W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2006

CONFORMANCE issues & resolutions -- based on new conformance section.

From: Gregg Vanderheiden <gv@trace.wisc.edu>
Date: Wed, 8 Mar 2006 07:31:56 -0600
To: <w3c-wai-gl@w3.org>
Message-ID: <000501c642b4$c0cf5ec0$ee8cfea9@NC6000BAK>
Here are the open items for Conformance - along with suggested closing
comments based on new conformance draft

 

Only two  open items at this point

 

Gregg

 

 

 

 

Mon Mar 6 23:53:40 CST 2006

  _____  

47 issues found. 



ID      Summary 


476 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=476>     suggested
improvements to conformance section 


-          Make sure it is clear that test suites don't guarantee
conformance or lack of it

CLOSE with comment - added comment that techniques and their tests are not
only way to conform. 


1324 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1324>  Remove
Baseline Technology , or limit Baseline Technology... 


-"Don't use baselines" - need to be sure to have good BASELINE doc

CLOSE with comment -  Baselines are needed.  Have created an improved
baseline document


1328 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1328>  Use UAAG1
as a Springboard 


- advice when CHOOSING A REASONABLE BASELINE

CLOSE with comment -  Put a section in baseline doc on how to choose a
baseline in order to provide good content for future UAAG work as specified


1361 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1361>  Level of
conformance seem very complicated 

- - Level of conformance being claimed
Is this section really necessary? Is it not too detailed? It does not seem
very clear to me - again very complicated.  'authored unit is defined as
"Some set of material created as a single entity by an author." What is a
single entity in web terms? A paragraph? A page? A set of pages? Should this
set of material be restricted to one author?
CLOSE with comment -    We have simplified this text and added definitions.

1362 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1362>  Scope of
conformance claims redundant? 

Scoping of Conformance Claims
Has this not already been covered in 'Conformance claims'? Should the set of
URIs should not cover this?
CLOSE with comment -    Yes - this has been removed as redundant.  

1437 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1437>  Supports
repair techniques to accommodate current user ag... 

It's a problem that no single user agent meets all of the UAAG 1.0 priority
1 checkpoints but I believe in a set of "repair techniques" that, as you
say, could be used by content authors who would like to create content that
not only meets WCAG 2.0, but that also makes up for the shortfall in current
user agents
CLOSE with comment -    Thanks for comment.   That is how we are proceeding.
User agent support should develop over time and help out here as well. 

1444 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1444>  Don't use
HTTP header for conformance information 

Wim Vanderbauwhede says:
The Section on Conformance Claims of the November draft of WCAG  2.0
contains the following Editorial Note:
"A question has been raised as to whether the information required in items
1-3 above should all be transmitted in the HTTP header or in some other
way."
The HTTP protocol is used for the transfer of a wide variety of content
types. It would therefore not be appropriate to include HTML-specific fields
in the HTTP header. Furthermore, this would require an extension of the HTTP
protocol specification. It would make more sense to add a field to the HTML
header.
CLOSE with comment -    Comment Removed - will not be requiring information
in HTTP header

 


1555 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1555>  Responses
to "how many levels of conformance" issue/summary 


All discussion on whether to have  2 or 3 levels of conformance. 

CLOSE with comment -    The working group has chosen 3 levels to provide the
best options for different users of the guidelines. 


1556 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1556>  Responses
to "baseline" issue/summary 


Comments of Support for Baseline

CLOSE with comment -   Are proceeding with baseline concept.  See improved
writeup.


1560 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1560>  Editorial
Notes in Conformance Section 


1) Use HTTP header or metadata to make claims

 

12)  <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1573> it is
difficult to imagine how GL 3.1 could ever be assured for community
contributed content even if the tool to create the community contributed
content conforms to ATAG. It would be better to allow delivery units to
exclude such authored units.

CLOSE with comment -   Claims can be made in many ways. We are not requiring
claims or the format for them.  Just the content. 

We have also removed the community contribution language. 

 

1573 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1573>  Reliance on
User Agent support is risky; provide repair t... 


Don't rely on user agents to be good.  Include repair techniques. 

CLOSE with comment -   Repair techniques will be included as advisory
techniques as they are identified. .

 


1574 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1574>  "Target
Users" should be better defined. 


Use 3 levels.   Define target users

CLOSE with comment -    The working group has chosen 3 levels to provide the
best options for different users of the guidelines.

Target users are discussed in the overview documents. 


1575 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1575>  Conformance
scheme too difficult to understand 


Make your conformance scheme and SC simpler

CLOSE with comment -   This has been simplified - as has the writeup .

