10% a year or more.

Translation, sadly, does not fit that bill (at least the way it is done by shady entrepreneurs). Sure, cheap translation is enough for sleazy characters to make some moolah and fund a lavish lifestyle, but at the expense of generating a lot more revenue than you would need to if you provided quality at heftier margins. Wheeldon raises the long-term idea of floating on the stock market within five years for £60 million. Another Dragon shoots this down out of hand and cites again the company’s dismal balance sheet. Cheap low-quality translation has not passed the smell test. “As a risk/reward ratio for me, it doesn’t stack up,” says one.

Nonetheless, Wheeldon does get a tentative offer from Dragon Duncan Bannatyne, of half the money (£125,000) for more than double the equity stake (9%) he sought. Which is one way of saying: “You, sir, are a fine purveyor of bollocks, but reality is far less rosy.”

In conclusion, the Dragons liked him as a salesman, but they were turned off both by the sector and the valuation, so they turned him down. Cheap translation strikes out.

3.- A New Hope: Capita Steps In

Okay, so that is that. The thing is that, only three months ago, the Financial Times reported that a private equity fund called Capita stepped in to buy the whole of ALS for £7.5 million. So the Cheap Translation Theorist might say: “Aha! Six and a quarter million pounds for ALS’s paltry revenue was not so crazy after all! Gavin Wheeldon secured one and a quarter million pounds more than he asked the Dragons for!”

However, two things have to be taken into account. First of all, the Dragons' Den clip was filmed several years ago, so you have to discount the erosive impact of inflation. Secondly, the private equity fund jumped in only after ALS secured the mega-juicy Ministry of Justice contract that has raised all the Sturm und Drang. This MoJ contract is reportedly worth £300 million over several (undisclosed) years, or £42 million a year, depending upon the source. So the details of how that breaks down revenue- and profit-wise totally skew the assessment of Wheeldon’s pie-in-the-sky valuation. The Capita buyout was a bet that ALS would execute successfully on a recurrent contract with a good government client, pure and simple. Moreover, the MoJ would represent more than 90% of the company’s revenue for many years to come. It would be like shooting fish in a barrel. Of course, that was if and when ALS executed efficiently on the contract, which is currently in doubt.

4.- Epilogue: What Is the Deal With the Cheap Translation/Sleazy Salesmanship Correlation?

The alpha and omega of the entire ALS debacle is the man himself, Gavin Wheeldon. His personality comes across forcefully in the Dragons' Den clip: smarmily charming, slightly sleazy, quick on his feet, evasive, and not overly analytical. It is easy to imagine him courting civil servants and cabinet ministers desperate to make budget cuts with ridiculous, pie-in-the-sky promises about 60% efficiencies. Sadly, these exchanges were not filmed. However, a simple Google check might have given the ministry's staff some pause.

Last week I visited Wheeldon’s Wikipedia page. After reading it, I assumed it had been vandalized, given the nationwide firestorm in the UK. But after checking the footnotes and hyperlinks, I realized that, OMG, those were actually things he said to journalists! Those are actual quotes from his mother!

The first pearl is from a Times profile entitled “How I Made It”. Wheeldon tells the newspaper how he secured his first fat contract:
I was ringing up and pretending I was this huge translation company when really it was just me in the back bedroom with a phone and PC. I won the contract and then thought: oh my God, how on earth do I deliver this?
Is there any chance that this is the modus operandi used to secure the mega-million-pound contract from the British government? Who knows? For me, in the mouth of the chief of a tiny company that is awarded responsibility for providing thousands of interpreters to the entire legal system of a large country, this sounds a lot like someone saying: “Mr. Excrement, may I introduce you to Mr. Fan?”

Sound harsh? Listen to his mother’s description of him as a child during an interview for his appearance in The Secret Millionaire: “My nickname for Gavin was our small Arthur Daley, my dad always said if he didn’t end up behind bars he’d end up making a fortune!”

