[ppml] 2005-1:Business Need for PI Assignments

[Stephen Sprunk]

Michael.Dillon at radianz.com said:
> In the end, I think that strict geographic allocations which
> follow the topology of the network, are the only workable solution.

Make up your mind -- do you want strict geographic allocations or do you
want addressing that follows topology?  Those are only the same thing for
very simple end user scenarios.

Real world example: picture an enterprise with 40,000 sites in the US. 
Should each of those sites be numbered from the local state/city pool? 
Should they be numbered from the nearest public Internet connection?  What
if a site in MT is connected to the Internet via four same-company hubs in
LA, NY, Hong Kong, and Brussels?  Please explain how _any_ addressing
model other than PI makes sense.

> Since we do have the space available to do geographic addressing
> by using another 1/8th of the IPv6 address space, why are we not
> seriously pursuing this avenue? I know that geopolitical addressing
> is sexier, i.e. the same old, same old, but if we had some serious
> work being done on geographical addressing that would be sufficient
> to push the geopolitical ideas to the bottom of the agenda.

The reason it's not being pursued is that (a) the topology does not follow
geography today and (b) providers have motivation to make it so.

It'd make for a nice Ph.D. thesis though.

S