Hi again, I was asked off-list if I could provide a unified diff showing the changes between version 1.0 and 2.0 of the proposed policy text. Attached! Tore * Tore Anderson

I would like to stress that version 2.0 of the proposal brings no substantial changes. The effective proposed policy remains the exact same. The changes since version 1.0 are all either editorial (language improvements, etc.), further removal of "dead"/obsolete policy text that was overlooked in version 1.0, and ensuring that the proposal does not

On Thu, Jul 4, 2013 at 1:48 AM, Tore Anderson <tore@fud.no> wrote:

Hi again,

I was asked off-list if I could provide a unified diff showing the changes between version 1.0 and 2.0 of the proposed policy text. Attached!

Excellent, that made it much easier to see the changes. I support this version. -- Jan

Hi, I also support it. Regards, Elvis Excuse the briefness of this mail, it was sent from a mobile device. On 12 jul. 2013, at 02:20, Randy Bush <randy@psg.com> wrote:

tore,

you did not succeed in making the proposal unappealing. so i still support it.

randy

Support this wholeheartedly. // Andreas Med vänlig hälsning Andreas Larsen IP-Only Telecommunication AB| Postadress: 753 81 UPPSALA | Besöksadress: S:t Persgatan 6, Uppsala | Telefon: +46 (0)18 843 10 00 | Direkt: +46 (0)18 843 10 56 www.ip-only.se Den 2013-07-12 02:19 skrev Randy Bush <randy@psg.com>:

tore,

you did not succeed in making the proposal unappealing. so i still support it.

randy

Dear colleagues,

The second version of the draft document for the policy proposal 2013-03 has been published, along with an impact analysis carried out by the RIPE NCC.

Support. D. Airspeed Telecom

While I expressed support for this proposal at the last RIPE meeting, and still have some sympathy for the proposal, I think valid objectionsn were raised at the last meeting. . As the free pool is gone - and there are no more 'unused addresses' to preserve - there is still a need for responsibe use of the IPv4 pool. Conserving the public resource of v4 is important to make the Internet work while migrating to v6. It was unfortunate that this discussion was cut short at the last meeting - as the original meeting plan allowed ample time for discussing this. So I do belive further duiscussion and carefull reflection of the consequences of this proposal at the next meeting is needed. Moving forward and concluding over summer holidays on such an important matter is not appealing to me. Hans Petter On Tuesday, July 16, 2013, Daniel Stolpe wrote:

On Mon, 15 Jul 2013, Donal Cunningham wrote:

Dear colleagues,

...

...

The second version of the draft document for the policy proposal 2013-03 has been published, along with an impact analysis carried out by the RIPE NCC.

Support.

D.

+1.

Daniel Stolpe

______________________________**______________________________** _____________________ Daniel Stolpe Tel: 08 - 688 11 81 stolpe@resilans.se Resilans AB Fax: 08 - 55 00 21 63 http://www.resilans.se/ Box 13 054 556741-1193 103 02 Stockholm

-- Hans Petter Holen Mobile +47 45 06 60 54 | hph@oslo.net | http://hph.oslo.net

Hello Tore, While I am curious for Michele's response to your mail too, here is my bit, as I feel I agree with him in terms of principles. First of all, I am not totally against your proposal; it cleans up the policy text reflecting today's circumstances to some extent and it makes a long document way more readable. This is also why I am writing back because I've read the draft policy document from the beginning to the end. My observation is that the new document proposed is not exactly a "registration" policy document. It more looked to me like a description of how address space management is done within RIPE NCC region "by the RIPE NCC". If I am not missing anything crucial, the main points described are: - The language is English, - How big an allocation can be, - That there is no PI anymore to be assigned directly from the NCC pools to End users (except for the IXPs) so all resources will only goto LIRs - And these blocks can be transferred between LIRs. The only bit about registration I see in the new text is section 4.0 Registration Requirements and it does not go more than saying details should be recorded in the database. So it does not contain any substantial information for registration or address management on the LIR's side. This is interesting as now with this proposed policy any End user's chance to get any IPv4 address space will be through an LIR and hope that these LIRs are responsible and know what they are doing. I would like to see some guidelines or at least principles mentioned in this document so the LIRs know their responsibility in terms of fair address management as well as the End Users so they know what to expect from these LIRs. This is what I would be expecting from a transparent documentation of a set of policies and principles that are still in place. We may not have too many specific policies to set for the few left-over resources but I would like to believe we still have "principles" towards the responsible management of these resources. In that sense Michele has a point and I argue that LIRs need to be guided for "good address management" even without the "conservation" principle as the top priority in the new IP world. This is missing in the proposed policy text for it to be considered as a helpful "registration" policy in my opinion. In practice, I can set up a new LIR now and ask for a new allocation and I may be someone who does not have any previous RIPE or RIPE NCC experience. If all I have is this document, I am not sure if it tells me enough about my responsibilities, while I will be a critical token in the EU address management and registration system by just becoming an LIR. My other concern is in regards to the transfers. As neatly put by the NCC Board in the Impact Analysis: --- As mentioned in previous sections, the policy proposal would negatively affect the ability of LIRs to engage in inter-RIR transfers, as the RIPE NCC’s service region would be the only one without a needs-based requirement for transfers. Implementation of the policy could expose LIRs to legal challenges under EU competition law. --- I think singling out the RIPE NCC region in the world of transfers may not be the best idea at this stage. Kind regards Filiz Yilmaz On 24 Jul 2013, at 10:52, Tore Anderson <tore@fud.no> wrote:

* Michele Neylon - Blacknight

...

As previously stated, I do NOT support the "no need" policy and cannot support this document.

IP addresses are a finite resource, as we all know, and obliging people to provide some level of justification makes sense.

The argument for "conservation" may no longer be valid, but there will always be a compelling argument in favour of good resource management, which I believe the policy covers.

RIPE should not remove the requirement to provide justification.

Hi Michele,

I doubt you'll find anyone in the working group who is against good resource management. I am convinced that the proposed policy is not in conflict with good resource management, otherwise I would never have proposed it. While I can obviously only speak with certainty for myself, I assume that the people who support the proposal feel the same way.

While it appears you believe that the proposal will bring about poor resource management, your message neglected to explain why or how. This makes it rather difficult for me to try to alleviate your concerns.

As Gert also pointed out recently, the main reason I believe that IPv4 would continue to be consumed responsibly under the proposed policy, is that the LIRs in the region are painfully aware that there is no more IPv4 to be had from the RIPE NCC. Should an LIR anyway decide to go on a "spending spree" with its remaining inventory, it would only end up hurting itself by expediting its own depletion date. The community will not be impacted - without a Common, there can be no Tragedy.

Best regards, Tore Anderson

Hello Tore, Thank you for your detailed e-mail. In order not to have people get lost between our inline comments, I will only attempt to respond to you taking quotes from your mail. Apologies if this distorts the logical thought process you had put in.

The language regarding registration requirements is not changed, so the proposal does not change anything - it merely upholds the status quo.

Those principle based registration and address management requirements were in the old policy between the lines, pointing to people notions like anti-hoarding practices through "need based" requirements. They may not ensure certain things 100% (policy is a policy, some abide some work around…) but surely they serve as gate-keepers for good behaviour and reflect the principles that this community engaged itself so far. So once you take those bits out, you also take those notions and principles out of the policy. That was my point and so I disagree that your proposal does not change anything. In fact it is bringing a big change, changing address delegation from "I want it because I need it on a network" to just "I want it and I am a member of the RIPE NCC so I can get it and maybe i won't even use it on a network".

What constitutes "good address management" is something left to the LIRs to decide for themselves.

I tend to disagree with this too. LIRs are members of the RIPE NCC but policies are developed by the entire RIPE Community, which is a bigger stakeholder. Those views should be reflected in the policy to be remembered and documented by the LIRs.

Example: As a LIR, you are today free to assign your last free IPv4 block to a group of spammers at the expense of having to turn down the Red Cross, perhaps because the spammers were willing to pay you more. Is that "good"? I'd say quite the opposite, but yet it is perfectly allowed by today's policy.

