May 25, 2007


Carnegie Mellon University has created a new version of the ever-present CAPTCHA that many of us use to sign up for accounts all over the web. We all hate it, but most of us understand that spammers would be worse than they are without it. reCAPTCHA is a program that requires the user to enter two words. One word is a known word from its database, and the other word is a scanned image from a book. Massive numbers of books are being digitized every day. Saved for posterity, and all that.

Computers can’t always read scanned pages very well. Often they make a valiant attempt, but they’ll read the word wrong. This reCAPTCHA method allows computers to get new words. If the user enters the correct first word, great. If they enter the second one, whatever book that word is from is one step closer to digital existence, and they can pass the word around a bit to make sure it’s right. That’s awesome.