Michael Dillon wrote:

We don't have consensus on this and the points raised by Sander Steffan in his email on the 13th of June, have not been answered yet. Here is Sander's email:

You know what, I'll indulge and repeat what I've been telling the past 2 RIPE meetings: Randy Bush: http://www.ripe.net/ripe/maillists/archives/address-policy-wg/2008/msg00 355.html - Transfering address space still assigned to end users Is not a part of this proposal. An additional proposal could accommodate this if desired. - Inter-region transfers Explicitly not a part of this proposal. An additional proposal could accommodate this if desired. Per Heldal: http://www.ripe.net/ripe/maillists/archives/address-policy-wg/2008/msg00 354.html - Keep demonstrated need criteria for address space by receiving LIR Fundamentally - what is more important? Keeping an accurate and usable database or setting limits to what people can and can not do without enforcement options? (One of the very few sticks the NCC currently has is denying people more allocations - if there's none more to be handed out, what do you expect to happen ?) I'm not averse to extra limitations in transferring space, what I DO disagree with is adding them to this current proposal. Time is not on our side. - Legal implications for RIPE NCC Anything we do, including doing nothing, has potential legal implications. - Viable plans for reclaiming space to continue with current policies for a significant time? The one viable plan I can come up with is to have ICANN buy back a lot of space using the money they got from opening the DNS root :) No transfer, reclaim or other policy can replace the global free pool as a sufficient resource for new allocations. - Possiblity of setting up LIRs for hoarding Is possible and viable today. Everyone can set up an LIR and get an initial allocation at the very least. RIPE NCC has very interesting presentations every meeting about the growth in membership. Jay Daley: http://www.ripe.net/ripe/maillists/archives/address-policy-wg/2008/msg00 366.html - Seller can choose who to sell to, not transparent Life isn't fair. Not knowing who buyer and seller are, or not knowing that a transaction has taken place is even less transparent. - Discrimination of LIRs in developing countries in the RIPE region What do you think would work and be less discriminatory? How about transferring address space to a needy LIR in a developing country as a (tax deductible !) act of charity? Not being 'legally' able to get more space discriminates against all who respect the rules. - Reclaim/reuse could be more efficient than transferring Given the fact that it requires more parties to actually do something, and they don't get a return for doing that, I don't see how that would work. - Faster depletion because of hoarding And nobody will hoard space if they're not entitled by RIR policy to transfer. Of course. - Legal implications for RIPE NCC See above. - Degradation of RIPE DB because of rival trading exchange databases ..is exactly what we're trying to prevent here. Rival databases only evolve if we set limits on what you can do with ours. If RIPE runs an accurate, usable and public database there's no business case for a rival database. Eric Schmidt: http://www.ripe.net/ripe/maillists/archives/address-policy-wg/2008/msg00 381.html - Breaks the current policy that unused address blocks return to RIPE NCC Which doesn't happen on any scale worth mentioning. Only when VERY significant effort is made by IANA/RIRs, space is returned. - Transfers open up more possibilities for abuse Not registering third party transfers open more possibilities for hidden abuse. ETNO: http://www.ripe.net/ripe/maillists/archives/address-policy-wg/2008/msg00 293.html - A transfer system can not ensure a process that is open, transparent and equitable Given any distribution scheme for a scarce resource, some people will always be disappointed. I'm not aiming for a perfect solution because there isn't one. Protocol DESIGN requires perfection, protocol REDESIGN requires pragmatism. - Negative impact on routing tables Any increase in the efficiency in which we use the v4 address space will increase fragmentation and thus have an impact on the routing table. This is unrelated to transfer policies. - Keep demonstrated need criteria for address space by receiving LIR See above. But again, all of this is very interesting stuff but has little direct relation to the text of 2007-08 or the rationale behind it. Best, Remco "The hottest places in Hell are reserved for those who, in a time of great moral crisis, maintain their neutrality." - Dante Any opinions expressed in the email are those of the individual and not necessarily of the company. This email and any files transmitted with it are confidential and solely for the use of the intended recipient and do not constitute an offer or acceptance by Equinix, Inc., Equinix Europe Ltd or any of their group entities to buy or sell any products or services in any jurisdiction. If you have received this email in error please delete this email immediately and notify the IT manager. This communication is sent on behalf of one of the European entities in the Equinix, Inc. Group. The ultimate holding company in Europe is Equinix Europe Ltd whose registered address is Quadrant House, Floor 6, 17 Thomas More Street, Thomas More Square, London E1W 1YW and the Company's registered number is 6293383. The registration details of other Group entities are available at www.eu.equinix.com

Michael Dillon wrote:

In your long list of points that you answered you forgot the most important one. IPv6 has lots of free addresses and it is increasingly better supported by vendors. Organizations who are concerned about an address shortage should not put their hope in transfer policies but should instead deploy IPv6 as soon as possible, and put tremendous pressure on vendors to fix the remaining issues.

