[ppml] Random v6 discussions (was Re: Policy Proposal: IPv4Transfer Policy Proposal)

[michael.dillon at bt.com]

> It stuns me that there are serious networking folks who don't 
> think we'll run out of v4 addresses.  But this is now 
> waaaaaay off the topic of policy, which is the purpose of this list.

Not at all. This is bang on topic for the policy list.

Clearly the current methods used for analyzing the data which
show IPv4 addresses running out at the IANA level in 2 to 3 years
are not sufficient. What other kinds of analysis do we need to do?
Perhaps just looking for a runout date is not sufficient but we
need some other measures as well.

And many people have claimed that there will be more free IPv4 
addresses as people dual-stack their networks. But this doesn't
make sense. Has anyone had the opportunity to analyse an ISPs
IPv6 deployment plans to figure when in the deployment process
they will begin to have an increase of free IPv4 addresses, and
how will that figure grow (or not grow) over what period of time?

This is definitely an area where we should consider soliciting
the help of the academic community to do the kind of studies that
will provide us hard facts to base our policies on. In fact, it would
be worthwhile for ARIN to increase its communication with the various
federal/national governments in the region to see if they can provide 
some support. Perhaps the FCC or CRTC will mandate ISPs to report
the data needed so that researchers can get access to analyze it.

--Michael Dillon