We’ve all come across a CAPTCHA, a challenge response test that web sites give viewers who are trying to register for an account, leave a comment, or perform some other task that might be vulnerable to spammers or bots. They are useful because they can differentiate human from machine (Completely Automated Turing Test to Tell Computers and Humans Apart… don’t ask me how “turing” became a “P” in that acronym).
They look something like this: 
These things are a minor nuisance, the price we pay to protect the sites we need from bombardment by unwanted traffic or use as a launching pad for spam attacks. According to researchers at the School of Computer Science at Carnegie Mellon University, “about 60 million CAPTCHAs are solved by humans around the world every day. In each case, roughly ten seconds of human time are being spent. Individually, that’s not a lot of time, but in aggregate these little puzzles consume more than 150,000 hours of work each day.”
What if the time spent solving CAPTCHAs could be harnessed for productive purposes? Thanks to reCAPTCHA, it can.
Carnegie Mellon is currently working with two organizations (the Internet Archive and the New York Times) to employ humans to decipher scans of text that are unreadable by OCR software (Optical Character Recognition). If your site uses reCAPTCHA, your users can contribute to a major digitization project. For details on how the technology works, click here.
This is the latest innovative effort to maximize productivity in a focused way by taking advantage of the reach of the web to congeal a distributed knowledge network. reCAPTCHA has tapped into existing knowledge and processes to build yet more knowledge through another process. All of us together are smarter than we are added up.
Brilliant work.
(Nod to Mikhail for the heads up about this technology.)
Interesting indeed. But do you know how this will work for blind users using assisstive technology? What if someone is using Jaws or similar speech recognition software to read everything that is on their screen: will the computer read the Captcha?
Good question.
From the CAPTCHA website, section on “Guidelines” for use:
The key to CAPTCHAs is that they must be automatically generated and nearly impossible to crack with a program. This creates impediments for the visually impaired user… a potential work around to give that person access is for the site to include contact information to manually arrange access.
Genius. Thanks for sharing this example of innovative thinking.
According to Wikipedia, the “P” in the acronym comes from “Public”:
Completely Automated Public Turing test to tell Computers and Humans Apart.
Thanks, Tom. Both for clarifying the point and displaying the difference between a fine historian (who does thorough research) and a middling one (who’s writing this comment).
Here’s a really thoughtful post on making CAPTCHAs more accessible: http://www.standards-schmandards.com/2005/captcha/
Thanks, Boone!
Boone, this is an interesting proposal. As it stands CAPCHA is problematic for any users with disabilities…Thanks for the link.
I just loaded a reCaptcha on our (soon to be unveiled) new portal page for Blogs@Baruch. It does have indeed support for the visually impaired; you can choose to listen to an audio file rather than to decipher an image… the file features some ambient sound, and a voice speaking 8 numbers, which you then have to enter into a field to verify your, um, humanity. Or at least your humanness.
For more on reCAPTCHA, in the context of an interesting discussion on artificial intelligence, see http://www.nytimes.com/2009/05/24/business/24novelties.html?th&emc=th