Re: The alt="" attribute

----- "Ian Hickson" <> wrote:
> If people do want to do actual research here, I would love to see more
> usability study videos of blind users using the Web without guidance,
> to 
> see how they actually interact with images. I think that that is the
> level 
> of research we need to really make more informed decisions.

I believe this statement to be born from a misunderstanding of usability.  Being able to show you videos of usability study participants is no more useful than the conjecture-and-anecdote based "research" you've referenced so far. The current state of usability work, as it relates to accessibility, is limited to a relatively small number of studies which were only qualitative in nature and seriously limited in scope. 

Gathering good, strong, quantitative data - which is the only kind that I'd regard as "actual research" - is going to be prohibitively time consuming, expensive, and unlikely to yield results that aren't already painfully obvious to those employed professionally in the usbaility and accessibility realm.  Numerous works have noted that an expert review performed by a skilled reviewer is just as likely (if not moreso) to find the bulk of usability problems in a system as an actual lab study [1]. "At present, the available research leaves us in doubt about the advantages and disadvantages of usability testing relative to other UEMs" [2].

This leads me to address your mention last week of "appeals to authority". While I can agree with your argument that statements shouldn't be accepted outright based simply on the identity of the person making the statement, it is also worth mentioning that these statements are born from sometimes decades of professional experience in the field of accessibility. I can't speak for others, but I can tell you that my 8-10 hour a day, 5-day-a-week interaction with disabled people largely negates the need for "research" into most accessibility issues - i.e. "alt text".

Asking for "research" seems like a smoke screen, in my opinion.  You either don't understand the nature of usability research, or you're using it as a way to say "Well, you didn't do any research so your argument isn't valid".  As a response, I'd challenge you to edit out every portion of the current draft which wasn't based on quantitative research.

Last, it bears mentioning that the type of "usability testing" you cite with blind users isn't going to lead to any conclusions that aren't already known.  For one thing, your results will be limited in their validity because they'll only be relevant to the types of images and alt text encountered by the user. Not only that but there are also other challenges such as the platform/ assistive technology used, the level of experience of the user and the nature/ severity of their disability.

Pointing a web cam at a blind person browsing the web is by no means "research" and is far less reliable than simply asking for (and listening to) advice from a skilled accessibility consultant.

1 - Chattratichart, Lindgaard. A comparitive evaluation of heuristic-based usability inspection methods (2008)
2 - Jacko, Sears. The Human-Computer Interaction Handbook.

Karl Groves
Senior Accessibility Consultant
703.637.8961 (o) 
443.889.8763 (c)


Received on Tuesday, 26 August 2008 15:25:43 UTC