On Sat, Apr 4, 2009 at 8:36 PM, Noel Chiappa <[email protected]> wrote: > > From: William Herrin <[email protected]> > >> That would make an MPLS tag an "address". Somehow I don't think so > > Why not? What characteristics of what you think of as an address are > > missing from the typical use of an MPLS tag? > > Um, because to me, an address, like its real-world prototype, says _where_ > you are, in some way that anyone can grok, independent of either who they, > where they are, or how they get to that destination. (Yes, NAT complicates > the issue for IPv4 addresses.)
Hi Noel, You can send mail to my mother via the postal address: General Delivery Dora, MO That address probably fails to reveal that she lives just north of Kitty Hawk NC. Defining address in terms of where you are, whether geographically or topologically, offers a very slippery definition. It's hard to give it any kind of rigor in the face of anycast, multicast and load balancing. > > Normally, I would love that idea. However, our specific request is that > > we define the terms locator and identifier. > > I don't see a conflict here. I'd like to a see a definition of locator and identifier which is rigorous and consistent both with our theoretical constructs and with existing network protocols. Giving the community a sloppy definition of identifier and locator would diminish us. Regards, Bill Herrin -- William D. Herrin ................ [email protected] [email protected] 3005 Crane Dr. ...................... Web: <http://bill.herrin.us/> Falls Church, VA 22042-3004 _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