I see your point but policy does not mean "policing". To my knowledge, RIPE NCC never sent agents to LIR premises to check if the justification they documented fit with the reality or not. Besides that the address space is assigned to Red Cross or to a spammer (so who the recipient is) is irrelevant and has always been in the eyes of the RIPE policy. But whether or not either of these networks really needed these resources was the main idea. And this still does not mean that policy is not going to be there to remind people that this is a common pool of resources we are dealing with and its management requires some degree of responsibility. We might as well then not have a policy document at all and just have a click, pay and get your resource button on the RIPE NCC website.

So while I'm all for discussing the codification of a set of principles in our policies telling our community to be "good", I'd rather not weigh down this proposal with the addition of a brand-new "principles" section

These are not brand new principles, need base was there and your proposal is taking this away. I agreed with you when you presented your proposal in Dublin and noted it is not efficient to request someone to breakdown their need in such detail as 3 months to 1 yr or so. Yes, agreed, this is bureaucratic, admin resource hogging etc and this needs a change. But the same policy should still make some sense in a region why an LIR would go request some resource from their RIR and would be given that in the first place. If I am an LIR and I go to the RIPE NCC and ask for a resource I should be needing it on *some* network, not to put in my IP collection drawer. Policy should be the guard, gate-keeper for this at the least, in my opinion.

...

* Implementation of the policy could expose LIRs to legal challenges under EU competition law.

If an LIR is willing to engage in unlawful anti-competitive practises, it will of course be exposed to legal challenges. This is certainly nothing new brought on by 2013-03 - the law trumps *any* address policy we can produce here, and that's how it's always been and always will be.

I am not a lawyer, but hypothetically speaking, if your proposal gets accepted there is some (again totally hypothetical…) potential that some LIRs may chose to rush and get whatever is left in the NCC pool (which is the rest of the last /8…) without really needing them, simply because they do not need to show any justification after all. They want it, they will get it and probably RIPE NCC will have to deal with some very serious First in First Out (granted) service scheme... And then these LIRs who were lucky to get all the space they wanted and the RIPE NCC could provide to them at the time, may chose to put these resources back in the market for others, end users or other LIRs. I would find this quiet an unfortunate environment for new entrants who are not LIRs yet at the time of your proposal gets accepted, but who really need the space and is left to get the resources from an LIR instead of an RIR whose job was the allocation of the resources in the first place. I believe thinking about the impact here in the light of competition laws like this may in fact have some ground. Kind regards Filiz Yilmaz On 24 Jul 2013, at 15:21, Tore Anderson <tore@fud.no> wrote:

* Filiz Yilmaz

...

My observation is that the new document proposed is not exactly a "registration" policy document. It more looked to me like a description of how address space management is done within RIPE NCC region "by the RIPE NCC".

If I am not missing anything crucial, the main points described are:

- The language is English, - How big an allocation can be, - That there is no PI anymore to be assigned directly from the NCC pools to End users (except for the IXPs) so all resources will only goto LIRs - And these blocks can be transferred between LIRs.

The only bit about registration I see in the new text is section 4.0 Registration Requirements and it does not go more than saying details should be recorded in the database.

So it does not contain any substantial information for registration or address management on the LIR's side.

Hello Filiz,

The proposal is a modification of the existing policy document (ripe-592), it is not a brand new document written from scratch. The language regarding registration requirements is not changed, so the proposal does not change anything - it merely upholds the status quo.

We can certainly discuss amending the current registration requirements, but to me this topic does not seem germane to the discussion of 2013-03.

...

This is interesting as now with this proposed policy any End user's chance to get any IPv4 address space will be through an LIR and hope that these LIRs are responsible and know what they are doing. I would like to see some guidelines or at least principles mentioned in this document so the LIRs know their responsibility in terms of fair address management as well as the End Users so they know what to expect from these LIRs. This is what I would be expecting from a transparent documentation of a set of policies and principles that are still in place.

We may not have too many specific policies to set for the few left-over resources but I would like to believe we still have "principles" towards the responsible management of these resources.

In that sense Michele has a point and I argue that LIRs need to be guided for "good address management" even without the "conservation" principle as the top priority in the new IP world. This is missing in the proposed policy text for it to be considered as a helpful "registration" policy in my opinion.

I have no problems being sympathetic towards a "good address management" principle, but the simple fact is that this is not written down in our current policy. What constitutes "good address management" is something left to the LIRs to decide for themselves. Example: As a LIR, you are today free to assign your last free IPv4 block to a group of spammers at the expense of having to turn down the Red Cross, perhaps because the spammers were willing to pay you more. Is that "good"? I'd say quite the opposite, but yet it is perfectly allowed by today's policy.

So while I'm all for discussing the codification of a set of principles in our policies telling our community to be "good", I'd rather not weigh down this proposal with the addition of a brand-new "principles" section, which I suspect would require a lot of back-and-forth word-smithing to make everyone happy. So if we do want such a section, let's add it in a separate proposal.

...

In practice, I can set up a new LIR now and ask for a new allocation and I may be someone who does not have any previous RIPE or RIPE NCC experience. If all I have is this document, I am not sure if it tells me enough about my responsibilities, while I will be a critical token in the EU address management and registration system by just becoming an LIR.

As a new LIR, all the IPv4 address space you are going to get from the RIPE NCC is a /22, or 1024 addresses - hardly a «critical token in the EU address management and registration system» by any stretch of the imagination. This is the status quo today, and this proposal merely upholds it.

I also find it rather hard to believe that an organisation willingly joins the NCC to become a new LIR, acquires an IPv6 allocation, and finally requests its first and only IPv4 /22 - only to find itself confused about how to go about using it in a sensible manner and end up assigning it away to an end user without any actual operational need.

...

* As mentioned in previous sections, the policy proposal would negatively affect the ability of LIRs to engage in inter-RIR transfers, as the RIPE NCC’s service region would be the only one without a needs-based requirement for transfers.

I feel this statement is somewhat disingenuous. The "odd RIR out" here is really ARIN, who is the only one to have a inter-RIR transfer policy that has a needs-based requirement that is also applied to the other region involved in the transfer.

The RIPE region does not have a inter-RIR transfer policy *at all*. The same goes for LACNIC and AfriNIC, AFAIK. So as things stand today, 2013-03 cannot possibly "negatively affect the ability of LIRs to engage in inter-RIR transfers" because such transfers are completely verboten in the first place.

That said, there has been an inter-RIR proposal (2012-02) on the table for a while now, but the community does not seem to want or care much about it. I think that this is important to keep that in mind when deciding on how much weight to give this argument against 2013-03.

APNIC does have a inter-RIR transfer policy, but it does not require the other region to have a needs-based requirement. It explicitly states in section 4.3 that for transfers where the recipient is in another region, any conditions on the recipient is up to the recipient's region to define. So (assuming for a second that 2012-02 has passed) 2013-03 will not have any negative impact on transfers between the APNIC and RIPE regions.

Another quite interesting thing to consider here is that APNIC initially had a transfer policy that did not require any need evaluation. It was only after the ARIN community changed their inter-RIR policy to explicitly require the other region to have "needs-based" policies APNIC added the need evaluation requirement to their general transfer policy. I think it is interesting to consider if yielding to the ARIN demand would actually be worth it, and APNIC actually provided some relevant data in this regard - on May 15th 2013, one of their reps explained to the APWG session at RIPE66 that there had been 11 inter-RIR transfers between ARIN and APNIC. APNIC depleted their IPv4 pool on the 19th of April 2011. Contrast this to the number of assignments that are being made in our region: The NCC reported that between depletion on the 14th of September 2012 and May 10th 2013, 251,254 assignments had been registered in the RIPE database.

If we do linear extrapolation of the above to get "per year" numbers, what this boils down to is that we ask the entire RIPE community to fill out, validate, and archive 386,327 ripe-583 forms per year to allow 5 LIRs to engage in inter-RIR transfers with the ARIN region. You be the judge if this passes your idea of a basic cost-benefit analysis. It certainly doesn't pass mine.

I'm not at all principally opposed to inter-RIR transfers, so if the author of 2012-02 would add a "need compatibility clause" to the proposed inter-RIR policy, I would have nothing against that. In other words, if it would be possible to make the need evaluation a voluntary thing that you only had to do if transferring addresses from the ARIN region - fine. Then those 5 LIRs, and those 5 only, could subject themselves to whatever demands the ARIN community might have, while the rest of us would be free of the IPv4 bureaucracy. Win-win.

...

* Implementation of the policy could expose LIRs to legal challenges under EU competition law.

