On Tuesday 2009-03-10 12:09 -0700, Sylvain Galineau wrote: > > And how can you possibly test for it given that any > > correct implementation is going to throw the rule out anyway? > > That seems circular. Should we test user agents under the > assumption that their implementation is correct ? I'm not quite > sure I understand why XHTML or HTML well-formedness should take > precedence over test coverage when that could allow a buggy > implementation to pass the test *and* select the attribute. What the spec requires is dropping the entire rule, which is a stronger requirement than what you're testing for (failing to match one selector within it), not a weaker one. -David -- L. David Baron http://dbaron.org/ Mozilla Corporation http://www.mozilla.com/Received on Tuesday, 10 March 2009 19:24:58 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:30:49 GMT