Re: Accessibility Procedure Comment

Laura Carlson wrote:
> 
> The procedure proposal was made in good faith to ensure that
> accessibility related issues within HTML5 are dealt with in an
> efficient and constructive manner utilizing the relevant resources and
> expertise available. We are not trying to change the rules and have
> proposed nothing to that effect. We have simply proposed a procedure
> to make the rules clearer.

I must be missing something that is extremely obvious to the people who 
are supporting this proposal.  You are not changing the rules, but are 
proposing a procedure (but not a rule?) that somehow would make things 
clearer.

Can you tell me exactly, and with precision, what this proposal, if 
adopted, would change in how we approach the summary attribute and 
making canvas accessible?

 From my perspective, on the summary attribute, Ian is the only one who 
has put forward a concrete and complete proposal, and that proposal:

  1) Approaches the subject with a goal of making HTML 5 the best
     solution for everybody, including people with disabilities, and
     including people without disabilities.
  2) Was based on the concrete use cases provided, and addressed the
     any and all accessibility functional requirements that were
     clearly identified.
  3) From my perspective #3 is largely redundant with #2 and #4, but
     I do believe the the use cases provided were listened to and
     that clarification was requested.
  4) It seems clear enough that the PFWG doesn't feel that what was
     provided was adequate, but it is equally clear to me that Ian has
     asked for input and has not received it.

In short, I don't see how the proposal fixes anything.  And I believe 
that the reason why that is the case is because the proposal assumes 
that there is only one "right" solution, and that the "right" solution 
is the one that we should be adopted.  I disagree on both counts.

This is not a physics problem, it is an engineering problem.  Worse: it 
is a social engineering problem.  In such problems, there often are 
multiple solutions, or no solutions.  And when there are no solutions, 
one must decide which constraint to violate and how to compensate for 
this violation.  Once you go down this path, you again find that there 
are multiple solutions and the selection criteria between these 
solutions is necessarily subjective.

Even if there was only one right solution, it may not be one that people 
are willing to follow.  I'll use shorthand for the next statement: 
suffice it to say that charset=iso-8859-1 and RFC 3023 are both works of 
fiction at this point in time.  I (or others) can explain what I mean, 
but the message is that a right solution that nobody follows is no solution.

The question we need to address is not whether the explicit marking of 
summary as obsolete is the right technical solution, but is it a work of 
fiction; i.e., is it advice that people are willing to follow?  If not, 
it may be both right and silly.  To find a solution that isn't silly, 
what we need at this point isn't more rules or more procedures or more 
use cases.  No, what we need is a simple, common sense, down-to-earth, 
concrete proposal that people can rally behind.

I could do a similar level of analysis for canvas accessibility, but 
this email is already way too long.  Suffice it say that I don't believe 
there is anybody who believes that what currently is spec'ed is ideal or 
meets all requirements.  It simply is the best concrete proposal that 
anybody has put forward to date.  And despite setting a rather low bar, 
to date nobody has provided any better ideas.  And that is not due to a 
lack of having a shared goal.  And not due to a lack of asking for 
advice from the PFWG.  And not due to a lack of interest in asking for 
clarifications.  And not due to a lack of asking what functionality is 
needed.

- Sam Ruby

Received on Tuesday, 28 July 2009 13:36:47 UTC