If an LIR is willing to engage in unlawful anti-competitive practises, it will of course be exposed to legal challenges. This is certainly nothing new brought on by 2013-03 - the law trumps *any* address policy we can produce here, and that's how it's always been and always will be.

Furthermore, it makes absolutely no sense to me for the community to demand that every single LIR and End User in the community uphold the "need" bureaucracy in an attempt to somehow shield or indemnify "bad" members of the community engaged in anti-competitive practises from legal repercussions.

...

I think singling out the RIPE NCC region in the world of transfers may not be the best idea at this stage.

This "world of transfers" has failed to materialise. The actual statistics show that the second-hand IPv4 transfer market is at most supplying 3-4% of last year's (pre-depletion) demand in our region. If we had completely removed the entire transfer policy today, I think that as far as the internet is concerned, tomorrow would have been business as usual. It seems that for most folks dealing with IPv4 depletion, CGN is the solution, not transfers.

I think Rob Blokzijl hit the nail straight on the head when he in the RIPE66 APWG session warned against getting bogged down with discussing the ins and outs of «little add-on policies like transfer». After all, this proposal is a large and rough clean-up - it will be much easier to fine-tune and polish the remaining policy afterwards (transfers, PI/PA merging, etc.). Like I've said before, this proposal is essentially the amputation of the policy limbs that died following IPv4 depletion, and any following cosmetic surgery is best left for later.

Best regards, Tore Anderson

Hello Tore and Gert, According to the current policy your main reason to request an IP address must be that you need it to use on a network. As soon as you remove this, that the IP address is actually needed on a "network" (network can be LIR's or End Users, does not matter), you make them a commodity. As soon as you do that within your policy, you stamp that they actually are money making assets to be turned into some value now or later and this practice is all fine according to the policy too. This means an LIR can request these resources to do whatever they want to do with them including requesting them not to be used on any network of their own or an End Users' at all but for some other reason. One of these reasons will be deliberately passing them on to a paying 3rd party who may or may not use them on any other network either. I think we all see this happening as one scenario. I do not think this is good practice or responsible address management. This is how I see this proposal will have an impact. I believe with this we are moving far away from a core principle; that we, as the RIPE Community, are dealing with networks and resources for networks. If you take the network notion from the address management policy, you are then dealing with some numbers separated with few dots with monetary value only. Rest assured I am aware of the developments in the industry and yes I know how the last /8 policy works too. In fact I was working at the RIPE NCC when it was at its earliest proposal stage and worked on it at various levels, from proposal, Impact analysis and resulting policy documentation within and across RIR regions. I remember there was also a strong argument in these circles, not too long ago, that as we reach the end of IPv4 pool, the more policy changes we make the more it will be disruptive to the system overall. And so in fact I am curious what is the rush of this proposal now that RIPE NCC is allocating from the last /8 and that soon it will be exhausted anyways and what is wrong with keeping what we have as policy as long as this last bit of /8 lasts. Maybe your proposal will be more suitable once everything has ran out and RIPE NCC does not have anything more to provide other than few recycled/returned (hard to imagine if that will be ever happen…) space. Finally Gert seems to get my point with his example of someone going through the effort of opening 5000 new LIRs. I am not saying this will happen but even someone going to through the effort of opening maybe 2 or three just to get more address space for the wrong reasons should be telling us something that our policy is missing some core notion, that notion being IP addresses are mainly there to be used on networks as soon as they leave an RIR pool. Kind regards Filiz Yilmaz On 24 Jul 2013, at 18:50, Tore Anderson <tore@fud.no> wrote:

Hi again Filiz,

...

In order not to have people get lost between our inline comments, I will only attempt to respond to you taking quotes from your mail. Apologies if this distorts the logical thought process you had put in.

No problem, I'll do the same. :-)

...

Those principle based registration and address management requirements were in the old policy between the lines, pointing to people notions like anti-hoarding practices through "need based" requirements.

Actually, you don't have to read between the lines to find the reason for "need based"/"anti-hoarding". It is explicitly stated in the policy, specifically section 3.0.3: «To maximise the lifetime of the public IPv4 address space».

And this of course made perfect sense. If the LIRs had been free to grab as much as they pleased with no checks and bounds, that would in all likelihood have caused the NCC's public pool of IPv4 space, and by extension the IANA's, to deplete many many years ago. It was all about delaying the Tragedy of the Commons as much as possible. Had we as a community done a much better job with IPv6, we could have avoided it from happening altogether, but that ship has sailed - the Tragedy occurred last September.

...

So once you take those bits out, you also take those notions and principles out of the policy. That was my point and so I disagree that your proposal does not change anything. In fact it is bringing a big change, changing address delegation from "I want it because I need it on a network"

There is a difference between a Service Provider and an Internet Registry. The former operates networks, the latter delegates numbers. Often a single organisation fills both roles, but this is not a requirement, just a convenience. You could easily run an LIR from your laptop or tablet in the local coffee shop - all you need is an e-mail account, really (and money to pay the membership fee). So when it comes to allocations, "need" isn't borne from "I need it on a network", but rather "I need it so I can assign it to an End User".

...

"I want it and I am a member of the RIPE NCC so I can get it

Get it from *where* and *who*, exactly?

...

...

What constitutes "good address management" is something left to the LIRs to decide for themselves.

I tend to disagree with this too. LIRs are members of the RIPE NCC but policies are developed by the entire RIPE Community, which is a bigger stakeholder. Those views should be reflected in the policy to be remembered and documented by the LIRs.

Clearly, if the RIPE Community makes a policy that says X, then the LIRs must do X. But *as of right now*, the policy does not define any principles of "good address management", hence, LIRs are free to do what they wish.

...

I see your point but policy does not mean "policing". To my knowledge, RIPE NCC never sent agents to LIR premises to check if the justification they documented fit with the reality or not.

Indeed, the whole "needs-based" system is entirely built on trust. The NCC is not the IPv4 Police (and I wouldn't want them to be either for simple cost/benefit reasons).

...

And this still does not mean that policy is not going to be there to remind people that this is a common pool of resources we are dealing with and its management requires some degree of responsibility. We might as well then not have a policy document at all and just have a click, pay and get your resource button on the RIPE NCC website.

Again, get that resource from *where* and *who*?

...

If I am an LIR and I go to the RIPE NCC and ask for a resource [...]

...the RIPE NCC necessarily say "sorry, we're fresh out". (Except if you didn't pick up your last /22 yet, see below.)

...

I am not a lawyer, but hypothetically speaking, if your proposal gets accepted there is some (again totally hypothetical…) potential that some LIRs may chose to rush and get whatever is left in the NCC pool [...] And then these LIRs who were lucky to get all the space they wanted

No, this is completely impossible, and to be honest I am starting to wonder if you are familiar with the "last /8 policy" at all, or if you are under the impression that 2013-03 proposes to remove it (it does not). In a nutshell, it is the following:

An LIR can get one (1) /22, no more, no less. Sizing according to "need" has *already been abandoned*; it doesn't matter one bit if the LIR's actual need is for 1 address or 1.000.000.000.000 addresses - the only thing you'll get, ever, is 1.024.

The only way an LIR could possibly get "all the space they wanted", is if they wanted 1.024 addresses or less. But in that case, they would be able to get it under today's policy, too.

Best regards, Tore Anderson

Dear Tore, I am aware of all the practical examples you have provided. You keep saying that everything but one thing is the status quo already but that one thing bares a whole notion and a principle (beyond maths) that I have been trying to express in my previous posts because I simply do care about it. Few points of clarification about my own thoughts so far instead of someone labelling them for me too: - Yes, I already agreed with certain things you are proposing, true. I've explained why. - I know what an allocation and assignment is. - I am not against transfers but I am against the practice of getting space now from the NCC just to be able to transfer it later. I do not call this a best practice and I would not like RIPE Policies encourage it. - But I find transfers extremely useful if they are to bring out some already allocated space out of some drawer. While the current system is based on trust, there is definitely enough of wording for what that trust is to be in the current text. It is my belief that your proposal is taking them out all together and leaving the resulting document a mere description of a system that is an IP market, removing the core essential that IPs are for networks in the first place. This is where my concern lies, not about removing the bureaucracy. You seem to say that you want to change the annoying step 6 in your John Q example, where everything else is going to stay the same anyways. ripe-583 may be an overkill today, agreed. It is just a form and a practice NCC applies though as a result of their interpretation of the policy. If it is a problem ask the NCC folks to come up with a more efficient form or a new scheme or totally have it removed even. Policy document does not talk about ripe-583 specifically anyway. But your proposal is removing way too many important principles from the policy text in order to fix an operational problem. In other words, the way it is written, the resulting text and so your proposal is taking it to the extreme, doing more than just removing the unhelpful bureaucracy in my opinion. This is not a matter of an other proposal that can be made after yours. This discussion is happening to note these about this proposal. I may be just one person who thinks we should still mention in the resulting document that RIPE Community cares about networks, that we do realise that IPs are for networks and that these are expected to be used with care and responsibility. And so this is my input to the current discussion. I hope it is now clear what I am opposing to and what not. Kind regards Filiz Yilmaz On 25 Jul 2013, at 09:32, Tore Anderson <tore@fud.no> wrote:

Filiz,

...