Who was Arthur Daley, you ask? Check Wikipedia. He was a character in a 1970s British TV show who is described as follows:

Arthur Daley, a socially ambitious but highly unscrupulous importer-exporter, wholesaler, used-car salesman, and anything else from which there was money to be made whether inside the law or not.

The entry goes on to note that “the name Arthur Daley has become synonymous with a dishonest salesman or small time crook.”

Jesus Christ…


Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Friday, March 16, 2012

Ray Kurzweil Was Right! Humans and Machines Are Merging!

This is the future predicted by our tech visionaries! The Singularity is just around the corner!

(Substitute "translator" for "homeless person" and "post-editor" for "wireless hotspot" and you have the dream dreamed by the Common Sense Advisory.)

Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International.

Thursday, March 15, 2012

CVs, Translation Databases, and Online Privacy


A very hard-working and experienced colleague discovered today that her very lengthy and distinguished CV had been copied lock, stock, and barrel by one of the little worms that populate the glorious online localization world. She took it pretty hard. She is very upset. I understand how she feels. It is like a violation of your privacy. Of your professional identity.

I had a similar experience a few years back. Out of the blue, I received this haughty message from some paper pusher in the European Central Bank telling me they were sorry to report that they would not consider me for something or other. I was a little peeved to be rejected for a position I didn’t even apply for. Which is a little weird and irrational, I know. Why should I give a flying flip that I was rejected for a position I didn’t even need or apply for?

The thing is, at those points, your lizard brain kicks in. After all, there is enough rejection in the world: social, sexual, professional, artistic… Who the heck needs unsolicited rejection? I mean really! Every single struggling actor says the same thing. “The hard part is the rejection.” Even successful actors get rejected more often than they are hired.

So I took it out on the snooty central bank drone (the attitude of some in-house financial translators, especially in northern Europe, is enough to raise your blood pressure…). I’m ashamed to say that I made insinuations about legal action just to get mine back. The thing is that I scared that poor, stupid woman into checking with the bank’s legal department to see whether there was a basis for action (of course, I had no intention of suing anybody). A few days later, a lawyer from the ECB got in touch with me to say that I had no basis for action because I had posted my CV on an online database. By then, the irritation I felt toward the original drone had subsided. But now the cause for embarrassment was double: the first was that I had scared the willies out of the original woman; the second one was that the lawyer was right. How can I complain if I uploaded my professional identity online? I basically gave CV thieves every single iota of information they needed to invade my privacy.

So, yes, there are all sorts of scumbags on the Internet. But what steps have you taken to protect not just your identity but the essentials of your personal biography? A few practical suggestions (believe me, I realize there is precious little of that in this blog):

1.- Consider password-protected pdfs for your online CVs.
2.- Do not publish personal information such as birth dates and addresses on your online profiles that are also used for verification of bank accounts (for people from Spanish-speaking countries, be very stingy with the use of your mother’s maiden name, which is used in other parts of the world for ID verification; a double-barreled last name is an easy way to impress people who don’t speak Spanish, but dangerous).
3.- Publish only a summarized CV. Maybe you can add a note that you will send a more detailed CV upon request.
4.- Give serious thought to Kevin Lossner’s frequent advice to shy away from ProZ.com. More successful translators have a modest online presence. There is a reason for that (and not just generational).
5.- Rethink in general your attitude toward an online presence. Become more sophisticated in the way you let it all hang out online.
6.- Do not swallow the current dogma about social media. It is neither as game-changing nor as indispensable for freelance professionals or businesses as the current wave of propaganda claims. Social media is basically cheap entertainment. Tweeting is not marketing, just as “Angry Birds” is not mental exercise.
7.- My personal little (infantile) rebellion against the algorithm: systematically lie on online databases and surveys that ask you for your information for free.
8.- Experiment with privacy and social media abstinence.

