W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2004

Re: General Usability DetecVS tor

From: Maurizio Boscarol <maurizio@usabile.it>
Date: Mon, 10 May 2004 14:56:21 +0200
Message-ID: <409F7BF5.8070002@usabile.it>
To: Gregg Vanderheiden <gv@trace.wisc.edu>
Cc: w3c-wai-gl@w3.org

Gregg Vanderheiden wrote:

>In discussions on our last call "usability" came up since it was a question
>from one of our reviewers
>How do we separate
>"Accessibility / Usability problems that are specific to disabilities" 
>"usability issues that are faced by all users"

Hello Gregg.

I think this is a tricky question. Usability cannot be referred to "all 
users", cause it depends on user knowledge, motivation, environment, 
goals, and so on.

Usability is not a general property of the site, but rather a property 
of the relationship (the interaction...) between a product (a site) and 
a specific user in a specific situation. This doesn't mean that we can't 
extimate general usability problems. We can in fact set up common 
scenarios for our site, and try to solve any problem that can cause 
trouble in that scenarios. There are also some common cognitive 
capabilities (perception, unique attention focus), and that capabilities 
should be takent into account.

But there are also some goal-related specific problem (and 
disability-related, of course). There are some domain-konowledge related 
problem. There are expertise-related issues, and so on.

Furthermore, to take the time as a reliable indicator of performance, we 
should have significant data, that in most usability test (with 3-8 
user) we can't have.
A task completing-time can vary a lot in non-disabled user. So it would 
be very rare (and expensive) to have good means and standard deviation 
in homogeneus testing groups, that need 15 and more user, and particular 
testing conditions.

Maybe we can try to set up some simple, standard scenarios, a common 
task-set (searching info, reading article, fillin up forms with personal 
data, browsing in categories, etc), and then build up a matrix in which 
we can compare different type of diabilities in common scenarios/task. I 
don't know if I am clear: I will try to do an example, if you like.

Best things,

Maurizio Boscarol
Received on Monday, 10 May 2004 08:59:21 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:49 UTC