According to the current policy your main reason to request an IP address must be that you need it to use on a network.

This is only the true for *assignments*, not *allocations*. This is a subtle, but important, difference.

...

As soon as you remove this, that the IP address is actually needed on a "network" (network can be LIR's or End Users, does not matter), you make them a commodity. As soon as you do that within your policy, you stamp that they actually are money making assets to be turned into some value now or later and this practice is all fine according to the policy too.

We have *today*, under our current policy:

- An allocation transfer policy that has no restrictions on monetary compensation being exchanged between the parties to a transaction; - A list of «Recognised IPv4 Transfer Brokers» published by the NCC; - A NCC-operated "lightweight self-service IPv4 broker" or "bulletin board" aimed at helping LIRs interested in trading find each other.

2013-03 takes no position whether or not all of this is a "good thing", nor do I. But let's face the facts, it's *already there*.

...

This means an LIR can request these resources to do whatever they want to do with them including requesting them not to be used on any network of their own or an End Users' at all but for some other reason. One of these reasons will be deliberately passing them on to a paying 3rd party who may or may not use them on any other network either.

This is *already possible* today; in fact, what you're describing above is pretty much the core function of an LIR.

Let me try to explain by making an hypothetical example:

Background: John Q. Public wants to make money on IPv4. He has a laptop and and e-mail account, but no networking equipment, nor an internet connection of his own; in fact, he's leeching off his local coffee shop gratis WiFi in order to get online. John then does the following:

1) Incorporates "JQP IPv4 Leasing Company" in his local jurisdiction. 2) Joins the NCC and pays the membership fee, thus becoming an LIR. 3) Requests and receives an IPv6 allocation, ignores it. 4) Requests and receives his "last /8" /22. 5) Sets up an auction or something like it, and finds the end user(s) willing to pay the most for [parts of] the /22. 6) Gets a completed ripe-583 form from the auction winner(s), forwards it to the RIPE NCC (if he does not yet have an AW), and saves a copy in an archive folder. 7) Assigns the address space over to the new end user(s) (i.e., registering ASSIGNED PA inetnum objects in the RIPE database).

So, again, this would be *already possible* and completely legitimate under our current policy. If John manages to get higher income from his customers then what he's paying the NCC in membership fees, then voila! John is making money off IPv4, without having ever forwarded a single IPv4 packet on behalf of his customers.

Again, 2013-03 takes no position whether or not this is a "good thing", but it is the status quo today, so when discussing the proposal, I think we should try to keep on-topic, i.e. discussing what the proposal actually *changes*, which in the above case is the removal of point #6 and nothing else:

John's customer won't have to fill out the ripe-583 form anymore. This is a good thing at least in one way, because it removes a bureaucratic overhead that would otherwise impose a work load on John, John's customer, and the RIPE NCC. I think you referred to it as "admin hogging" in an earlier message and I couldn't agree more.

Your concern is that by removing point #6, it becomes possible that the winner(s) of John's auction does not "need" the space, yet are able to receive it. I concede that this is theoretically possible. But I remain unconvinced that this is an *realistic* problem - I simply don't see why someone would bid on address space that they don't "need", especially when they have to outbid a bunch of others who *do* have potentially quite desperate "need".

And even if such actors do exist and do outbid all the folks that do "need", keep in mind that *today* 1) the entire system is trust-based and that 2) John has absolutely no incentive to undertake a deep investigation into whether or not his customer's ripe-583 form is truthfully filled out or not - if it looks credible enough, his job is done. He could also make a habit of forwarding the forms to the NCC to get "official" stamps of approval.

...

I do not think this is good practice or responsible address management.

Maybe, maybe not - but it is what we have today. If we want to ban paid IPv4 allocation transfers and assignments because it's these do not constitute "good practice or responsible address management", then we as a community can do that, but please let's keep that discussion in a separate proposal.

...

And so in fact I am curious what is the rush of this proposal now that RIPE NCC is allocating from the last /8 and that soon it will be exhausted anyways and what is wrong with keeping what we have as policy as long as this last bit of /8 lasts.

Your definition of "soon" differs wildly from mine, or you haven't done the math. Allow me:

- Number of /22s in 185.0.0.0/8 (sans the /16 for IXPs): 16320 - Number of LIRs that held allocations outside 185/8 at the time of depletion (these are the only ones who can get their last /22 as an additional allocation): 7912 - Number of reclaimed space held by the NCC as of today, in /22s: 903 - Number of reclaimed space held by the IANA as of today, divided by 5 RIRs, in /22s: 3731

In other words, there is a total of 13042 /22s to be given out to new LIRs under the last /8 policy (and this number keeps increasing as the IANA and the NCC continues to recovers space). As of today, there are 959 LIRs who hold *only* a /22 from 185/8. 314 days have passed since IPv4 depletion.

So how long will the last /8 last us based on the current usage rates and available space?

13042 / (959 / 314 * 365) = 11.7 years

I truly hope that we've gotten IPv6 properly out the door by then...

To answer the second question, "what's wrong with keeping what we have": About half of the current policy text is describing out of date concepts, is self-contradictory or otherwise wrong, and that it imposes a bureaucracy on the community that demands that hundreds of thousands forms be filled out every year, all in the name of delaying something which has already happened.

I mean, you even said it yourself in an earlier message: «Yes, agreed, this is bureaucratic, admin resource hogging etc and this needs a change» - and I could not agree more, this is *exactly* why I have proposed 2013-03.

...

Finally Gert seems to get my point with his example of someone going through the effort of opening 5000 new LIRs. I am not saying this will happen but even someone going to through the effort of opening maybe 2 or three just to get more address space for the wrong reasons should be telling us something that our policy is missing some core notion, that notion being IP addresses are mainly there to be used on networks as soon as they leave an RIR pool.

I believe Gert's point was that you can open 5000 LIRs in this manner *today*, under our *current* policy. 2013-03 brings no change. Therefore, this is not a valid argument against 2013-03.

If you want to close this loophole, then that's food for another policy proposal. Opposing 2013-03 won't do anything about it, one way or the other (nor would supporting 2013-03).

(That said, in the current data there's no evidence of someone attempting to use this loophole in the described manner. Personally I believe that the NCC's membership fee is a sufficient deterrence against it.)

Tore

* Tore Anderson

...

I may be just one person who thinks we should still mention in the resulting document that RIPE Community cares about networks, that we do realise that IPs are for networks and that these are expected to be used with care and responsibility.

I can tell you with absolute certainty that you are not the only one. This statement describes me, too, and for the record I operate both a network and an LIR.

Uhm, apologies, I misread your original statement there Filiz. So to clarify, I do *not* think 2013-03 needs to add a statement saying that the RIPE Community cares about networks and so forth. After all, the current document makes no such statement either. What I thought you said first was that *you* was one person in the RIPE Community that cares about networks and so forth. That was what I wanted to agree with because I care about networks too. :-) In any case, I would be quite willing to discuss a separate proposal, independent from 2013-03, that added a statement along those lines. Tore

...

I am not saying this will happen but even someone going to through the effort of opening maybe 2 or three just to get more address space for the wrong reasons should be telling us something that our policy is missing some core notion

the issue is not policy. the issue is that ipv4 is essentially gone and folk will do whatever they can to get that last pathetic little bit. we should not be writing kinky and complex policies to try and prevent every bit of stupid greed we can fantacize. ipv4 space is gone but we will get to live with the silly policies and then have to clean them up as we are having to do now. randy

On Jul 25, 2013, at 6:42 AM, Nick Hilliard <nick@inex.ie> wrote:

Also they can still serve a useful purpose as the registries of allocated space.

