RE: QED & Marshall McLuhan

[Apologies for long posting. Executive summary: David Poehlman is right
about the fundamental distinction between accessiblity and understanding;
there are intrinsic reasons why the accessible non-text web will be a subset
rather than the whole; the WAI seems to have its plate full at this time
anyway; can we all be nice please. - WM]

> please consider how you would feel, excluded from most common 
> activities,
> including the www, and finding a job.

There are many situations which may lead to someone - even a reader like
myself - from feeling 'excluded from most common activities, including the
www' or to having difficulties 'finding a job'. Being a non-reader is one
such situation. Being poor is another. People who are poor cannot access the
www because they cannot necessarily get access to it, and even where they
can, they cannot get guaranteed regular access to it. This is arguably worse
than the situation of the non-reader confronted by a screenful of text which
they are not capable of processing. This is the situation of the reader who
is unable to actually get to the screenful of text in the first place.

Ought the WAI confront this issue of the economic inaccessibility of the
web, and campaign for computers and internet connections to be distributed
to all human beings at birth? It would be nice, but seems self-evidently
unlikely, at least for the WAI at this time.  Ought the WAI confront the
issue of the inaccessibility of the web on technical grounds, where a site
has been designed in a restrictively non-standards-compliant way such that
the access software used by a blind user cannot make sense of it? Of course
it should and so it does. Where does the non-reading community stand in
between these two poles? On which side of the line is it? While my previous
postings may have suggested that the non-reading community ought stand on
the same side of the line as the blind and visually impaired community, I am
no longer so sure.

The first place I looked for an answer to this question was the WAI mission
statement as quoted on the WAI homepage:

"The W3C's commitment to lead the Web to its full potential includes
promoting a high degree of usability for people with disabilities."

This seems clear enough. The WAI's commitment is to making sure that the web
is as usable as possible to all. It is a commitment to the proper structure
and presentation of the information stored on the web. It in no way makes
any demands whatsoever on the complexity of that information itself, except
asking that it be presented in a form appropriate to context, and it
presupposes that any extra software required to translate the standard
'view' of a given web page into a form more appropriate for a particular
individual is in place, just as it assumes that the computer and net
connection required to access a web page at all is also in place.

Since the vast majority of the information available on the web is in text
form, the problem of the non-reader is a particularly hard one, since it
involves treading on some highly dangerous ground. Fine. Let's tread on it.

Complicated scientific PhD theses may be intelligible to no more than a
couple of hundred people world-wide. When confronted with a page like that,
we are all, in a sense, non-readers. Here is an example, (with apologies to
Dr Jan Vroonhof):

'The Classical Antifield-BRST Operator and the Variational Bi-complex'

http://www.math.ethz.ch/~vroonhof/scripties/intro_phys/intro.html

of which I can make neither head nor tail.

(If you are a top mathematical physicist then this will not be a good
example for you. Try here instead:
http://www.csse.monash.edu.au/cgi-bin/postmodern )

What is to be done with such material? What is to be done with any material?

As David Poehlman has stated, the fundamental distinction that needs to be
drawn is that between access and understanding. It is clear that web
technology may be used in order to provide a more complex and potentially
fruitful experience space for those with reading problems or cognitive
difficulties, and it is also clear that Jonathan's peepo site is one of a
few pioneering works in this field. It is not at all clear that there is a
case for the construction of a non-text version of every web page; moreover
it is not at all clear that this would be possible or even desirable. After
all, at bottom, all that text is is a sequence of pictograms that may be
converted into words which a given human being may or may not comprehend. A
page full of text is just a page containing an awful lot of those
pictograms, and Jonathan's suggestion, taken to its extreme, would appear to
be that no page ought contain this level of pictogram complexity. This would
be self-evidently absurd, since logically it would appear to be demanding
that no-one posted any page at all ever in any language that anyone else
didn't understand. All web pages, by this yardstick, are irretreivably
inaccessible.

