Re: Proposal: remove aria-describedat from the ARIA 1.1 specification

Deborah,

Is it important to show the relationship to the extended description or
could an implied grouping work?

<div role="group" aria-label="Periodic Table">
<img src="foo" aria-describedby="baz" title="periodic table">
<div id="baz">This is a picture of a the periodic table as it would appear
in 1900</div>
<details>
<table>
...
</table>
<summary>Extended Description</summary>
</details>
</div>

Rich

Rich Schwerdtfeger



From:	Deborah Kaplan <dkaplan@safaribooksonline.com>
To:	Richard Schwerdtfeger/Austin/IBM@IBMUS
Cc:	Steve Faulkner <faulkner.steve@gmail.com>, DPUB-ARIA
            <public-dpub-aria@w3.org>, WAI Protocols & Formats
            <public-pfwg@w3.org>
Date:	11/05/2015 11:19 AM
Subject:	Re: Proposal: remove aria-describedat from the ARIA 1.1
            specification



On Thu, 5 Nov 2015, Richard Schwerdtfeger wrote:
> That was not clear to people in the dub discussions we had at TPAC.

I am sorry about that. We tried to make it very clear in the
requirements documents but obviously did not succeed.

> aria-describedat does not solve that either.

DPub Has no opinions about which solution gets proposed, as long as one
does. We understand that the only way in which aria-describedat would
have solved the problem was by implicit association; it was being
implied that it would be used to mean extended description only, which
is not rational, since that was not part of the definition.

> So you want a description AND an extended description. ... or is it that
you want to have a description that can reside at a different
> location than what is in the page?

Regular describedby clearly has a use case of being something that
content creators need need for descriptions which automatically show in
the page, because people are using it that way in the wild. Extended
description is definitely something else -- serving the use case also
served, for images only, by longdesc.

Deborah Kaplan

>
> Rich
>
>
> Rich Schwerdtfeger
>
> Inactive hide details for Deborah Kaplan ---11/05/2015 11:04:54 AM---What
the digital publishing industry was requesting in thi Deborah
> Kaplan ---11/05/2015 11:04:54 AM---What the digital publishing industry
was requesting in this particular case was a single mechanism f
>
> From: Deborah Kaplan <dkaplan@safaribooksonline.com>
> To: Steve Faulkner <faulkner.steve@gmail.com>
> Cc: Richard Schwerdtfeger/Austin/IBM@IBMUS, DPUB-ARIA
<public-dpub-aria@w3.org>, WAI Protocols & Formats <public-pfwg@w3.org>
> Date: 11/05/2015 11:04 AM
> Subject: Re: Proposal: remove aria-describedat from the ARIA 1.1
specification
>
>
_________________________________________________________________________________________________________________________________________

>
>
>
> What the digital publishing industry was requesting in this particular
> case was a single mechanism for extended descriptions which always means
> extended description, and only means extended description.
>
> Deborah Kaplan
>
> On Thu, 5 Nov 2015, Steve Faulkner wrote:
>
> > arai-flowto is another mechanism that could be used to provide a
relationship between content and description
> >
> > http://www.w3.org/TR/wai-aria-1.1/#aria-flowto
> >
> > --
> >
> > Regards
> >
> > SteveF
> > Current Standards Work @W3C
> >
> > On 5 November 2015 at 16:20, Deborah Kaplan
<dkaplan@safaribooksonline.com> wrote:
> >       On Thu, 5 Nov 2015, Richard Schwerdtfeger wrote:
> >
> >
> >             You could link it via aria-describedby to whatever you are
describing.  This way you have an explicit extended
> >             description.
> >
> >
> >       But aria-describedby is for non-extended descriptions, standard
> >       descriptions that can't be associated with standard controls.
Admittedly
> >       the formal description of the element is ambiguous, in that it
> >       explicitly clarifies that a description is more verbose than a
label,
> >       but currently all AT (I believe) treats aria-describedby as a
> >       non-extended description which should be placed in the normal
flow and
> >       not reported as special. Is the proposal that aria-describedby
should
> >       always mean extended descriptions, and anything which is less
verbose
> >       than an extended description belongs in a label? If that's the
case, the
> >       formal definition of aria-describedby will need to be clarified,
and
> >       screenreader manufacturers will need to be asked to change the
behavior
> >       accordingly (e.g. to treat aria-describedby like longdesc, not
like
> >       aria-labelledby).
> >
> >       Deborah Kaplan
> >
> >             Rich Schwerdtfeger
> >
> >             Inactive hide details for Deborah Kaplan ---11/05/2015
10:07:45 AM---On Thu, 5 Nov 2015, Richard Schwerdtfeger
> >             wrote: > but I b Deborah
> >             Kaplan ---11/05/2015 10:07:45 AM---On Thu, 5 Nov 2015,
Richard Schwerdtfeger wrote: > but I believe this
> >             addresses the requirements of
> >
> >             From: Deborah Kaplan <dkaplan@safaribooksonline.com>
> >             To: Richard Schwerdtfeger/Austin/IBM@IBMUS
> >             Cc: WAI Protocols & Formats <public-pfwg@w3.org>, DPUB-ARIA
<public-dpub-aria@w3.org>
> >             Date: 11/05/2015 10:07 AM
> >             Subject: Re: Proposal: remove aria-describedat from the
ARIA 1.1  specification
> >
>
>________________________________________________________________________________________________________________________________________

>
> >             _
> >
> >
> >
> >
> >
> >             On Thu, 5 Nov 2015, Richard Schwerdtfeger wrote:
> >
> >             > but I believe this addresses the requirements of
> >             > the digital publishing industry.
> >
> >             One of our biggest concerns was that this extended
description be
> >             something that could be semantically identified AS an
extended
> >             description. Was this addressed in discussions at TPAC?
Based on the
> >             email that has gone by, I have not seen anything that will
distinguish
> >             extended descriptions in <details> elements from anything
else in a
> >             <details> element.
> >
> >             Deborah Kaplan
> >
> >
> >
> >
> >
> >
> >
>
>
>