Re: Requirements for Accessibility Description Language (ADL?)

At 08:46 AM 2000-11-11 -0800, William Loughborough wrote:
>At 11:48 AM 11/11/00 -0500, Al Gilman wrote:
>>"the response of the system should be predictable by the user"
>
>It's early on a Saturday morning and I've been doing other stuff but isn't 
>that a "principle" that should get into the guidelines in some kind of 
>Priority 1 sense?
>
>It's sort of covered when we talk about not making page leaps or 
>too-short-timed switchings and magnifying images causing loss of resolution 
>and descriptive text for movies that isn't synchronized with what's being 
>described.
>
>I guess what I'm positing is that in addition to being "predictable" it 
>should ideally be "controllable" in the sense that CSS has a "!important" 
>feature allowing the last cascade to be where God intended it.

Yes, sufficient (approximate) observability into the future of the interaction
process, together with controllability from the actual UI, even when the UI
has
been adjusted to work around device or user limitations or dis-preferences,
are
both required.  It is even possible to trace both these to a root requirement
for the closed-loop stability of the user's view of the interaction process. 
Both the observability and controllability requirements can be derived from
this root requirement, as you can't deliver the closed loop stability of the
process without some apropriate information and some appropriate control.

An important story to spread at the upcoming CUU conference is the following:

"The required closed-loop stability can be enhanced either by better
information or by better control.  In fact, to have a 'universal' level of
robustness in your service offering, it is not enough that the default
presentation of the service (as information plus action opportunities) be
informative and controllable; but that there be safety margin on both fronts. 
There must be easy-to-use methods of getting "we try harder" levels of either
information or control as the user chooses."

Note that it is not sufficient to look only at the user's view.  While I am a
big fan of making "author proposes, user disposes" a pervasive protocol, the
sustainability of practices that we propose is dependent on these practices
forming a framework for win-win transactions.  The art of the deal is the name
of the game.  The consumer and provider views both have to meet appropriate
standards.  The art of the deal is in setting up a protocol which rates as a
win from either perspective.

>
>AG:: "This stuff is part of accessibility, not in addition to 
>accessibility. Accessibility is not limited to static constraints that can 
>be evaluated in the context of a single page 'document.'"
>
>WL: For us to have any "softness" on whether "usability" is absolutely 
>essential to "Accessibility" just won't do. To exactly the extent that 
>usability is compromised so is accessibility. The most frustrating obstacle 
>to information access is spread across a population that's way more than 
>"half the planet" - all the same problems people have with using things are 
>just as significant for blind folks as they are for me. While on the steep 
>part of the learning curve, it's easy to roll backwards.
>

But we have to have 'softness' in how we deal with usability.  Usability is a
matter of better and worse, and how much better and how much worse.  Some ways
to achieve 'better' rate recognition as "reasonable accomodations," and others
don't.

Al  

>--
>Love.
>                 ACCESSIBILITY IS RIGHT - NOT PRIVILEGE
>

Received on Saturday, 11 November 2000 13:22:15 UTC