W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2004

2.1 operable with keyboard (was: Request for review)

From: Phill Jenkins <pjenkins@us.ibm.com>
Date: Wed, 18 Aug 2004 09:30:44 -0500
To: <w3c-wai-ig@w3.org>
Message-ID: <OFF217CBA3.98B2B3E5-ON86256EF4.004DB85A-86256EF4.004FB8CC@us.ibm.com>
>At 03:54 PM 17/08/2004 +0100, Andy Budd wrote:
>>Guideline 2.1 Make all functionality operable via a keyboard or a 
>>Is this not a little subjective? For instance, if a web browser doesn't 
>>support tabbing between links or form elements, it's beyond the site 
>>developers ability to do anything about it.
>I take this one to discourage practices such as using onmouseover, 
>onmouseout as the sole means of triggering an event.  Using the more 
>general onfocus or onblur if you must use events to trigger items on the 
>web page.  Though being a tad more explicit about this might be good.

I believe the guidance in the HTML Techniques needs to be very specific. 
The assumption that the browser is UAAG 1.0 compliant needs to be taken 
into account.  For example, onMouseClick is fired by most browsers when 
pressing the keyboard Enter key.  Should onMouseOver and onMouseOut also 
be triggered by the browser when tabbing to and from the element?  Why 
not?  Why put the burden on the author?  We shouldn't ask to have it three 
ways; keep the event in the spec but ask the author to not use that event 
and then also ask the browser to provide access to that event anyway.  The 
XHTML spec, WCAG 2.0 and UAAG 1.0 need to be considered as a set, and if 
there is contradiction, then it's probably WCAG 2.0 that may need to be 

In my opinion, WCAG 2.0 guideline 2.1 is good in principle (of course we 
all want all functionality to be operable via a keyboard), but in 
implementation it is most likely the browser that provides the keyboard 
access to the event.  I believe the real problem is with the event spec in 
XHTML.  For example, why was onFocus and onMouseOver both specified? Is 
there some real good reason to be able to specify both?  In other words 
did the spec writers expect onFocus to work with either the keyboard, or 
the mouse, etc. etc. and onMouse to *only* work with the mouse?  From an 
accessibility point of view, why did they do that? 

Phill Jenkins
IBM Research - Accessibility Center
Received on Wednesday, 18 August 2004 14:31:17 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:29 UTC