Re: Wade through the archives (was Re: another renumbering question)

[Matt Crawford]

> 1) Picking numbers at random is easy, but you need a good random
> generator, and besides you hit the birthday paradox with 2^^19 entries,
> i.e. less than a million. There are way more than 1 million potential
> sites.

You don't care about the existence of any pair of duplicates -- only
about a duplication with a site you eventually expose your private
address space to.  If everyone picks an unbiased random site id, and
you connect in that way to N sites, the chance of a duplication is
only around N^2 / 2^19.

If you reserved half the space for the number-tsar to allocate
uniquely and half for random picks, the chance of hitting a
duplication problem (including by connecting to someone with the same
site id as your site and connecting to two which chose the same) is
not more than about N^2 / 2^18 -- 0.1% if you connect 16 sites -- and
far less if you and they are mostly from the unique-numbered camp.
--------------------------------------------------------------------
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]
--------------------------------------------------------------------