Well, assuming they don't shoot themselves in the head by attempting to use their registries as a weapon against their customers.

With regard to 2013-03, we know that:

1. the RIRs will no longer be the dominant source for injecting liquidity into the available addressing pool

Agreed.

2. demand for ipv4 address space over time will continue to increase

Demand for IPv4 will increase as long as the price is competitive relative to the alternatives. Right now, the price of IPv4 is wildly distorted by the regulatory regime imposed by the RIRs so stuff like CGN and IPv6 is (seen to be) too expensive. At some point, the regulatory regime that is 'injecting liquidity' _IS_ going to be unable to do so, and the price of IPv4 is going to skyrocket with a corresponding decrease in demand as alternatives become more appealing. The only question is when.

3. there will be a market for ipv4 address space.

There already is.

The ipv4 address market will work if the market stays liquid, but 2013-03 creates an environment for full deregulation of the addressing market with almost no control mechanisms for handling problems

AFAICT, 2013-03 simply says "get out of the way and let the market work." The main concern appears to be that Evil Greedy Speculators will descend upon RIPE like locusts, gobbling up the remaining RIPE IPv4 free pool in a blink of an eye, and not putting that address space into actual use. I tend to agree that is a risk. However, as has been pointed out, there is little stopping that now (well, other than penalizing honesty). Regards, -drc

On 25/07/2013 16:31, Tore Anderson wrote:

For what it's worth, the motivation behind 2013-03 has nothing to do with "the market" whatsoever.

perhaps it's not the intention, but the consequence of the proposal is full market deregulation.

What this Evil Greedy Speculator needs to do is to register a few thousand LIRs, since each new LIR is entitled a single /22. Today, each of those LIRs must demonstrate "need", but this is merely a formality - if you have an intention of assigning one (1) IPv4 address to some End User within 12 months, you qualify. I seriously doubt that this formality is what has prevented the Evil Greedy Speculators from having done this already.

the current ripe ncc pricing structure means that each IP address has a nominal capex value of €3.70 + third party costs. So the nominal residual value of the remaining .88 /8 is about €55m + a real headache when you realise that you also need to convince some company registration office that they should be ok about 15000 new company registrations if they wouldn't mind thankyouverymuch. Microsoft paid $11 per integer in 2011 for ~660k addresses. So right now, opening up bazillions of LIRs doesn't look like a particularly good option. [that said, maybe we should put forward a policy to ask RIPE to create a restful api for opening new LIRs, and to plump up Axel's medical insurance policy for when he needs to sign his name 15000 times. Just thinking out loud here...] Nick

On 25/07/2013 21:50, Milton L Mueller wrote:

I am detecting a contradiction in your argument: you say excessive costs associated with fees and registrations WILL stop people from hoarding under the current system, which underprices address blocks. But at the same time you are arguing that if we remove needs assessment, droves of "hoarders" will pay excessive and rising costs to buy up large quantities of number blocks they don't really need.

I'm not arguing for/against the policy and any concern I have relates to market speculators sucking up transfers from existing holders rather than new LIR allocations - which by any reasonable analysis is currently too painful for anyone to bother abusing on a large scale.

The most reasonable way to get out of this contradiction is simply to admit that hoarding will have a very limited impact on the remnants of the IPv4 address space due to existing market conditions.

I don't believe any of us is in a position to make an assumption of this form.

The price of v4 blocks will rise as scarcity increases. People who pay that money have strong incentives to move the blocks to people who can use them to generate revenue. Possession of number blocks is widely distributed now, so no one can corner the market.

no-one can corner the market in terms of overall possession of assigned address, but my concern is about speculators reducing liquidity in the market by hoovering up transferred addresses rather than getting lots of tiny new allocations from the lirs. Because there is relatively little liquidity in the market at the moment, this probably isn't very difficult or very expensive to do, yet it could have a significant effect on pricing and availability.

The real hoarders are the legacy holders who are not releasing their addresses as widely as possible now. Why? Because needs assessment prevents them from transaction with many willing buyers. If you were really concerned about hoarding, you wouldn't be talking about some hypothetical future state, you'd be talking about the situation we are in right now. 30% of the v4 address space is hoarded.

I mentioned this as a relevant factor in one of my emails yesterday.

Hoarding is an economic phenomenon, driven by economic incentives. If you want to argue it will take place, don't just assert it, provide a coherent economic analysis.

"Please provide more convincing hand-waving". Right :-) Nick

* Nick Hilliard

The ipv4 address market will work if the market stays liquid, but 2013-03 creates an environment for full deregulation of the addressing market with almost no control mechanisms for handling problems

Hi Nick, I'd like to know: - What kind of problems are you foreseeing? - What kind of control mechanisms exist in today's policy, that are removed by 2013-03, will forestall these problems?

[There is a 24-month reallocation freeze, but I don't think this is going to work because it will not affect the reality of real world allocation transfers. Consequently, people will flout it, and then we will need to remove it from the policy mechanism because it's causing the registry function to break down.]

s/24-month reallocation freeze/requirement to demonstrate need/ Does this argument work equally well? If no, why not?

Liquidity will be generated almost entirely by existing address space holders releasing existing address space into the market. So the interesting question is: just how much slack space is there out there?

Probably impossible to answer accurately. The best indication I could come up with is to 1) compare pre-depletion allocation rates with post-depletion transfer rates, making the assumption that the actual demand has at least not decreased, and 2) compare supply with demand at the RIPE NCC's listing site. Today's numbers: Pre-depletion IPv4 RIR->LIR allocation stats (20110914-20120725): - allocations: 1980 - addresses: 36052992 Post-depletion IPv4 LIR->LIR allocation transfer stats (20120914-20130725): - transfers: 70 (3.54% of pre-depletion) - addresses: 1189888 (3.30% of pre-depletion) RIPE NCC IPv4 Transfer Listing Service stats: - requested: 15043584 - available: 225280 (1.50% of requested) Tore

* Nick Hilliard

On 25/07/2013 15:44, Tore Anderson wrote:

...

- What kind of problems are you foreseeing?

address hoarding,

I would actually claim that this is not difficult to pull off under today's policy either. If you want to set up a LIR to hoard address space, you can team up with someone who does have "need". (There are several buyers on the Listing Service who are seeking large blocks, perhaps a good place to start looking.) Make a proposition to them to loan them any address space you are able to find and buy on the market for free. With their ripe-583 forms in hand, your "hoarder LIR" has the required need to get approval from the NCC to receive allocation transfers. When the time comes that you need the addresses yourself, terminate the loaning contracts and reassign/transfer to yourself as appropriate.

leading to reduction in market liquidity,

The numbers I posted in my previous message suggest that the market is only supplying 3-4% of actual demand (assuming last year's demand is a good indicator of this year's). I'd say that this is not a particularly liquid market to begin with, making it hard to see how any kind of further marginalisation of whatever little liquidity is there now can be a significant issue to the community when looking at the bigger picture.

leading to profiteering.

It's perfectly possible and sanctioned by today's policy for LIRs to make a profit from IPv4. I gave an example to Filiz earlier about how you can, under today's policy, set up an LIR that only exists to lease out IPv4 blocks to whatever End Users are willing to pay the most. We were talking about the /22 allocation from the "last /8" then, but the procedure would work equally well for transfers - simply line up End Users and gather their ripe-582 forms, find space on the market and buy it, assign whatever you bought to whoever is first in line (or willing to pay the most). Lather, rinse, repeat. When a certain allocation has been in your possession for 24 months, you may also choose to transfer it away, or just keep leasing it out of course. Do whatever maximises your profits. You're not in any way breaching current policy, after all, you're essentially just operating an LIR. I'm not saying this is "good" or this is "bad", all I'm saying this can be done today, it's not something new that 2013-03 brings to the table.

...

