On 2012-03 -25, at 16:53, Noah Mendelsohn wrote: > > On 3/25/2012 3:37 PM, Tim Berners-Lee wrote: >> x 303 -> y means "y is a description of x" and therefore y is an >> information resource. > > My point is: that's a perfectly coherent definition for 303 in principle, but > I don't read RFC 2616 as saying that. I read RFC 2616 as saying, "If you were > interested in x, then y is something that you might want to >see also<". Now > if (the representation of) y happens to be an RDF document that happens to > make statements about x, terrific. I'm fine encouraging that idiom. I'm a bit > nervous about saying: "by the way, if y were to turn out to be anything other > than such a description, than that 303 is certainly an error". Where are the > normative specs supporting that conclusion?
We are trying to write them now, I guess, where there are hole in what is there.. The 303 should be defined in the thing welcome up with, including the its range being IR. > > Noah >
