[ppml] Proposed Policy: eGLOP Multicast Address Assignments -not accepted by AC as formal policy proposal

[michael.dillon at bt.com]

> I should add that the same proposal has been submitted to RIPE and  
> APNIC, and I presented on
> it to APNIC yesterday.

All the more reason for ARIN to reject it.

The RIR system was set up so that each region can set up policies
relevant to their region without having to deal with global issues.
Global policy comes from the IETF, IANA, ICANN.

Why do people persist in trying to create global policies by shotgunning
a proposal to every RIR? And why do they wail and moan when we don't
march in lockstep with the other RIRs to further the submitters' own
private goals.

It's a poor strategy. Much better to go to one RIR, get the policy
passed, get it implemented, then go to another RIR with a better policy
that builds on the lessons of the first. 

In any case, even if APNIC passes the policy, ICANN and the NRO can
still block it from being implemented if it exceeds the charter of the
RIRs. ARIN just decided to do things differently by stopping it at the
proposal stage by noting that the RIR role is first defined by RFC 2050
which explicitly leaves out multicast. That is an important issue which
you should really have addressed FIRST with the IETF and ICANN.

--Michael Dillon