W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > October to December 2001

Re: OnChange in dropdown combo box used for navigation

From: Al Gilman <asgilman@iamdigex.net>
Date: Tue, 18 Dec 2001 18:18:21 -0500
Message-Id: <200112182306.SAA2478935@smtp2.mail.iamworld.net>
To: "STELZER,MELINDA (Non-HP-PaloAlto,ex1)" <melinda_stelzer@non.hp.com>, w3c-wai-ig@w3.org

At 04:42 PM 2001-12-18 , STELZER,MELINDA (Non-HP-PaloAlto,ex1) wrote:
>I was reading the discussion below concerning two different ways of handling
>pages that use a dropdown, or "combo box" control as a navigation tool.  In
>one method, the user must press the form button to cause the browser to go
>to the new destination, and in the other method, the "OnChange" javascript
>event causes the choice to take effect as soon as the selection is made.
>
>I am trying to assess whether using the OnChange event to activate dropdown
>navigation controls is accessible or not.  According to Charles, Jaws users
>often have problems with it.  
>
>There is a work around: pressing "ALT-DownArrow" will produce the effect of
>pressing the mouse button and holding it, and allow the user to subsequently
>browse the dropdown with the arrow keys, without activating the OnChange
>event.  Freedom Scientific says their users generally know this technique.
>Is it true?  
>
>Do other assistive tools have problems with the OnChange method of
>activating dropdown navigation?
>

AG::

Users do have problems with the page changing when they didn't expect it.
This
technique has been a recurring source of trouble as reported on the WebWatch
mailing list.  I don't know how many of those reporting trouble were using
Jaws
or not, and how many had ALT-downArrow available to them and just didn't
realize they needed to use it at that point in their browse.

The view from the User Agent Accessibility Guidelines Implementation Report is
that this workaround is working around a mis-implementation of onChange by IE,
and not the basic behavior that one should expect of that event.

This would seem to indicate that this technique unsafe, because of the
widespread use of IE.  

Many users need keyboard activation access who don't need screen readers,
so it
is not a question of the assistive technology having a problem.  Keyboard
access should work without resort to assistive technology.

Please consult the User Agent Accessibility Guidelines Implementation Report. 
Find "onChange" in
<http://www.w3.org/WAI/UA/implementation/report-cr2-max-rating-list.html>ht
tp://www.w3.org/WAI/UA/implementation/report-cr2-max-rating-list.html.

Al

>-----Original Message-----
>From: Charles McCathieNevile
[<mailto:charles@w3.org%5D>mailto:charles@w3.org]
>Sent: Sunday, December 09, 2001 7:19 AM
>To: David Woolley
>Cc: w3c-wai-ig@w3.org; cturne@essex.ac.uk
>Subject: Re: Fw: [webwatch] Fw: [htmldesigners] What's the difference
>visually?...
>
>
>Yes, what happens for a mouse user is that they click on the box (equivalent
>in explorer to alt-down arrow) and then scroll up und down the list until
>they have selected the item tehy want, at which point they release.
>
>Usually there is no visual difference - the following button is often
>provided, but a javascript frees the user from having to click it.
>
>I think that the problem is not with HTML, but with browsers that force an
>onChange event to be fired when you try to select with the arrow keys,
>instead of waiting for you to move out of the combobox before registering
>whether you have changed the selected item.
>
>I don't know how many browsers have this problem, although I have noted it
>is
>a common experience for users of Jaws with IE.
>
>cheers
>
>Charles McCN
>
>
>On Sun, 9 Dec 2001, David Woolley wrote:
>
>  > > kind of comboboxes which only activate when you press some button
>  > after
>  > > them, as opposed to the ones which activate as soon as you press
>  > cursor
>  > > down.  I mean what would a mouse-user do with these different kinds of
>
>  This sounds like a case of Document Object Model (popularly called
>  JavaScript) onchange events being used.  The change wouldn't be
>  recognized until the mouse button was released.  It's not an HTML
>  problem, but a designer problem (designing only for clicking with a
>  mouse).
>
>  Plain HTML will not do this.
>
>
>-- 
>Charles McCathieNevile   
<http://www.w3.org/People/Charles>http://www.w3.org/People/Charles  phone: +61
409
>134 136
>W3C Web Accessibility Initiative    
<http://www.w3.org/WAI>http://www.w3.org/WAI    fax: +1 617
>258 5999
>Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia
>(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex,
>France)
>  
Received on Tuesday, 18 December 2001 18:06:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:13:59 GMT