Re: Action-2036 aria-keyshortcuts

Well, that is what the text saying Authors should not …

A browser is a software application. Within the application is your web page which becomes part of the application. The events hit there first and then bubble up so if you steal a key combination for a keyboard shortcut it won’t make it up to the browser function that uses it. … But that is after the Operating system receives them first. 

The screen reader hooks all the key events at the OS level. Some they grab, some they do not. If you have focus on an aria-enabled widget or you are in an ARIA “application” screen readers are programmed to allow all the key commands to go through with the exception of switching into a virtual cursor mode such as in the case of JAWS. However, if they don’t relinquish the keys you will never see them as they steal them at the operating system level. 

It is good that VoiceOver and ChromeVox steal key sequences that you are unlikely to use but there is no guarantee that the author won’t choose them. The author must make sure that they choose keys that do not conflict with them. … and again this is also an authoring issue. 

We have put specific authoring guidance in aria-keyshortcuts so that authors will be made aware of these general guidelines front and center. James Craig also wanted these in there.  

Rich 


> On Apr 7, 2016, at 3:03 PM, White, Jason J <jjwhite@ets.org> wrote:
> 
> 
> 
>> -----Original Message-----
> 
>> WRT user agent conflicts, most browsers use ctrl+s for saving the page, but
>> lots of online editors use ctrl+s to save the online document in the cloud. This
>> does not create any loss of user agent functionality because it is very easy and
>> efficient to use a browser menu to navigate to the user agent function.
> 
> If you handle a key event in a script and prevent the default action, is this guaranteed to override whatever the user agent assigns to that key combination? If so, then I would agree with Matt that it's a question of author guidance.
> 
> Some screen readers, on the other hand, aren't good at avoiding key combinations likely to be used by applications, which is a reason for using role="application" where necessary and for users to be aware of "pass-through" commands. I think this issue deserves a cautionary note in the spec.
> 
> VoiceOver under OS X and ChromeVox are particularly good at assigning commands to key combinations that applications are unlikely to use.
> 
> 
> ________________________________
> 
> This e-mail and any files transmitted with it may contain privileged or confidential information. It is solely for use by the individual for whom it is intended, even if addressed incorrectly. If you received this e-mail in error, please notify the sender; do not disclose, copy, distribute, or take any action in reliance on the contents of this information; and delete it from your system. Any other use of this e-mail is prohibited.
> 
> 
> Thank you for your compliance.
> 
> ________________________________

Received on Thursday, 7 April 2016 22:12:01 UTC