[whatwg] Accesskey in Web Forms 2

[Matthew Raymond]

Ian Hickson wrote:
>>I don't think RENDERING of the access key should be optional.
> 
> It has to be. Google is a user agent. You can't require that google render 
> access keys, that makes no sense. :-)

   ??? Google is a user agent???

>>>I also think that there should be an accesskey value which is basically
>>>"auto", and which picks a non-clashing access key based on the element
>>>content.
>>
>>What's the purpose of this? Are there really situations where there are
>>so many access keys that the webmaster doesn't want to bother with
>>keepingtrack of them and just wants them auto-assigned?
> 
> I meant clashing with the browser's existing access keys, not clashing 
> within the page.

    Some user agents may have access keys spanning most of the alphabet. 
I'm concerned that this kind of solution does little to solve the 
fundamental problem, which is the fact that most browsers implement HTML 
access keys in such a way that they conflict with the browser's own 
access key space. Remember that we're limited to only a few dozen 
characters, and when you consider the keys the browser already uses, 
that number is tiny.

    Perhaps the best way to handle access key conflicts between the 
webpage and the browser is to simply prompt the user when they press the 
conflicted access key in question.