Re: Failure technique: Implementing inappropriate technique is a failure

Loretta et al,
Maybe a general non-specific technique  may be open-ended and  non-testable but every item listed in my  last email is testable. 
Justification: Failing to comply with H2 also gives a less than optimal experience for AT-users. Yet H2 has been included as a technique.
If such duplication (refer H2) is regarded as "a clear violation of the requirements described in a success criterion", I contend that  all the cases listed below involves extra "accessibility" markup that duplicates content and is a clear violation of a level-A SC. As a AT-user I am adversely impacted by over-use / improper application of accessibility techniques. A developer has to fix the issues only once. An AT-user  is subjected to the effects every time he visits the page and needs to suffer through it.

And these are all real-world cases I have encountered. I have pushed developers and project managers to get them fixed. Documented techniques will empower me. These are not 'usability' or personal preferences but real accessibility issues. Believe me.   
Well than each of the following can be included as: 
- a failure of a technique and hence of an SC, or
- an individual technique (like h2), or 
- a note  within an existing technique
(Probably the least desirable option is to include it as a note within an existing technique because it will be lost in the text).

I suggest the following  specific failure techniques for consideration:
1. About table caption and summary:
i. Failure technique: Table caption duplicates  / repeats the table’s title marked up with h<n> tag That is placed above the table
ii. Failure technique: Table summary attribute duplicates / repeats the table’s caption or  title marked up as a heading 
2. About INPUT buttons:
Failure technique: Setting an alt or title attribute for  <input type = "button", "submit", or "reset">
3. About FIELDSET:
Failure technique: Improper use of FIELDSET:
This should cover:
- Use of a FIELDSET with a LEGEND for the form’s title (say, “Registration Form”) as a wrapper for an entire form.
- Use of a FIELDSET (without a LEGEND as a wrapper for an entire form simply for the border that FIELDSET creates (Note: This is mentioned in  a sort of indirect way in H71, where it says “LEGEND should be the first element in a FIELDSET)

4. About longdesc
Failure technique:Text alternative for an image provided by the longdesc attribute is invalid
This should cover:
- The text alternative rendered via the longdesc merely a duplication of the value of the alt attribute; i.e. -it contains no additional information  
- The value of the longdesc attribute is  a text string and not an htm file 
- The value of the longdesc points to a non-existent htm file 

5. About skip nav link
Failure technique: Placing a ‘skip navigation’ link  above a set of footer links at the end of a page
(And the target of the link is the bottom of the page just after the footer links!) 

6. About form labels:
Failure technique: Inserting an off-screen label that duplicates a visible text label that is not marked up as a label 

Thanks,

Sailesh Panchang (MS, ASQ-CSQE)

Tel 571-344-1765
Accessibility Services Lead at www.deque.com

--- On Thu, 7/29/10, Loretta Guarino Reid <lorettaguarino@google.com> wrote:

From: Loretta Guarino Reid <lorettaguarino@google.com>
Subject: Re: Failure technique: Implementing inappropriate technique is a  failure
To: "Sailesh Panchang" <spanchang02@yahoo.com>
Cc: public-comments-wcag20@w3.org
Date: Thursday, July 29, 2010, 8:48 PM

On Tue, Jun 15, 2010 at 7:41 PM, Sailesh Panchang <spanchang02@yahoo.com> wrote:
> Hello WCAG-WG,
>
> Time and again I come across situations where an inappropriate technique has been employed.Often these lead to duplication or unnecessary verbosity. This in itself creates a problem  for real users who depend on AT and accessibility  for their daily Web access needs.
> Developers are reluctant to correct these saying it creates more work and question which technique prohibits over-usage of techniques. They tend to brush aside such issues as usability issues.
> Can the WG take a stand on this and maybe list this issue as a failure?
> Consider:
> i. Use of summary attribute and / or  off-screen caption for a simple data table that repeat the title (heading) of the table that is marked up with an h-tag just above the table.
> ii. Use of a FIELDSET with a LEGEND for the form’s title (say, “Registration Form”) as a wrapper for an entire form.
> iii. Use of a FIELDSET (without a LEGEND as a wrapper for an entire form simply for the border that FIELDSET creates
> iv. Use of a title attribute say for INPUT type=submit even when it has a clear value attribute. H91 says ‘value’ is sufficient. Rendering of title is user-agent and AT-dependent and creates a mismatch between  what sighted user sees and what is exposed to AT in some browsers.
> v. Use of longdesc for images that  repeats the alt-value or is broken.
>
> Thanks,
> Sailesh Panchang
> Centreville VA
> Tel 571-344-1765
>
================================
Response from the Working Group
================================

Thank you for your comments. While we agree that many of the failures
you have suggested will result in a less than ideal experience for AT
users, failures are common authoring practices that would result in a
clear violation of the requirements described in a success criterion.

We feel that a general failure about using techniques inappropriately
would not be testable and that many of the suggestions you've made
here may fit in better within the context of education, outreach and
training materials.

However, if you have additional examples where the inappropriate use
of a technique presents a significant barrier to accessibility, please
feel free to submit them.

If you have specific suggestions for how some of this advice might be
incorporated in the context of existing techniques (ex. as a note,
example or as an advisory technique), please let us know.


Loretta Guarino Reid, WCAG WG Co-Chair
Gregg Vanderheiden, WCAG WG Co-Chair
Michael Cooper, WCAG WG Staff Contact


On behalf of the WCAG Working Group

Received on Friday, 30 July 2010 17:01:47 UTC