[arin-ppml] Draft Policy ARIN-2013-3: Tiny IPv6 Allocations for ISPs

[Owen DeLong]

On Apr 6, 2013, at 19:58 , John Curran <jcurran at arin.net> wrote:

> On Apr 5, 2013, at 9:37 AM, Seth Mattinen <sethm at rollernet.us> wrote:
> 
>> I see no reason to have a policy motivated strictly by fees to remain after fee changes that may or may not negate it, and to determine that it should go back through the PDP. Otherwise we're just cluttering the NRPM with irrelevant policy.
> 
> Seth - 
> 
> The ARIN Board has often lowered the fee schedule as the
> financial circumstances allow, and in general this has been
> independent of number resource policy changes.
> 
> The Revised Fee schedule is much improved over the existing 
> fee schedule with respect to organizations being able to 
> cost-effectively make use of IPv6 (as it is lower than the 
> existing fee schedule's smallest IPv6 category today.)

Unless you are an end-user in which case adding IPv6 is still
a (potentially significant percentage) increase over your IPv4
fees.

> If the community does not support 2013-3, then smaller ISPs 
> will still have the current option of receiving a /36 of IPv6 
> (if they wish); this will put them in the x-small fee category 
> of $1000 per year. 
> 
> If the community supports 2013-3, then some very small ISPs will 
> gain an option of taking a /40 allocation of IPv6 space, which 
> will put them in the xx-small fee category with fees of $500 per 
> year.
> 
> If very small ISPs receiving /40 of IPv6 space does not make good 
> technical sense, then the community should not support the draft 
> policy... it's that simple. (These ISPs will still have lower fees 
> under the Revised Fee schedule, just not as low as they might have 
> had otherwise...)

It's not that simple because we have to balance differing tradeoffs
as a result of the fee structure.

If we don't support the policy, then we create a disincentive to deploy
IPv6 at all among these smaller ISPs.

If we support the policy, then we create an incentive to do things which
 do not make good technical sense in order to avoid creating a
disincentive to deploy IPv6.

The current fee structure creates a Sophie's choice in policy. IMHO,
the board seriously erred in creating this situation and should do
whatever is possible to have a correction ready to present prior
to the discussion of 2013-3 in Barbados.

One possible way to distinguish ISP size for IPv6 would be to use
annual gross revenue instead of total address holdings. I don't know
if that would be an improvement or not. I realize it would be more
complex to calculate and enforce. I think it would likely be more fair.

I would like to hear about other creative ideas on ways to measure
size and/or appropriate metrics to be used for fee determination.
We have a lot of smart people in the community. I don't think we
should look at the dysfunction of what we have always used and
simply shrug it off to "but that's how we've always done it."

Owen