In fact, all this thread by me started with my saying effectively that - ILNP Locator identifies a subnet. So, it can also be interpreted as ID. - So, again with Locator, we have semantic overloading. Please, please, please. We shouldn't screw up (by semantic overloading) new terms. A locator of a node denotes the node's location and in case that node changes its location it has to get a different locator. A locator of a subnet (what is this anyway ? isn't it rather an identifier ?) denotes the subnet's location and in case that subnet changes its location it has to get a different locator. Heiner -----Ursprüngliche Mitteilung----- Von: Dae Young KIM <dy...@cnu.ac.kr> An: Tony Li <tony...@tony.li> Cc: rrg@irtf.org; Noel Chiappa <j...@mercury.lcs.mit.edu> Verschickt: Sa., 12. Jun. 2010, 3:51 Thema: Re: [rrg] Terminology straw poll On Sat, Jun 12, 2010 at 9:52 AM, Tony Li <tony...@tony.li> wrote: > > >> No, I'm talking in the MAC layer. Be in the L2 network. >> >> In MAC L2 network, there's only one number type, MAC address pointing >> to a MAC node. >> >> There's no such thing as MAC interface in MAC L2 network. > > > Does the MAC address 'point'? It seems like a MAC 'address' is nothing more > than a flat, globally unique namespace. > > Certainly if I'm on a L2 network and I change my physical attachment, I can > change my location without changing my MAC address. > > Yet I can't do that with my postal address. > > If we look to the dictionary, we find: > > address |əˈdres; ˈaˌdres| > noun > 1 the particulars of the place where someone lives or an organization is > situated : they exchanged addresses and agreed to keep in touch. > • the place itself : our officers went to the address. > • Computing a binary number that identifies a particular location in a data > storage system or computer memory. > > > So it seems like location is necessarily a fundamental component of the > conventional definition of an 'address'. > > I humbly suggest that in fact, the MAC address is not, in actuality, an > address at all and is simply an identifier. Yes, you can forward on it, but > then you can forward on any global flat namespace if you put your mind to > it. Full human names, anyone? > > Regards, > Anthony Joseph Li (Full human name) > Location: ??? You ask DY Kim at the entrance to Corea, and theoretically, if it's a FQNM, they will locate me for you. You don't have to know my postal address beforehand. Actually, we have social security number, which has no locational structure. You give the number at entry, and they will lead you/packets to me. I mean, as foreigner, you don't have to bother with the internal locational structure of the address. It's their job, not yours. In fact, all this thread by me started with my saying effectively that - ILNP Locator identifies a subnet. So, it can also be interpreted as ID. - So, again with Locator, we have semantic overloading. - LISP ID (EID) is routable within a site. So, it can also be interpred as (intra-site) Locator. - So, ID can also be semantically overloaded. - So, semantic overloading may not be the real problem with the IP address. Something else might have been to blame. -- DY _______________________________________________ rrg mailing list rrg@irtf.org http://www.irtf.org/mailman/listinfo/rrg
_______________________________________________ rrg mailing list rrg@irtf.org http://www.irtf.org/mailman/listinfo/rrg
