Re: Acessibility of <audio> and <video>

Leif Halvard Silli wrote:

> VoiceOver is based on WebKit - aka Safari.

VoiceOver works with Safari/WebKit, but it is not based on Safari/ 
WebKit.

> This leads to this strange contradiction: While Apple's Safari  
> supports @longdesc, their Safari based screen reader doesn’t. (The  
> same contradiction applies for the screenreading support in Opera on  
> Mac OS X - which also uses VoiceOver.)

Even if VoiceOver was based on Safari, this is a flawed argument. Just  
because a property is in the DOM doesn't mean that the browser knows  
how to map it to an external API. By the logic of your statement, I  
could also add a custom "myAttribute" attribute to the DOM and then  
state that the value of "myAttribute" wasn't accessible.

> Anyway, my point was to say that the support for @longdesc is much  
> better than Lachlan claims simply because all the mayor browsers, on  
> which the screen readers usually are based, they support it.

The DOM is pretty freeform, so all browsers also support @myAttribute,  
too.