Re: @aria-describedat at-risk in ARIA 1.1 heartbeat draft

Alex, three was interest in SVG in having this function. Gerardo Capriel
from Benetech, at that time, worked with George to bring the
aria-describedat feature to the aria working group.

Behavior work was something we were targeting for ARIA 2.

Honestly guys. longdesc was included in HTML4 for years and browsers
supported it for years. I have stated that I am not a huge fan of it as it
needs visual indication that there is more information there but
programmatically it is not a new invention. Rendering the indication should
be a requirement of the host language. The ARIA spec. can suggest this via
a MAY or a SHOULD but that is as far as it goes.


Rich Schwerdtfeger



From:	Alexander Surkov <surkov.alexander@gmail.com>
To:	John Foliot <john@foliot.ca>
Cc:	Dominic Mazzoni <dmazzoni@google.com>, James Craig
            <jcraig@apple.com>, WAI XTech <wai-xtech@w3.org>, "Michael[tm]
            Smith" <mike@w3.org>, Daniel Weck <daniel.weck@gmail.com>, "Ted
            O'Connor" <eoconnor@apple.com>, "W3C WAI Protocols & Formats"
            <public-pfwg@w3.org>
Date:	12/10/2014 01:05 PM
Subject:	Re: @aria-describedat at-risk in ARIA 1.1 heartbeat draft



That's where we disagree I think. If a feature in HTML doesn't exist in
other host languages but still useful there then imo it doesn't mean that
ARIA should host it. I don't have a nice solution though, I think the
solution will be feature dependent. Maybe we should start a new spec for
accessibility features and not only for stuff like universal longdesc that
will define semantics, UI and probably behavior. Maybe IndieUI is something
we should look at.
Thanks.
Alex.

On Wed, Dec 10, 2014 at 10:43 AM, John Foliot <john@foliot.ca> wrote:
  (Adding Public PF to the cc list per Rich's request)





  Alex,





  This then brings us to the crux of the problem: if we have an
  accessibility feature in HTML that does not exist in other host
  languages, is it reasonable to suggest that we ensure that feature can be
  implemented via ARIA?





  In more specific terms (and I already know the counter-arguments from
  Apple) if @longdesc is deemed a useful and needed accessibility feature,
  and we have existing content creators requesting this functionality, then
  why would we not want to have an ARIA equivalent? I read nothing in the
  draft spec that suggests that browsers need to support aria-describedat
  in any fashion differently than the current support for @longdesc, of
  which Mozilla is already doing a better job than some other browsers
  (with the current native support via context menu).





  I'll also note that it is worth mentioning that one of the requirements
  for @longdesc was that there be no mandated visual encumbrance, so the
  desire to not have a mandated impact on the GUI is already well
  understood and respected. I see no conflict here.





  (I will however continue to suggest that the end user SHOULD have the
  ability to apply a visual hint when either @longdesc or @aria-describedat
  is used, as users should always have the final say on how they get their
  content delivered to them. I have stated on numerous occasions that I
  believe this should be a Settings preference switch)





  JF








  From: Alexander Surkov [mailto:surkov.alexander@gmail.com]
  Sent: Wednesday, December 10, 2014 10:24 AM
  To: John Foliot
  Cc: Dominic Mazzoni; James Craig; WAI XTech; Michael[tm] Smith; Daniel
  Weck; Ted O'Connor



  Subject: Re: @aria-describedat at-risk in ARIA 1.1 heartbeat draft





  Hi, John. I don't. I think ARIA can be useful in any HTML-like
  environment, if you create custom controls in SVG then ARIA is perfect
  for that.


  Thanks.
  Alex.





  On Wed, Dec 10, 2014 at 10:16 AM, John Foliot <john@foliot.ca> wrote:


  Hi Alex,





  Once again, I ask you, do you believe that ARIA is for HTML only? Does
  the language-agnostic part of ARIA no longer apply?





  I totally support using Native semantics (etc.) over ARIA supplied
  whenever possible, but when the host language lacks a feature required
  for a11y, then I maintain we should be able to provide it via ARIA.





  If this is an incorrect understanding of the role of ARIA within the
  larger eco-system, then I think it requires further clarification, as the
  resources I have previously referenced seem (to me) to back up my
  understanding.





  Cheers!





  JF





  From: Alexander Surkov [mailto:surkov.alexander@gmail.com]
  Sent: Wednesday, December 10, 2014 10:11 AM
  To: Alexander Surkov; Dominic Mazzoni; John Foliot; James Craig; WAI
  XTech; Michael[tm] Smith; Daniel Weck; Ted O'Connor
  Subject: Re: @aria-describedat at-risk in ARIA 1.1 heartbeat draft





  Hi, Janina. I don't have clear use case of the universal longdesc in my
  mind so I'm not sure where the feature should be. If it's supposed to be
  used on HTML elements then HTML spec should be a right place to host it.




  In general I just don't support the idea to let ARIA to have UI dressing
  since it's all about semantics and I wouldn't want to change that.


  Thanks.
  Alex.





  On Tue, Dec 9, 2014 at 11:59 AM, Janina Sajka <janina@rednote.net> wrote:


  Do I understand that you correctly?

  If our intent is a general feature for the whole world, than you say
  "Put it in HTML?"

  If our intent is a feature for accessibility, would you say "ARIA is
  OK?"

  Is this correct? Is this your view?

  Just walking througha  clarification here ...

  Janina



  Alexander Surkov writes:
  > It looks like I should comment too. I think that if aria-describedat is
  a
  > nice feature for any element and all users can benefit of it then it
  should
  > be part of HTML5 spec.
  > Thanks.
  > Alex.
  >
  > On Mon, Dec 8, 2014 at 9:15 AM, Dominic Mazzoni <dmazzoni@google.com>
  wrote:
  >
  > > On Mon, Dec 8, 2014 at 9:04 AM, John Foliot <john@foliot.ca> wrote:
  > >
  > >> I think as well that your characterization of "dissent" w.r.t. Gecko
  and
  > >> Blink
  > >> is, shall I say, somewhat exaggerated, but (again) I think we should
  ask
  > >> these
  > >> actors directly, and neither you nor I assume anything.
  > >>
  > >
  > > Just to be clear, then, I officially object/dissent to the language
  "User
  > > agents should provide a device-independent mechanism to allow a user
  to..."
  > > used anywhere in the ARIA spec, because I feel the user agent
  directly
  > > providing to all users a user-level feature based on an ARIA
  attribute is a
  > > radical departure from the rest of the ARIA spec.
  > >
  > > Resolutions I would be happy with include:
  > > * Change the language so that aria-describedat is mapped to native
  > > accessibility APIs only, like the rest of ARIA
  > > * Or, make it part of HTML5 and take ARIA out of the name
  > >
  > >
  > >


  --

  Janina Sajka,   Phone:  +1.443.300.2200
                          sip:janina@asterisk.rednote.net
                  Email:  janina@rednote.net

  Linux Foundation Fellow
  Executive Chair, Accessibility Workgroup:       http://a11y.org

  The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
  Chair,  Protocols & Formats     http://www.w3.org/wai/pf
          Indie UI                        http://www.w3.org/WAI/IndieUI/