[arin-ppml] 2014-14, was Internet Fairness

John Springer springer at inlandnet.com
Fri Dec 26 18:01:26 EST 2014

Hi John,

Thank you for the clear statement of opposition. Please allow me to 
address the points you offer inline.

On Wed, 24 Dec 2014, John Santos wrote:

> Oppose 2014-14
> 1) /16 is not "small"

This has actually been mentioned before, by several commentators. The 
problem with "big" and "not "small"" is that they require reference to a 
datum, which WRT to 2014-14 has not been provided. Owen Delong provided 
a fair attempt to come to grips with what big or small actually mean as 
percentages of the number and size of transfers that have occured since 
the STLS policy was adopted in 2009, here:

I can provide the attachments, if you wish.

I welcome analyses that put different metrics forward. In this analysis, 
transfers between /16 and /24 constitute 17.47% of the address space 
transfered. That is smaller than the 82.53% represented by transfers 
between /16 + 1 /32 and /11, but some commentators feel that it is not 
small enough for the unspoken datum. Several thresholds have been 
proposed. A rewrite is underway and input as to threshold is most welcome. 
/20 is trending near the mean, IMO.

> 2) The problem the proposal purports to solve hasn't actually been
>  demonstrated. "ARIN staff [...] is spending scarce staff time on needs 
> testing of small transfers."

You pose an interesting procedural question. In our considerations of 
advancing prop-204 to Draft Policy, the question of the problem statement 
was specifically discussed and it was agreed that the community had spoken 
"loud and clear" that request turn around times are too long. We can 
quibble about the author's linkage, but we didn't ignore this and the 
motion to advance passed.

Certain AC actions may be petitioned, 
https://www.arin.net/policy/pdp.html, PART THREE PDP PETITION PROCESS, but 
I don't think failure to convince everyone that the problem statement is 
clear is one of them. In fact, the petition process appears to be 
specifically directed at failure to advance things. There does not appear 
to be provision to petition advancement of a policy proposal to Draft 
Policy, which is what you seem to be suggesting here.

> Obviously, doing the necessary checking
>  requires staff time, but is it a significant amount?  Is it taking much
>  longer than it used to?  Is it costing ARIN a lot of money in staff
>  wages and overhead to do these assessments, or is it lost in the noise?

If 17.47% is not small, is it significant? There definitely was a lot of 
comment from the community in Chicago that it is taking too long to turn 
requests around. What is the datum for a lot of money? I'll leave it to 
the AC staff to specifically answer the financial questions you pose.

> 3) This proposal not only eliminates needs testing for qualifying transfers, 
> but also removes the requirement for the recipient to sign an RSA.

To the extent that this is the case, it was not intended and is being 
dealt with in rewrite.

> 4) Rearranging the IPv4 deck chairs

I understand the impulse to belittle and the frustration of some network 
professionals with certain proposals, even categories of proposals. But 
I personally am not at liberty to ascribe much weight to an appeal to 

> 5) Pie

As I said below, I don't think we should be able arbitrarily to torpedo 
any community member's proposed policies without good reason.

John Springer

> On Tue, 23 Dec 2014, John Springer wrote:
>> Hi PPML and Randy and Steven,
>> Subject change and sorry for the top post.
>> WRT ARIN Draft Policy 2014-14, Removing Needs Test from Small IPv4 
>> Transfers, this started out as "ARIN-prop-204 Removing Needs Test from 
>> Small IPv4 Transfers on 16 April 2014. At the 15 May 2014 ARIN AC 
>> teleconference, the motion to move the proposal to a Draft Policy was 
>> passed unanimously. Prerequisite to this action was agreement among the AC 
>> present that, inter alia, the proposal had a clear problem statement.
>> It might still.
>> So whatever other failings 2014-14 may have, a unclear problem statement 
>> would seem, at least by the AC's definition, not to be one of them.
>> As far as the why, ARIN is a community of often polarized interests. The 
>> majority does not, and equally importantly, should not, automatically get 
>> to quash all things it does not agree with. Obversely, minorities, even 
>> despised ones, have the right to work for incremental change in their 
>> interests and receive a fair hearing. This would appear to include the 
>> right to rational discussion, even if irritation sometimes shows up.
>> The shepherd's job is made more difficult by a lot of this talking in code. 
>> Clearly this is a beloved behavior, so I won't say cut it out. But. In the 
>> case of 2014-14, I don't know if it is going to be enough to say 'I don't 
>> like pie', or 'I don't like pie because pie sucks', or 'I don't care. I am 
>> never going to like pie', or even 'I am so much smarter than you, that you 
>> don't even know pie'. Please, no one take this personally.
>> Shepherds are currently contemplating rewriting 2014-14 to accomodate 
>> objections even though some objections more resemble the above. I am not 
>> completely optimistic about either the rectitude or the efficacy of this 
>> move, but am thinking and working on it.
>> If anyone might care to comment on the following three choices, I would be 
>> grateful:
>> 1)	Abandon 2014-14 entirely because... (Don't say pie.)
>> 2)	This part of it is clearly wrong because..., do this to fix it.
>> 3)	Advance it. I haven't heard any convincing opposition.
>> TIA and in reverse to everyone for the comments and the courtesies.
>> John Springer
> -- 
> John Santos
> Evans Griffiths & Hart, Inc.
> 781-861-0670 ext 539

More information about the ARIN-PPML mailing list