Seriously though, von Ahn is a creative genius. His work focuses on harnessing humans to do the work that machines are bad at. Recaptcha is just one of his projects; he's also developed collaborative games to help tag images (if you've ever used Google image search, you know that there's big room for improvement in this area).
I expect to see more big things from him in the future. CMU FTW.
One thing I'm confused about. If the computer's have failed to recognize the letters, then how can the scanned text work as a CAPTCHA? Either the website already knows what the letters are, in which case the user isnt adding any value. Or it doesn't, in which case it has no way of verifying whether the user is a bot or not. Am I missing something?
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 typically a two part CAPTCHA. One is a word that the machine already knows, and the other is a word that the machine is unsure about. You only need to get the known word completely correct to authenticate yourself.
I tried to sign up for a Twitter account yesterday, and couldn't get past the captcha gate despite about 15 attempts. I wonder if this has something to do with it.
I use CAPTCHAs almost daily, as the PTO's website requires them. Just for fun, I guessed which of the two words was the "real" one, and deliberately mistyped the second one. It let me in. So now some text is being misinterpreted because of me...
Welcome to 2008, Ms. McArdle and Mr. Hutchinson!
Seriously though, von Ahn is a creative genius. His work focuses on harnessing humans to do the work that machines are bad at. Recaptcha is just one of his projects; he's also developed collaborative games to help tag images (if you've ever used Google image search, you know that there's big room for improvement in this area).
I expect to see more big things from him in the future. CMU FTW.
One thing I'm confused about. If the computer's have failed to recognize the letters, then how can the scanned text work as a CAPTCHA? Either the website already knows what the letters are, in which case the user isnt adding any value. Or it doesn't, in which case it has no way of verifying whether the user is a bot or not. Am I missing something?
Thanks google
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 typically a two part CAPTCHA. One is a word that the machine already knows, and the other is a word that the machine is unsure about. You only need to get the known word completely correct to authenticate yourself.
These sorts of ideas restore my faith in humanity's genius. Thanks for posting this link Megan.
I tried to sign up for a Twitter account yesterday, and couldn't get past the captcha gate despite about 15 attempts. I wonder if this has something to do with it.
No, that was probably God just doing you a favor...
I use CAPTCHAs almost daily, as the PTO's website requires them. Just for fun, I guessed which of the two words was the "real" one, and deliberately mistyped the second one. It let me in. So now some text is being misinterpreted because of me...