 


1590 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1590>  clarify
when it's acceptable for content's default presen... 


It should be ok if default doesnt meet if you can adjust it to meet.

CLOSE with - SC have ability to turn off or adjust built into them rather
than conformance. 


1598 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1598>  3 levels of
success criteria 


Explain that not all are 3 levels

CLOSE with comment -   language added to say this specifically .

 


1623 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1623>  3 Levels of
conformance


Comment - people will think these 3 levels are same as WCAG 1 

CLOSE with comment -   Language has been added to clarify differences.  They
are similar in many respects - different in others.  .

 


1630 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1630>  Intranet
example of 4.2 needs rewording 


Transferred to 4.2


1702 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1702>  Date when
conformance claims were made should be required 


Include DATE in conformance claim

CLOSE with comment -   Date is now required .

 


1723 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1723>  Baseline
section is convoluted and confusing 


Need Baseline WhitePaper

Need guidance on how to make good baseline

CLOSE with comment -   Baseline section cleaned up, organized, and
structured.  A whitepaper on baseline has been drafted. .

 


1724 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1724>  Conformance
claims should include baseline definition 


Need baseline in conformance claim

CLOSE with - Baseline is in conformance claim now. 


1725 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1725>  Comments
about use of UAAG as baseline 


How do you know if user agent is conformant?

CLOSE with - You don't know anything about the user agent of a person.
Therefore these standards make certain assumptions about user agents - but
most of those assumptions will be made by those who set the baseline
technologies. 


1726 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1726>  Indicate
that baseline information is not yet available i... 


Intro says techniques include baseline info - but it isn't there

CLOSE with comment -   Baseline information is there - but not explicit.
Language in intro to Understanding WCAG 2.0 will clarify this .

 


1727 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1727>  WCAG
guidelines should not specify what should or should ... 

we don't think WAI should appear to be dictating what public policy makers
should or shouldn't include in their work. Make the best recommendations you
can and let others decide how to use it. You are preparing guidelines... not
policy.

CLOSE with comment -   WCAG does not specify baseline or policy.    The
guidelines are structured to maximize ability for policy makers to use it
effectively in different ways. .

 

1728 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1728>  Scoping
allows sites to scope out all multimedia 

Scoping is a very scary concept. With scoping it is possible to scope out
all the multimedia (e.g. training packages) that are critical to users and
claim full conformance with the rest of the site even though the rest of the
site is not much use without the multimedia. Scoping should be handled very
carefully.

CLOSE with comment -    This is specifically excluded from scoping.  But
sections of a site could be scoped out if policy makers so choose to allow
that. 

 

1758 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1758>  Define
Level 1 as "minimum necessary and sufficient condi... 


Comment about how to define level 1 and 2 so WG has less discretion.  

CLOSE - level descriptions are descriptive - not prescriptive. 


1759 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1759>  Clearer
statement needed about baselines 


Lots of different comments and advice about baselines  and BASELINE WRITEUP-


CLOSE with comment -    See new draft baseline description   .

 


1760 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1760>  AAA
conformance 


Group the items for AAA conformance 

CLOSE with comment -    Multiple attempts to group the level 3 items were
made but conformance in this fashion did not work out. 


1761 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1761>  Issues with
"delivery unit" 


Is delivery unit the right term?   Something less confusing? 

CLOSE with comment -    We agree.  Have changed it to WebPage


1762 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1762>  "Should be
testable" vs "are testable"

Conformance: "The working group believes that all success criteria should be

testable".
 CHANGE  "SHOULD BE' to ARE TESTABLE 
CLOSE with comment -     thanks - done. 

1763 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1763>  Software
for download should be covered by guidelines 


Don't exclude downloaded software

CLOSE with comment -    This clause has been removed. 


1764 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1764>  Conformance
claims for content aggregators 


Advice on how to handle Aggregated Content 

CLOSE with comment -    See new language which allows use of original author
claim or aggregator claim. 


1778 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1778>  Are fully
conformant user agents required? 


Wcag is useless if it requires user agents that conform.

CLOSE with comment -    WCAG does not require fully conformant User Agents 


1787 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1787>  Don't base
conformance on use of ATAG tools 


Preposterous to do this since ATAG doesn't force conformance

CLOSE with comment -    This has been removed. 


1799 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1799>  Meaning of
conformance levels 


Complaints about how we define the three levels

CLOSE with comment -    Levels are chosen to allow best use of the document
- not to be same as 1.0.  Definition reflects our current understanding
which is more complicated unfortunately - but that is  because our 1.0
approach was simple but inaccurate and based on a false impression.
/assumption. 


