WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Hidden Items read out by Voiceover and Talkback


From: Birkir R. Gunnarsson
Date: Mar 26, 2015 7:59PM

My experience was directly related to display: none not being
sufficient for VoiceOver )(we were developing some accessible modals
for a client).
Putting aria-hidden="true" along with the display: none solved the
problem .. at least for VO versions prior to 8.2 (have not tested with
the latest and, I hope, greatest).
I definitely recommend giving it a shot over display: none and
visibility: hidden being used together.
At least I have not seen that bug recently, though one never knows
with all the versions and updates going on out there.

On 3/26/15, James Nurthen < <EMAIL REMOVED> > wrote:
> VO certainly shouldn't read content which is hidden using display:none
> although I have found a number of cases recently where it does. One of
> these was even a <script> element where the JavaScript source was getting
> read by voiceover.
> I have a few bugs to log.....
> On Thu, Mar 26, 2015 at 6:31 AM, Patrick H. Lauke < <EMAIL REMOVED> >
> wrote:
>> On 26/03/2015 13:25, Robert Fentress wrote:
>>> I haven't played around with this much, but here is a post from Roger
>>> Johansson that suggests visibility:hidden, in addition to
>>> display:none, may sometimes also be necessary.
>>> http://www.456bereastreet.com/archive/201205/hiding_visible_
>>> content_from_screen_readers_with_aria-hidden/
>>> Following the link trail further back, you see a Gez Lemon post that
>>> spells out the details. That piece is from 2007 and focuses on JAWS
>>> and Window-Eyes, so I don't know how relevant it is to VO on iOS, but
>>> worth taking a look at, maybe.
>>> http://juicystudio.com/article/screen-readers-display-none.php
>> I tested this recently (at least the display:none not being enough and
>> visibility:hidden also being required) in recent Window-Eyes version, and
>> it seems that's an antiquated bug that's been fixed.
>> https://github.com/twbs/bootstrap/pull/15755
>> https://twitter.com/jitendravyas/status/562940090553733121
>> I still suspect that the original problem here is something far simpler
>> (correct CSS not actually being applied correctly), but until we can get
>> confirmation (via remote debugging/inspection of the actual code in
>> question), there's not much more we can try to do here... (particularly as
>> it doesn't seem to be a reproducible problem).
>> P
>> --
>> Patrick H. Lauke
>> www.splintered.co.uk | https://github.com/patrickhlauke
>> http://flickr.com/photos/redux/ | http://redux.deviantart.com
>> twitter: @patrick_h_lauke | skype: patrick_h_lauke
>> >> >> >>
> > > >

Work hard. Have fun. Make history.