RE: Granularity of conformance claims

Allow me to quote from an email sent by contractor defending his work after 
I critiqued his horribly inaccessible site.  Mind you, this vendor 
understands that accessibility is an issue.  My main point in posting this 
here is to provide hearsay evidence that vendors will try and use WCAG as a 
"Chinese menu" -- picking and choosing among what they want.  And this is 
with the current WCAG.  Charles' observations are quite on the mark.  We 
don't dare weaken the A/AA/AAA levels!

The names will remain anonymous to protect the guilty...

> There is nothing in these guidelines which prevent having an alternate
> site.  In fact, it is encouraged.  The W3C/WAI Web Content Accessibility
> Guidelines themselves explicitly tell developers to create an alternate
> page when the current page doesn't "transform" well.  As it stands,
> our site meets a large number of the priority 1, 2, and 3 checkpoints
> outlined in the guidelines.  Also, the graphics version is needed to
> satisfy the requirements regarding people with learning disabilities to
> help them associate ideas.  The text only version designed for blind 
users
> might violate their rights.

I have written back saying that "required as a last resort" is more 
accurate than "encouraged" and that it does not matter how many Priority 2 
and 3 checkpoints are addressed if even ONE Priority 1 checkpoint is 
missed, the site will remain inaccessible.

Actually I went on in detail about a number of points.  This particular 
work in progress is in real trouble because the vendor is generating the 
"text-only parallel version" (yes, I have emphasized that this approach is 
misguided) from a Java applet!  My main goal has been getting them to 
understand that EVERYTHING they do with Java, by definition -- and 
including the text-only pages, is not accessible!  [heavy sigh...]


On Wednesday, July 21, 1999 9:37 PM, Charles McCathieNevile 
[SMTP:charles@w3.org] wrote:
> It seems to me that there is nothing to stop developers claiming partial
> conformance in any way they want, but the actual conformance section of 
the
> document says effectively that the group thinks there are three different
> types of conformance, which signal three different levels of 
accessibility,
> and do not offer any official recognition for any other type of claim.
>
> Effectively this allows the developers of web content to say all manner 
of
> things about what they have done to improve the accessibility of their 
web
> content, and to say that they claim to have attained (or exceeded) a
> particular level of accessibility according to the WCAG 1.0 out of a 
small
> number of possible levels. Having a very small number of possible levels,
> with simple natural descriptions of what they mean, allows for a much 
easier
> comparison and requirement setting than having to weigh the relative 
value of
> each checkpoint against each other.
>
> Which seems to me a good argument for leaving the status quo alone.
>
> Charles McCN

Received on Thursday, 22 July 1999 10:42:44 UTC