best practices in site testing:

----- Original Message -----
From: "John Foliot - bytown internet" <foliot@bytowninternet.com>
To: "WAI-IG" <w3c-wai-ig@w3.org>
Sent: Wednesday, December 18, 2002 8:21 AM
Subject: RE: Rockville, MD- Seeking low vision users for testing federal
website



For what it's worth, the Canadian Federal Government offers the
following
testing/verification steps for GoC web sites:

* Use an automated accessibility tool and browser validation tool.
Please
note that software tools do not address all accessibility issues, such
as
the meaningfulness of link text, the applicability of a text equivalent,
etc.
* Validate syntax (e.g., HTML, XML, etc.).
* Validate style sheets (e.g., CSS).
* Use a text-only browser or emulator.
* Use multiple graphic browsers, with:
- sounds and graphics loaded,
- graphics not loaded,
- sounds not loaded,
- no mouse,
- frames, scripts, style sheets, and applets not loaded.
* Use several browsers, old and new.
* Use a self-voicing browser, a screen reader, magnification software, a
small display, etc.
* Use spell and grammar checkers. A person reading a page with a speech
synthesizer may not be able to decipher the synthesizer's best guess for
a
word with a spelling error. Eliminating grammar problems increases
comprehension.
* Review the document for clarity and simplicity. Readability
statistics,
such as those generated by some word processors may be useful indicators
of
clarity and simplicity. Better still, ask an experienced (human) editor
to
review written content for clarity. Editors can also improve the
usability
of documents by identifying potentially sensitive cultural issues that
might
arise due to language or icon usage.
* Invite people with disabilities to review documents. Expert and novice
users with disabilities will provide valuable feedback about
accessibility
or usability problems and their severity.

(Source: http://www.cio-dpi.gc.ca/clf-upe/6/tools-outils2a_e.asp)

Interesting to note is that this list *is* vendor neutral.  While Bob's
points are certainly with merit, I would suggest that if the page "makes
sense" in the text only browser (as recommended above) that most if not
all
screen reading technologies will be able to access the page content.

JF



> -----Original Message-----
> From: w3c-wai-ig-request@w3.org [mailto:w3c-wai-ig-request@w3.org]On
> Behalf Of Nissen, Dan E
> Sent: Wednesday, December 18, 2002 8:08 AM
> To: WAI-IG
> Subject: RE: Rockville, MD- Seeking low vision users for testing
federal
> w ebsite
>
>
>
> Hi!
> I see a whole lot of criticism of what is a pretty minimal
> description of a
> part of an activity that is definitely going to be better than
> not doing it.
> The stick seems to be all some of you know how to do.  How about
> the carrot
> and see if we can encourage people to start down this road
> without setting a
> standard none of us can meet?  No way all the discussed
> environments need to
> be tested if the AT follows the standards and the web site is
> also designed
> to the standards.
>
> The expectations are way up there and the criticism is pretty quick on
the
> draw.
>
> Best regards,
> Dan
>
> -----Original Message-----
> From: David Poehlman [mailto:poehlman1@comcast.net]
> Sent: Wednesday, December 18, 2002 6:18 AM
> To: Joe Clark; WAI-IG
> Subject: Re: Rockville, MD- Seeking low vision users for testing
federal
> website
>
>
>
> any testing which reaches the rong conconclusions and passes them off
as
> correct is bad.
>
> ----- Original Message -----
> From: "Joe Clark" <joeclark@joeclark.org>
> To: "WAI-IG" <w3c-wai-ig@w3.org>
> Sent: Tuesday, December 17, 2002 10:42 PM
> Subject: Re: Rockville, MD- Seeking low vision users for testing
federal
> website
>
>
>
> > As others have mentioned, this is not the right approach to testing
> > website accessibility.  At best it tests one narrowly-defined aspect
> > of accessibility
>
> ...which nonetheless needs testing.
>
> > at worst it risks reinforcing any bad practices
> > you may have - such as authoring to browser behaviour at the expense
> > of presenting the website contents clearly
>
> ...which you have no evidence they are doing.
>
> > Both JAWS and Window-Eyes deal with one particular disability
>
> ...which nonetheless requires accommodation, and these are the two
> most popular ways to do it.
>
> > Both are themselves inaccessible to many users, by virtue of cost
> > and the prerequisites required to install them
>
> ...which is irrelevant and a tiresome albatross hung around the
> necks of the accessibility "movement." By this reasoning, no
> adaptive technology should be developed if it cannot be handed out
> for free to everyone who could possibly use it.
>
> If you disagree with the planned testing of actual disabled users,
> don't participate in it. But we need more such testing, and, as I
> argue in my book, even sub-optimal testing of disabled users beats
> the heck out of none at all.
>
> --
>
>   Joe Clark  |  joeclark@joeclark.org
>   Author, _Building Accessible Websites_
>   <http://joeclark.org/access/> | <http://joeclark.org/book/>
>
>

Received on Wednesday, 18 December 2002 09:17:11 UTC