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

"Charles McCathie Nevile" <chaals@yandex-team.ru> wrote on 06/21/2013
04:29:55 PM:

> From: "Charles McCathie Nevile" <chaals@yandex-team.ru>
> To: Richard Schwerdtfeger/Austin/IBM@IBMUS, w3c-wai-gl@w3.org,
> public-html-a11y@w3.org, "David MacDonald" <david100@sympatico.ca>,
> Cc: "'Steve Faulkner'" <sfaulkner@paciellogroup.com>,
> mick@nvaccess.org, kirsten@can-adapt.com
> Date: 06/21/2013 04:30 PM
> Subject: Re: HTML5 alternatives to table summary don't work in
> current browsers, and  Screen Readers
>
> On Fri, 21 Jun 2013 17:05:17 +0200, David MacDonald
> <david100@sympatico.ca> wrote:
>
> > Hi Richard
>
> > A small correction to your take on what I said in my post. I actually
did
> > not make a recommendation to return to table Summary. I am simply
> > documenting that as we enter into recommendation status, the advice we

> > are providing to web authors fails WCAGs conformance requirement of
> > accessibility support. And the example techniques listed currently
don't
> > help blind folks... even though web authors trust us to give them
useful
> > advice.
> >
> > There are certainly well documented disadvantages (and advantages) to
the
> > Summary attribute but until AT catches up on replacements, and they
have
> > had several years to do so, we are looking at another of those awful
> > gaps that work on paper but not in the real world for blind folks.
> >
> > From: Richard Schwerdtfeger [mailto:schwer@us.ibm.com]
> [...]
> >> 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.
>
> Do you have some timeline for that?
>
They are working on it as we speak. Beyond that I am not in a position to
share Freedom Scientific product release date.

> >> I have copied Mick Curran at NVDA and hopefully they will also correct
> >> the problem with IE and Firefox.
>
> In my experience the NVDA guys generally do a good job.
>
yes they do.

> [...]
> >> We can't be writing specs. based on proprietary assistive technology
> >> defects.
>
> Actually, part of the HTML5 revolution was that instead of writing things

> that ought to work, we should be relying on what *does*.
>
> The "proprietary technology" whose defects could derail us are the
> fundamental products people are relying on. In the absence of viable
> alternatives, and without fixing those products, we're not ready to claim

> that we have produced a spec that is actually useful to anybody.
>
I appreciate that but this is a bug. We had one feature in ARIA that we
added for a work-around in a design flaw in JAWS that was fixed in JAWS. It
is something that I have regretted ever since.

> That may only be a temporary setback, but the lesson of the past is that

> temporary might last a generation (or their opportunity to get a useful
> education). We should be careful about charging ahead and saying stuff
> works, even while we don't want to try and stop real progress.
>
Sure. If we have a bug though let's report it and get it fixed. That is
happening now and I am glad David reported it. In the future, if you have
bug with an AT over ARIA support please let me know directly. I may be able
to help. I don't always monitor this list.

Best,
Rich

> cheers
>
> Chaals
>
> --
> Charles McCathie Nevile - Consultant (web standards) CTO Office, Yandex
>        chaals@yandex-team.ru         Find more at http://yandex.com
>