[arin-ppml] Revised: ARIN-2017-4: Remove Reciprocity Requirement for Inter-RIR Transfers

Fri Sep 8 23:54:30 EDT 2017

> On Sep 7, 2017, at 17:06 , Scott Leibrand <scottleibrand at gmail.com> wrote:
> 
> IMO we're overthinking this.  The problem is simply that not all companies can get access to the IP addresses they need to run their businesses. 

This simply isn’t true. The problem is limited only to IPv4 addresses. There is no shortage of IP addresses in any region.

> The fix is to enable transfers from ARIN to all other regions, and between as many regions as are willing to participate. 

This is laughable. There are 7 billion people on the planet. There are a total of 3.2 billion unicast addresses in the total IPv4 space.

No matter how you shift around the shell game here, you’re still short by half the world population of even being able to give each person a single IPv4 address.

It doesn’t matter how you move things around, IPv4 simply doesn’t scale to a global internet. It was never intended to do so.

The only fix is to go to a protocol (e.g. IPv6) that has a large enough address space to actually run a global network.

> Only LACNIC and AfriNIC don't allow such transfers yet, and if we can make it easier for them to allow transfers at least in the important direction (to their regions) then the transfer market will take care of making sure everyone who needs addresses can get them.  So any policy, including the draft policy as written, that allows unidirectional transfers to LACNIC and AfriNIC solves the problem.

Why are transfers to their regions (the largest remaining free pools) the important direction? This seems utterly illogical to me in the face of current facts.

No policy will solve the actual problem short of a policy that somehow gets everyone to IPv6 faster (and believe me, if I thought that were possible in policy, I’d have written it long ago). The shell game we are talking about doesn’t fix anything. It simply moves the shortages around in favor of those with the greatest economic resources.

