Re: Is there an html or WCAG definition of "modality" as it relates to keyboard behavior in a modeal dialog?

GV:  Hmmm.   the dialog box sounds like a context to me.  and moving out of the dialog box using standard navigation commands (arrows or tab-key) would seem to be a ‘change of context due to change of focus” which would be a WCAG SC failure.     explicit commands (like hot keys or command key combinations ) would not violate it - but moving focus around in a context should not jump them out of that context.  At least that is my take.  If you want a ruling  you might run this past the committee as a whole.  I added the chairs to this email for their information. 


> On Apr 15, 2015, at 7:46 PM, Matthew King <mattking@us.ibm.com> wrote:
> 
> On 15 April 2015 at 20:22, Birkir Gunnarsson wrote:
> > It is not clear to me that allowing user to get to the address bar is
> > a definitely violation of 2.4.3
> 
> Steve Faulkner <faulkner.steve@gmail.com> wrote on 04/15/2015 02:40:19 PM:
> > Why would it be considered a violation?
> > If the modal dialog does not allow interaction with the HTML 
> > document content it is scoped to, then it is doing its job no? It 
> > should not disallow interaction with UI outside of the document 
> > (such as browser UI)
> 
> The problem is when keys used to navigate a dialog, e.g., tab and arrows, allow focus to leave the modal dialog. It is roughly analogous to an experience that would allow tab to move focus to the start button and system tray from a modal in a Windows desktop app.
> 
> There are keys in all browsers that support navigation to the browser chrome just like there are keys in all desktop operating systems that support navigation to the chrome. Web aps should not block navigation to the chrome with such keys. But allowing focus to wander outside a modal using the keys that are used to navigate the modal creates serious problems.
> 
> The problems are particularly hard hitting for low vision and screen reader users. I would imagine they are equally vexing for many users with cognitive disabilities.
> 
> I strongly support the idea of an "author SHOULD" or even an "author MUST" in the language of the new modal property to address this issue.
> 
> Matt King
> IBM Senior Technical Staff Member
> I/T Chief Accessibility Strategist
> IBM BT/CIO - Global Workforce and Web Process Enablement 
> Phone: (503) 578-2329, Tie line: 731-7398
> mattking@us.ibm.com
> 
> 
> 
> From:        Steve Faulkner <faulkner.steve@gmail.com>
> To:        Birkir Gunnarsson <birkir.gunnarsson@deque.com>, 
> Cc:        "WAI Protocols & Formats" <public-pfwg@w3.org>
> Date:        04/15/2015 05:09 PM
> Subject:        Re: Is there an html or WCAG definition of "modality" as it relates  to keyboard behavior in a modeal dialog?
> 
> 
> 
> 
> On 15 April 2015 at 20:22, Birkir Gunnarsson <birkir.gunnarsson@deque.com <mailto:birkir.gunnarsson@deque.com>> wrote:
> It is not clear to me that allowing user to get to the address bar is
> a definitely violation of 2.4.3
> 
> Why would it be considered a violation?
> 
> If the modal dialog does not allow interaction with the HTML document content it is scoped to, then it is doing its job no? It should not disallow interaction with UI outside of the document (such as browser UI)
> 
> As a reference, Chrome implements the HTML5 dialog element <http://www.w3.org/TR/html51/semantics.html#the-dialog-element>, when in a modal state the user can only interact with the modal dialog content of the document, but is able to interact with the browser UI.
> 
> If you open http://s.codepen.io/stevef/debug/dPxgZq <http://s.codepen.io/stevef/debug/dPxgZq>? in chrome , you can check the keyboard interaction of its dialog element implementation.
> 
> --
> 
> Regards
> 
> SteveF
> HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/>
>