I didn't forget that - I just think it's irrelevant for this discussion. If you need v4, you need v4. Given the fantastic migration options for moving from v4 to v6, you'll need v4 until you don't depend on it anymore to have your traffic delivered. Migrating to v6 means that you either run dual stack for a long time or have somebody else run dual stack for you. Best, Remco Any opinions expressed in the email are those of the individual and not necessarily of the company. This email and any files transmitted with it are confidential and solely for the use of the intended recipient and do not constitute an offer or acceptance by Equinix, Inc., Equinix Europe Ltd or any of their group entities to buy or sell any products or services in any jurisdiction. If you have received this email in error please delete this email immediately and notify the IT manager. This communication is sent on behalf of one of the European entities in the Equinix, Inc. Group. The ultimate holding company in Europe is Equinix Europe Ltd whose registered address is Quadrant House, Floor 6, 17 Thomas More Street, Thomas More Square, London E1W 1YW and the Company's registered number is 6293383. The registration details of other Group entities are available at www.eu.equinix.com

Current policies only work because there's a carrot associated with good behaviour (allocations). When there is no more carrots there's no

Per Heldal wrote: point

trying to make any kind of rules unless we're prepared (technically and legally) to enforce them. Thus 2007-08 is pointless. We could just as well turn RIPE's IPv4 WHOIS registry into some kind of best-effort-run self-service-robot. All that remains is a requirement to be able to identify the user of any address-resources that IANA has allocated to RIPE.

In your opinion, once we run out of space, any policy at all is pointless. I do not agree. Most people don't need a carrot (or a stick, for that matter) to cooperate nicely. What we do need is some non-intrusive rules that are easy to adhere to. The need for an accurate and usable database will remain, combined with an organisational structure for maintaining that. We've got the NCC doing that now - I think we'd like to keep that going forward.

What I meant was to obtain reasonably reliable numbers which show how much address-space might become available for re-use given various schemes and/or pricing. I.e. for the RIRs to perform surveys towards existing LIRs and known legacy-allocation-holders as businesses analyse their markets. Policies and work to deal with a transfer-market is a waste of time and resources unless numbers are significant. Growth is king, and if v4 can't provide for growth it'll go elsewhere.

I would very much like to see some figures as well, but is it relevant for the decision at hand? We have no policy for registering address space other than 'fresh' allocations, and we're going to run out of that space. So we need some replacement policy before that time. Or be content that as far as v4 is concerned, we can burn the database on a DVD and use that as the final distribution of IPv4, ever. I don't think that's what we want, hence 2007-08.

Except the suggested policy explicitly removes any regulation on the receiving LIR. Now you can't just get an initial allocation, but have the blessing to acquire 1000s with no questions asked until the existing requirements to document continued need of current blocks kicks in. Do you expect the NCC to go after those in retrospect, and that it will easier to enforce than a requirement to document need-for-space prior to the acquisition?

Well, LIRs can already do that when they buy another LIR. Which is a cumbersome transaction and not very transparent from a registration perspective. If you want a genuine 'need-for-space' we should reclaim all IPv4 space today and have IANA hand out addresses on a daily basis, DHCP-style. It will of course kill aggregation and a host of other things, but it's as far as you can go from the conservation perspective. If you think that without enforcement there can be no rules, I'd hate to live in your neighborhood. The reward for good behaviour is of course, that your neighbors don't consider you to be an a.....e. The Internet is an infrastructure based on cooperation - appearances count. Best, Remco Any opinions expressed in the email are those of the individual and not necessarily of the company. This email and any files transmitted with it are confidential and solely for the use of the intended recipient and do not constitute an offer or acceptance by Equinix, Inc., Equinix Europe Ltd or any of their group entities to buy or sell any products or services in any jurisdiction. If you have received this email in error please delete this email immediately and notify the IT manager. This communication is sent on behalf of one of the European entities in the Equinix, Inc. Group. The ultimate holding company in Europe is Equinix Europe Ltd whose registered address is Quadrant House, Floor 6, 17 Thomas More Street, Thomas More Square, London E1W 1YW and the Company's registered number is 6293383. The registration details of other Group entities are available at www.eu.equinix.com