- From: Charles McCathieNevile <chaals@opera.com>
- Date: Sat, 12 Nov 2005 23:04:08 +0100
- To: "John Foliot - WATS.ca" <foliot@wats.ca>, "'Rinit Satishkumar Shah'" <shahr3@cs.man.ac.uk>, w3c-wai-ig@w3.org
- Cc: Jukka <jkorpela@cs.tut.fi>
Hi Rinit, John, all,
On Friday my friend John Foliot wrote:
> The debate on the pros and cons of accesskeys is, by now, well
> documented. While there *are* arguments for using them, on balance I
> believe that they can in fact cause more harm than good, and say so
> every chance I get. Should you care, we have written extensively on the
> topic at WATS.ca. I urge you to at the very least give the arguments
> some serious consideration:
And, as John knows, I usually follow along saying "Actually, access keys
are good, but the specification in HTML added a informative suggestion
about invocation that was a dumb idea and led to lots of people finding
the implementations were not very good".
I urge you, having read John's well-written material, to read Jukka
Korpela's article 'Improving accessibility with accesskey in HTML forms
and links' at http://www.cs.tut.fi/~jkorpela/forms/accesskey.html and Gez
Lemon's article 'Access Key Companion' at
http://juicystudio.com/article/access-key-companion.php and to give the
arguments *for* using accesskey serious consideration.
I take it that your intended project is in fact meant to alleviate some of
the problems described in existing implementations.
I also suggest you look at the Opera and iCab implementations. iCab has
the luxury of leaving all keys open, so you just press the relevant key.
And it shows what the access keys are (in most cases). Opera uses a single
command to move to acccesskey mode (default is shift-esc, but my personal
configuration uses slash instead) and then you can select any key. Both
implementations have room for improvement, but are well ahead of the more
common approaches used in reducing the problem of conflicts.
Just wait until we get Compound Document Formats using accesskeys in
several different source documents ;-)
I don't think that XHTML2 will not have an accesskey mechanism, although
it is likely to be specified slightly differently than the HTML 4 version.
There is no reason why the implementation in content should not be
compatible though.
cheers
Chaals
--
Charles McCathieNevile chaals@opera.com
hablo español - je parle français - jeg lærer norsk
Peek into the kitchen: http://snapshot.opera.com/
Received on Saturday, 12 November 2005 22:04:36 UTC