Re: Proposed Text for Section 1.3 (was Re: Meeting tomorrow)

Tools are selected by people who want to do a review. I had hoped we would
have some more input from tool developers, to give us a broader range of
reviews. I have a coule of reviews in process at the moment - looking at
different types of tool including image editors, a couple of SMIL tools, and
Word processor type tools. Actually I was looking for things that showed what
can be done (and how) - I don't think there is much value in a review of a
tool that doesn't meet any checkpoints at all.

I think this information is important to provide some further idea on how to
check conformance - the issue that was raised in the Proposed Recommendation
review comments, as well as to provide some ideas about how tools can meet
checkpoints. Among other things, this allows us to provide reference
implementations for how to meet particular checkpoints without having to
produce complete reference implementations. I don't think there is any
disagreement that creating a triple-A tool is a lot of work, and I don't
think the Working Group is likely to be able to produce one as a
demonstration. So working on this is important.

In doing it, and particularly in working on an RDF scheme to describe the
results, I have come to the conclusion that it is valuable to be able to
describe conformance to relative priority checkpoints by including
information about how a tool meets individual checkpoints of WCAG. However I
feel that it is not necessary to have those checkpoints in ATAG - they
are already available from a W3C recommendation, and even in checklist
format. Familiarity with both specifications is required in order to do a
good review, but then our guidelines make it clear that familiarity with WCAG
is necessary to implement them anyway. Knowing a tool well seems to be the
most important requirement, and then being able to think about how a tool
satisfies the requirements listed.

Essentially, these type of reviews can be done by anyone, and I think it is
valuable to people looking to implement the specification if there are
reviews done by the people who wrote it, and the more the better. This is
analagous to a magazine reviewing software - they generally don't pretend to
review every tool available.

I have already explained why I think that reviews are only valuable if they
can be reproduced, which requires "naming names", but I don't see that as a
particular problem - all the information is publicly available anyway.

Charles McCN

On Fri, 26 Nov 1999, Kynn Bartlett wrote:

  I dunno, I can see where William is coming from here.  Do we have plans
  to do a comprehensive survey or just a survey of a few selected tools?
  If so, how will these be selected?
  
  --Kynn

Received on Saturday, 27 November 1999 04:37:23 UTC