[Bug 23384] ARIA: <script>, <style>, <head> etc should not have aria-hidden="true" as default.

https://www.w3.org/Bugs/Public/show_bug.cgi?id=23384

--- Comment #1 from steve faulkner <faulkner.steve@gmail.com> ---
(In reply to Leif Halvard Silli from comment #0)
> http://www.w3.org/html/wg/drafts/html/master/dom.html#sec-strong-native-
> semantics
> 
> Currently, <script>, <style>, <head> (etc) have gotten aria-hidden="true" as
> a strong, native semantic.
> 
> As consequence, if an author makes - via CSS - any of these elements
> visible, AT users will not get the content since the aria-hidden="true"
> semantics would not be removed by a change of the CSS.
> 
> Also, moving thes elements from the strong to the weak table would not
> change much since, even then, users would not get the content unless the
> author made sure to change the default aria-hidden status from "true" to
> "false".
> 
> I would argue that nothing is gained by giving script, style, head etc
> aria-hideen="true" as a default value. Rather, I think HTML5 should just
> considere these elements as hidden, in the traditional display:none meaning
> of the word.
> 
> As small detail for <head> is that it contains <title>, and since the
> aria-hidden state applies to the children as well, this woiuld in theory
> hide the title element.
> 
> One could ask for the usecase for making script/style/head visible. And yes,
> it will probably not be done very often. But the fact that is is so seldom
> done also means that there is no problem to solve - thuse we don't need to
> say that these elements should have aria-hidden="true" as default.

Hi leif, I agree and had flagged this previously (can't find where at the
moment), browsers do not show this content by default to users and already
handle that effectively in the acc layer. Also note that aria-hidden removes
content from the acc tree, setting display:block on one of the elements you
listed will no add the content back in the acc layer, so it will be visible,
anyway have removed aria-hidden mapping, if edit is acceptable please close the
bug

-- 
You are receiving this mail because:
You are the QA Contact for the bug.

Received on Sunday, 29 September 2013 09:18:59 UTC