[arin-ppml] Fee structures for ARIN

[Owen DeLong]

On Oct 28, 2011, at 7:56 AM, John Curran wrote:

> On Oct 28, 2011, at 2:35 PM, William Herrin wrote:
> 
>> On Fri, Oct 28, 2011 at 10:03 AM, William Herrin <bill at herrin.us> wrote:
>>> I would also be willing to pay greater fees were those fees calculated
>>> something like this:
>>> 
>>> (  # of /24 IPv4 blocks held
>>> + # of /48 IPv6 blocks held
>>> + # of ASNs held)
>>> / total of the same across all ARIN registrants
>>> x ARIN's annual budget
>> 
>> I would also support a version of this formula which, for now, omits
>> IPv6 from the equation. There's something to be said for progressive
>> taxation which encourages healthy behavior by reducing its cost.
> 
> Presumably, legacy block holders currently receiving services without 
> charge would begin paying these fees upon signing the LRSA?  I imagine
> a few /8 and /16 address holders might suggest that the services they
> require do not cause ARIN expenses even close to the suggested fees?
> This becomes even more important if you omit IPv4 from the equation.
> 
I think you mean if you omit IPv6 from the equation.

Frankly, I oppose this structure in general. ARIN's cost to administer
address space are related to three factors:

	1.	The number of resources for which ARIN is maintaining
		records.

	2.	The number of organizations conducting interactions with
		ARIN.

	3.	The number of add/update/remove transactions that ARIN
		is processing.

Nowhere in the above list does the size of the address space have
any direct effect. Size combined with nature of usage (ISP vs. end
user) may have some second-order impacts and swip vs. rwhois
may also have some impact.

I would not oppose modifying ISP fees such that those who use
rwhois pay less than those who use SWIP.

I do oppose treating end users the same as ISPs from a fee perspective
because end users as a general rule have very few interactions with
ARIN beyond requesting space and the occasional POC update (very
occasional). Additionally, end-users are generally not using the addresses
they receive in a "resale" manner and often are not making any sort of
direct profit from the possession of the addresses themselves, vs. the
ISP whose bread-and-butter is the services that they can provide using
the addresses.

Owen