From: Charles McCathieNevile In lieu of a proper position paper I can offer the following as a starter: In my work with the Authoring Tools Accessibility Guidelines[atag] it has been clear for some time that assessing how actual tools met the requirements was important to doing the work. This has three aspects: 1. Demonstrating that the Guidelines Specification could be implemented 2. Providing exposure to a range of techniques for implementation, which helps in producing the document "Techniques for Authoring Tool Accessibility"[techs] and in "reality-checking" the guidelines drafts. 3. Getting an idea of how standardisation in the area progresses, and what work remains to be done. As part of this work it became clear that there was a real need for detailed information on how to test for conformance to the specification, which has now begun[eval]. It also became clear that there was a need for a good way to express testing results, which was a factor in beginning the work on the EARL project[EARL] at W3C - a reusable metadata vocabulary for expressing conformance. An early approach to this work is available[conf]. [atag] http://www.w3.org/TR/ATAG10 [techs] http://www.w3.org/TR/ATAG10-TECHS [eval] http://www.w3.org/WAI/AU/ATAG10-EVAL/ [conf] http://www.w3.org/1999/11/conforms Charles McCN