Re: How important is aria-hidden, when already hidden with CSS?

Thanks everybody for the clear, relevant replies. I'm concluding that for
mobile responsive, I will use CSS alone to hide content. For user
interactions that show and hide content, I will try the HTML5 hidden
attribute and use it as a selector for CSS.

This matters for performance. I have alternate layouts for a small piece of
content: one layout for large screens, and one for small screens. The two
layouts are semantically different, so I want to show one and hide the
other. This is straightforward with a CSS media query. If instead I had
needed to toggle aria-hidden, then it would have required the
implementation effort of JavaScript to detect window resize events, and the
runtime overhead of modifying the DOM.

In light of this use case, would it be correct to change "must" to "should"
in the ARIA spec?

(Begin quote...)
    Some assistive technologies access WAI-ARIA information directly
through the DOM and not through platform accessibility supported by the
browser. Authors must set @aria-hidden="true" on content that is not
displayed, regardless of the mechanism used to hide it.
(...End quote)

http://www.w3.org/TR/wai-aria-1.1/#aria-hidden


On Dec 1, 2014 7:08 AM, "Steve Faulkner" <faulkner.steve@gmail.com> wrote:
>
> these results from last year may be helpful:
> of the SR's tested chromewvox is the only one that relied solely on the
DOM (though has changed as DOM is apparently too limiting as an
accessibility API) On the other end is VoiceOver that does not interpret
the DOM directly at all.
>
> Screen reader support for hidden content
> http://www.html5accessibility.com/tests/hidden2013.html
>
> --
>
> Regards
>
> SteveF
> HTML 5.1
>
> On 27 November 2014 at 00:42, Mitchell Evan <mtchllvn@gmail.com> wrote:
>>
>> I have content with CSS display:none; visibility:hidden. The WAI-ARIA
spec says add aria-hidden, to ensure support support for assistive
technologies (AT) that access the DOM directly.
>>
>> For practical accessibility: which AT accesses the DOM directly?
>>
>> For conformance: is the CSS alone enough to meet WCAG success criteria?
>>
>> @mitchellrevan
>
>