On Thursday 02 December 2004 00:11, Robert Ewald wrote:
Replying to myself...
> On Wednesday 01 December 2004 18:10, Lubos Lunak wrote:
[snip]
> >  Besides, this looks a
> > lot like accesskeys. Wouldn't it be simpler if you only extended
> > accesskeys support to also assign some keys to elements which don't have
> > any accesskey assigned directly in the html page?
>
> It does indeed sound much like accesskeys, and in fact I have been
> "inspired" by the code dealing with access keys.
>
> The problem I see here: the access keys occupy a certain range, and in the
> worst case occupy all 26 characters, so if there are more elements visible
> there are not any characters left to span more elements with a second
> character. To give an example: 25 access keys are defined: 'a-y' then we
> could use 'za' for the first link with no access key, 'zb' for the second
> and so on. If 'z' would have been also defined as an access key by the
> webpage, no space would be left for links without access keys.
>
> That is why I think a seperate function  would be better.

After thinking a bit more, the seperate function is not such a good idea, it 
should really be merged with accesskeys. It would confuse all users that want 
to interact with a form and links via the keyboard. Also form elements should 
be accessible, even if no accesskeys are defined. The collision remains 
though. We could choose to ignore the accesskey definition in the webpage 
altogether. Thats a quick and dirty sollution to the problem, but I am not 
sure if it is a good idea.