[arin-ppml] Policy Proposal: A Modest Proposal for an Alternate IPv6 Allocation Process

[lar at mwtcorp.net]

My analysis of an IPV6 deployment, for my company, is "painful and 
expensive". There are still
hurdles in the mixed IPV4-IPV6 environment for VOIP and "tons" of work to 
learn to design and
deploy an efficient IPV6 WAN.
  
I am a small ISP that is geographically spread out. Lots of grass, small 
number of people. I feed
other yet smaller ISP's. I had planned on assigning each one a /48  and 
dangle the carrot that
as long as I was their provider they would never have to renumber.
They will still resist but I think I can nudge them along.

A /32 will probably last me for many years, maybe for ever. A /48 will not.

I have been forced to renumber IPV4 4 times in the past 12 years. The last 
one took two years to fully implement.
I have lost customers every time.

I had tentatively planned on starting deployment in spring of 2010. (A lot 
of engineering remains
to be done.) This proposal raises the spector of being forced to renumber at 
some point in the future.
If adopted this is a deal breaker as far as implimenting IPV6 for the next 
few years. I will whip IPV4
as long as it looks like it can crawl. Sorry just the truth.


Larry Ash
Network Administrator
Mountain West Telephone
400 East 1st St.
Casper, WY 82601