W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2012

Re: approval

From: Karl Groves <karl@karlgroves.com>
Date: Tue, 21 Feb 2012 10:26:31 -0500
Message-ID: <CABScKPC0PVv38e0e9V_=Y6j0Oeg2Z=71G5-9DFAM5QNH011kBA@mail.gmail.com>
To: joshue.oconnor@ncbi.ie
Cc: David Woolley <forums@david-woolley.me.uk>, WAI Group <w3c-wai-ig@w3.org>, Meliha Yenilmez <melihayenilmez@yahoo.com>
On Tue, Feb 21, 2012 at 3:46 AM, Joshue O Connor CFIT
<joshue.oconnor@ncbi.ie> wrote:
> Karl Groves wrote:
>>
>> On Mon, Feb 20, 2012 at 5:02 PM, David Woolley
>> <forums@david-woolley.me.uk>  wrote:
>>
>>> To find out if it is actually accessible, you need to find people with
>>> lots
>>> of different disabilities and perform a usability survey on them,
>>> allowing
>>> them to use their own browsers and any assistive technology that they
>>> use.
>>
>>
>> Do you do this?  Really?
>>
>> With all due respect, this is exactly the type of attitude that
>> perpetuates the impression that accessibility is nebulous, expensive,
>> and difficult.
>
>
> How exactly? All David seems to be saying is that usability testing with the
> widest range of users is a great way of testing your 'real world'
> accessibility. I agree with him. For sure, it can be expensive and time
> consuming but it isn't _impossible_ and will reap rewards in the long run
> that far outweigh it's initial cost.
>
> Not least because the experience of watching people with disabilities using
> their website stays with the designer and informs their approach to
> accessibility with any future projects they undertake.
>

IMO, you've already answered the question. : "All David seems to be
saying is that usability testing with the
widest range of users is a great way of testing your 'real world'
accessibility."

"Widest range of users". Lets look at what such usability testing requires:
1) Recruiting of participants (preferrably, of course, those who fit
defined personas).  To get the "widest range of users", you need to
include a number of users with different disabilities.  For each of
those participants you need to consider their operating system and
their assistive technologies.  For screen reader users, do they also
use a braille output device?  What version of screen reader? What
brand & model of braille display?  What operating system? What browser
& version?
2) Combine the above with the logistical challenges of testing with
PWD.  First, PWD often don't drive, so you're sunk in the water unless
you live in an area with great public transportation - unless of
course you're going to hire someone to go get participants.  But if
you get them to your facility, then you have to deal with creating a
test environment that mirrors their own work/ home environment.
Alternately, you can go to them.  Do you have a portable lab?

This all takes a long time and a lot of effort. At the bare minimum
you have to spend time *finding* these participants, determining the
test tasks, scheduling the sessions, performing the sessions,
collating the notes gathered from the sessions, etc. etc.  This type
of effort is not free.  Even if you do all of this stuff in-house
using in-house participants you still have actual time spent doing
this work by employees.  Their fully actualized cost is real money out
of their employer's pockets.

I won't argue that usability testing is the hands-down best way to
know whether real people can successfully perform tasks on a system.
What I will argue is that gathering accessibility data with usability
studies is inefficient and there are plenty of other methods that
should be utilized first.

Karl
Received on Tuesday, 21 February 2012 15:27:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 21 February 2012 15:27:02 GMT