> From: Chris Grundemann <[email protected]> > I am told that Jon Postel once said: > "A name indicates what we seek. An address indicates where it is. A > route indicates how we get there."
He seems to have been quoting John Shoch, "A note on Inter-Network Naming, Addressing, and Routing" (IEN-19, 1978): http://postel.org/ien/txt/ien19.txt which says exactly the same thing; RFC-760/791 (the source of the 'Postel' quote) include a reference to Shoch at that point in the text. > It appears that the attempt here is to restate it as: > "An Identifier indicates what we seek. A Locater indicates where it > is. A route indicates how we get there." Well, sort of. The 'identifier' defined in this thread is a subset of 'names', as Shoch defined them - and we continued to use other kinds of names, in addition to identifiers, to indicate "what we seek" (most specifically, DNS names). But the rest of your restatement is accurate. > I am not convinced that an Address by design or definition combines > identity and topology. Yes IP addresses have come to be used for both > but that is not necessarily required by their (original/default) nature. Err, the 'original nature' of 'IP addresses' goes back to the very earliest versions of TCP/IP (in fact, to back before it was even split into separate TCP and IP protocols), and even in those early versions it both i) identified the entity at the far end, and ii) indicated where it was. Noel _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
