Accessibility and anti-robot testing

One thing that PAS78 hopefully will bring to more people's attention is the use of graphics for anti-robot testing... Also known as the obscured numbers in a graphic that you have to type into a box to confirm your registration or if your a complete marketing-speil word-monkey then 'Captcha' or 'Turing testing'.
PAS78 raises the issue, quite rightly, that if a user has any type of sight impairment the use of this technology could in fact disable the use of these forms unless there is an alternative method available.

Hotmail does include an audio version of it's Captcha image, so that a user can listen to the numbers being read, and then enter them into the verfication form field (although if you have ever attempted to listen to one of these it's incredibly diffcult to inteprete what the hell is being said as they have distorted the sound to deter robots with sound analysis!)

Google has gone to the extent of using SMS to send a user a passcode, which is a novel idea, but SMS isn't completely accessible yet either so that isn't a solid solution - and let's face it - not exactly cost effective either if you have to set up an SMS gateway etc etc.

Unfortunately at the moment there is no surefire ideas on how to implement accessible Captcha methods... Virtually all methods can be hacked and using these methods of human detecting inevitably will be difficult to use for some users... A 'code' based system such as those mentioned above may be difficult for dyslexic users; The common 'distorted image' method is difficult for those with visual impairments; and so on and so on...

So what is the solution? Well I'm not about to answer that here but do make sure you consider these users next time you implement some form of anti-robot testing in your next application... Create an audio version if you are going to use the recognised 'distorted image' method, try to avoid complex logic and always consider ALL of your users!

related link

PAS78 : What's all the fuss about?

So PAS78 has been released now and the dust is starting to settle somewhat but my initial impressions are of some disappointment.

Although finally we have some solid statistics and analysis to convince our clients that developing a website with the broadest range of users in mind is welcoming, to say the very least, there is nothing new in the DRC's documentation - it's been the accepted norm that the W3C's Web Content Accessibility Guidelines are the guidelines that we should all be following for some time now.

So why all the fuss? Well that's something I still can't determine... I mean I genuinely hope that the government has decided to take more of a stance and enforce the Disability Discrimination Act of 1995, with regards to digital media formats, and was hoping that the PAS78 documentation might actually be an enforcement more than a general overview but I'm afraid that's exactly what it is.

So the conclusions are the same as before: consider what the site should do and for whom; consider who you are going to get to create it; consider how to enforce the guidelines; and consider how this can be tested.

So much hype; so much expectation; so little new information.

There are some valid points that I have tried to advocate on all projects I've worked on however:

Never rely solely on automated WCAG checking tools such as WatchFire and the like; try to actively envolve disabled people in your testing

Consider how disabled people might access your website - there is no substitute for working directly with people with a range of disabilities

Remember that most disabled people don't use assistive technology but rely on standard browser features such as text resizing

Compliance isn't all about design or the code you use - consider the way you present content and the wording you use

Consider whether the same functionality provided by client-side technologies such as javascript and flash can be achieved on the server before the data is presented to the user.

I'll be delving deeper into the guielines on a regular basis and providing my views and opinions on solutions to some of the issues raised so watch this space!

Check out the W3C's Web Content Accessibility Guidelines

Meeting with the Shaw Trust

I had the privelage of meeting some representatives of the Shaw Trust today regarding their work with disabled web users. Very very interesting and although I've been advocating accessibility guidelines for over a year now they certainly opend my eyes to some new ideas... The Shaw Trust works directly with disabled web users so has a really good understanding of exactly how they use the internet.

Our meeting coincided with the launch of the DDA's PAS78 Guidelines (which I'll be reviewing soon!) and between us we managed to raise some very interesting points:

Accesskey's and whether they are good are bad... What?? Yes we actually came to the conslusion that pre-defining acceskeys is a bad idea... For example what happens if they conflict with a usres browser shortcuts? Or, what if you've set a user defined shortcut in your browser so that ALT+G takes yo straight to Google? Ahhhh not so great now are they? The W3C guidelines state that defining accesskeys is a good thing so it'll be intersting to see what PAS78 makes of them.

There's no denying they are a great tool but for now I suggest you create a set of session variables which allows a user to define their own accesskeys and indeed allows them to be "turned on" or "off" accordingly.

related link