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

Re: Scripting Techniques first draft

From: <Becky_Gibson@notesdev.ibm.com>
Date: Tue, 14 Sep 2004 16:28:35 -0400
To: w3c-wai-gl@w3.org
Message-ID: <OF95203DD0.9F090DED-ON85256F0F.00663B65-85256F0F.00706E3D@notesdev.ibm.com>
Hi, Matt, 
great start on the Scripting techniques.  I do have a couple of questions, 

I'm confused about the scripting techniques and whether or not a page that 
uses them still needs to work with scripting turned off?  My impression 
was that if a developer was using scripting techniques they were fairly 
certain that scripting would be available in the browser or AT.  Perhaps 
this is my misunderstanding?  I do note that you have indicated that the 
working group is looking into this.    The forms  examples refer to 
supporting non-script capable browsers as a reason for not using a 
technique.   I agree that if an action can be done without JavaScript, it 
is best not to rely on it, but then why bother with the JavaScript version 
at all? Often the reason for JavaScript is to do some calculation or 
function that can not be achieved without it.  Take the window.open 
example, one reason for using JavaScript is so you can add parameters from 
the current page or form to the url being opened. 

When I validate a link example using onactivate, I get that it is not 
valid XHTML 1.0 transitional.  We should make sure that all of our 
techniques documents refer to the specification we are using in the 

 Also, when I try to create an onactivate example: 
<a href="somelocalpage.html" onactivate="window.open('somelocalpage.html', 
'some title');" >Open somelocalpage</a>. 
Firefox just executes the href and opens the window in place.  IE will 
open the link in a new window but it does so as soon as I tab to the link. 
This needs further investigation.
I think that most browsers now support onclick for a link for keyboard 
accessibility so Technique 2.4 may need rewording.  I have seen issues 
with Mozilla calling a function twice if there is both an onclick and an 
onkeypress.  I'll double check and post an example if appropriate. 

Regarding onfocus - I think it can be used wisely.  I submitted a 
JavaScript /CSS technique for calling attention to an error message when 
client side validation of a form fails (
 I use the object.focus() method to specifically set focus to an error 
message so the user knows what form field to fix.  The screen readers pick 
up on the focus change and speak the error message to the user, the next 
tab press takes the user to the field that needs to be updated.   It works 
in WindowEyes, JAWS, and HPR but it does, however, rely on JavaScript and 
CSS being available. 


Becky Gibson
Web Accessibility Architect
IBM Emerging Internet Technologies
5 Technology Park Drive
Westford, MA 01886
Voice: 978 399-6101; t/l 333-6101
Email: gibsonb@us.ibm.com
Received on Tuesday, 14 September 2004 20:28:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:50 UTC