W3C home > Mailing lists > Public > w3c-wai-eo@w3.org > January to March 2003

"What is Web Accessibility?" slide

From: Al Gilman <asgilman@iamdigex.net>
Date: Thu, 06 Feb 2003 12:13:36 -0500
Message-Id: <>
To: w3c-wai-eo@w3.org
Cc: pjenkins@us.ibm.com, Charles McCathieNevile <charles@sidar.org>

Judy asked for comments on the slide content as follows:

Our latest version, as of today's discussion, and on which I'll be asking
for feedback on from the WAI CG (Coordinating Group):

What is Web Accessibility?
An accessible Web is one that can be used by people with disabilities, and
also benefits other users of the Web. Web accessibility includes:

- content that is presented in a way
          * that can be perceived, operated, navigated, and understood by
people with disabilities;
- software used to access the Web
          * that can be used by people with disabilities,
          * and that works with technologies that some people with
disabilities use
- software used to build Web sites
          * that can be used by people with disabilities,
          * and that supports creation of accessible Web sites;
- technologies that underly the Web (HTML, CSS, SMIL, XML, etc)
          * that support accessible Web content and software.


Yes, I have serious heartburn about this slide.

What the WAI provides is knowledge about what it takes for the Web to be
accessible, where "accessible" in this context means "accessible to people
with disabilities."

Stating that that is what "an accessible Web" *is* (implying outside of the
discourse conducted on the W3C site) is an affront to the autonomy of every
reader of this material.  Don't go there.  It is placing a gratuitous
barrier in the way of having the reader sign on to follow our lead.  Making
the message inaccessible.

There is also a problem that Wendy raised: the breakout that follows
reflects a backward-looking view of what the Web is.  This still reflects
the Web one can browse today, but it does not describe the Web that the
people PF talks with are building technologies for.

But the real problem is not this obsolescence in the "what is the Web" story
assumed.  More fundamental than this accidental obsolescence is the
unfortunate mixing of 'how' into 'what.'  Break these out and tell both
stories.  "*What* an accessible Web *is* (in the context of the WAI
publications) is a different matter from *how* to have an accessible Web
(which is the topic of the WAI publications).

The public controls the "Journalist's Agenda" of topics: who what when where
why and how.  If we haven't decomposed our story enough to distinguish among
these primitives, we are not speaking in the common rhetorical vernacular.
It's not just simple words, it simple story lines and roles within the plot
that will make our story accessible.

If we in the WAI are not re-factoring [see Extreme Programming] our 
of our customer, the Web; and re-grouping our efforts around a forward-looking
map of where the action is in creating the to-be Web, we will always be at 
one technology generation behind in our assistive provisions, and we won't even
be able to keep up with maintaining that, to boot.

What we need on this point is something more or less like the following 

"When we in the WAI talk about an 'accessible' Web, we mean "a web that is
accessible to people with disabilities."

What's the difference?  We don't claim that this is what an 'accessible' 
Web *IS*,
i.e. what this language should mean in any and all discourse contexts, but 
we take a humbler posture and _describe_ for the reader a short-cut form 
that we
will habitually use because we need to refer frequently to this concept whose
plain-language statement is a nested prepositional phrase.  But that nested
prepositional phrase employs a syntactic form that is controlled by the vast
majority of speakers of English World-wide, and in the expanded adjective 
form, it does not depend on any Terms of Art -- it uses terms that are current
in the common parlance in their common senses.

I am not sure if you want to include the development below in your overview 
It is how one should teach this patch of usage conventions, however:

accessible [Webster] ::= readily grasped (to a user)

[aside, to be developed in a footnote, but must be understood to see why
the next is said as it is.]
the Web is both a medium in which information is delivered and actions 
accepted from users, and a corpus of services engaged in those 
activities.  Users don't distinguish between the medium and the 
message.  [Here message == service, wherein web documents afford an 
information-transmittal service.]

an accessible Web is a Web where:

- the user can grasp the information that web content is trying to tell 
them,   limited only by their ability to understand, not by any artifacts 
of the medium of expression.

- the user can attain any outcome that Web applications are trying to offer 

- the user can remain confident that they are directing the course of the 

a Web which is accessible to people with disabilities is one where the 
above three qualities are sustained in the presence of human conditions 
impairing functions often employed in the human-computer interaction of the 
user engaged with the above services.

a Web which offers equal access to people with disabilities is one where 
the performance on the above three axes: complete delivery of information, 
access to outcomes, and effective command of the process -- is subtantially the
same for people with disabilities and for people without disabilities.


To make what we mean by various terms clear, we

- must not make it sound as though we assume people should always equate
accessibility with PWD accessibility.  I.e. recognize and affirm the general
sense of 'accessible.'  Distinguish _our_ use of 'accessible' from what
'accessible' 'is' already to our audience.  There is a great line on this in
the current CACM in the article by Philip Agre "P2P and the Promise of
Internet Equality[1]." He points out that technologists turned social
advocates are singularly naive in failing to start with an assessment of how
the public _already_ views the issue where they wish to sway public opinion.
We have fallen into the same trap on this one.  We don't define the topic.
We have to engage with the topics that the public perceives.

- distinguish between 'accessible' and 'equally accessible' because the 
bulk of experience falls within the difference between these two.

Those are the basic must-haves in this phase of our exegesis.

In addition, this slide degrades our exposition of what it is for the Web to
be accessible (to PWD) -- which is simply a matter of the user's experience.
This slide mixes in a deployment of "what it takes for the Web to be 
based on a backward-looking model of what _the Web_ is.  This is bad on two
counts.  Don't mix the statement of what it is to be accessible -- which is in
in operational performance terms in the user's experience -- with methods for
attaining this outcome.  Those have to be on two different slides.

The bottom four bullets address "how the structure of the WAI Technical 
Activity addresses the issue of ensuring that the Web will be accessible."

That is a very different issue from what "an accessible Web" 'is.'  And we 
should not make that the question, anyway.  Make it "The accessible web we 
are talking
about _here_ is <explanation/>."


I am copying Phill and Chaals because they did yeoman duty in the QuickTips 

[1]  Communications of the ACM February 2003/Vol 46, No 2, p. 39.
Received on Thursday, 6 February 2003 12:13:44 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:29:33 UTC