W3C home > Mailing lists > Public > wai-xtech@w3.org > January 2006

RE: agreement: user disposes; disagreement: author proposes [was: Re: When actions speak louder than words]

From: John Foliot - WATS.ca <foliot@wats.ca>
Date: Mon, 2 Jan 2006 11:14:31 -0500
To: "'David Poehlman'" <david.poehlman@handsontechnologeyes.com>
Cc: "'Charles McCathieNevile'" <chaals@opera.com>, <wai-xtech@w3.org>
Message-ID: <074c01c60fb7$9e890710$6501a8c0@bosshog>

David Poehlman wrote:
> John,
> Charles does have a point though, what if the user agent needs a
> specific key binding bbecause it does not support the technology
> which replaces access key?

S.O.P. - user agents that do not support a particular element or
attribute ignores said element or attribute.  Look at <q> in Internet
Explorer...  End users live with it, or upgrade accordingly.

When you seek to bind a key, you end up with the same mess that we have
with accesskey today.  Which key David?  And what about conflict
resolution?  Accesskey has no means of conflict resolution today, which
is one of the reasons that I continue to advocate not using accesskey
(and thankfully today I am not alone). There are other reasons too, but
I will stay on topic... You've read and heard my rants in the past.

The ACCESS element and @role and @key attributes are newly proposed for
XHTML 2; user agents that will support this new Recommendation will
conceivably support these concepts natively - this is yet again another
reason to get this right the first time.  I argue that the real need for
@key does not exist - that the reasoning for it's inclusion is flawed.

John Foliot  foliot@wats.ca
Web Accessibility Specialist / Co-founder of WATS.ca
Web Accessibility Testing and Services
Phone: 1-613-482-7053  
Received on Monday, 2 January 2006 16:14:44 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:51:31 UTC