Crowdsourcing

rowdsourcing seems to be all the rage these days—as if machine translation were not enough to scare all of us sleepless.

Wikipedia defines "crowdsourcing" as "a neologistic compound of Crowd and Outsourcing for the act of taking tasks traditionally performed by an employee or contractor, and outsourcing them to a group of people or community, through an 'open call' to a large group of people (a crowd) asking for contributions."

This would be a great definition if it did not hide the basic concept underneath the word contribution. A contribution may be paid or unpaid, but crowdsourcing implies that contributions are unpaid.

It is important to bear in mind that crowdsourcing is different, very different from team work. Take, for example, the recent case where a Dan Brown blockbuster was translated into Swedish faster than you can say "smörgåsbord." This did not qualify as crowdsourcing, because a team of professional and paid translators and editors worked on it. This is teamwork, as so many of us have done so many times.

Work for free or next to nothing is what some people who have other sources of income do "for a cause" or because they are in it for their hearts, "not for the money." A new variation is shown here http://blog.cubeofm.com/how-i-reduced-translation-costs-of-200-articl where someone pays translators with some type of funny money.

We do not blame people who resort to crowdsourcing. Let's be honest, who in their senses would pay for something that could be had for nothing or nearly nothing—and legally, too? This is not the same as talking the natives into working for mirrors and booze: People who provide translation services free of charge have reached a certain degree of intellectual development and are supposed to know what they are doing. The question should be: Do people who resort to crowdsourcing know what they are doing?

Many people would say "yes." After all, it is all translation, isn't it? But we are not so sure. There are some reasons why crowdsourced translation cannot replace professional work, a few of which are summarized below.

Most people who participate in crowdsourced projects are ready to contribute small portions of work. This means that jobs have to be sliced thin, salami-like, and participants will take as many slices as they think they are ready to chew.

Of course rising a large enough crowd can be difficult, and volunteers will tend to concentrate in "interesting work"—like they usually do. In many countries, it is not difficult to get people who will translate "good literature" or poetry for nothing. A few experts will translate important texts into their own languages, because they believe they are doing a favor to culture. Dozens of Harry Potter enthusiasts will band together to beat the publishers to the bookshelves—but this is more of a teenager prank than crowdsourcing proper.

Now try to have those people translate a few hundred credit reports or patents over the weekend and see what happens.

Some texts are beyond crowdsourcing because they are confidential. By the way, how many texts have you translated that are NOT confidential? Do you think any client will crowdsource patent work? Alpha Chemical may not be willing to part with a few thousand dollars to pay a translator, but we bet it would be more than pleased to pay twice the same amount to keep a translator's mouth shut.

Most of the work we two have been doing is covered by NDAs and we could have made a quite a few dollars by keeping some parties posted concerning certain facts that came to our knowledge during the course of our work.

A rose, by any other name, would smell as sweet... or so does the bard say. But translators know it is not true. We are expected to be consistent in our translations: you simple cannot call something a bucket on one page and a pail in the next and woe is the translator who makes that mistake. It is hard enough to control terminology in a one-translator job and jobs handled by more than one translator require glossaries and editing by as few editors as possible, to make sure a rose does not appear by any other name but "rose."

Is this possible in crowdsourcing? Yes, of course it is. But can you imagine setting up a team of unpaid editors to make sure the translation is consistent? On a second thought, even a well paid editor would probably refuse to handle the mess.

Will all crowdsourced work be crappy? Not necessarily. Or, at least, not all of it. There are and have been excellent amateur translators and the crowd may have quite a few of them—especially if the cause is felt to be worthy or otherwise appealing. Considering the restrictions discussed above, on the average the translation may be quite decent. Or not—it is all a matter of luck: a lottery.

On the other hand, a lot of paid translation is incredibly bad—especially in the lower rungs of the fee ladder.

Some of the people who resort to crowdsourcing would never pay for a cent's worth of translating services, either because they don't have the money or because they think there is a better use for their cash. Before crowdsourcing came into fashion, those people either would have nothing translated, or resort to unedited machine translation or to the services of some neighbor of good will.

How about the people who used to pay translators? How many of them will switch to crowdsourcing? We believe many of them will give it a try. Why not? What is free is good. But will they make crowdsourcing their standard way of securing translations? We believe that will not be the case. One experience should be enough to change their minds, considering the limitations pointed out above.

Some translators advocate a campaign to "educate the client." We do not think such a campaign would work. Whatever arguments we can put forward will meet with the counterargument that it is free. And the "free" argument wins most discussions. In addition, since we will be defending our own income in the process, our credibility will be very low. Unfortunately, clients will have to learn from their errors.

On the other hand, crowdsourcing is a fad and it will fade away like all fads, whether we do something about it or not.

Truth be told, the worst type of prospective client is not the one who resorts to crowdsourcing, but the one that resorts to blackmail: People who say "we have got a mind to resort to crowdsourcing, to cut costs, but if you would charge a more reasonable fee..." or, worse, the agency that says "the client wants to use crowdsourcing, but ..." The correct answer to this is a very polite "crowdsource it and be damned", said with a sunny smile in your face.

No, sorry, not that—although it would be quite appropriate! But we advocate a strong reply of the "Oh, well, that is not my decision. But it is your company, you know, and if you want to run the risk, that is OK by me." Never submit to blackmail.

Translators are not the only ones to struggle against crowdsourcing. Can you imagine the nightmare photographers have been through? Professional cameras get cheaper and cheaper, and storing photos on the internet is easier than walking. Since most people don't pay their debts by selling photographs, any extra dollar is welcome by the ham.

To be honest, when we started writing this article, we had no idea of how widespread the problem was. But we found this http://www.wired.com/wired/archive/14.06/crowds.html and, all of a sudden, we felt we were not alone. Just like Mark Harmel, some of us will have to rethink their careers and change their paths. That may not be a bad thing, after all.

The point is, in addition to not reducing our already-too-low fees and telling cliets to crowdsource and be damned, what should we do?

We cannot say what you should do. But we can tell you what we have been doing in the face of this and other challenges. It is a two-pronged approach.

Translation is one of the few activities that can be performed well by an amateur. For instance, it is difficult to become a brain surgeon without going to medical school. But look at cabinet-making, for instance. There are excellent self-taught amateur cabinet-makers, and you will often see people who proudly show you a fine piece of woodwork they have crafted with their own hands. But a professional cabinet-maker, like a professional translator, is something different. And we make all efforts to stress the PROFESSIONAL in our activities.

Far too many translators believe that being able to translate well is enough. No, it is not. Many amateurs translate well. Only the professional can handle every job professionally. An amateur tries to makes things right; the professional ensures that things cannot go wrong.

We try to acquire a strong command of the tools of our trade, and we try to be knowledgeable with all types of translation problems so that we can prevent and solve them. We advise clients on translation strategies and often help them save money by showing a different and better way to do something. And we try to keep learning.

But that is far from sufficient, perhaps useless, without the other prong: the client must know we are pros. That does not mean that we bad-mouth our competitors. Expressing unfavorable opinions of colleagues is unprofessional. We simply and quietly show we have done a good job. A job far better than the client could get from a bunch of amateurs. Value for money.

There are plenty of good causes around and there is nothing wrong with donating your time and professional capabilities to one of them. Very much on the contrary.

However, be careful with NGOs where there is plenty of money for everything, except for paying translators. Before donating your work, check who else is working for the same organization on a pro bono basis. If other people are getting paid, why shouldn't you be?