Owen

> 
> -Scott
> 
> On Thu, Sep 7, 2017 at 3:19 PM, Chris Woodfield <chris at semihuman.com <mailto:chris at semihuman.com>> wrote:
> Replying to myself, I decided to look up the population proportions mentioned in my last email:
> 
> North America - 7.79%
> South America - 5.68%
> Africa - 16.36%
> 
> So if one were to use numbers similar to these - the average formula doesn’t make much of a difference for LACNIC, and actually qualifies AFRINIC for a far larger share of space than the straight average. 
> 
> I’m wondering what, if any, types of metrics might exist for measuring demand for resources instead of population? Or does that run afoul of the concept of Internet access as a worldwide human right?
> 
> -C
> 
>> On Sep 7, 2017, at 2:50 PM, Chris Woodfield <chris at semihuman.com <mailto:chris at semihuman.com>> wrote:
>> 
>> Thinking more about the use of an average distribution in the proposal, I’m wondering if this accurately reflects the issue. 
>> 
>> The distribution of IP addresses by IANA to the various RIRs is only inequitable if it results in a clear difference in the ability of an entity in different regions to acquire IP address space. We don’t need the same number of allocations in each region - if nothing else, the allocations should roughly reflect regional populations - but it should be no more difficult for a party in Africa or South America to acquire IPv4 resources than it is for a party in North America, Europe, or Asia to do so. To the extent that this is not the case, we owe the community action to correct.
>> 
>> The question then becomes - does the lack of a transfer policy from ARIN to these regions make it substantially more difficult to acquire space on the transfer market today? I’d argue that to the extent that doing so requires transferring to the space to the local RIR, then the answer is YES, as from my point of view, the bulk of transfer market supply is from allocations in the ARIN region (resellers on the list who are in a position to comment, please keep me honest and speak up if that isn’t the case).
>> 
>> This is somewhat mitigated by the current case that both LACNIC and AFRINIC still have space to allocate, while ARIN does not. But shower term point-in-time facts shouldn’t drive far-reaching policy decisions IMO.
>> 
>> As such, I support the goal of the policy, but I believe that the calculation used to determine qualifying RIRs could be tweaked. Could we compare allocation percentages to world population, perhaps?
>> 
>> -C
>> 
>>> On Sep 7, 2017, at 2:27 PM, Cj Aronson <cja at daydream.com <mailto:cja at daydream.com>> wrote:
>>> 
>>> David.. I agree with your very well written summary.  I just feel that the mathematical formula to determine when the transfers have to start being reciprocal is not needed.  
>>> 
>>> The reason why I feel that way is that we're computing something that was said earlier, "To go below the global average, the RIR above the average and closest to
>>> it would need to lose 81,871,002 more addresses, which at the current rate
>>> (14,592 lost per month) would take 5,620 months (468 years)." 
>>> 
>>> It seems like we're spending time computing something that is not likely to happen.. I would surely hope we are done with IPv4 within the next 468 years  :-)   
>>> 
>>> 
>>> Thanks!
>>> -----Cathy
>>> 
>>> 
>>> {Ô,Ô}
>>>   (( ))
>>>   ◊  ◊
>>> 
>>> On Thu, Sep 7, 2017 at 2:46 PM, David Farmer <farmer at umn.edu <mailto:farmer at umn.edu>> wrote:
>>> Cathy,
>>> 
>>> Yes, in some ways it would be more straight forward to just say LACNIC and AFRINIC are allowed an exception to the reciprocity requirement.  However, that policy would contain only the facts of the situation.  Whereas this policy contains quantifiable reasoning why LACNIC and AFRINIC are exempted from the reciprocity requirement and why APNIC and RIPE are not. 
>>> 
>>> To be honest, I didn't want the reciprocity requirement in the original transfer policy to being with, because of the optics of this very situation with LACNIC and AFRINIC.  However, I didn't push the issue with the original transfer policy because I knew it would be several year before LACNIC and AFRINIC got to the point of approving a transfer policy of any kind. So, when this issue with LACNIC and AFRINIC came up I thought obvious thing to do was to eliminate the reciprocity requirement all together. However, I really like this compromise as well as the reasoning that comes with it. 
>>> 
>>> There is absolutely no reason for transfers with APNIC and RIPE to not be on a reciprocal basis. However, with LACNIC and AFRINIC I feel there should be room for some nuance. LACNIC and AFRINIC have received the short-end of the stick, so to speak.  There was no conspiracy or wrongdoing that caused this result, but it is a stark fact when you look at the numbers. I therefore believe these facts should afford LACNIC and AFRINIC some latitude to decide for themselves how best to move forward. 
>>> 
>>> In the long-run I totally believe LACNIC and AFRINIC should approve reciprocal transfer policies. However, we need to give them room to decide this for themselves, it is arrogant and inconsiderate of the facts for us to dictate a reciprocal transfer policy to them.  If they feel they need to start with a one-way transfer policy, there is logic to such a strategy, and the current facts seem to justify at least some caution on their part.   
>>> 
>>> Finally, the numbers show we have more than enough room to be magnanimous in this situation, I believe we should give LACNIC and AFRINIC room to maneuver, and choose their own way forward. 
>>> 
>>> Thanks.
>>> 
>>> On Wed, Sep 6, 2017 at 4:10 PM, Cj Aronson <cja at daydream.com <mailto:cja at daydream.com>> wrote:
>>> Okay so this formula.. does it just give us Afrinic and Lacnic right?  So why don't we just say that?  Since there are only 5 RIRs it seems that maybe a formula isn't needed?
>>> 
>>> 
>>> {Ô,Ô}
>>>   (( ))
>>>   ◊  ◊
>>> 
