W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2003

Re: screen reader testing

From: David Poehlman <poehlman1@comcast.net>
Date: Thu, 24 Apr 2003 08:06:32 -0400
To: Charles McCathieNevile <charles@sidar.org>, "Alexander, Dan" <Dan.Alexander@mdx.com>
Cc: w3c-wai-ig@w3.org
Message-id: <00b701c30a59$f31c9690$6501a8c0@handsontech>

Good points Chals,  I would add that the differences in the way screen
readers interact with and provide the information to the user may not even
be a question of compatibility, but one of style. If this is repetitious, it
may be due to the style in which my screen reader provides info to me <grin>
At any rate, one form of rendering is not enherantly better than another so
The best thing to do is to get information from the screen reader
developpers about what they support and how they suport and at what version
level if you want to have that on hand to use during your design.

----- Original Message -----
From: "Charles McCathieNevile" <charles@sidar.org>
To: "Alexander, Dan" <Dan.Alexander@mdx.com>
Cc: <w3c-wai-ig@w3.org>
Sent: Wednesday, April 23, 2003 9:35 PM
Subject: Re: screen reader testing



Screen readers are like browsers - there is probably a dominant one at
any given time, but it can change pretty fast. In addition, if you code
accoding to the quirks of product A, you are betting that you don't need
other clients than users of product A, and that you will learn the new
quirks when there is a new version. And new versions come out every
month or so for some screen reader or other - much like browsers.

Testing for screen reader (and browser) compatibility is a good idea
after you have ensured that you are working to standards, so that any
system you haven't tested but which works better according to the agreed
standards will still work for you. It's one thing to say that your site
isn't easy to use with Braillesurf 1.0 because it is not capable of
recognising modern standards. It is another to say that you can't
actually work with standards-compliant sytems because you only coded for
(the obsolete browsers) IE 5.5 and Netscape 4.77.

I would suggest testing with as many screen readers as you can afford,
both with very skilled users (who can tell you a lot about how
professionals use a screen reader), and with beginners (presumably like
you and certainly like me) who can show what happens when a new user
starts up for the first time. If that is too much, I would suggest
making sure that you are meeting standards and not breaking anything for
basic users.

just two cents worth

Charles McCN

Alexander, Dan wrote:

>I've noticed a lot of differences between screen readers in the way they
>read the content. This makes for a particularly troublesome problem in
>testing because, not only are we testing for browser compatability but also
>screen reader compatability. I was wondering if there has been any market
>research done as to which screen reader is the most commonly used? Which is
>the best to test on?
>
>I would appreciate any help you can provide.
>
>Dan Alexander
>
>
>
Received on Thursday, 24 April 2003 08:08:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:09 GMT