W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 2005

RE: Agenda + [2.4] CORRECT version of 2.4 proposal

From: Jason White <jasonw@ariel.its.unimelb.edu.au>
Date: Thu, 5 May 2005 13:42:19 +1000 (EST)
To: John M Slatin <john_slatin@austin.utexas.edu>
cc: Gregg Vanderheiden <gv@trace.wisc.edu>, w3c-wai-gl@w3.org
Message-ID: <Pine.GSO.4.58.0505051331290.13858@ariel.its.unimelb.edu.au>



On Wed, 4 May 2005, John M Slatin wrote:

>
> Interesting point.  The current wording of 2.4 L3 SC1--
>
> <blockquote> 1. When content is arranged in a sequence that affects its
> meaning, that sequence can be determined programmatically. [I]
> </blockquote>

A preliminary point: as worded here, this looks like a constraint on how
the content is encoded in the delivery unit, not on how the content is
expressed. With this wording, it belongs under 1.3, not 2.4, as it merely
requires the structure to be exposed in the delivery unit by specifying
that it be determinable programmatically. Worse, this is redundant with
guideline 1.3 level 1.

>
> Was *intended* (by me; the wording is mine) to exclude the situation
> Gregg envisions here, where the content is arranged in such a way that
> there *is* no default linear order.  (I presume that at some point the
> user must be offered choices about what path(s) are available; how much
> information is provided about the paths will vary from situation to
> situation, as (for example) in Michael Joyce's hypertext fiction
> *afternoon, a story( (1987) or Stuart Moulthrop's *Victory Garden*
> (1991), both composed using the hypertext authoring tool Storyspace
> (which predates the Web by a number of years).

If that's the intent, then simply require that there be a sequence (or
reading order) in the content which is more meaningful than all the
others. That may not be testable in every situation, needless to say; but
I think the programmatically determinable part belongs under 1.3 and is
already covered there, unless you think it ought to be made more explicit;
but then, table structure falls under 1.3 anyway and that's a prime
example.

>
> So I don't think we're trying to prevent authors from creating webs of
> linked nodes in which users choose where they want to go and there is no
> default order (or if there's a default it's not in any sense a preferred
> order-- it's just a default).sequence affects meaning: I think we're
> trying to find a way tell authors who *don't* mean to create such webs
> what they have to do to help user agents and Ats avoid really stupid
> mistakes that would create unnecessary confusion for the user.

In that case, why isn't it adequately covered by 1.3? If on the other hand
you want to exclude the type of content described above which fails to
possess a single, most meaningful, reading order then it would be better
to write a more carefully worded (and testable) criterion along these
lines and introduce it at level 2 or level 3 of guideline 2.4. This would
be a constraint on the structure of the content, not on how it is encoded
to make it explicit to user agents.
Received on Thursday, 5 May 2005 03:42:41 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 23:39:37 UTC