Re: what is the spec telling authors about missing link/@rel?

On Thu, 2009-09-03 at 11:27 -0700, Edward O'Connor wrote:
> >> I think it's defining the semantics of a nonconforming case.
> >
> > That makes no sense to me. Why tell authors semantics
> > of nonconforming cases?
> [...]
> > i.e. that some documents are non-conforming but
> > their semantics are of interest not just
> > to tag-soup-consumers but to authors as well?
> 
> As an author, I'm interested in the semantics of documents that I write,
> regardless of whether or not I manage to obey all of the conformance
> criteria. So I think the spec is sensible in this case.

If the semantics are specified in the case of a missing link/@rel
attribute, I don't see how "must be present" is necessary for
interoperability (cf RFC2119). "should be present" is
more appropriate.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
gpg D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E