Let us step back from the extreme, and say instead that each page contains
an implied context of the audience which it is intended for. In the case of
the WAI site, this would be web and accessibility professionals together
with anyone interested in the subject of web accessibility. The subject of
web accessibility is made up of complex notions and constructs that, as with
the rest of the work of the W3C, demand carefully worded texts in order to
properly delineate them. There is an unashamed sense of intellectualism in
those texts that positively excludes not merely the non-reader but also the
limited vocabulary reader, or, as we have seen from some recent reports in
the press, the axe-grinding knee-jerk-reaction journalist on a deadline. All
of these are members of groups that cannot be guaranteed to have understood
any given text regardless of how it is presented to them and require some
form of precis instead; meanwhile the WAI site quite rightly and unashamedly
demands a certain level of cognitive ability, to say nothing of reading
ability, on its users, who get to enjoy the original texts in all their
polysyllabic glory; anything we don't understand is our own problem. (Or the
list's, or the web's, depending on which bit it was). The same goes for the
rest of the W3C site.

Checkpoint 14.1 of the content guidelines are extremely clear on this point:

'14.1 Use the clearest and simplest language appropriate for a site's
content.'

It would be highly inappropriate for the W3C site to be so clear and simple
that it was fully accessible to a non-reader, since the content of the W3C
site is precisely focussed on an information system created with and by one
of the more complex text-based  constructions ever created by humanity.
Certainly there could be a non-text based section of the site, but the site
itself would actually fail in its mission were it to accede to Jonathan's
demand to 'face the music and stop copping out.' That is, in any case, a
harsh demand with which to attack the W3C, which, in terms of accessibility,
is already having to deal with the far larger problem of actually soluble
accessibility issues around the blind and visually impaired, or around the
widespread deployment of near-HTML (ie not HTML), or around the widespread
built-in structural gotcha where a fool in a suit hires a fool in a
pony-tail to be their web guru, and the fool in the pony-tail takes
advantage of the ignorance of the fool in a suit to hide their own ignorance
and makes false claims like 'accessibility is too hard, too expensive and
too time-consuming' or whatever.

In terms of solving these pressing problems, there is much to be done, and
it is being done. In terms of providing complete web access to non-readers,
there is no music to face and no copping out. Dealing with accessibility
involves confronting some harsh realities. Non-experts can't understand the
article referenced above, blind people can't see, and non-readers can't
read. Solving accessibility problems demands that we accept these realities
and do our best to maintain the high degree of usability for all that
underlies the whole of the www. That degree of usability will vary from
person to person, and meanwhile, nothing we can do will ever make a
non-expert understand that article, make a blind person see, or make a
non-reader read. The success of the WAI is based on its realistic approach
to the issues and selective application of resources to actually solvable
problem domains, and the problem that Jonathan is presenting us with is not
a solvable problem domain, not, at least, in the terms that he seems to be
presenting it.

As we know, 'accessibility is right, not privilege'. Understanding, however,
*is* privilege. Those of us who are lucky enough to be able to read, to have
access to computers and the www and so on, in addition to having the time to
play with it all a bit, may also be lucky enough, like Jonathan, to get to
spend time working on ways of creating non-textual interfaces to *some* of
that information available online. However, I can conceive of no arguments
suggesting that it will ever be possible to create a non-textual interface
to *all* of that information. Moreover, I would assert to the contrary that
the set of information so complex it can only be expressed in textual form
constitutes too large a proportion of the information so available for it
ever to be possible.

Perhaps a future clarification of the scope and mission of the WAI might
preempt these problems with which Jonathan so eloquently presents us, by
making the distinction between access and understanding more clearly.
Perhaps the content guidelines could do the same. At some time in the future
it might be appropriate for an additional working group of some sort to look
at the issue of the non-text web and the extent to which some of the
text-based web may be filtered so as to be presented without words. However,
this last will involve co-operation and mutual understanding between the
parties involved, both of the limitations of the exercise and of the
limitations of the organisation involved. Recriminations, bad feeling, and
the reposting of personal correspondence to a mailing list probably do
little to foster such aims.

Yours wordily,

Wayne Myers
Interactive Software Engineer
BBC Digital Media
http://www.bbc.co.uk/education/
0181-752-6116

Received on Monday, 7 June 1999 14:04:57 UTC