On fre, 2003-05-30 at 02:45, Hans Kruse wrote: > I actually see a lot of value in the /56 proposal; I really like the > simplicity of creating the /56 from any MAC-48 in the network. It > accomplishes the uniqueness property without requiring central > registration, and should serve organizations up to considerable size very > well. And it readily discourages the notion of "make up a prefix" for > temporary or (temporarily) disconnected networks.
I have been lurking in this discussion by reading it as a newsgroup over at gmane.org. (I tried to post too, but even though gmane.org said it was posted, I do not think it reached the mailing list.) It seems to me that things would be easier by letting people invent their own numbers by whatever method they prefer. Then, if they care whether the numbers are unique, they can register the numbers they picked in one or more registries. This way they can be assured that other well-behaved people will not use the same numbers, and at the same time it is clear to everyone that the numbers are not /guaranteed/ to be unique. Later when two organisations need to connect to each other directly and the numbers conflict, I bet the organisation that registered its numbers has a good chance of being the one not having to renumber... By the way, I am not very fond of the MAC-address method. If we are into misusing other registries, we could turn a phone number into hexadecimal (including the international prefix.) 12 digits can be stored in 40 bit... /Benny -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
