W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2006

Text-based Captchas

From: Matthew Smith <matt@kbc.net.au>
Date: Wed, 03 May 2006 17:52:40 +0930
Message-ID: <44586850.3060201@kbc.net.au>
To: WAI Interest Group <w3c-wai-ig@w3.org>

(Apologies to anyone who gets this as a cross-posting!)

Greetings

I have been working on producing a simple solution to replace graphical
Captchas that require nothing of the user agent, other than support for
HTML forms.

The biggest accessibility issue with my system would be (I believe) for
people with cognitive difficulties.  It wouldn't be too hard to make it
multi-lingual, although I'd probably prefer to use a database backend if
that were the requirement.

One caveat would be the use of homophones in the answers - this could
confuse those who are listening to rather than looking at the page.

My original attempt was written in Perl (my language of choice), but I
have now re-written the solution in PHP.

Appropriate links are here:
<http://www.smiffysplace.com/blog/archives/169-php-text-captcha>

Please feel free to use, abuse and generally criticise.  I'd be
especially interested to hear of any simple algorithms that might be
used to circumvent it.  (It would be easy enough to block excessive bad
attempts from one IP address within a specified time period, to block
brute-force methods.)


Cheers

M

-- 
Matthew Smith
IT Consultancy & Web Application Development
Received on Wednesday, 3 May 2006 08:22:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:24 GMT