If you've tried to leave a comment on this journal recently, you've seen that an anti-spam measure called captchas has been implemented there.

Captchas prevent computer-generated spam messages from being left in people's journals, albeit at the cost of inconveniencing real people a bit of their time, and in the case of blind people may prevent access altogether. I think cruel and unusual punishment should be brought back for losers who create the spam in the first place, and am glad that captchas were invented to block them.

So far captchas seem to be doing a pretty good job, but apparently some spammers are now beginning to hire people in third-world countries to manually insert spam where they have been implemented. Sigh.

In any case, the genious who invented captchas has come up with a way to put all the nonproductive time spent entering captchas to better use (as well as the time spent by the aforementioned third-world work-arounds). He calls this project reCAPTCHA, and I just *love* the idea.

Instead of just the plain old single captcha, now, two captchas appear. One is a control captcha, the other is a real image of word from a book that's been scanned for digitization, but one which the computer has not been able to figure out through OCR (optical character recognition) because the letters are smudged, or run together, or otherwise indistinguishable from one another.

This way humans can enter the word the computer can't determine (even if they're blind - reCAPTCHA has an audio file component), and thus help giant library digitization projects. It's only a few seconds for any given person, but thousands of hours of effort when aggregated, and it's for a good cause. Tens of millions of words will be corrected this way. That's just brilliant!

(This is not a hint that JournalScape should move to reCAPTCHA, but of course if it did that would be cool :-)