Re: HTML5 alternatives to table summary don't work in current browsers, and Screen Readers

Janina,
Thanks for your response.
Indeed I had  participated in that  discussion 5 yearas ago on "summary" attribute with  my 2 cents.
And that is the point. Five years down the line as David's analysis and Charles' comments  reveal there are any hardly workable alternatives supported reliably by browser-At combinations.
Charles refers to  a 'generation' gap with ref. to  user agents playing catch-up in implementing specs and that is based on real world experience.
That's why I advocate  a practical approach: do not deprecate something that works.
And the reference to 'hodge podge' of  attributes: Well it did go through the process for getting into the  specs,like other attributes in the specs, right?  
And about ARIA: Sure it can be used to augment HTML or SVG or whatever. But the guidance for using ARIA says (not me ... I am simply trying to learn and practice what is documented), that authors should use native elements and attributes first. And when they hit a roadblock then turn to ARIA. Or is my understanding wrong?
Thanks,
Sailesh Panchang

--------------------------------------------
On Fri, 6/21/13, Janina Sajka <janina@rednote.net> wrote:

 Subject: Re: HTML5 alternatives to table summary don't work in current browsers, and  Screen Readers
 To: "Sailesh Panchang" <spanchang02@yahoo.com>
 Cc: w3c-wai-gl@w3.org, public-html-a11y@w3.org, "Richard Schwerdtfeger" <schwer@us.ibm.com>, "Steve Faulkner" <sfaulkner@paciellogroup.com>, mick@nvaccess.org
 Date: Friday, June 21, 2013, 4:38 PM
 
 Hello, Sailesh:
 
 
 A few comments in line below ...
 
 Sailesh Panchang writes:
 > >>Let's stick with a consistent set of APIs
 (ARIA) that developers can go to as much as possible for one
 stop shopping. 
 > >>"back to the hodge podge of attributes that
 were thrown in at the end of HTML 4's release." 
 > 
 > The 'summary' attribute is like any attributes defined
 for  various HTML elements meant to aid accessibility.
 
 > Not sure why this is being singled out.
 
 I suggest your tense is incorrect. This is not a case of
 "is," but of
 long ago "was."
 
 This debate goes back many years. PF objected to removing
 table summary
 as early as Aug 2008, and the debate was quite old even
 then:
 
 http://lists.w3.org/Archives/Public/public-html/2008Aug/0213.html
 
 As I pointed out in my last message, the decision to drop
 summary on
 tables was made by the HTML-WG according to their decision
 policy--and
 it was made a long time ago as well, namely in April 2011:
 
 http://lists.w3.org/Archives/Public/public-html/2011Apr/0091.html
 
 Please note their decision announcement includes their
 reasons. One may
 disagree with their reasons, as many of us in PF and in the
 HTML-A11Y
 Task Force did disagree. Nevertheless, this was the WG
 decision.
 
 If you believe it should be put back, I would suggest you
 file a bug
 against HTML 5.1. That will insure this attribute is
 reconsidered.
 
 > ARIA is designed for making custom elements accessible
 that HTML elementss and attributes are unable to handle and
 not replace features that are AT-supported and serve users.
 
 
 
 ARIA is host language independent. This means it provides AT
 the
 opportunity for consistent feature support regardless of
 what host
 language is used. We need ARIA for HTML, but we also need it
 for SVG,
 for instance.
 
 > Not sure why something that works to make content
 accessible as demonstrated by David's analysis is being
 thrown out of the  window.
 
 
 It's not. Table summary was remove from HTML 5 long long
 ago. We failed
 to get it back in.
 
 Meanwhile, the WG decision referenced above argued that
 table summary
 was unneeded, because we were supposed to be able to use
 caption
 instead, as defined in the HTML 5 spec to this day.
 
 Well, it didn't work then, and it
 doesn't work now. That's what's going out the
 window--because we have a better replacement in the wings,
 and it will be
 specified in a PF published FPWD quite soon.
 
 > Making AT and browsers support new techniques that have
 been authored now is not 'fixing a defect' but an
 enhancement. 
 > Deprecating attributes that are well supported by AT is
 a retrograde step. 
 
 
 Indeed. I hope my explanation helps.
 
 Janina
 
 
 > Thanks,
 > Sailesh Panchang
 > 
 > --------------------------------------------
 > On Fri, 6/21/13, Richard Schwerdtfeger <schwer@us.ibm.com>
 wrote:
 > 
 >  Subject: Re: HTML5 alternatives to table summary
 don't work in current browsers, and  Screen Readers
 >  To: w3c-wai-gl@w3.org,
 public-html-a11y@w3.org
 >  Cc: "Steve Faulkner" <sfaulkner@paciellogroup.com>,
 mick@nvaccess.org
 >  Date: Friday, June 21, 2013, 10:48 AM
 >  
 >  
 >  David,
 >  
 >  
 >  
 >  I saw your post that you
 >  think summary should be reinstated because of a
 JAWS and/or
 >  NVDA have a defect (it works in VoiceOver) does
 not warrant
 >  going back to the hodge podge of attributes that
 were thrown
 >  in at the end of HTML 4's release. 
 >  
 >  
 >  
 >  Let's stick with a
 >  consistent set of APIs (ARIA) that developers can
 go to as
 >  much as possible for one stop shopping. 
 >  
 >  
 >  
 >  I have alerted Freedom
 >  Scientific of the defect and they will fix it. I
 have copied
 >  Mick Curran at NVDA and hopefully they will also
 correct the
 >  problem with IE and Firefox. 
 >  
 >  
 >  
 >  Mick, please see the
 >  following link:
 >  
 >  
 >  
 >  http://lists.w3.org/Archives/Public/w3c-wai-gl/2013AprJun/0089.html#start89
 >  
 >  
 >  
 >  We can't be writing
 >  specs. based on proprietary assistive technology
 >  defects.
 >  
 >  
 >  
 >  Rich
 >  
 >  
 >  
 >   
 >  
 >  
 >  
 >  Rich Schwerdtfeger
 
 -- 
 
 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/

Received on Sunday, 23 June 2013 17:46:32 UTC