Chair pre-review of the issue-204 change proposals

http://dev.w3.org/html5/status/issue-status.html#ISSUE-184

"Exempt ARIA attributes from the rule that prohibits reference to hidden 
elements"

While the call for Alternate or Counter Proposals has not timed out yet, 
the proposals appear to be stable, and the Chairs are expediting the 
processing of this issue in order to get to the resolution of issue 30.

  - - - - - - - -

At the present time, we have two change proposals:

http://www.w3.org/html/wg/wiki/ChangeProposals/AllowAriaReferHidden
http://www.w3.org/html/wg/wiki/ChangeProposals/ARIA_Can_Only_Refer_To_Hidden_Content_With_Specific_Restrictions

First a review based on format:

AllowAriaReferHidden has all of the required sections and content.

ARIA_Can_Only_Refer_To_Hidden_Content_With_Specific_Restrictions
(hereafter referred to as ARIA_CORTHCWSR) has all of the right sections, 
but the details section doesn't contain the required information and, in 
fact, contains extraneous information.  It appears that all of the 
material currently in the details section can be retained as it would 
fit in the rationale section.  It is merely the description of the 
change being proposed that is missing.

Unless this flaw is corrected by April 18th, the chairs will not accept 
the ARIA_CORTHCWSR proposal, and will instead proceed to a call for 
consensus on the AllowAriaReferHidden proposal.

Next, a review based on content:

Both proposals seem to agree that the following statement in the current 
specification (post revert) is either inaccurate or incomplete:

> It would similarly be incorrect to use the ARIA aria-describedby
> attribute to refer to descriptions that are themselves hidden. Hiding
> a section means that it is not applicable or relevant to anyone at the
> current time, so clearly it cannot be a valid description of content
> the user can interact with.

The AllowAriaReferHidden proposal proposes a change that it asserts to 
be consistent with the ARIA specifications.  As previously discussed, it 
is not clear what the ARIA_CORTHCWSR proposal proposes to remedy this 
situation.

The ARIA_CORTHCWSR proposal also makes a number of claims based on 
process and scope.  Without knowing what is actually being proposed, it 
is hard to evaluate this argument.  If aria-describedby were the only 
ARIA attribute referred to in the existing document, then perhaps this 
could argue for removal of the paragraph cited above.  But this is not 
the case.  In fact, there is an entire section (numbered 3.2.7 at the
moment) on WAI-ARIA, and was the subject of a number of decisions,
including:

http://lists.w3.org/Archives/Public/public-html/2011Apr/0091.html
http://lists.w3.org/Archives/Public/public-html/2011Mar/0161.html
http://lists.w3.org/Archives/Public/public-html/2011Mar/0005.html

It is not clear why the HTML WG cannot refer to part of another group's 
specification.  If the point being made is that there is attempt being 
made to change the semantics of "aria-hidden" then this point needs to 
be expressed better.

The following statement needs to be backed up by evidence:

> This is in fact consistent with how browsers are implementing this
> today.

Based on this, the process and scope concerns should either be enhanced 
to include new information sufficient to cause the previous decisions to 
be reopened, or these concerns should be removed from the proposal.

Finally, an editorial comment on the following statement:

>> While the HTML5 specification can "allow" other standards and
>> specifications, even those as closely linked to the HTML5 work effort
>> as ARIA is, to do as it chooses with regard to defining conformance
>> and parsing requirements, it cannot prescribe however how those other
>> standards or specifications must operate.

This is a poorly worded run on sentence.  The "it" in "to do as it 
chooses" is unclear.  It is not clear what "allow" refers to here.  We 
suggest you reword this sentence.

- Sam Ruby

Received on Wednesday, 28 March 2012 11:26:11 UTC