W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2004

Re: CONFORMANCE - SUMMARY OF BUGS WITH RECOMMENDATIONS

From: Joe Clark <joeclark@joeclark.org>
Date: Mon, 1 Nov 2004 18:58:18 +0000 (UTC)
To: WAI-GL <w3c-wai-gl@w3.org>
Message-ID: <Pine.LNX.4.60.0411011842170.2379@aristotle.multipattern.com>

> Comment: This is part of a larger question or suggestion that we create a
> new document titled something like "WCAG 2.0 Development Notes". This
> document would parallel the structure of WCAG and provide notes as to the
> rationale of the use of specific words and terms, the placement of elements
> at various levels, etc.

Great-- yet another piece that harried and befuddled authors would need to 
plug into the puzzle. WCAG needs fewer documents, not more. The user 
testing of the WAI site proved that people could not handle the existing 
panoply of incomprehensible documents.

Plus it's grandstanding. I think it's fair to say that the guidelines are 
more important than the curious and fascinating anecdotes of how they were 
put together. I assume that Participants in Good Standing will want their 
photos on that page, too, sort of like a bad corporate site.


> 2)       include a sentence in the guidelines that says "The Working Group
> believes that provisions at all 3 levels are important or essential for some
> people.  Thus the old descriptions of "impossible to access" for Level A,
> "difficult to access" for Level AA, and "somewhat difficult" for Level AAA
> are no longer used.  Instead we define the three levels as above."

You do not have consensus that three levels are needed. The current WCAG 2 
drafts are full of insane combinations like *no Level 1* but 2 and 3, or 
the more predictable *no Level 3* but 1 and 2. I think the guidelines with 
only Level 1 and Level 3 guidelines are a particularly nice touch.

It's been explained before on this list, but since I was the one 
explaining it, Gregg, who acts like some kind of hereditary monarch 
anyway, went right ahead and ignored it. The three levels of WCAG 1 were 
too confusing. Matt May pointed out that the DRC study in the U.K. found 
no sites whatsoever that met Priority 3. Kynn Bartlett and others have 
argued that WCAG 1 AAA compliance is actually impossible due to the 
preponderance of unresolved until-user-agents clauses.

Plus, as I told you already, *you don't have enough guidelines for three 
levels*. "Core and Extended" was a good idea at the time, but for some 
godawful reason, Gregg more or less unilaterally decided that WCAG 1's 
structure was pretty good and only needed a wee bit of tweaking.

WCAG 1's structure stinks and needs to be thrown out. Clear?

> Summary: This is a series of comments mostly dealing with a request to
> explain why we have some core and extended guidelines. These comments are
> overcome by events.

False. You have no consensus that three levels are needed and an 
avalanche of facts proving they are not.

> The rest had to do with a request that we not abandon the A, AA, AAA, or 
> that we directly map our results into them in some fashion.

And since that result directly maps onto Gregg's biases, Gregg made sure 
it went straight through.


> " The Working Group believes that provisions at all 3 levels are important
> or essential for some people.

An unworkably vague and self-contradictory claim. Essentially, all 
guidelines are important; if so, then no guidelines can be more important 
than others, so you can't even have *two* priority levels, let alone 
three.

Do you even think about the implications of these things before you 
publish them?

> 3)       Recommend that we open a new bug item titled "Cumulative
> Conformance Policy, Description, and Labeling comments "and that we just
> list items that we should keep in mind which are not specific
> recommendations that have to do with any particular parts of our current
> wording. For example, from this checklist I would pull the following
> concepts.
>
> -          Make it clear what the relationship between the new levels and
> the old A, AA, AAA are [368]

Gregg's three-level bias is clear here.


> -          It is helpful to have as much agreement between the old WCAG 1
> levels and the new WCAG 2 levels [368]

Gregg confirms my impression that he believes WCAG 1 merely needs to be 
tweaked.


> -          In naming the levels (A, AA, AAA or whatever you use) think about
> cross-cultural and cross language as well as screen reader readability.

No. We're working in English, and screen-reader exception dictionaries 
aren't our problem. "Core" and "Extended" are pretty solid English words.


> -          Consider not using A,AA and AAA so that people don't
> automatically map them when they are different[368]

How about "Consider not using A, AA, and AAA because we have no basis to 
set out three conformance levels"?

> 2) Then close this bug.

Not quite yet, I'm afraid.

> Summary: Suggestion was that any scope claims in Metadata be done in URI
> identification.
>
> Comment: This appears to be a good idea, as much as we are able to. However,
> a scope claim may talk about all of the archives and that may not be a
> single URI, maybe that's a range of URIs. However, even in this case, it is
> better to use a URI than a vague descriptor such as "all archives" or
> "except for the archives" since it is not clear what that would constitute.

Right. So in my database-driven Web site, with, let's say, dozens or 
hundreds of new items added every day (e.g., Amazon user reviews), I'm 
supposed to continuously lard up my <head> element with <meta> elements 
listing every URL?

How's that gonna work?

I think it's great we're trying to make some of this information 
machine-readable, but HTML just isn't set up for that. Now, if you'd like 
authors to use only XHTML 1.1 with a certain as-yet-published module, 
please say so. And when Internet Explorer X for Windows 2020 finally 
supports it, we can have a nice party at our retirement villa.


> 1000 <[34]http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=1000>   AAA vs
> Triple-A on purpose?
>
> Bug in full:  There was some confusion as to whether the slight difference
> in naming convention between conformance claims in WCAG 2.0 vs. 1.0 was
> purposeful:
> conformance at level AAA vs. Conformance Level "Triple-A"
>
> Comment:   Change was made to make it easier for screen readers.  We have
> gone back to AAA for now

Screen-reader exception dictionaries aren't our problem. Why aren't you 
using <abbr>AAA</abbr> in the first place?

Also, won't it come as a surprise that not everyone in the People's 
Republic of WAI agrees with Dear Leader's decision that we shall have 
three conformance levels?

> (1)     . reword  L2 SC in conformance to read:
>
> "Level 2 success criteria:
>
>  1)  Increase accessibility through one or both of the following:
>
> *    further facilitating the ability of user agents to provide accessible
> content additional facilitation of user agent based accessibility

This isn't even English.


> *    recommends content and/or presentation that provides direct
> accessibility without requiring users or their user agents to do anything
> different from users without disabilities  "

Like, I dunno, *not see* or *not hear*?

-- 

     Joe Clark | joeclark@joeclark.org
     Accessibility <http://joeclark.org/access/>
     Expect criticism if you top-post
Received on Monday, 1 November 2004 18:58:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:32 GMT