- What kind of control mechanisms exist in today's policy, that are removed by 2013-03, will forestall these problems?

historically, two things: the presence of policies which make it difficult to transfer large amount of address space from many different sources,

Could you point out which part/passage of the policy you're referring to here? I am under the impression that if you have approval from the NCC for, say, a /8, there's no language in the current policy that makes it more difficult to transfer 16384 /22s from 16384 different sources, compared to transferring 1 /8 from 1 source. [I have absolutely no problems seeing that performing lots of small purchases is wildly more difficult than performing one large one, but for reasons that has nothing to do with any current policy language. Locating all the sellers and striking a deal with all of them, for starters.]

secondly, the fact that IP addresses had little market value due to the fact they were effectively being dispensed for $notmuch by the rirs. I.e. lots of effort for little return.

But - this already changed when the RIPE NCC ran out of space and implemented the last /8 policy last September. I don't see what change is brought by 2013-03 in this regard? Could you point to the changed/removed policy language in question?

...

s/24-month reallocation freeze/requirement to demonstrate need/

Does this argument work equally well?

it's the same argument, which means that we need to understand that it is a bad idea to put any faith in the 24-month reallocation freeze rule as a means of regulating the market, or to believe that the rule will stay there as a long term fixture.

So just to make sure I have not misunderstood you here, you are also accepting the following argument to be equally valid (or invalid) as the one you wrote above? "we need to understand that it is a bad idea to put any faith in the requirement to demonstrate need rule as a means of regulating the market, or to believe that the rule will stay there as a long term fixture." Tore

Hi, Op 25 jul. 2013, om 21:55 heeft Jan Ingvoldstad <frettled@gmail.com> het volgende geschreven:

I'm bewildered, confuzzled, and wide-eyed, yes, nearly astonished, that Tore's points don't seem to penetrate the fog.

I've read this proposal a few times, and as far as I can tell, most if not all of the alleged counterpoints to 2013-03 are _not_ counterpoints to 2013-03, but to something else.

I may have misread the proposal, and therefore I request that Filiz, Michele and Nick please point out exactly which points _introduced_ or _altered_ with 2013-03 they disagree with, so that I, and other confused people, can have a better chance of understanding your arguments.

As chair I can only agree to this. There is a lot of discussion going on, which is a good thing: it shows that the community (which includes all of you) actually cares about this and is involved in the decision making process. This makes chairs happy :-) But I have seen lots of cases where Tore asks people to point out which part of his policy proposal causes concern, and I haven't seen any answers to that. This makes an honest discussion about 2013-03 very difficult. I realise that there are many things in the current policies that people might want to change, but trying to put everything into one policy proposal doesn't work. If you have concerns that are not directly caused by the changes that 2013-03 proposes then please discuss them in a separate thread (and change the subject line accordingly please) and we can turn them into separate policy proposals if necessary. So for the sake of an open, transparent and honest discussion about 2013-03 please include references to existing policy that is removed by 2013-03 or to specific sections of 2013-03 itself. When a need for improvement is identified providing suggestions for text would of course also be appreciated. Thank you, Sander Steffann APWG co-chair

Hi Malcom, and thank you for your detailed feedback! I'll be trimming your message heavily in order to try to respond to your main points in order to avoid doubling the size of the message - if you feel I have neglected to respond to something I ought to have, please let me know.

if we abandon the core objective, as 2013-03 would do, we're basically closing the door on any discussion of what those alternatives might be.

I do not agree with this at all. The IPv4 policy document is a living thing, and anyone may suggest changes to it at any time. For example, the possible addition of a section describing "Principles" has been discussed earlier, and 2013-03 would not prevent this in any way. Indeed, such a proposal could be proposed and adopted independently of 2013-03.

Finally, as noted by the RIPE NCC, this takes away a key message the RIRs have used to justify their independence from government. That may not be sufficient reason to oppose the policy if it is otherwise necessary, but it does seem to me to indicate caution and counsel against acting precipitously.

This is one of those things where I feel the message needs to be updated to match the new realities on the ground. Regardless of any change brought by 2013-03, the RIPE NCC's role in IPv4 address distribution is pretty much over, the "last /8" policy being its final death cramp, so to speak. It was claimed earlier that the proposal would change the NCC into a "pay here to buy your IPv4 block today" type of outlet, and I don't doubt that some government could potentially make similar claims in an attempt to undermine the NCC's role in IPv4 address distribution. However I have full faith in the NCC's ability to adjust its argumentation to explain that this is simply not the case, as its role in IPv4 address distribution is pretty much unchanged by 2013-03. That said, I would be more concerned about keeping or making certain policies for the purpose of "staying in power" instead of due to a belief it is "the right thing to do". If anything would undermine the community's validity, IMHO, it would be this.

However this argument is fallacious. The Conservation policy, even as stated, expresses *two separate* policy objectives: 'fair distribution', and maximising the lifetime of the public address pool. Depletion means that reality has superseded the second objective, but not necessarily the first.

So my first question to Tore is: "Why should 'fair distribution' of addresses between users no longer be considered an overarching objective of IPv4 address management?"

For NCC->LIR and NCC->End User delegations, it has already been abandoned. For example: - LIR A requires an allocation to serve 10000000 nodes - LIR B requires an allocation to serve 1000 nodes - End User C requires an assignment to serve 1000 nodes These would (both with current and 2013-03 policy) get 1024, 1024, and 0 addresses, respectively - even though the NCC currently has enough addresses to give all three every single address they requested. In my view, this is clearly unfair to A and C, but it's what we have today. For LIR->End User delegations, I question that 'fair distribution' has ever existed, or if it ever did - if it does now. I'll demonstrate with another example: Your LIR has a /23 left of unassigned address space. Three End Users are knocking on your door, requesting a /24, a /23, and a /22. All fully justified. How does our current policy guide your LIR to be "fair", let alone *enforce* this fairness?

2. "Fair distribution" establishes a basic goal for IPv4 address management policy. Other policies exist to pursue that goal. If we abolish the goal, we not only abolish those other, secondary policies (which may indeed be out of date)

How does the secondary mechanisms (I'm guessing you're referring to things like the AW, 80% rule, and "documented need") actually ensure "fairness", these days? Going back to the previous argument with the three End Users - the only way "fairness" could work in the first place, is that your LIR could just give out everything that was asked by covering their "expenses" with fresh allocations from the NCC's free pool. But that doesn't work anymore.

but we also deny ourselves the right to introduce new, updated policies to pursue that goal.

3. The proponents of 2013-03 have argued on this list that the existing policy is unsupportable because, in a post-depletion world, RIPE NCC has lost the means of enforcing it. It is certainly true that post-depletion it will have lost the means by which it has traditionally enforced it. But that doesn't necessarily mean that the policy should be abolished: it might alternatively suggest that the required "reality adjustment" is that the community develop new policies that would give the RIPE NCC alternative means of enforcing it in the post-depletion world, where IPv4 addresses continue to be an important resource.

I just don't buy the argument that 2013-03 makes the IPv4 policy immutable. This isn't APNIC prop-103. If 2013-03 turns out to be a terrible mistake, it could even be reverted word by word. Last year, I reverted 2009-03 *precisely* in this manner (proposal 2012-06). If you believe that the current policy provisions that are responsible for enforcing "fairness" are no longer effective (if they ever were), then we might as well take them out completely. If we come up with a clever replacement, then we can discuss that in a separate proposal, but if we fail to come up with an effective replacement, what good will it do us to keep the inefficient stuff around? (Surely not the bureaucratic overhead?)

Certain governments are keen to end the independence of the RIRs and move them from being run by and on behalf the community to being made directly answerable to governments. In practice, this would mean the abolition of RIPE (or at least, its policy making function) and perhaps having the RIPE NCC board appointed by an intergovernmental body, likely the ITU.

