W3C home > Mailing lists > Public > wai-xtech@w3.org > April 2012

Re: Comment on ARIA 1.0: Problematic conformance requirement for aria-hidden

From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Date: Fri, 27 Apr 2012 15:48:43 +0200
To: Steve Faulkner <faulkner.steve@gmail.com>
Cc: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>, PFWG Public Comments <public-pfwg-comments@w3.org>, W3C WAI-XTECH <wai-xtech@w3.org>
Message-ID: <20120427154843404173.40087789@xn--mlform-iua.no>
It might be that it needs to be clarified whether one must use 
@aria-hidden="true" *also* when one uses @hidden. However, in 
principle, I do not agree that clarification is needed: 
aria-hidden=TRUE is in that case *not* needed. Why not? Because ARIA 
says that the host language feature should be used, rather than the 
ARIA feature, whenever there is one. And in the case of @hidden, then 
is written in HTML5 that it implies aria-hidden="true".

I think it is much more crucial what ARIA says about the situations 
where @hidden is *not* used but the element still is intended to be 
hidden: Should ARIA, then, say SHOULD rather than MUST?

I personally think that it perhaps should say SHOULD and not MUST.

We must consider that HTML5's @hidden is not intended to be used on 
elements that  are only temporarily hidden or for content that is 
hidden only for presentational reasons.

The problem, however, is that ARIA says that one, in these situations - 
the temporarily hidden situations - should use aria-hidden=true. Why is 
that problematic? Well - very many authors are much more proficient 
with CSS than with JavaScript. And with CSS, it is easy to temporarily 
hide things. However, if one adds aria-hidden="true" to a temporarily 
hidden element, then one must use javascript to set it to 

I personally know, with myself, that I am not going to use 
aria-hidden="true" on all temporarily hidden elements.

Leif H Silli

Steve Faulkner, Fri, 27 Apr 2012 14:13:59 +0100:
> Hi leif, I am saying that if display:none or html5 hidden is used to 
> hide content aria-hidden should not be needed or used.
> from some cursory testing that is the case with chromevox which use 
> the DOM dorectly and does not use accessibility APIs that is the case.
> regards
> Stevef
> On 27 April 2012 13:57, Leif Halvard Silli 
> <xn--mlform-iua@xn--mlform-iua.no> wrote:
>> Steve Faulkner, Fri, 27 Apr 2012 11:01:49 +0100:
>>> hi Ben,
>>>>How does this impact what ARIA should say here, in your view?
>>> I think that we should not be adding additional burden and complexity
>>> for authors with aria-hidden.
>>> aria-hidden confuses me.
>>> I think that html5 hidden and CSS display:none should be functionally
>>> and semantically the same and that both should imply aria-hidden=true
>> I think the intention of ARIA's MUST is that authors must/should do
>> element[aria-hidden=true]{display:none}
>> rather than
>> element{display:none}
>> Do you suggest that ARIA 1.0 should say "SHOULD" instead of "MUST"
>> w.r.t. aria-hidden="true"?
>> Leif H Silli
>>> On 27 April 2012 09:29, Benjamin Hawkes-Lewis wrote:
>>>> On Fri, Apr 27, 2012 at 9:21 AM, Steve Faulkner wrote:
>>>>> Have not been following this discussion closely, but just wanted
>>>>> to provide info about an implementation detail.
>>>>> In all supporting browsers CSS display:none is applied to elements
>>>>> with the hidden attribute.
>>>> Thanks Steve.
>>>> They are following the HTML5 spec in that default styling detail.
>>>> How does this impact what ARIA should say here, in your view?
> -- 
> with regards
> Steve Faulkner
> Technical Director - TPG
> www.paciellogroup.com | www.HTML5accessibility.com | 
> www.twitter.com/stevefaulkner
> HTML5: Techniques for providing useful text alternatives - 
> dev.w3.org/html5/alt-techniques/
> Web Accessibility Toolbar - 
> www.paciellogroup.com/resources/wat-ie-about.html 
Received on Friday, 27 April 2012 13:49:23 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:51:46 UTC