1811 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1811>  Determining
appropriate baseline may be too burdensome fo... 


Baselines may make WCAG useless - and 508 will become the standard

CLOSE with comment -    Baselines are the only way to deal with the "until
user agent' issue that plagues 1.0 and limits its useful life.   We could
find no other way to address the problem.


1823 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1823>  Note that
some GL have no level 3 SC 

Note: Some guidelines do not contain level 1 success criteria, and others do
not contain level 2 success 
criteria.
>>Not all contain level 3 success criteria either - I would add this to the
sentence
CLOSE with comment -    Section has been reworded to reflect this.

1840 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1840>  Conformance
levels should not depend on how accessibility is achieved 

The document might explain (not part of SC definition) that accessibility
can be attained through:
  a. markup, scripting, or other technologies that interact with or enable
access through user agents, including assistive technologies

  b. the design of the content and presentation

CLOSE with comment -    very good point.  This is made clear in the "how to
meet" documents now. 


1855 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1855>  baseline
and conformance not well explained 


Many terms are not clearly defined.   Baseline, terms form Dev Indep

CLOSE with comment -    This has been rewritten to be clearer


1860 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1860>  conformance
levels aren't based on improvements to access... 

 Feels that SC are at levels based on machine testability rather than user
need. 
CLOSE with comment -    They are not based on machine testability.  They are
based on what is needed at most basic level and what is most important as
well as what can be applied to all web sites. 

1865 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1865>  interaction
of content negotiation and baseline 

Regarding the statement, "If multiple representations can be retrieved from
a URI through content negotiation, then the conformance claim would be for
the delivery unit that is returned when no negotiation is conducted ...." it
would seem that the results of the content negotiation might be dependent
upon the baseline. If I understand what is meant by content negotiation, it
could include, for example, browser identification. So if the baseline on an
intranet, say, used Safari, and the delivery unit was set up to provide
highly conformant content when the browser identified itself as Safari and a
lesser conformance level when it did not, then the optimal level of
conformance would be achieved on baselined workstations. But this process
would appear to involve content negotiation. Clearly you would want to make
a conformance claim based on your 99% of workstations that were baselined,
rather than on a few oddball workstations attached to the network to meet
special requirements.
 
CLOSE with comment -     That isn't quite how it works.  Negotiation is on
capabilities.   If the 99% can negotiate they will get just what they need.
If content negotiation doesn't work - then the accessible form has to be
there or it won't be reachable. 

1868 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1868>  A, Double-A
and Triple-A, not A, AA, and AAA 

4 examples should read: Level A, Double-A or Triple-A instead of: Level A,
AA or AAA and should be consistent throughout
 
CLOSE with comment -    Fixed. 

1869 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1869>  target
audience requirements that conflict with univeral ... 


DON'T include target audience in conformance claim 

CLOSE with comment -    This has been changed to only refer to language etc.



1870 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1870>  conformance
claim examples list SC level instead of confo... 

In this section and <http://www.w3.org/WAI/intro/wcag20-baseline>, the
success criterion are stated (E.g. ..conforms to W3C's WCAG 2.0, Conformance
Level 1..) instead of the conformance requirement (E.g. Level A, Double-A or
Triple-A).
CLOSE with comment -     thanks .  Fixed now.

1871 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1871>  example of
scoped conformance claim 

The section headed "Conformance claims" handily gives three text examples of
such claims. It would be 
useful to see at least one example of a "scoped" conformance claim.
CLOSE with comment -      DONE

1872 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1872>  What if SC
are not applicable? 


State EXPLICITLY that you pass if not relevant - (e.g  1.2 if you have no
multimedia)

CLOSE with comment -      DONE

 

 

UNRESOLVED

 


1290 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1290>  Cumulative
Conformance Policy, Description, and Labeling ... 


- Big long list - all done except SEPARATE LOGO FOR EACH CONF LEVEL A,AA,AAA

 


1786 <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1786>  Suggest
using EARL for conformance claims 

conformance section should refer to the possibility of using EARL to
provide machine-readable conformance claims.
 


Gregg

------------------------

Gregg C Vanderheiden Ph.D. 
Professor - Depts of Ind. Engr. & BioMed Engr.
Director - Trace R & D Center 
University of Wisconsin-Madison 
< <http://trace.wisc.edu/> http://trace.wisc.edu/> FAX 608/262-8848  
For a list of our list discussions http://trace.wisc.edu/lists/

The Player for my DSS sound file is at http://tinyurl.com/dho6b 

 <http://trace.wisc.edu:8080/mailman/listinfo/>  

 

 
Received on Wednesday, 8 March 2006 13:32:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:45 GMT