2013-03 is specific to IPv4, and does not make any difference to other number resources such as IPv6 and ASNs. While I have no difficulty understanding that governments would have liked to have the RIR role, I guess I fail to see how they could possibly get that role in IPv4 address distribution now. I mean - what addresses would they be distributing? :-) Tore

On Thu, Aug 1, 2013 at 8:38 AM, Tore Anderson <tore@fud.no> wrote:

Having slept on it, I have a suggestion that hopefully will make us reach common ground. How about, instead of removing the old Conservation goal completely, we rewrite it as follows:

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

That looks sensible enough to me, it doesn't appear to add red tape. :) -- Jan

On 01/08/2013 07:38, Tore Anderson wrote:

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

Sounds good to me! One minor nit: can you define "fair"? Nick

On Saturday, July 27, 2013, Randy Bush wrote:

http://www.ietf.org/id/draft-resnick-on-consensus-02.txt

may be of help

definitely for me the following are open issues for me -1 how it will affect inter region transfers - if this is important to the community -2 how will it affect reclamation of address space by RIRs - and is this important -3 how wil it affet the critics of the RIRs 1 - presently it will not be simple to establish an inter-RIR transfer policy with Arin. Personally - based on previous discussion at Ripe meeting and other places - I do not think the Arin communicty will change opinion on this, at least until they run out. This could be an issue as the "hidden reserves" of "legacy" assignments are in the Arin region. 2 - I think reclamation by the RIPE NCC will decay over time. 3 - the traders will be happy - he ITUs and critical governments will probably make more noice. how much of an issue is this to us? I liked the proposal initialy - but I do not buy the initial argument that since we have run out, conservation is not important. Since we have run out of unused addresses conservation is in many ways more important - unless we belive v6 can take over. Maybe this proposal will increase the number of transfers for money - thus putting more focus on conservation (all businesses tend to focus on conserving cash) It is not clear to me that theese aspects are well undrestood by the majority, but ai might be wrong. -hph -- Hans Petter Holen Mobile +47 45 06 60 54 | hph@oslo.net | http://hph.oslo.net

On Aug 1, 2013, at 6:00 PM, Tore Anderson <tore@fud.no> wrote:

APNIC and ARIN happens to share compatible inter-region transfer policies. Yet, the number of inter-region transfers going from the ARIN region to the APNIC region is negligible - at RIPE66 the total amount of transfers was reported to be 11 (5 per annum, if extrapolating linearly).

ARIN Inter-RIR Transfer statistics are available here: <https://www.arin.net/knowledge/statistics/transfers.html> Presently there have been 16 transfers in the period of time of that compatible policy has been in place (approximately 1 year); the blocks transferred range in size from /24 to /15, and are listed on the transfer statistics page.

So the notion that there is some big hidden reserve sitting around in the ARIN region, ripe for the taking and ready to transform our current our state of scarcity into one of abundance (if we only pass an inter-region transfer policy!), seems extremely implausible to me. At least it has not worked out that way for the APNIC region.

I am not asserting any relation between the ARIN/APNIC transfer rate to the desirability for IP transfers to those in the RIPE region, but provide the above information solely for a factual basis of the discussion since the question was raised. FYI, /John John Curran President and CEO ARIN

On 03/08/2013 15:17, Milton L Mueller wrote:

...

I liked the proposal initialy - but I do not buy the initial argument that since we have run out, conservation is not important. Since we have run out of unused addresses conservation is in many ways more important - unless we belive v6 can take over.

I don't understand why people confuse the elimination of needs assessment with the elimination of conservation.

Because they're linked. The requirement of justifying and documenting is a lever for us today, to get serious cooperation from our clients in ordre to get their needs correctly evaluated, and /24s (or more) not being wasted in careless (and preventive) over-subscriptions of adresses. This helps us preventing a rush to our allocation depletion and thus, from requesting the last /22. And thereafter, to create a new LIR to satisfy these needs we would let run. Am I wrong if I suspect commercial teams would happily distribute large bunches of undocumented adresses to their asking customers if the Ripe allowed it. As far as I can see, commercials do appreciate to be able to satisfy their customers requests.

On Sun, Aug 11, 2013 at 03:45:22PM +0000, Milton L Mueller wrote:

But if they have to pay a price that reflects the relative scarcity of the resource they have a much stronger incentive to conserve that the alternative situation you describe. "Commercials" cannot simply provide all the addresses their customers want when they cannot replenish their stock without paying a rising price.

We are talking about 1,75 € per address and per year. I guess this wouldn't stop any commercial from reselling to a client who asks for it, and certainly can pay for it, if conservation is abandonned. Best regards, Sylvain

On 01 Aug 2013, at 19:27, Tore Anderson <tore@fud.no> wrote:

* Nick Hilliard

...

On 01/08/2013 07:38, Tore Anderson wrote:

...

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

can you define "fair"?

Well *that* is the million dollar question, isn't it. In a state of scarcity, what is "fair"?

I do not know it is such a million dollar question. Scarcity (remember the switch from classful delegations to CIDR…) had been a (maybe perceived) issue in the past too. Basically we dealt with it by putting our trust on the RIRs that they would ensure that through policies which bared "justification of the need". In other words, this was always the case, RIRs had been made the judge of this "fairness" through their community built policies. Now the real problem here 2013-03 is suggesting to remove this judgement of fairness role from the RIPE NCC in the RIPE region, by removing the justification of need from the policy, without coming up with a real alternative, while obviously sense of fairness is still important to various members. Filiz

This is a question that neither I, Malcom, 2013-03, nor ripe-592, presume to have an answer for.

However, I understood the crux of Malcom's objection to be that if we remove this stated objective of "fairness", then we lose our principles in the process, and it becomes hard to add it back later (presumably along with a precise definition of "fair").

Therefore I was hoping that retaining this sentence (which is there in today's policy as well) would help move the discussion forward in the direction of consensus.

Tore

Could RIPE NCC chime in on if such a small change with no real world implications would allow for reduced, or no, re-evaluations? I agree that implications and intentions should be laid out clearly, but I agree with Gert even more that at this stage, some feedback can be.. annoyingly frustrating.. Richard Sent by mobile; excuse my brevity.

Hi, On 02 Aug 2013, at 11:56, Tore Anderson <tore@fud.no> wrote:

* Filiz Yilmaz

...

Scarcity (remember the switch from classful delegations to CIDR…) had been a (maybe perceived) issue in the past too. Basically we dealt with it by putting our trust on the RIRs that they would ensure that through policies which bared "justification of the need".

In other words, this was always the case, RIRs had been made the judge of this "fairness" through their community built policies.

Now the real problem here 2013-03 is suggesting to remove this judgement of fairness role from the RIPE NCC in the RIPE region, by removing the justification of need from the policy, without coming up with a real alternative, while obviously sense of fairness is still important to various members.

The relevant community built policy that defines "fairness" at this point in time is the "last /8 policy", with its strict "one /22 per applicant" rationing. This does not go away with 2013-03, so in pragmatical terms, the implementation of "fairness" would remain.

Tore, i do not see how it remains. I gave the example of two new members getting each a /22 if your proposal gets accepted before: One is requesting it to use it on a network soon. The other is requesting it to sell it in 24 months because they cannot sell it before - your proposal leaves this requirement. I dont think this is fair and this will happen if NCC does not ask for justification for need or some word on that a network exists for the IPs to be used on.

The stated philosophical goal of fairness does go away with the current version of 2013-03 though, as Malcom identified. While I think this makes little *practical* difference as long as the "last /8 policy" remains, I concede that on a philosophical level it could be considered problematic to remove the explicit fairness objective from the policy.

This is why I have suggested to roll a new version of the proposal that would leave the fairness goal intact. This way we can maintain both the philosophical objective of fairness and the actual implementation of it, without conflicting with 2013-03's main goals which is to remove the LIRs' bureaucratic overhead and to clean up old and out of date policy text.

I agree with the parts of your proposal that are removing the overhead on LIR level. But I do not see it is much of an overhead to be able to say "i need a last /22 because of this network here.."

Would such an amendment make the proposal more appealing or at least acceptable to you? If not, what else is needed?

Pls see above and few previous mails. Before NCC really runs out and still providing "allocations" I advocate for keeping the justification for need or some kind of commitment from the requester that the space is to be used on a network shortly and not to be sold to a 3rd party. I am happy if the secondary market brings already allocated space to those who need it. But the space that NCC still holds should goto those in need, not to an organisation to sell it to the one in need. Again, I am fine with the other cleanup your proposal is doing: removal of txt on AWs, further allocations etc..

As Gert mentioned earlier, changing the proposal text at this stage of the PDP isn't a trivial thing to do,

I should have missed this but as far as i know a review phase can be repeated or extended with an updated text. Happy to see the new text. And for those who find some of the feedback frustrating - as the word keeps coming up and being used here and there... Well, welcome to a bottom-up process. Ideas and perspectives vary, we have the process in place to hear all those ideas so nobody here is to be hushed or discouraged to speak. From my point of view, there had been views here that I do not fully agree but I encourage them too, because an open discussion for these issues is the right tool and this is what PDP is for. Filiz

but if it turns out to be unavoidable, I would at the very least want to avoid having to do it more than once.

