it doesn't work, yet

Kelly Pierce told the story of a compatibility failure between a
search server at the American Printing House for the Blind and
the Lynx browser.  The two products diverged after the search
server was carefully selected only after compatibility testing.

First, this reaffirms why we need some organization such as the
WWW Consortium involved in accessibility.  We need a
compatibility plan that works, but which leaves different tool
developers free to make lots of decisions within the constraints
laid down in the compatibility plan.  The pre-competitive
technology argument is just "right" for universal access.

If only there were a W3C compatibility test suite that both
developments were using, and it enforced a sufficient model so
that the services needed were actually covered.

This sad tale is a cautionary one, because it shows that the
Consortium is not yet effective in assuring basic
interoperability across the Web, never mind special concerns for
adaptive technology and people with disabilities.

There are two components to a solution, as I see it.

On the one hand, we need all the abstract models used to define
dialogs, data, and information connections to be connected
formally to one another so that there is one, connected web of
meta-stuff rationalizing all that we call "the Web."  [This may
exist and I just don't understand it; I am still learning my way
around.]

On the other hand, the Consortium may need to take a fresh look
at its bias toward "technology, not policy" and tilt a little
more toward enforcement behavior so that the industry achieves
adequate self-regulation.  This is not just for special needs,
this is the amount of torque applied to the proposition "you
should conform to Consortium recommendations..." across the
board.

-- Al Gilman

Reference:

To find Kelly's post, start at the webwatch home page at

 http://www.teleport.com/~kford/webwatch.htm

Received on Tuesday, 2 December 1997 11:32:14 UTC