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

[Matthew Kaufman]

On 4/7/2013 6:41 PM, Paul Vixie wrote:
> ...
>
> Matthew Kaufman wrote:
>> ... we have clearly heard that we can't give everyone with a /32 of
>> IPv6 the same low price otherwise ARIN wouldn't have enough money to
>> operate. ...
> can you hum a few bars? that is, i didn't clearly hear that.

Quoting John Curran from 28 March 2013 2:04 AM:
"Owen - It's not 300 ISPs; it really has to be all ISPs with the same 
holdings
and with /32 as the lower bound, then that will be the _majority of all 
ISPs_.
Even with some aggressive efficiencies baked into ARIN's operating 
costs, the
fees would be $1500 or more per year as a result.  This is truly a 
question of
trying to achieve the lowest fees for these smaller providers, and being 
able
to have some amount of stratification allows their fees to be lower than the
average otherwise."

I am making the small leap here that since John believes it would be 
"$1500 or more per year as a result" best-case that numbers like 
$500/year (the new XX-Small rate that I'm referring to as "the same low 
price" above) would bring in about 1/3rd of what ARIN "needs". But I 
don't think that's much of a leap.

>
>> ... We've heard several times that it wouldn't do to have all of the
>> orgs holding IPv6 paying $500/year for that, because it wouldn't bring
>> in enough money.
> this also. (somebody from ARIN, speaking for ARIN, said this?)

See above. Don't even need to make the above leap to figure out that 
$500/year "wouldn't bring in enough money".

>
>> On 4/7/2013 2:05 PM, Paul Vixie wrote:
>>> i've been here nine years now and i've been looking for the avoidable
>>> costs or the self-creating monetary vacuum that you're talking about
>>> here. i havn't found them. i don't think we'd like (you, me, anybody) a
>>> nearly-volunteer system without the controls, outreach, and policy
>>> process we're all getting from the RIR system in its current form.
>> Well, some of the policy process and nearly all of the outreach are
>> more than I would want to pay for, personally.
>>
>> Along with the entire travel budget.
> that's a short-sighted view. the thing we live on is round, and
> telepresence doesn't reach to the hallways. the RIR system and the
> larger internet governance community that the RIR system is part of,
> meets all over the world. as someone who just crossed the 85K mile
> threshold in the three months since january 1, i'm the first to admit
> that we all travel too much. but the travel budget is there for good
> reason. (as is the policy and outreach budgets.) if you think that stuff
> is all silly or crazy, that helps me understand why you think a
> mostly-volunteer org could do the important part of what ARIN does
> (maintaining the database) but it also puts a discount on all of your
> observations since it's so completely unrealistic.

I'm not sure why the entire address-holding community needs to be 
funding this outreach and worldwide face-to-face meetings for 
governance. Never mind that the idea that the governance can't be done 
over email just shows how useless this whole Internet thing is.

>
>> But even if it is true that ARIN is being exceptionally frugal, the
>> draft policy is bad policy and we're only talking about it because
>> ARIN doesn't want to ("can't") simply charge everyone the x-small
>> price and be done with it.
> i think that arin is reasonably but not exceptionally frugal. and i'm
> listening carefully to your comments about the fee schedule.
>

thank you

>
> Matthew Kaufman wrote:
>
>> ... Or, even better, a flat price for all. But of course there *would*
>> be some opposed, because for every choice it isn't "fair" to someone.
> i think it's safe to say that any fee schedule will be deemed unfair by
> someone. that doesn't mean we should balance the perceived unfairness.
> rather, it means we have to form and then follow a theory of objective
> fairness.

or not, and just have it be unfair. but either way, having ISPs get 
smaller than /32 of IPv6 is a bad idea.

>
>> This is just like the US tax system... nobody thinks that any of the
>> models are "fair". Flat tax? Unfair to the poor. Higher taxes for the
>> rich? Disincentive to working harder and unfair to the rich. You can't
>> win.
> we're way off the topic of arin policy at the moment, but i'm concerned
> that you think the progressive tax system is a disincentive to working
> harder. no matter how high the next tax bracket up might be, you keep
> more if you earn more. my incentives are based on what i earn and keep;
> are yours based on perceived fairness?

well, the US tax system is certainly an incentive to move my income from 
salary to long-term capital gains, but yes, we're way off the topic here.

>
>>> I don't see it so much as extracting extra revenue so much as
>>> differentiating the allocation of the total costs.
>> Those are exactly the same thing. One just sounds better than the other.
>>
>> The idea is that you find the people who are willing to pay more, you
>> figure out how to charge them more. In this case, you take some of
>> that "extra" revenue and use it to drop the prices for the people who
>> aren't willing to pay as much. Classic pricing theory covers this, and
>> I would expect no less from any organization. ...
> this is... completely wrong. the ARIN fee schedule is based in no way on
> willingness to pay more.

Willingness, ability, same thing. Since I started writing this reply, 
John has chimed in to note that if everyone paid the same it would need 
to be $2800/year. So ARIN has happily found a few ISPs that are willing 
to pay so much more than some others can pay as little as $500.

>
>>> I don't believe for a second that ARIN is increasing its overhead
>>> just to ensure the money gets spent. I have reviewed the ARIN public
>>> financials several years in a row and I believe that ARIN does a
>>> great deal for the community at a very reasonable cost to said
>>> community.
>> We've had offers to do it for less.
> can you be more specific? has someone offered to maintain ARIN-like
> registry records for you, at a lower cost? got a URL? if someone out
> there can do it for less it may simply be that their real business model
> is in trading addresses and the registry function is a sunk cost for them.

I wish my email archive searching was this good... way back when, in the 
original formation of the RIRs days, several people volunteered to run 
free address registries out of the goodness of their heart (and I'm sure 
a few others had ulterior motives as well). I can't remember who right 
off hand though.

>
>> [the ARIN Board] apparently can't figure out how to run ARIN in a
>> world where everyone with a /32 pays the kind of fees that these
>> "extra small" ISPs claim they can afford. That's not a numbering
>> policy problem.
> i think it is a numbering policy issue. in an address space with 4
> billion /32's, the dominant cost to the community of any allocation will
> be the global routing table slot not the allocation, but the need for
> size categories still exists, simply because larger allocations (when
> justified) are a more conservative use of those routing table slots.

Yes. Which is a good reason for all ISPs to have as few chunks 
(preferably one) of as big as they need long term and to use them 
wisely. In IPv6, "wisely" doesn't mean "crowd down to the bottom /40 or 
smaller of a /32" however.

> that's a settled issue, though you're welcome to reopen it with a policy
> proposal of course. from the settled policy on this topic, fee schedules
> are constructed.

But then constructed in a way that opens the address allocation policy 
again. It happened once when /36 allocations were added (motivated 
mostly) because of the fee schedule, it might happen again now when /40 
needs to be added because of the fee schedule,... when will it end?

Matthew Kaufman