[arin-ppml] IPv4 Transfer Policy Change to Keep Whois Accurate
owen at delong.com
Wed May 11 18:37:22 EDT 2011
On May 11, 2011, at 2:13 PM, Mike Burns wrote:
> Hi Owen,
>> Sorry, I call it as I see it. I think abandoning needs-based resource management
>> is an abandonment of the responsibility of stewardship in managing the resources.
> The needs requirement is a hold-over from the soon-to-be-past.
The needs requirement is as valid post run-out as it was prior to runout. There is no
reason in my mind that need is any less important in reallocation than it was in allocation.
> A needs requirement was mandatory when it came to allocating addresses from the free pool.
> I hope I don't need to go over why this is so.
It is at least as mandatory for maintaining some level of sanity in a transfer world as well.
> That requirement of stewardship was there to ensure that addresses were allocated to where they would be used efficiently.
Among other things, yes. More accurately it was to ensure that early address acquisition
by parties without need did not unfairly prevent or increase the difficulty/cost/other metric
of later address acquisition by parties that didn't have present need.
> Another requirement of proper stewardship, to my mind, is not making rules where they are not needed.
Agreed. If I fell that needs-basis was unneeded in a post run-out environment, I'd be the
first to champion the idea. However, that isn't the case.
> To maximize freedom to members of the community, isn't the lightest touch absolutely necessary for order what is demanded of the steward?
Indeed it is. Hence my belief that the relatively light touch of needs-basis must remain there to avoid
the heavy handedness of the invisible hand wielded by greedy power-brokers attempting to control
a market for their own interests over that of the community.
> In a post-exhaust world, human greed will ensure address are used efficiently, so the stewardship need for justification can be eliminated, as it's raison d'etre has ceased to exist.
No, it won't. It will ensure addresses are used to the greatest monetary benefit of the people who have them, whether
that is in the community's interest or not.
> On the other hand, in a post-exhaust world where IPv4 addresses may increase in value, it is easy to see that conflict between address rights claimants is likely to rise.
> This will naturally put stress on Whois that hasn't existed to that extent in an age of "free" address availability.
This statement is less convincing.
> We have witnessed in the MS/Nortel transaction that ARIN policy is not completely aligned with the law, and the shabby way ARIN scurried to slap an ARIN-approved sticker on the transaction has cost ARIN trust. Surely you see that if Microsoft didn't luckily have an ARIN-assessed need which was in line with the prior agreement negotiated with Nortel, that ARIN would have been faced with a legal transfer which could not be accurately reflected in Whois under current policy. Other transactions like this have occurred and will likely increase in frequency as we move through total exhaust of the free pools.
Repeating this invalid assertion doesn't make it any more accurate than it was the first time you uttered it.
When you assert other such transactions, are you referring to the 10 for which John Curran provided anonymized
data, or, are you asserting that there are others which have occurred outside of ARIN? If so, can you cite examples
or provide any documentary proof of such a claim?
> Should reliance on Whois accuracy suffer as a result, the door is open to private registries whom network operators can go to to find more accurate information about routing authority.
I am utterly unconvinced that an ad-hoc pile of competing registries none of whom have any obligation to
maintain uniqueness vs. what other private registries would somehow provide something the community
would find more valuable. Guaranteed Uniqueness is the true value of IP number registrations when you
get right down to it. In an ad-hoc private registry system, guaranteed uniqueness is likely one of the first
casualties sacrificed on the altar of greed. Additionally, it remains unclear how ISPs would go about determining
which of disparate and disagreeing registries was "more" accurate than the others.
With the RIR system, we have a set of known rules enacted by the consensus of the community for the benefit
of the community. The fact that one court wasn't as aware of those rules as would be ideal and that there are
others who seek to circumvent those rules for their own benefit is not a convincing argument for the idea that
we should utterly abandon the rules.
As I said before, to me, this seems akin to legalizing bank robbery in the hopes that the thieves would pay
income tax on their stolen money.
>>> I think I could characterize your opposition better by saying that you believe the danger of hoarding and speculation outweigh the risk to whois accuracy.
>> I think that the risk to whois accuracy is relatively low and is a spectre or boogeyman
>> argument used by those with more libertarian economic views than my own.
>> OTOH, I think that the risks to the community that come with an abandonment of
>> needs-basis and other secondary aspects of address resource management that
>> spring from that is an extreme risk to the community and transitions the fundamental
>> nature of the internet from the greatest tool ever developed for the democratization
>> of communication to yet another tool for media exploitation.
> What is the "extreme risk" you mention, if not the impact you feel would come from hoarders and speculators?
Hoarding and speculating are one extreme risk.
Extreme disaggregation is another.
The movement of the internet at its foundation from its current status as a tool for the
democratization of communications to another resource to be exploited by those with
the greatest wealth is a third.
> Is not my proposal in favor of more freedom for the members of the community?
Anarchy is the greatest form of freedom possible. It does not lend itself well to creating guaranteed uniqueness
among competing economic interests.
> What's more, my proposal will provide an incentive to bring legacy addresses under RSA through increasing the rights of RSA holders to hold and transfer addresses.
At the price of rendering virtually every purpose of the RSA moot.
> I think the value of bringing address space under an agreement where no agreement existed before should be considered as another positive effect of my proposal.
Only to the extent that the agreement itself has value to the community. By stripping the community value of the
RSA, you are, in effect, making the agreement more palatable to those who prefer not to be part of the community,
but, you are also eliminating the benefit to the community at the same time.
> To me, the objections are: Fears of a free market >> Risk to whois + getting more legacy space under RSA.
> This decision comes down to how much do you fear a free market?
While I don't think that's the entire argument or even a particularly accurate framing of that portion of the
argument, I would say that the history of free markets does give one plenty to fear. Especially when you
consider that history in situations of truly finite (even for a short time) resources. (Tulips anyone?)
Other things that play a factor are the idea that the community has placed rules on address distribution for
good reasons that benefit the community. As a member and as a representative of the community, I cannot
favor the elimination of those rules just to benefit a handful of vocal opponents. While I may be the most
vocal opponent of your proposal and/or your position, there has been significant pushback from other
members of the community.
More information about the ARIN-PPML