RE: Validity

I think it is important to keep the focus on the generalized case, not
the specific technology issue. 

 

What happens when valid code creates issues for AT? 

 

I think it is fine for this group to point fingers at the AT community,
but it won't serve the end users very well. Even when everyone works
really hard to work together, there are only specific opportunities for
changes to be made. This issue is amplified in smaller AT companies with
limited resources to make changes. Assuming that changes can and will be
made, in what seems like a quick turn around from an AT vendor
standpoint, where does that leave the end user? 

 

This does not relieve the pressure on tool makers or the AT vendors. As
Roberto is eager to point up, this issue is covered in the other working
groups. 

 

The question for WCAG is what should authors do in these cases. Simply
pointing to the AT vendors or tool makers does not fundamentally help
the end user. Authors should be allowed to employ techniques when valid
code will not solve the issue. 

Cheers,
Bob


------------------------------------------------------------------------
-
bob regan | macromedia | 415.832.5305

________________________________

From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On
Behalf Of Neil Soiffer
Sent: Friday, November 04, 2005 10:32 AM
To: w3c-wai-gl@w3.org
Subject: Re: Validity

 

I'm surprised that the arguments for allowing invalid documents seem to
revolve around a technology.  My understanding is that the guidelines
for WCAG 2.0 are suppose to be technology-independent.  Is it the case
that other guidelines were accepted or rejected or placed at a certain
level because some technology did not support the guideline?  It is
certainly not hard to imagine that non-valid documents in other
technologies (including non-valid flash) prevent proper functioning of
the document in some UA, which affects all users of the document.

 

The HTML or XHTML techniques document is the proper place to address
technology-specific issues such as using an embed tag.  As Gez mentions,
perhaps some weasel words that allow for non-valid documents to be used
until UAs conform to a spec would be a compromise.  Personally, I don't
think that is necessary -- I see the "embed" argument as a baseline
issue ("our baseline is HTML browsers that allow the use of <embed>") ,
but obviously, others see differently.

 

I also want to add my name to the list of people who have asked for the
criteria for deciding what goes in level 1 or level 2.  How can there be
rational debate on the subject without a policy on this?  I have a vague
memory this was discussed before and such a criteria for what goes in
levels 1, 2, and 3 were agreed upon, but I don't remember what they
were.

 

Neil Soiffer
Senior Scientist
Design Science, Inc.
neils@dessci.com <mailto:neils@dessci.com> 
www.dessci.com <http://www.dessci.com> 
~ Makers of Equation Editor, MathType, MathPlayer and MathFlow ~

 

 

> ----- Original Message -----
> From: "Gez Lemon" <gez.lemon@gmail.com <mailto:gez.lemon@gmail.com> >
> To: "Bob Regan" <bregan@macromedia.com <mailto:bregan@macromedia.com>
>
> Cc: "WCAG WG mailing list" <w3c-wai-gl@w3.org
<mailto:w3c-wai-gl@w3.org> >
> Sent: Friday, November 04, 2005 9:13 AM
> Subject: Re: Validity
>
>
>
> Hi Bob,
>
> On 04/11/05, Bob Regan <bregan@macromedia.com
<mailto:bregan@macromedia.com> > wrote:
> > I believe there needs to be a mechanism whereby an author may deviat
> > from valid code for reasons specific to accessibility for the audien
> > defined in their own baseline.
>
> That would be a great start to working towards some kind of
> compromise, which I believe was the intent of guideline 4.1 before the
> meeting in Brussels. Would validity be considered acceptable if there
> was an exception for instances where non-conformance improved
> accessibility without degradation to other users?
>
> Best regards,
>
> Gez

 

Neil Soiffer
Senior Scientist
Design Science, Inc.
neils@dessci.com
www.dessci.com
~ Makers of Equation Editor, MathType, MathPlayer and MathFlow ~