[ppml] Combining Forecasts

[John Curran]

At 9:03 AM -0400 8/30/07, Keith W. Hare wrote:
>How accurate to we need to be?  The projection that we are likely to run
>out of available IP addresses sometime in 2010 should be enough
>motivation to push IPv6.

I doubt if a few months of uncertainty either way would
matter for purposes of policy consideration, whereas a
difference of a few years is definitely relevant since one
wants to calibrate any policy changes against the need.

For example, if we're looking at 2015 as an expected IPv4
depletion date, it's possible that IPv6 could have a sizable
deployed base by then and be fairly mature (due to various
initiatives and use in private network contexts)   There
could even be an agreed-upon set of functional transition
mechanisms by such time.  The overall IPv4 policy might
simply be to maintain status quo till the forecasted date.

If one considers a different scenario with a 2012 depletion
date, we've got a different situation and need to explore how
to increase conservation and encourage IPv6 deployment.

Using an IPv4 depletion date of 2009 strongly suggests looking
into aggressive conservation, proactive IPv6 deployment,  and
reclamation/reuse of legacy space, and assessing the usability
of reserved space, and any other measure which maximizes
IPv4 effective utilization without destroying the routing system.

So, the accuracy of the projections matter, since businesses
generally want the smallest possible change to the existing
system which still allows us to keep it all running.

/John