RE: Please review latest requirements draft before Wednesday, 22 January

Andi Snow-Weaver writes:
 > 
 > There is definitely something confusing about the description of
 > "Technology-specific Checklists" in section 5. It states that the checklist
 > "must include technology-specific checklist items that address every
 > success criterion in the guidelines." But Section 3.2 states that "for a
 > given technology, it is not necessary to provide Techniques for every
 > checkpoint if the checkpoint is not applicable to the technology and the
 > technology is designed to be used with another technology."
 > 
 > I read 3.2 to mean that not every checkpoint will apply to every
 > technology. If the checkpoint doesn't apply, then neither do the the
 > success criteria for that checkpoint. So then how can we require, in
 > Section 5, that there must be technology-specific checklist items for every
 > success criterion in the guidelines?

Here are a few suggestions. First, it may be recalled that we plan to
write "core techniques" covering those success criteria of which the
implementation is not dependent on the features of any specific
technology. In each checklist, we could combine checklist items from
the core techniques with technology-specific checklist items so that
the entire set of guidelines is covered in the resulting checklist.

If we decide to set an absolute requirement that every checklist must
cover every checkpoint in the guidelines, then this would be
tantamount to a decision that "dependent" technologies such as CSS,
which are intended to be used in conjunction with other formats, could
not have their own checklists. For example, there could be an (X)HTML +
CSS + DOM/Ecmascript checklist but neither a CSS nor a DOM/Ecmascript
checklist alone. I think there is still an unresolved question as to
whether that is the result we want. For the purpose of developing the
Techniques schema we don't have to resolve these issues, so work on
the standard format for techniques can proceed apace even if the finer
details regarding checklists haven't been completely worked out. The
most important point, I would argue, with this draft of the
requirements document is that we agree on what should be included in
the techniques and the rough details of the output formats we plan to
support.

Tomorrow's meeting will, however, provide an opportunity to elucidate
these issues and to arrive at a disposition concerning the
requirements document. We can of course attempt to resolve these
issues at the meeting, but if further work is needed then the working
group may decide to proceed with publication of the requirements after
inserting an appropriate editorial comment at a suitable point in the
document.

I think the most important point is to ensure that we have a solid
basis on which to develop the schema for techniques, and that some of
the issues
concerning the output formats can be safely postponed as long as they
don't affect the source XML format currently under development.

Received on Thursday, 23 January 2003 03:30:00 UTC