Re: Proposal for Guideline 2 as well as a proposal to trim WCAG 2.0 to 3 guidelines (won't william be glad?)

On Sun, 7 Jan 2001, Charles McCathieNevile wrote:

> I am not sure if there is a need to seperate out compatibility - it is very
> closely related to device-independence. But somehow it feels right to me like
> this, so I would be happy either way. If we are going to have a "use W3C
> technologies where available", or "use the most accessible technology
> available for a task", would that go in guideline 4?
That would seem to be the logical place for it. I agree with Charles'
intuition that separating out compatibility works better. Alternatively
one could write the full statement of my proposed guideline 1 in such a
way as to incorporate both compatibility/interoperability and
device-independence. The drawback here is that the number of checkpoints
under guideline 1 would then increase to at least 9, a situation which
would, of itself, be likely to attract objections.

Also, the concept of user agents as "devices" is problematic, as
it is contrary to established usage. On my laptop computer I have two user
agents installed (Emacs/W3 4.0 and Lynx 2.8.2, to be specific). If, for a
particular purpose, I switch from one of these user agents to the other
(by closing down one program and executing another), under the proposed
definition I would suddenly be using a different device to access the web,
a highly counter-intuitive idea, given that I would still be using the
same hardware, the same operating system, the same speech-enabled software
environment, etc. By "device" I think we should mean hardware (for
example, a screen of a particular resolution, a printer, a mobile
web-connected appliance, a dynamic braille display, a telephone, a digital
television set, etc.).

Nevertheless, it would remain possible to insert words or definitions into
the proposed guideline 1 that would allow it to encompass compatibility
with various user agents, without re-defining well understood terms such
as "device". One could even construct a technical term that would
encompass devices and user agents, but this might add unduly to the
complexity of the document.

Comments? Suggestions? Criticisms?

Received on Sunday, 7 January 2001 18:57:30 UTC