Just wanted to add the comment, since I didn’t specify it explicitly, that I’m not trying to claim that the accessibility of this particularly CAPTCHA is all that fantastic — it’s pretty good, but there are serious problems. I’m just saying that it’s a neat idea.
In case you don’t already know, “CAPTCHA” is an abbreviation for “Completely Automated Turing Test To Tell Computers and Humans Apart.” From an accessibility perspective, they tend to have significant problems — and I’m not going to try and claim that this one is perfect. However, it is very thoughtfully done, and has a very interesting additional feature which I appreciated.
I ran across this via Stumbleupon. Unusually, rather than finding it because I was busily stumbling around, I actually became aware of it because I was trying to create a new account. The interesting CAPTCHA is called “reCAPTCHA.”
Specifically, the concept behind it (explained thoroughly on the reCAPTCHA site) is to gain value from user input in CAPTCHA texts.
Most spam protection systems are based on nonsense words, random strings of letters, or obscured text. Anything, fundamentally, which might be difficult for a computer to identify.
What the folks at reCAPTCHA observed was that scanning old books provides a wealth of resources in the realm of obscured text which can’t easily be understood by computers. To solve this problem, they pasted together the needs of a CAPTCHA and their scanning process to create a service which helps them identify these unknown texts.
Obviously, there’s an immediate problem: if the computer has already failed to identify the text, how do you test whether a human has read it correctly? Simply speaking, you don’t.
Instead, reCAPTCHA provides two words for the user: one they know, and one they don’t. The known word is the Turing test — the unknown word creates a source for the computer to identify the word they didn’t know.
About 60 million CAPTCHAs are solved by humans around the world every day. In each case, roughly ten seconds of human time are being spent. Individually, that’s not a lot of time, but in aggregate these little puzzles consume more than 150,000 hours of work each day. What if we could make positive use of this human effort? reCAPTCHA does exactly that by channeling the effort spent solving CAPTCHAs online into “reading” books.
But if a computer can’t read such a CAPTCHA, how does the system know the correct answer to the puzzle? Here’s how: Each new word that cannot be read correctly by OCR is given to a user in conjunction with another word for which the answer is already known. The user is then asked to read both words. If they solve the one for which the answer is known, the system assumes their answer is correct for the new one. The system then gives the new image to a number of other people to determine, with higher confidence, whether the original answer was correct.
It’s unlikely I’ll implement it, I’ll confess. The fact that it’s delivered via an iFrame and the simple nature of a CAPTCHA go against my generally preferences in web development. However, should I be in a situation where I need to implement one — this will certainly be a strong candidate! (And even stronger if they fix their accessibility issues.)