Re: updated change proposal for issue 129 ARIA in HTML5

hi stephen,

If a user skims the page content using a screen reader, for example by
moving from heading to heading using the h key, he/she will get the whole
hgroup content, because the whole content is no a single heading with
multiple lines.

The only time the subheading is masked is in regards to the outline
algorithm.

a mechanism which exposes the semantics of a heading and subheading is best
for accessibility, not one that collapses them.

>The default stylings of h2 and h3 also have no meaning out of context.

so? what we are discussing is what is conveyed by the default visual styling
of 2 headings of differing ranks put together.

if styling conveys no semantics to the visual user when  hgroup is used then
the styling of the hx of highest rank should be applied to all hrgoup
content.

the other point which I think is an important one, is that how headings (or
any content) is presented to end users should not be mandated in the HTML5
spec, it is not for graphical user agents, why should the graphical user
agents default mappings to accessibility APIs remove information that can be
used by downstream user agents to present content to users?


regards
stevef


On 2 February 2011 11:32, Stephen Stewart <carisenda@gmail.com> wrote:

> Hi Steve,
>
> Great, though I think maybe I'll try once more and apologies for
> clogging the list :)
>
> As I understand it the "semantics" of the design of a headline, other
> than it is a headline, it is that one line is the mainline, the rest
> are not required to get the gist of the headline. Thus in the example
> I have provided 'Clinton Acquitted' is enough; if you are skimming you
> can get that part and miss the underline but still retain the subject
> of the article. hgroup is giving you exactly this mechanism for
> accessibility, the ability for an accessibility client to mimic the
> skimming of a document which headline design is meant for. If you wish
> to get the whole headline, read the whole hgroup. In this area I think
> hgroup is actually a very successful element wrt accessibility.
>
> Your new example doesn't help me, "Foo Bar" has exactly no meaning
> therefore it is likely that any conclusion could be drawn from it. The
> default stylings of h2 and h3 also have no meaning out of context.
>
> Sorry to be a bore,
> Stephen
>
> On 2 February 2011 10:32, Steve Faulkner <faulkner.steve@gmail.com> wrote:
> > thanks Stephen,
> >
> > 2 points:
> >
> > 1. The example text content is not meant be indicative, I have changed it
> to
> > <hgroup>
> > <h2>Foo</h2>
> > <h3>Bar</h3>
> > </hgroup>
> >
> > 2. there are internal semantics conveyed visually
> > h2 has a default rendering that differentiates it from h3
> > and this is usually augmented by the use of CSS to further differentiate
> a
> > heading and subheading.
> >
> > thus the heading/subheading semantics are lost to accessibility clients.
> >
> > regards
> > stevef
> >
> >
> > On 1 February 2011 23:46, Stephen Stewart <carisenda@gmail.com> wrote:
> >>
> >> Hi,
> >>
> >> The section on hgroup[1] caught my eye, it seems to me to use a bad
> >> example:
> >>
> >> <hgroup>
> >> <h2>The problem</h2>
> >> <h3>Analysis</h3>
> >> </hgroup>
> >>
> >> This seems to define two sections, Problem and Analysis ("The problem
> >> Analysis" is very clumsy so it seems more likely to be two headings).
> >> hgroup is for headlines which associate n+1 H1-6 elements with one
> >> section, not two. If I were to substitute a more correct example I
> >> think most of the problems disappear in this section as per the
> >> author's own logic:
> >>
> >> <hgroup>
> >> <h2>Clinton Acquitted</h2>
> >> <h3>Perjury, obstruction charges defeated</h3>
> >> </hgroup>
> >>
> >> Taking away the layout that should read as "Clinton
> >> Acquitted<beat>Perjury, obstruction charges defeated", there are no
> >> internal semantics to convey other than the order of the words.
> >>
> >> [1]
> >>
> http://www.w3.org/html/wg/wiki/ChangeProposals/ARIAinHTML5#basis_for_defining_h1_to_h6_element_that_does_have_an_hgroup_ancestor
> >>
> >> HTH,
> >> Stephen
> >>
> >> On 1 February 2011 12:12, Steve Faulkner <faulkner.steve@gmail.com>
> wrote:
> >> > Hi all,
> >> > the html accessibility ARIA subteam have discussed the counter
> proposal
> >> > provided by the html5 editor.
> >> > We have updated our change proposal[1]
> >> >
> >> > We have removed any out of scope issues (for this proposal) and have
> put
> >> > issues of an editorial nature (in this proposal) to one side.  This
> does
> >> > not
> >> > mean that we do not consider editorial changes are required, but that
> >> > these
> >> > can be dealt with during last call.
> >> >
> >> > We have essentially boiled the change proposal down to what we
> consider
> >> > the
> >> > most important issues that need to be resolved prior to last call.
> >> >
> >> > [1] http://www.w3.org/html/wg/wiki/ChangeProposals/ARIAinHTML5
> >> >
> >> > --
> >> > with regards
> >> >
> >> > Steve Faulkner
> >> > Technical Director - TPG
> >> >
> >> > www.paciellogroup.com | www.HTML5accessibility.com |
> >> > www.twitter.com/stevefaulkner
> >> > HTML5: Techniques for providing useful text alternatives -
> >> > dev.w3.org/html5/alt-techniques/
> >> > Web Accessibility Toolbar -
> >> > www.paciellogroup.com/resources/wat-ie-about.html
> >> >
> >> >
> >>
> >>
> >>
> >> --
> >> Stephen Stewart
> >> carisenda@gmail.com
> >
> >
> >
> > --
> > with regards
> >
> > Steve Faulkner
> > Technical Director - TPG
> >
> > www.paciellogroup.com | www.HTML5accessibility.com |
> > www.twitter.com/stevefaulkner
> > HTML5: Techniques for providing useful text alternatives -
> > dev.w3.org/html5/alt-techniques/
> > Web Accessibility Toolbar -
> > www.paciellogroup.com/resources/wat-ie-about.html
> >
> >
>
>
>
> --
> Stephen Stewart
> carisenda@gmail.com
>



-- 
with regards

Steve Faulkner
Technical Director - TPG

www.paciellogroup.com | www.HTML5accessibility.com |
www.twitter.com/stevefaulkner
HTML5: Techniques for providing useful text alternatives -
dev.w3.org/html5/alt-techniques/
Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html

Received on Wednesday, 2 February 2011 11:49:16 UTC