Stop Spam. Publish Books.

Can you really stop the global fight against spam AND help to publish books online? Yes, using a nifty website called reCAPTCHA…

reCAPTCHA is a neat little idea from School of Computer Science at Carnegie Mellon University in Pittsburgh. Everyday internet users encounter CAPTCHA boxes when filling in web forms. The box asks the user to decipher an odd-looking series of letters from an image and enter it as plain text, this prevents automated ‘bots’ from filling in a form and causing spam, as humans are much better at recognising letters than software.

As if to prove this point, the OCR software that ‘digitises’ scanned pages into plain text sometimes can’t read the pages of a scanned book. Typefaces can be feint or smudged, in short it needs a human to decipher it. So, what if you could take undeciphered words, serve them up in a CAPTCHA form, ask the web user to tell you the word and then serve it back to the digitisation project corrected? Well that’s precisely what reCAPTCHA does, currently helping the Internet Archive to digitise scanned pages of books. So upload reCAPTCHA to your website today, kill off spam and make books available online. Awesome.