RE: Question re: WCAG2.0, Requirement 3.3.2

Jonathan,
Yes, I do agree with you that a screen reader or magnifier having to guess 
is not sufficient.  Sorry, I was just trying to explain that screen 
readers often do try to guess based on proximity when they can't 
programmatically determine the label, not that it is a sufficient means of 
compliance.  Sighted folks try doing the same thing too, but are also 
sometimes confused when the label text is inconsistently placed (e.g., 
below or above the control). 

I sure would welcome some more support from the browsers to better 
highlight the focus on the control and label, especially after the author 
goes to the effort of correctly marking up the form.    Style sheets can 
help here too, but again it just more work for the developer when I think 
it more efficient and effective when implemented in the browser, allowing 
the user to have more control, not the author of the site.

Regards,
Phill Jenkins, 





From:   Jonathan Avila <jon.avila@ssbbartgroup.com>
To:     w3c-wai-ig@w3.org
Date:   06/16/2011 10:01 AM
Subject:        RE: Question re: WCAG2.0, Requirement 3.3.2
Sent by:        w3c-wai-ig-request@w3.org



[Phil wrote]
Ø  Screen readers today can determine which is the correct label by either 
guessing or by the author correctly marking up the label so that it is 
programmatically determinable (recommended), 
 
I have to disagree with this.  A screen reader guessing at a label is not 
sufficient and often not correct – I see this fail almost every day.
 
Explicit labels, titles, ARIA or some other assistive technology supported 
definitive way is required for WCAG 2 conformance to 1.3.1 or 4.1.2.  
Implicit labeling or guessing is not sufficient in my review of WCAG 2.
 
Jonathan
 
From: w3c-wai-ig-request@w3.org [mailto:w3c-wai-ig-request@w3.org] On 
Behalf Of Phill Jenkins
Sent: Thursday, June 16, 2011 9:51 AM
To: w3c-wai-ig@w3.org
Subject: Re: Question re: WCAG2.0, Requirement 3.3.2
 
I think we also need to be asking what is the responsibility of the 
browser and AT, including the magnifiers, not just the screen readers. 
 and also how the content is re-flowed for smaller mobile displays.  And 
internationalization (right to left languages) needs to be considered too. 


Screen readers today can determine which is the correct label by either 
guessing or by the author correctly marking up the label so that it is 
programmatically determinable (recommended), regardless of where the label 
is visually.  So, a screen magnifier could also determine and speak (if it 
has that feature) and correctly position the view pane to display the 
label and the checkbox or radio button. Reflow and transformation and 
translation tools can use this too.   

I agree that best practices should usually (always?) but the label in a 
consistent place. I'm just trying to also add to the discussion the 
responsibility of the users agent/browser and any assistive technology and 
not try (which I believe is impossible) to solve all the considerations 
with just author markup and CSS. 

Regards,
Phill Jenkins, 
IBM Research - Human Ability & Accessibility Center
http://www.ibm.com/able

http://www.facebook.com/IBMAccessibility

http://twitter.com/IBMAccess

http://www.linkedin.com/in/philljenkins