pageoftext and recaptcha

by Hugh

Two great little aps:

Page of Text
Nice simple little aps are such a pleasure. Page of Text is a brilliant little wiki environment – so easy and clean. I’m not quite sure what it’s good for, but if you want a page of text on the net, without having to host anything or do any fancy formatting (say, directions to your birthday party, or, maybe to make an editable list of players for your rugby match), page of text by Gordon McCreight is just what you need. It’s a perfect little gem of undetermined usefulness. (Speaking of useful, the wikiclock is run on a page of text platform).

ReCaptcha
ReCaptcha uses the human brainpower put into solving captchas to good use – and helps digitize public domain books.

recaptcha

If you run any forums or anything like that, you know what a pain spammers are, and in parallel, what an annoyance captchas are (captchas are usually those little weird numbers/letters you need to fill in to prove you are not a spammer). Captchas are either too good (keeping many legit people out) or too bad (letting all the spammers in). Usually both.

I can’t vouch for ReCaptcha’s effectiveness, but I can sing the praises of brilliance behind it. Firstly, there is an “audio” captcha included for those who can’t see so well (and the many frustrated eagle-eyed people who get foiled by really “good” captchas). Secondly, they estimate 60 million captchas are “solved” every day, at about 10 seconds per “transaction”, resulting in 150,000 hours of work each day, or 54 million person*hours of work a year! What if, instead of wasting that aggregated effort, captchas were used to do something useful? Say, helping to proofread badly-scanned, public domain digital books? That’s what reCaptcha appears to do, though I still can’t quite figure out how. Here’s what they say:

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.

In any case, it’s worth having a think about what other such otherwise wasted human processing time could be used to productive ends, and what sorts of projects might benefit from such an approach. I imagine that reCaptcha’s system could be used for other things, tho looking thru their literature, it appears they’re not GPL.