- From: Tim Boland <frederick.boland@nist.gov>
- Date: Thu, 16 Dec 2004 16:41:41 -0500
- To: w3c-wai-au@w3.org
I don't understand why the "rationales" are designated as "normative".. From QA SpecGL 22 Nov 04 WD 1.1 Good Practice C: "Normative content is the prescriptive part of the specification, whereas informative content is for informational purposes and assists in the understanding and use of the specification. Content includes all sorts of different forms — not only descriptive prose, but illustrations, examples, use cases, formulae and other formalisms. .... Conformance of implementations is defined by and measured against normative content." I do not think the Rationales are testable; they aid in understanding of the specification, and therefore, should be informative, if one follows the Good Practice C mentioned previously. Also, it needs to be mentioned explicitly in Section 3 what it means to "meet a checkpoint" in Section 3. I thought it meant satisfying the success criteria associated with that checkpoint as prescribed by Section 3 (passing all the relevant tests associated with that success criteria?) Perhaps Section 3 material should be moved to before the Guidelines, since it mentions priorities and a reader might want to know this material before accessing the Guidelines, and conformance is an important subject that should be "up front" Thanks and best wishes, Tim Boland NIST PS If an offering claims conformance to Level AA ATAG2.0 by immediately passing all the Level AA tests first, does it also conform by default to Level A (assumed to pass the Level A tests by default, as a way of saving testing effort and resources)? I assume not from Figure 1 in Section 3?, but this is not explicitly stated.. What constraints are there on the various levels (subdivisions)? PPS A list of changes from ATAG1.0 to ATAG2.0 needs to be included (probably as an appendix), as well as whether the specification allows extensions PPPS Any deprecated features from ATAG1.0 to ATAG2.0 (see previous) need to be identified; if there are any, then ATAG2.0 needs to define how to handle them
Received on Thursday, 16 December 2004 21:41:58 UTC