Von Ahn starts from the premise that the ten seconds taken to enter a ‘traditional’ word verification is ‘time wasted’. Well, of course, it isn’t: it’s time spent productively protecting oneself from spammers. He then goes from there to claim that reCaptcha harnesses this ‘wasted’ time to a productive end, i.e. to interpret words from scanned documents that can’t be deciphered by character recognition but can be by humans.
He doesn’t dwell on the fact that, by making us type two words rather than one, he has doubled (at least) the time consumed. And with no direct benefit to us, the users – the benefits, often commercial, accrue entirely to the organisations who are, effectively, using us as unpaid, unwitting slave labour.
And that’s without mentioning Google’s implementation, on Blogger, of their own product, which must rank high in ineptitude even by their standards. Because, as far as I’ve seen, a lot of the time it doesn’t even use real words! At least with the old wv, though the character strings weren’t ‘real’ words, they behaved as if they were. Now we get a random string of letters, so heavily disguised that you need a lot more than the presumed twenty seconds before getting them wrong.
And of course, that introduces the most farcical flaw. Many people have reported that they’ve switched off wv on Blogger, out of sympathy for other people’s frustration with the advanced Mensa test it’s become. So as a result, we don’t get the spam protection, and reCaptcha doesn’t get its translation. Just how lose-lose can you get?