reCaptcha: Stopping spam while digitizing books

[app103]

We have all seen captcha text...everywhere.

Carnegie Mellon University has come up with one with a bit of a twist that they call reCaptcha.

While helping to stop spam and ensuring that a human is actually submitting a response, it is also helping to digitize books. (currently helping Internet Archive)

To archive human knowledge and to make information more accessible to the world, multiple projects are currently digitizing physical books that were written before the computer age. The book pages are being photographically scanned, and then, to make them searchable, transformed into text using "Optical Character Recognition" (OCR). The transformation into text is useful because scanning a book produces images, which are difficult to store on small devices, expensive to download, and cannot be searched. The problem is that OCR is not perfect.

reCAPTCHA improves the process of digitizing books by sending words that cannot be read by computers to the Web in the form of CAPTCHAs for humans to decipher. More specifically, each word that cannot be read correctly by OCR is placed on an image and used as a CAPTCHA. This is possible because most OCR programs alert you when a word cannot be read correctly.

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.

The service is free, and they even have one specifically for protecting your email address on a website.

http://recaptcha.net/

[mouser]

cool 

[Wordzilla]

I just gave it a try at: http://recaptcha.net/learnmore.html

In one case (among others) i was given the image that should read: relations ascrib, and when I intentionally entered: relati0n5 a5crib (by substituting o, s with 0, 5), it still passed.

That wasnt gonna give site owners anti-spam assurance was it? -- while the bot does not have exact idea how letters should read, neither does the verification!

Here's an idea: instead of giving out two captcha string challenges with indeterminate open sesames, it displays the first string with definite answer, followed by the second one with some uncertainty at the verification end -- whose purpose is to help to digitize books.

[RLee]

That was my question as well: if the users are helping digitize books (meaning the service doesn't know the correct answer), how does it stop spam?

[app103]

Here's an idea: instead of giving out two captcha string challenges with indeterminate open sesames, it displays the first string with definite answer, followed by the second one with some uncertainty at the verification end -- whose purpose is to help to digitize books.

-Wordzilla (July 27, 2007, 01:58 PM)

That is exactly what it is supposed to be doing.

In one case (among others) i was given the image that should read: relations ascrib, and when I intentionally entered: relati0n5 a5crib (by substituting o, s with 0, 5), it still passed.

-Wordzilla (July 27, 2007, 01:58 PM)

Could be that it thought that was a very 'human' error and let you slide. It could be a really '1337' captcha. 

What would happen if you typed something totally messed up instead? Would it still pass?

Try it using some non-obvious substitutions, like instead of 0 for o and 5 for s, try 3 for o and 7 for s.