[arin-ppml] Accusation of fundamental conflict ofinterest/IPaddress policy pitched directly to ICANN

Owen DeLong owen at delong.com
Mon May 2 22:39:35 EDT 2011


On May 2, 2011, at 5:25 PM, Mike Burns wrote:

> Hi Jonathan, answers inline.
>  
> >What is broken about ARIN is that scandalously large numbers of netblocks do not have valid POCs, for example. The stewardship of Whois leaves a lot to be desired.
>  
> >What steps would a commercial entity take to resolve this that RIRs cannot?
>  
> A commercial entity could recognize justification-free legacy transfers, and many legacy transfers happened  but were not reflected in Whois because of this.

This assumes that the recording and legitimization of such transfers is in the community interest.
I would argue that the identification and reclamation of these blocks is the more appropriate
action.

> Not that the RIRs cannot do this, but they have not. It could well be that changing conditions will cause a realization of the legal realities of these transfers, and induce an RIR community to change policy to turn that registry into a title agency, reviewing evidence of property transfers without reference to the need to transfer networks or customers, or justify a need.
>  
They have not because community consensus has been that they should not (except in the
APNIC region).

>  
> >Competitive pressures would help to finally decide who controls these addresses and allow them to be transferred to those who would pay for them.
> >Leaving other equally "worthy" entities with less money unable to acquire space despite their need?
>  
> I meant competitive pressures between registries, but sure, poor entities couldn't afford expensive IP addresses.
>  
Today, this is not an issue as the RIRs operated as non-proffit organizations do not charge
excessive fees. There are entities that cannot afford addresses even under the current
situation (I'm involved in some), but, generally speaking, the current scope of that issue
is much smaller than it will become under a competitive for-profit structure.

> >Network operators don't really have much of a choice in accessing Whois information to determine the rights to advertise addresses, and competive registries.
> >I'd argue that network operators are the very ones who give the RIR their "power." I also don't see why you seem to claim that they can't? Tell me, what's stopping them from using whatever registry they want?
>  
> I couldn't agree more. I don't think there really is anything stopping them from using whatever registry they want. I think they have slim pickings in registries now, and use what they have, but they don't trust it unilaterally. If a registry comes along that offers network operators some insurance against conflict or abuse claims, my bet is they would avail themselves of it rather than turn down business.
>  
There is nothing preventing a business from attempting to offer those services today.
They cannot coordinate their registrations with the existing RIR system, but, they can
offer those services. If they can attract enough of the internet, then the RIR system
would, de facto, become irrelevant.

> >What is broken about ARIN is that their transfer policies are more restrictive than APNICs, and that will cause a flow of addresses out of ARIN and into APNIC.
> >Could you explain why you think this?
>  
> All else being equal, I believe buyers and sellers will migrate towards the freer market. The justification is like a transaction cost. Transactions will go to where the cost is lowest.
>  
There is no current possibility for addresses to flow out of ARIN to APNIC as there is
no policy supporting such a transfer. Even the APNIC transfer policy only supports
transfers within the APNIC region. Any transfer of addresses administered by ARIN
would require the consent of ARIN. There is currently no policy to support ARIN giving
said consent.

The less restrictive policy allows APNIC to make a mess of their own address space,
but, it does not allow the world to migrate from the other registries with policies more
consistent with proper stewardship and the goals expressed in RFC-2050 to the
less restrictive policy regime in APNIC.

>  
> >Not to speak for him but you did say " Competitive pressures would help to finally decide who controls these addresses and allow them to be transferred to those who would pay for them."
> 
> >Were you not suggesting that the folks with the most money would be the ones who got address space registered to their name and the others would be out in the rain?
> 
> >Let me know if I misunderstood.
>  
> Yes, you misunderstood, I was talking about competitive pressures between registries. If a registry becomes freer, like APNIC did, it puts competitive pressures on the other registries to free their markets up,  that's what I was trying to say. Sorry if I was unclear. But yes, eventually it will cost money to get IP addresses. I guess the richer you are, the more easy it will be to acquire addresses in the post-exhaust world. I don't see any honest way around this, but at least this will serve to free up huge swaths of address space currently on the sideline. Wouldn't this buy some time to finally get that orderly V6 transition in motion?

It wasn't what he intended, but, it is also the inevitable result of what he
is proposing.

Under the current system, the community in each region (where community is
literally defined as anyone in the world who has an email address and chooses
to participate) can define policies for that region with limited risk of requests
for address resources being "policy shopped" to the most favorable registry.

What Mike is proposing is that the theoretical race to the bottom in terms
of market pressure creating the least restrictive possible set of policy in
favor of "sell 'em all, let the almighty currency sort them out" approach to
address distribution would somehow be in the best interests of the community.

Owen


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20110502/ba12ca39/attachment.html>


More information about the ARIN-PPML mailing list