RE: model for applying WCAG 2.0 to WCAG2ICT using a the concept of “objects of assessment”

Yes/no evaluation makes little sense, and speaking as someone who has been overseeing such work for years now, has no value.  Basically you get a list of compliance gaps, or no list at all of things found.  If you use a reliable test process and find no issues, you should be considered compliant unless proven otherwise.  I’ve had many folks ask me how perfect does compliant mean and I always answer that we need to see that accessibility problems are reduced to “bugs”, e.g. we developed this but made a few mistakes, but overall almost everything is compliant.  You can treat “bugs” then as you would any other bug, and remediate it on a schedule.  What we don’t want to see is whole sections with multiple accessibility issues, or absolutely no attention paid to accessibility at all.  If you have software with 50 screens, a reasonable conformance level should be no more than a couple errors total.  I’m sure that is a movable percentage for various folks, but the reality is get it as right as you can considering that software development does sometimes deliver bugs.  At one point in my previous life I proofread Braille for the National Library Service.  Braille proofreaders actually were required to fix even things that were incorrect typos in the print materials being translated.  To this day I still can’t figure out why blind people were so sensitive to minor typo errors as a group.  I’d apply the same concepts to conformance—test it until no more bugs are found, and consider it compliant until proven otherwise.  This approach does require a consistent, reliable test process however, which is definitely something needed.






From: Peter Korn [mailto:peter.korn@oracle.com]
Sent: Thursday, October 04, 2012 2:34 PM
To: Gregg Vanderheiden
Cc: public-wcag2ict-tf@w3.org Force
Subject: Re: model for applying WCAG 2.0 to WCAG2ICT using a the concept of “objects of assessment”

Gregg,

This is a really excellent document.  I need to re-read it a few more times to fully digest it.

I would really like to see much of this text adapted to be part of our final document deliverable, as it gives so much more context and understanding to readers of what we are are doing and why - far beyond the too little we gave in the first public draft of "applies directly as written... replacing "web page" with..." + occasional notes.


I remain concerned with how things would sort out in practice for our 4 unresolved SCs, particularly for software.  I want to review some of our examples<https://sites.google.com/site/wcag2ict/cross-cutting-issues-and-notes/user-interface-context/ui-context-examples> with this approach, and see how that feels (e.g. Bypass Blocks - what is "repeated content" in software that would be bypassed?  What constitutes a software title that describes the topic or purpose?  Is "Skype" or "Pidgin"  "Adobe Acrobat 9 Pro" or "NetBeans IDE 7.0" a title for software that sufficiently "describes the topic or purpose"? (of four random apps on my computer at the moment - which respectively do IM and audio and video conferencing, IM via a bunch of different protocols than Skype, read & author PDF documents, and a developer tool for creating any of a huge variety of types of programs)? ).


I also remain concerned with how the white paper treats Conformance.  Our work statement directs us to describe "the extent to which WCAG Conformance<http://www.w3.org/TR/WCAG20/#conformance> is meaningful to non-Web ICT" (emphasis added).  Yes/No evaluation software support for the individual SCs is extremely difficult.  One cannot test all of the UI permutations possible in many software situations.  Yes, this situation isn't unique to non-web software; it also arises in web apps.  AND we are seeing EvalTF struggling with that very question, and heading now in the direction of a report of what was found rather than a simple Yes/No evaluation.

I continue to believe it is not only appropriate - but also within our remit as stated in our Work Statement - to go beyond finding word substitutions for determining "what WCAG 2.0 Conformance means in the context of non-Web ICT".


I will try to provide more detailed feedback on the various sections of your otherwise excellent white paper in the survey for tomorrow.


Peter


On 10/3/2012 8:14 AM, Gregg Vanderheiden wrote:
REPOSTING THIS WITH A LINK RATHER THAN AN ATTACHMENT (since the attachment was stripped off for some)


Hi All,

I finished my writeup evaluating all of the SC,  looking for consistency,  and proposing an approach to resolving the final 4 plus the conformance requirements based on the concept of "object of assessment".

It is attached.

the abstract is below

nite.

G

Use this link to download the document : http://goo.gl/Shf8d


This whitepaper is provided to help in the discussion of how to apply WCAG 2.0 to non-web content and software in a manner equivalent the way WCAG 2.0 was designed to be applied to web content.  It starts with a discussion of a concept of “objects of assessment” and then shows how this can lead to a better understanding both of WCAG 2.0, and how to apply it to non-web ICT.   It shows that such an approach leads to both an agreement with the 34 provisions the WCAG2ICT task force has already reached consensus on.  But it shows how the WCAG2ICT decisions can be explained by a couple simple rules rather than as 34 individual decisions.
It also leads to a resolution for the final 4 provisions as well as the WCAG Conformance requirements.  This resolution comes from a better understanding of what we are assessing on 3 of the 4 and how they are different from the others (leading to our problem in resolving them).  A resolution to the 4th is also proposed.  The paper concludes with some observations and a full summary (listing each provision) and showing what the solutions would look like in place.
(As a bonus the summary also shows what the task force's suggested global replacement of  electronic documents with “non-embedded content” would look like – thus closing one of our action items).

(see page 12 for a 1 page summary of the recommendations, then read paper for rationale)


Gregg
--------------------------------------------------------
Gregg Vanderheiden Ph.D.
Director Trace R&D Center
Professor Industrial & Systems Engineering
and Biomedical Engineering
University of Wisconsin-Madison

Technical Director - Cloud4all Project - http://Cloud4all.info
Co-Director, Raising the Floor - International
and the Global Public Inclusive Infrastructure Project
http://Raisingthefloor.org   ---   http://GPII.net









--
[Oracle]<http://www.oracle.com>
Peter Korn | Accessibility Principal
Phone: +1 650 5069522<tel:+1%20650%205069522>
500 Oracle Parkway | Redwood City, CA 94065
[Green            Oracle]<http://www.oracle.com/commitment>Oracle is committed to developing practices and products that help protect the environment