Re: Preparation for 27 April teleconference [Was Re: AGENDA: W3C WAI User Agent Telecon 27 April 2000]

At 10:18 PM 2000-04-25 -0400, Ian Jacobs wrote:
>Hello Working Group,
>
>As starting points for discussion at tomorrow's teleconference, 
>please consider the following comments and proposals.
> 
>   Proposal:
>
>   1) Leave 2.1 checkpoint text the same.
>      ("Make available all content, including equivalent 
>        alternatives for content.")
>   2) Require that for content known by specification to
>      be for users (including information in style sheets),
>      that a document source view does not suffice.

While this may feel good as a principle, I have a problem with the "which
content is meant for users" concept.  Let me correct that.  I think the WAI
and the W3C should have a problem with that distinction, as it creates
fatal flaws in e-commerce, not just disability access.  It is just the
users of minority views [such as people with disabilities] who suffer the
consequences first.  [Canary in mineshaft...Universal Design...]

There should be no markup, ever, that is completely beyond the user's
discovery.  It can take N steps to expose and explain it, but it should be
reachable somehow.  This is a key element of the information architecture.
If it isn't self explanatory, the explanation should be discoverable by an
obvious process.  With the Web at our disposal as a way to wire in layers
of backup, there is no excuse for less.

I would like to check this with the PF WG for a formal position.  May I
register a dependency and promise a report on this?

The tactical problem with this split is that it points at a body of
literature which is simply not clear on this point.  To put this language
in the guidelines invites a large rathole of deferred argument.  The
strategic problem is that the distinction should not exist in the ideal
case, so why insert it now?

"What is for display" is view-specific.  Not document-information-generic.

"What is for the user" is not a valid concept in the Universal Access
architecture.  It is a residue of "view chauvenism;" someone's assumption
as to what view the user is using.  All the properties are informative, and
may be exposed in the over-the-wire encoding as text or (where available)
in a friendlier transform of that encoding.

Al

>   3) A document source view (or better) satisfies the requirement 
>      of making content available when otherwise difficult
>      (e.g., style sheets, script source) or when it is not
>      possible to know from the markup language specification
>      which content is meant for users.
>   4) The "document source" view is not a view of the
>      document object (the structured navigation view). The
>      user should find for example, raw script and style
>      sheet data in the source view.
>
>-- 
>Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
>Tel:                         +1 831 457-2842
>Cell:                        +1 917 450-8783
>

Received on Wednesday, 26 April 2000 10:04:36 UTC