E-mail List Archives

Re: Wai Aria how useful?


From: Birkir RĂșnar Gunnarsson
Date: Jul 27, 2010 2:06PM


My point regarding the ability of NVDA to recognize certain things and
its potential as a usability screen reader, is mostly that if NVDA
support a given feature it is available to any user who really needs
it, free of charge. In Iceland we support a lot of users and have come
across web browsing problems with other major screen readers, in those
situations we download NVDA for the user and show the user how to use
it as a stop gap while the other SR is being fixed. This way we can
definitely say a user that has to access page X can do so
independently. If the alternative was to, say, by Jaws for a Hal user
who cannot access page X or document Y, that is an expensive and
lengthy procedure and we cannot really justify that to our purchasing
department, and hence we cannot say that the page is accessible with
screen reader Hal (this is merely a hypothetical scenario, feel free
to insert any other combination of screen readers herein).
We have been doing evaluation on NVDA with a lot of common user tasks
and would be happy to share our findings with you off-list if you
think it would be useful. Your sample of an accessible pdf form has
helped us tremensously so if I/we can repay the favour just ask, But
that is getting a bit off topic.
I feel usability testing is to estimate whether the user can, through
current software or free software changes (upgrade to browser/open
source screen reading) access and work with an object. I am far from
being an expert so this may be a wrong approach to usability theory.
Perhaps the WebAim survey is representative of the types of users who
would be likely to encounter, and interact with, Aria content though.
It is very likely that the majority of users who are elderly or not
very computer savvy will come across many Aria pages any time soon,
though even the most likely pages for an infrequent user to serve
would be social sites, such as FaceBook, I suppose.

On 7/27/10, Phil Teare < <EMAIL REMOVED> > wrote:
> Anyone know a screen-reader I can use on a mac that recognizes Aria live
> regions?
> Cheers
> Phil
> Phil Teare,
> On 27 July 2010 20:05, < <EMAIL REMOVED> > wrote:
>> " NVDA is rapidly growing in popularity. Whilst it lacks some of the
>> functionality that other screen readers provide, it's increasingly used as
>> a
>> backup option and in time I think we'll see it gain primary ground as
>> well.
>> Jaws is undoubtedly the most popular primary option, but the
>> WebAIM
>> survey reports that 49% of people use more than one screen reader. Of the
>> alternatives, NVDA is the most popular with nearly 26% of people choosing
>> it.
>> http://webaim.org/projects/screenreadersurvey2/"
>> I am very familiar with the results of the WebAIM survey. I was one of the
>> respondents. I think that the survey supports my view that NVDA in most
>> cases it is not the best platform for user testing. Only 3% of respondents
>> listed NVDA as their primary screen reader. I believe that user testing
>> should be conducted using the user's primary screen reader and browser
>> combination unless you are testing for use in a closed environment where
>> the
>> user will not have a choice of platform, including their OS, AT and other
>> software.
>> I believe that there is a risk in assuming that the population of survey
>> respondents is fully representative of screen reader users in general. My
>> gut instinct is that the population may have been a bit more tech savvy.
>> For
>> example 10% of the respondents reported that they do not have a
>> disability.
>> I would be very surprised to find that 10% of JAWS users do not have a
>> disability. I would also expect that 10% is heavily concentrated among
>> accessibility experts. Even among the 90% of respondents who reported a
>> disability I would not be surprised to learn that there was a
>> disproportionate representation of folks who could be considered
>> technology
>> and/or accessibility experts. I am not suggesting that the results of the
>> survey are invalid, just that some of the more surprising findings such as
>> use of multiple screen readers, and recent updates of AT may not truly
>> reflect the screen reader user population in general.
>> NVDA is a good tool for troubleshooting problems. It helps us confirm when
>> problems are related to the AT and not the application, document, or
>> system
>> under test. We are starting a pilot program to evaluate NVDA and/or SAToGo
>> as a supplemental assistive technology for our staff who use a screen
>> reader. Since most of our staff who use a screen reader are not
>> technologists and most could be considered "average" in their computer
>> skills, it will be interesting to get their feedback on the usefulness of
>> a
>> second screen reader.
>> Mike Moore