Re: Self-descriptive assertions

Hey Stuart,

On Thu, Mar 25, 2004 at 04:59:35PM -0000, Williams, Stuart wrote:
> Hello Mark,
> 
> > Fair enough.  What I meant to say was that from a message 
> > sender POV, describing their RDF/XML document as text/plain 
> > is one way to avoid communicating the graph.  If any 
> > recipient does extract the graph, then that's "sniffing", and 
> > "bad" per the TAG finding on authoritative metadata[1].
> 
> Hmmm... that's not quite what the finding says... last para section 3.2 [2]
> 
> "The Internet media type asserts "this is X", not "process this as follows."
> Representation metadata does not constrain the receiving agent to process
> the representation data in one particular way."
> 
> "this is X" meaning this is an instance of media-type X.

Sorry, I misspoke.

Extracting the graph isn't the bad part.  What's bad is believing that
the recipient was trying to communicate the graph.

Mark.