I have personally decided to heed Kevin’s frequent admonitions and not renew my ProZ membership this year when renewal time comes around. I will also erase my data from the site (although I suspect that ProZ’s terms of service are now like those of Facebook and they have a right to greedily hoard my personal information until the Sun collapses upon itself). ProZ.com is now for all intents and purposes mainly a price comparison tool that reduces thousands of people to a single per-word rate. And, secondly, it is also a comparison database for information thieves. As in so many online social websites, YOU are the product. On ProZ.com, you are fishing, gutting, icing, packaging, shipping, and exhibiting yourself on a fishmonger’s counter so that the mass of shady online middlemen and agencies can peruse you more easily. With the amazing downside that ProZ isn’t even free.

Once again, as occurs in almost every one of my blog posts, we have to return to Jaron Lanier and You Are not a Gadget. We have to abandon the childish, naïve notion that we can live our lives publicly and online in the same way that we live our non-digital lives. Access to a database of thousands of freelancers is a double-edged sword, as I wrote in Whenever I Hear Agency Owners Griping About Translators

We have to grow up and become more sophisticated in the way we use the Internet (and the way in which we let the Internet use us). We must rethink the way in which we take our personalities and upload them to a database for the easier consumption of sleazy advertisers and crappy agencies. We must think twice before we jump into ProZ.com and make it easier for Low Quality Translation companies to reduce us all to the lowest common denominator.

We have to grow up, period. The dogmas about online connectivity bringing forth a new world are nothing more than the post-modern incarnation of simplistic, medieval beliefs about the end of the world.


Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Wednesday, March 14, 2012

“Another Fine Mess”: The Applied Language Solutions Debacle




The whole Applied Language Solutions (ALS) catastrophe reminds me of a Simpsons episode in which Homer is put in charge of handling the response to a nuclear meltdown. Inevitably, everything goes south immediately. While scrambling desperately for a solution, Homer grabs the manual, suddenly realizes he doesn’t understand the first thing about nuclear physics, and slowly begins to grasp how desperate his situation is. He finally exclaims in exasperation: “Who would have thought a nuclear reactor could be so complicated?” You can sort of imagine ALS CEO Gavin Wheeldon crying in his office while he wonders where he went wrong.

In a nutshell, the British Ministry of Justice (MoJ) decided to assign responsibility for finding interpreters for all courts in England and Wales to this single company. Previously, every individual court sourced its own linguists via a national register of vetted interpreters.

The problem is that the company that won the contract, Applied Language Solutions, was totally unprepared for the changeover. The new deal imposed by the ministry included a sharp reduction in the per hour rate earned by interpreters, from £30 to a sliding rate of £22-£16. Furthermore, compensation for the time spent traveling and other costs incurred were severely slashed. All in all, an already difficult task which was woefully underpaid was made completely unattractive with one stroke of a bureaucratic pen.

The thing is that the system collapsed almost immediately. Many interpreters simply decided to boycott the new system in protest against the pay cut. Hearings were suspended because ALS failed to fill many slots. Lawyers and judges complained about missed deadlines and the level of qualification of the people supplied by ALS. The overall perception is one of generalized chaos up and down the green and pleasant land.

These are some of the lessons to be drawn from the affair:

A massive contract was dumped on a tiny, inexperienced company incapable of providing the service. It is quite evident that, in this case the need to make hasty cuts to the ministry's budget led to some very, very bad decisions. First of all, news reports indicate that Gavin Wheeldon’s ALS only made £7 million in revenue last year. The Ministry of Justice contract is worth £45 million per annum, i.e.,almost seven times as much. ALS, therefore, is sort of a tiny mosquito that suddenly swallowed two pints of blood and is about to pop like some grisly water balloon.

