On Dec 5, 2007, at 9:29 PM, Noel Chiappa wrote:
My impression is that most people who are worried about the "routing
about to collapse" are worried about a number of things, one of
which is about
that giant distributed computation going unstable (e.g. Ts going
kind of knee).
Maybe it's not a problem, and we don't need to worry about it, but
is that it's so complex, and so little studied/understood, that it's
be sure there's no problem lurking out there.
Of course, it's hard to prove a negative, especially about large
complex distributed systems. OTOH, there's some argument to be made
(and I've made it before) that there's no obvious knee in the BGP
performance curve. I'm not aware of a persuasive argument to the
No dispute whatsoever regarding your point about the lack of study of
large scale Internet routing dynamics (and implicitly, the need for
rtgwg mailing list