[ppml] Re: [arin-announce] NRO Response to ITU Comments on the Management of Internet Protocol (IP) Addresses

[Michael.Dillon at radianz.com]

> The full text of the NRO response is available at:
> 
> http://www.nro.net/documents/nro17.html

The response notes that IPv4 allocations were done
differently in the past due to the tchnology available
at that time and acknowledges that the previous
rules of allocation were unfair when viewed from
a viewpoint of balancing resources among nations.

However, the response doesn't mention that the
national allocation system for telephone numbers
is a direct result of the telecommunications
technology of the time and the way that early
telecom networks were structured. I think that
it is important to note that E.164 country codes,
early IPv4 allocation rules and current IPv6
allocation rules are all driven by technology
and not by intergovernmental politics. It is purely
a coincidence that phone numbers are allocated in
national blocks. If the phone number allocation
system was reinvented today, this probably would
not be the case because the technology of voice
telecommunications today is so different. You can
catch a glimpse of this at
http://www.freeworlddialup.com/advanced/peering_numbers

Having said that I would like to note that the 
current IP (v 4 and 6) allocation system does
support the notion of geographical allocations.
We call it the RIR system. Perhaps someone can
find real technical justification for a more
finely-grained system than just the 5 major
continents but I suspect that such a system
would end up ignoring national boundaries
by both aggregation and splitting. For instance
Luxembourg, Amsterdam and Brussels are in three
countries but it is hard to see any technical
justification for a telecom system that doesn't
treat them as the same locality. On the other hand
Moscow and Vladivostok are in one country but
are clearly not in the same locality. If ITU really
wants to see some geographically based addressing
for IPv6, then it needs to do a lot of technical
committee work before bringing it forward at the
policy level.

--Michael Dillon