Properties for evaluations

Hi.

A while ago we talked about the kind of things we want to say about
conformance - that a checkpoint is met at a certain level, or is not
applicable, or is not met, or is not tested.

ER have discussed doing this with greater granularity to a document - how to
say it for diffferent parts of a document, and I am thinking that you should
be looking at Xpath as a way of defining those parts.

AU has been looking at patial evaluation based on the fact that there are
several different tests that need to be done on an Authoring tool (or piece
of content) to determine if it meets all the relevant guidelines. As a
convenience we are trying to reduce the number of tests, and to describe what
those tests themselves cover. For example, some tests are one-for-one testof
a checkpoint, some can cover several checkpoints at once (most obviously
where there are redundancies, and there are a couple in WCAG and one in ATAG
by even the most conservative estimates), and some checkpoints require
several (which/how many) tests.

I would be interested to find out what properties people think are important
to allow testing to be proceduralised. Some things to think about are
comparing different tests of the same thing, and combining test of different
things to get an overall result.

So Who, when, and extra comments have already been put up as important. There
are probably more...

Cheers

Charles

-- 
Charles McCathieNevile    mailto:charles@w3.org    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative                      http://www.w3.org/WAI
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053
Postal: GPO Box 2476V, Melbourne 3001,  Australia

Received on Monday, 4 September 2000 05:01:23 UTC