a. To the best of my understanding, nothing in the RDF syntax impliesthat the URI scheme must be the HTTP URI scheme, common as that is.b. Some of your arguments here seem to depend on that (e.g. your point 3 ).c. My conclusion from a. is that any Semantic Web tool, e.g. a SPARQLprocessor, should usually be deemed broken if it behaves differentlyon an HTTP URI than it does on any other URI.Bob