"Alec H. Peterson" wrote:
--On Thursday, February 7, 2002 17:29 +0100 Brian E Carpenter <brian@hursley.ibm.com> wrote:
Cathy,
Yes, I understand why people are nervous about repeating historical mistakes. What I was clumsily trying to express is that objectively, I don't think we need to be particularly nervous about whether we give out /32 or /35 at the large end, or /48 or /51 (say) at the small end - yes I know that's a factor 8 in each case, but we do have that much slack in terms of available space. What is very nervous-making is the risk of people believing that they need, and can get, PI prefixes, and a lot of the last day's discussion is right on the mark about the risks there.
So perhaps we should focus on the question of who will get PI allocations before we worry about how large those allocations should be?
Along those lines, perhaps we should figure out how sites are going to multi-home, since that will (well, should) have a huge bearing on who gets PI alloocations.
Indeed. That is exactly why the IETF has a MULTI6 working group. Brian