>>> On Wed, Sep 6, 2017 at 12:35 PM, ARIN <info at arin.net <mailto:info at arin.net>> wrote:
>>> The following has been revised:
>>> 
>>> * Draft Policy ARIN-2017-4: Remove Reciprocity Requirement for Inter-RIR Transfers
>>> 
>>> Revised text is below and can be found at:
>>> https://www.arin.net/policy/proposals/2017_4.html <https://www.arin.net/policy/proposals/2017_4.html>
>>> 
>>> You are encouraged to discuss all Draft Policies on PPML. The AC will evaluate the discussion in order to assess the conformance of this draft policy with ARIN's Principles of Internet number resource policy as stated in the Policy Development Process (PDP). Specifically, these principles are:
>>> 
>>> * Enabling Fair and Impartial Number Resource Administration
>>> * Technically Sound
>>> * Supported by the Community
>>> 
>>> The PDP can be found at:
>>> https://www.arin.net/policy/pdp.html <https://www.arin.net/policy/pdp.html>
>>> 
>>> Draft Policies and Proposals under discussion can be found at:
>>> https://www.arin.net/policy/proposals/index.html <https://www.arin.net/policy/proposals/index.html>
>>> 
>>> Regards,
>>> 
>>> Sean Hopkins
>>> Policy Analyst
>>> American Registry for Internet Numbers (ARIN)
>>> 
>>> 
>>> 
>>> Draft Policy ARIN-2017-4: Remove Reciprocity Requirement for Inter-RIR Transfers
>>> 
>>> Version Date: 6 September 2017
>>> 
>>> Problem Statement:
>>> 
>>> AFRINIC and LACNIC are currently considering one-way inter-RIR transfer proposals. Those RIR communities feel a one-way policy a policy that allows network operators in their regions to obtain space from another region and transfer it into AFRINIC and LACNIC may best meet the needs of the operators in that region.
>>> 
>>> ARIN staff, in reply to an inquiry from AFRINIC, have formally indicated that ARINs 8.4 policy language will not allow ARIN to participate in such one-way transfers. The staff formally indicate to AFRINIC that the word reciprocal in 8.4 prohibits ARIN from allowing ARIN-registered space to transfer directly to AFRINIC (in this context).
>>> 
>>> ARIN as a community should recognize that other RIR operator communities have different needs than we do. We should recognize that:
>>> 
>>> - network operators in AFRINIC in LACNIC have need to obtain space in the market;
>>> 
>>> - have reasons they think are important to not allow two-way transfers; and
>>> 
>>> - we should understand that the history of the RIR system has led to LACNIC and AFRINIC having multiple orders of magnitude less IPv4 address space than ARIN does.
>>> 
>>> Policy statement:
>>> 
>>> Add the following sentence after the first sentence of NRPM 8.4:
>>> 
>>> Inter-RIR transfers may take place to an RIR with a non-reciprocal inter-RIR transfer policy only when the recipient RIR has an IPv4 total inventory less than the average (mean) of the IPv4 total inventory among all of the RIRs.
>>> 
>>> Timetable for implementation: Upon the ratification of any inter-RIR transfer policy at another RIR that is one-way as described in the problem statement.
>>> _______________________________________________
>>> PPML
>>> You are receiving this message because you are subscribed to
>>> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net <mailto:ARIN-PPML at arin.net>).
>>> Unsubscribe or manage your mailing list subscription at:
>>> http://lists.arin.net/mailman/listinfo/arin-ppml <http://lists.arin.net/mailman/listinfo/arin-ppml>
>>> Please contact info at arin.net <mailto:info at arin.net> if you experience any issues.
>>> 
>>> 
>>> _______________________________________________
>>> PPML
>>> You are receiving this message because you are subscribed to
>>> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net <mailto:ARIN-PPML at arin.net>).
>>> Unsubscribe or manage your mailing list subscription at:
>>> http://lists.arin.net/mailman/listinfo/arin-ppml <http://lists.arin.net/mailman/listinfo/arin-ppml>
>>> Please contact info at arin.net <mailto:info at arin.net> if you experience any issues.
>>> 
>>> 
>>> 
>>> -- 
>>> ===============================================
>>> David Farmer               Email:farmer at umn.edu <mailto:Email%3Afarmer at umn.edu>
>>> Networking & Telecommunication Services
>>> Office of Information Technology
>>> University of Minnesota   
>>> 2218 University Ave SE        Phone: 612-626-0815 <tel:(612)%20626-0815>
>>> Minneapolis, MN 55414-3029   Cell: 612-812-9952 <tel:(612)%20812-9952>
>>> ===============================================
>>> 
>>> _______________________________________________
>>> PPML
>>> You are receiving this message because you are subscribed to
>>> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net <mailto:ARIN-PPML at arin.net>).
>>> Unsubscribe or manage your mailing list subscription at:
>>> http://lists.arin.net/mailman/listinfo/arin-ppml <http://lists.arin.net/mailman/listinfo/arin-ppml>
>>> Please contact info at arin.net <mailto:info at arin.net> if you experience any issues.
>> 
>> _______________________________________________
>> PPML
>> You are receiving this message because you are subscribed to
>> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net <mailto:ARIN-PPML at arin.net>).
>> Unsubscribe or manage your mailing list subscription at:
>> http://lists.arin.net/mailman/listinfo/arin-ppml <http://lists.arin.net/mailman/listinfo/arin-ppml>
>> Please contact info at arin.net <mailto:info at arin.net> if you experience any issues.
> 
> 
> _______________________________________________
> PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net <mailto:ARIN-PPML at arin.net>).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-ppml <http://lists.arin.net/mailman/listinfo/arin-ppml>
> Please contact info at arin.net <mailto:info at arin.net> if you experience any issues.
> 
> _______________________________________________
> PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> http://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20170908/1853c4d7/attachment.htm>