[TECHS] Review of Test #63 - propose rejecting this test

As part of the Techniques Task force I have reviewed Test #63: input 
element, type of "text", must have default text. [1]

I propose rejecting this test since default text is no longer necessary 
for accessibility of input elements of type text. As long as the input 
element is associated with a label there should be no user agent issues. 
In fact, the HTML technique for this, Place-holding characters in empty 
controls [2] has been marked as deprecated.  Since all common user agents 
correctly handle input elements with no default value, this technique is 
not required for WCAG conformance. 

Other issues:
Currently the test is associated with Guideline 3.1, Ensure that the 
meaning of content can be determined [3] and no success criteria are 
listed.  It seems to fall under Level 3 #6: There is a statement 
associated with the content asserting that the Strategies for Reducing the 
Complexity of Content (the following list) were considered. Specifically 
under the section heading "Instructions and operable content ". 

The written procedures for this test are incorrect. The procedures outline 
providing alt text for input fields of type image which is not the focus 
of this test.  I am not proposing corrections for this test since I am 
proposing rejecting it. 


[1] http://www.w3.org/WAI/GL/WCAG20/tests/test63.html
[2] http://www.w3.org/TR/WCAG20-HTML-TECHS/#emptycontrol
[3] http://www.w3.org/TR/WCAG20/#meaning


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