RE: fragment identifiers

> RDF, we are *very* concerned about the pervasive use of fragment
> identifiers in RDF.  Our intermediaries don't see the fragment id, and
> therefore cannot take action based upon the resource being accessed.

Can you please elaborate on how you believe RDF to be related to
*accessing* a resource?

The *web* is about accessing resources.  The *semantic* web is about
*describing* resources, even those which cannot be accessed.  Describing
does not and should not require accessing.

If you are making RDF depend on GET at all, I perceive a serious
misunderstanding.