Computer- and network-driven efficiencies were promised which turned out to be non-existent. The idea was that, by not having to source their own interpreters, individual courts would save time and money. However, while it makes sense hat court staff would spend less time on this task, it is not evident that this represented a monetary cost for taxpayers. The ALS case thus seems to me to constitute a pretty obvious case of the failure of computerized centralization to generate the magic efficiencies promised by some databases. Of course, simple incompetence on the part of ALS and general lack of preparation on all parts could also be a factor here. But, in any case, the MoJ has no one to blame but itself. The idea that the amount of inefficiency in the old system was such that the simple introduction of a database and call center would allow ALS to slash interpreter rates from £30 to £16 is so disingenuous as to be either the product of sheer stupidity or the tacit hope that by outsourcing the service, the Ministry would be able to ride out the outrage from the interpreters. (Considering the reports that Tory MP Crispin Blunt, the undersecretary for prisons and youth justice, has stated that interpreters were grossly overpaid, stupidity seems to be jockeying for first place here.) 

A very complex ecosystem that worked fine was clumsily replaced with a hastily prepared and crude database that focused solely on costs at the expense of qualifications. The old system sounds like a well-functioning market that was simply obliterated by the ham-fisted decision to abolish it and move to a single, monopolistic operator (ironic for the party of free-market liberalism). Moreover, ALS, as a monopolist was entitled to draw rents, in the form of profits, from a system that previously worked fine without that monopoly. Seriously, how twisted is that? The deal is more reminiscent of Bourbon France in the 18th century than free-market Britain in the 21st.

Court interpreting is a difficult profession that probably was already underpaid. By seeking aggressive pay cuts of 30% or 40%, the MoJ and ALS probably pushed the per-hour compensation below what most qualified professionals were willing to accept for a demanding and sometimes depressing job. To use the language of free-market economics, the market cleared at £30. Anything below that point probably triggered not so much a strike or a walkout but a repricing by the market of other freelance activities. Suddenly, maybe taking on 10,000 or 12,000 more words a month from an agency did not sound so bad, considering you don’t have to pay for fuel and you can stay home on a rainy day. (Look up the term "opportunity costs," Crispin, you half-witted product of centuries of inbreeding.) Thirty quid might just convince you to drag your weary bones 100 kilometers to help out at a deposition. Twenty-two quid…? Maybe not so much.

The sleaze factor in the translation sector should not be overlooked. Relative newcomers like ALS that are only focused on revenue should immediately raise concern. Mr. Wheeldon is an attractive, sharp man who can probably make a convincing presentation, but due diligence seems to have been sadly missing. The company repeatedly bleats that it was created nine years ago in the founder’s bedroom. While Wheeldon’s entrepreneurial spirit should be commended, it is a little childish to be impressed by that sort of “origin story.” Exhibit A: There are reports that one Czech interpreter registered her pet rabbit in the company’s database. Jajo the Rabbit was apparently welcomed with open arms into the ALS database. (Jajo, by the by, has a Twitter account.) 

All in all, the ALS catastrophe is a perfect storm created by a spectacular collision between the urgent need among developed governments to cut costs willy nilly, on one hand, and the onrushing train of the “anything goes,” “quality bad, four legs good” translation philosophy that is currently in vogue, on the other. Mr. Wheeldon of ALS thinks, like so many cutting-edge l10n entrepreneurs, that translation is a commodity, so the interpreting slot at a murder inquiry can be filled by any warm body at half price. In this case, since a rabbit is warm blooded, he fit the bill quite nicely. Perhaps in a few years technology will have evolved so much that reptiles will also fit the bill. Who knows? The world is changing so quickly... 

Is this McLocalization's Waterloo? Yeah, right. You bet. My prediction is to expect more debacles like this in the future.

In the Simpsons episode referenced above, Homer finally averts the meltdown by playing “eeny, meeny, miny, moe.” Perhaps the time is approaching when Mr. Wheeldon will have to resort to this solution. The problem is that solving this mess is way more complicated than shutting down a nuclear reactor.


Further reading

To read more on the ALS issue, visit any of the following links:

Mr. Wheeldon puts on a brave face in the middle of the storm.


Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Tuesday, March 6, 2012

Press Release: Lionbridge Technologies Announces Innovative Freelance Rate Structure


As part of its mission to bring better service at every-day low prices, Boston-based Lionbridge Technologies (NASDAQ:LIOX) has unveiled an innovative pricing structure designed to align the remuneration of its freelance workforce with the company’s long-term interests. As of April 1, 2012, the company will now tie the rates it pays to its freelancers with the performance of its shares on the stock market. The announcement was made by Didier Hèlin, the company’s VP for vendor management, from an undisclosed location using capital letters cut out by hand from the headlines of Sunday’s Boston Globe which were then glued onto a blank piece of company stationery and mailed to several major news outlets.

For illustrative purposes, a freelancer who earned $0.12 per word of Spanish to English translation in 2007 will now earn $0.06 per word from now on, based on the hair-raising 50% plunge in the company’s market capitalization over the past five years. “We feel it is a very transparent way of linking up our dismal performance as a listed company with the ever-decreasing rates we pay freelancer vendors,” stated chief executive officer Rory Cowan after being tracked down using a court-ordered ankle monitor. “Why should our schmuck shareholders be the only victims of my incompetent leadership?”

The company’s stock—currently rated unanimously by Wall Street analysts as “a worm-ridden dog that should be taken out behind a shed and put out of its rabies-infested misery”—closed yesterday at a price deemed by one amused onlooker as “as close to a penny stock as you can come without actually costing less than one dollar.”

When asked if that meant freelancers would profit if the company’s stock defies the private projections of Wall Street analysts and actually rises, Lionbridge PR VP Mitzy Wallerstein laughed hysterically and blurted out: “Are you kidding?! If we ever pull out of this death spiral, OF COURSE we are going to change this policy. For a freelancer, working with us is tails you lose, heads you lose even worse.”


When reached for comment, self-proclaimed translation visionary and part-time trainspotter Renato Beninatto described the announcement as an “intriguing development” and the potential “first step in the most incredible corporate turnaround story since Ford Motors rolled out the Edsel, Coca-Cola unveiled New Coke, Napoleon Bonaparte returned from Elba, and Hitler invaded Russia.”



Disclaimer: None of this is actually true. This post was written purely for satirical purposes. LIOX CEO Rory Cowan does not wear an ankle monitor, as far as this blog is aware.


Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.

Friday, March 2, 2012

Cheap Translation Will not Save a Single Starving Somali Child

Have you come here to play Jesus
To the lepers in your head?
--U2, "One"

There is basically just one way of cutting costs in translation. That is because the central cost in any translation process is human labor.

Do I need to spell it out? The only way to cheapen translation is to slash the rates you pay translators with the cruel zest of Ming the Merciless. That’s it. That’s the only game in town. Go downtown. Force your freelancers into urban poverty.

So, basically, if you have ANY sort of business plan that involves making translation cheaper, you are not Henry Ford or Steve Jobs or Thomas Edison or the guy who invented Post-Its. You’re just another low-rent hack who wants to profit from the reserve army of the unemployed.

I will say it one more time: Cheap translation will not save a single Somali child from starving. If you want to save Somali children, go to bleeping Somalia with a 50-pound bag of rice. DO NOT fly your fat ass on a jumbo jet to a localization conference 1,500 miles away from home to give the same hacky speech you’ve given in Madrid, Los Angeles, Santiago de Chile, Tokyo, and London. 

Cheap translation is only an incredibly unimaginative way for some second-rate “entrepreneur” to turn a quick buck while providing bad service. And, as the ALS debacle proves, it is often a recipe for disaster rather than a vehicle for generating value. Dressing it up under the guise of some sort of dubious technological advance does not make it any better. Or you any less culpable.


Miguel Llorens is a freelance financial translator based in Madrid who works from Spanish into English. He is specialized in equity research, economics, accounting, and investment strategy. He has worked as a translator for Goldman Sachs, the US Government's Open Source Center, and H.B.O. International. To contact him, visit his website and write to the address listed there. You can also join his LinkedIn network by visiting the profile or follow him on Twitter.