Most of us who are on the net for years might remember the SETI@Home project where your computer’s idle time was used to analyze radio telescope data in the search for ET. The great thing behind this idea was that unused resources (your computer doing nothing) were leveraged into something beneficial. Even if you don’t believe in extra terrestial life, the spin off projects that used the same concept (Cancer research, furter refining the value of PI) all had it’s value.
Yesterday I ran into something similar. reCAPTCHA, developed by the School of Computer Science at Carnegie Mellon University, is a CAPTCHA system which actually is useful. According to the reCAPTCHA website, each day more than 60 million CAPTCHA’s are solved daily which corresponds to roughly 150.000 hours of work each day. That of course is an awful waste of resources being wasted just to prevent comment spam on websites so over at Carnegie Mellon they came up with the following idea.
Everyday, books are digitized by OCR programs to preserve them for the future. Unfortunately, OCR programs aren’t perfect and due to the quality of the source they make mistakes in interpreting the text. Having these mistakes fixed by a human takes a lot of time and money.
reCAPTCHA solves this problem (and the problem of preventing comment spam) by presenting a user with two scanned in words. For one of these scanned words, the answer is known. The user is asked to type in both words and the reCAPTCHA system assumes that if the known word is entered correctly, the answer to the unknown word is probably also correct. It will store this answer and present it to other people until it is confident enough the original answer is right.
The scanned books used by the reCAPTCHA system are from the Internet Archive which are mostly Creative Commons licensed books. Hopefully soon all these books can be enjoyed in digitized form by all of us.
If you are interested in integrating reCAPTCHA into your website, check out the reCAPTCHA resources page where you will find plugins for WordPress, phpBB, Drupal etc. Also available are programming libraries for languages like PHP, Python and Ruby and a reference to the web-based API.
Most of us who are on the net for years might remember the SETI@Home project where your computer’s idle time was used to analyze radio telescope data in the search for ET. The great thing behind this idea was that unused resources (your computer doing nothing) were leveraged into something beneficial. Even if you don’t believe in extra terrestial life, the spin off projects that used the same concept (Cancer research, furter refining the value of PI) all had it’s value.
Yesterday I ran into something similar. reCAPTCHA, developed by the School of Computer Science at Carnegie Mellon University, is a CAPTCHA system which actually is useful. According to the reCAPTCHA website, each day more than 60 million CAPTCHA’s are solved daily which corresponds to roughly 150.000 hours of work each day. That of course is an awful waste of resources being wasted just to prevent comment spam on websites so over at Carnegie Mellon they came up with the following idea.
Everyday, books are digitized by OCR programs to preserve them for the future. Unfortunately, OCR programs aren’t perfect and due to the quality of the source they make mistakes in interpreting the text. Having these mistakes fixed by a human takes a lot of time and money.
reCAPTCHA solves this problem (and the problem of preventing comment spam) by presenting a user with two scanned in words. For one of these scanned words, the answer is known. The user is asked to type in both words and the reCAPTCHA system assumes that if the known word is entered correctly, the answer to the unknown word is probably also correct. It will store this answer and present it to other people until it is confident enough the original answer is right.
The scanned books used by the reCAPTCHA system are from the Internet Archive which are mostly Creative Commons licensed books. Hopefully soon all these books can be enjoyed in digitized form by all of us.
If you are interested in integrating reCAPTCHA into your website, check out the reCAPTCHA resources page where you will find plugins for WordPress, phpBB, Drupal etc. Also available are programming libraries for languages like PHP, Python and Ruby and a reference to the web-based API.
Today, 