Best regards, Tore Anderson

On Sun, Aug 4, 2013 at 3:41 PM, David Farmer <farmer@umn.edu> wrote:

On 8/1/13 12:27 , Tore Anderson wrote:

...

* Nick Hilliard

On 01/08/2013 07:38, Tore Anderson wrote:

...

...

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

can you define "fair"?

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

Uhm, if that has been the definition of fairness, then it certainly has NOT been enforced or used in the past few years. Perhaps never. It is, as you state later on, one of those hollow "jesters". Furthermore, I believe that now that everyone's operational need can no

longer be meet, a state of scarcity, that fairness is doubly important. How does verified operational need provide fairness in a state of scarcity? If someone without verified operational need were to receive Internet number resources, presumably through a transfer, and you have verifiable operational need that can no longer be meet; it would add insult to injury that someone without that verifiable operational need receives Internet number resources when you can't.

This is how Internet number resources have been handled for years; organizations without verified operational needs have received Internet number resources, some in huge quantities. One could easily argue that this is one of the root problems with former Internet number resource handling. Fortunately, IPv6 came to the rescue.

Therefore, verifying operational need for transfers, still provides some minimal amount of fairness to those that are not going to receive Internet number resources.

If the verification is going to be at the level it used to be, that fairness is so minimal that you can't poke a stick at it. It is, at best, an illusion. While this is a start, just saying fairness is necessary, is a hollow

jester, without verified operational need or some other replacement mechanism to provides the fairness you are saying is necessary.

If you want to introduce verification of the operational need, I suggest you write your own proposal to the RIPE community, so that we can discuss and handle the proposed changes accordingly. But I cannot, for the life of me, understand what this has to do with 2013-03, sorry. -- Jan

On Aug 4, 2013, at 10:59 AM, bmanning@vacation.karoshi.com wrote:

On Sun, Aug 04, 2013 at 04:35:25PM +0200, Jan Ingvoldstad wrote:

...

On Sun, Aug 4, 2013 at 3:41 PM, David Farmer <farmer@umn.edu> wrote:

...

On 8/1/13 12:27 , Tore Anderson wrote:

...

* Nick Hilliard

On 01/08/2013 07:38, Tore Anderson wrote:

...

...

+Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.;

can you define "fair"?

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

This is how Internet number resources have been handled for years; organizations without verified operational needs have received Internet number resources, some in huge quantities.

One could easily argue that this is one of the root problems with former Internet number resource handling.

Fortunately, IPv6 came to the rescue.

Pragmatically, there is zero chance of verification of operational need for anything larger than a /96 in IPv6 space.... So the rules for v6 allocation actually are fairly close to the original v4 allocation policies.

The concept of verified operational need arose in times of scarcity, when there was -no- other option. Those times have past and its not clear to me why there remains this slavish devotion to having an unlicensed regulator second guess the viability of a given operational model.

Hi Bill, Out of curiosity, about how many Class As would you say had been assigned before that scarcity came to be widely recognized, and the concept of "verified operational need" was first articulated? At that time, was promptly doling out all of the remaining Class As on a first first-come-first served basis *not* considered an option? If not, why not? If the prospect of creating an operating environment in which critical number resources were collectively, indefinitely controlled by a relatively small minority of the total number of operators who could such resources to similarly good use (and whose commercial/existential viability would forevermore be contingent on the terms of their access to someone else's number resources) was not considered a viable option back then, what makes it any more palatable now? Granted, both the numerator (resource-haves) and the denominator (have-nots) numbers have grown several orders of magnitude over the ensuing years, but how exactly does that matter if the resulting fraction remains broadly unchanged? Curiously, TV

As usual, YMMV.

/bill

On Aug 5, 2013, at 9:15 AM, bmanning@vacation.karoshi.com wrote:

On Sun, Aug 04, 2013 at 03:35:25PM -0400, Tom Vest wrote:

...

On Aug 4, 2013, at 10:59 AM, bmanning@vacation.karoshi.com wrote:

...

On Sun, Aug 04, 2013 at 04:35:25PM +0200, Jan Ingvoldstad wrote:

...

On Sun, Aug 4, 2013 at 3:41 PM, David Farmer <farmer@umn.edu> wrote:

...

On 8/1/13 12:27 , Tore Anderson wrote:

...

* Nick Hilliard

On 01/08/2013 07:38, Tore Anderson wrote: > >> +Fair use: Public IPv4 address space must be fairly distributed to the >> End Users operating networks.; >> > can you define "fair"? > I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

This is how Internet number resources have been handled for years; organizations without verified operational needs have received Internet number resources, some in huge quantities.

One could easily argue that this is one of the root problems with former Internet number resource handling.

Fortunately, IPv6 came to the rescue.

Pragmatically, there is zero chance of verification of operational need for anything larger than a /96 in IPv6 space.... So the rules for v6 allocation actually are fairly close to the original v4 allocation policies.

The concept of verified operational need arose in times of scarcity, when there was -no- other option. Those times have past and its not clear to me why there remains this slavish devotion to having an unlicensed regulator second guess the viability of a given operational model.

Hi Bill,

Out of curiosity, about how many Class As would you say had been assigned before that scarcity came to be widely recognized, and the concept of "verified operational need" was first articulated? At that time, was promptly doling out all of the remaining Class As on a first first-come-first served basis *not* considered an option? If not, why not? If the prospect of creating an operating environment in which critical number resources were collectively, indefinitely controlled by a relatively small minority of the total number of operators who could such resources to similarly good use (and whose commercial/existential viability would forevermore be contingent on the terms of their access to someone else's number resources) was not considered a viable option back then, what makes it any more palatable now?

Granted, both the numerator (resource-haves) and the denominator (have-nots) numbers have grown several orders of magnitude over the ensuing years, but how exactly does that matter if the resulting fraction remains broadly unchanged?

Curiously,

TV

/8, pre or post classfull addressing? pre-classful, all there was were /8's. Kind of like all there are now are /64s in v6 land.

when ~10% of the total v4 pool was allocated, scarcity triggered the creation of classful addressing, which gave us /16 and /24 space in v4-land.

at ~40% of the total v4 pool, scarcity created CIDR and started on developing v6-land.

I'd suggest that with the viablity of IPv6, that the remaining v4 pool is simply vestigal and its getting far more traction than it deserves. We are clearly not in a position of scarcity when we hand out, to individual devices, the functional equivalent of the -entire- v4 pool, raised to the 32nd power.

That is the most profligate waste of address space I have ever seen and these petty "verified operational use" arguments seem farcical and almost hypocritical in comparison.

/bill

Hi Bill, I agree that, *once* the viability of IPv6 has been established widely, including as a stand-alone alternative to IPv4 for new entrants who have no IPv4 of their own*, IPv4 issues will indeed become vestigial. In the mean time, I interpret your use of "profligate waste" to mean something like "possessing exclusive right to use (of addresses) in quantities that vastly exceed 'operational need,' or capability to put to productive use." I could understand a general distaste for "waste" of this kind on purely aesthetic grounds, but given your visceral reaction I can only assume that you believe that such waste has a cost -- and since exclusive of possession of "profligate" quantities of number resources can be privately beneficial to the possessor under certain conditions, I'm tempted to assume that the costs of profligacy that concern you are the costs that are imposed by the "profligate" on everyone else, in the form of needlessly accelerated scarcity and/or exhaustion of resources that could otherwise be put to productive use by others... If that interpretation is not too far off-base, then I guess you must know something about the state of IPv6 usage* that I don't know -- or maybe, your definition of "waste" excludes anything that has been paid for, even if the purchased/idle stockpiles serve no other purpose than to exacerbate demand/scarcity among others... ? TV

On 8/4/13 09:35 , Jan Ingvoldstad wrote:

On Sun, Aug 4, 2013 at 3:41 PM, David Farmer <farmer@umn.edu <mailto:farmer@umn.edu>> wrote:

On 8/1/13 12:27 , Tore Anderson wrote:

* Nick Hilliard

On 01/08/2013 07:38, Tore Anderson wrote:

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

can you define "fair"?

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

Uhm, if that has been the definition of fairness, then it certainly has NOT been enforced or used in the past few years. Perhaps never.

There has allows been a need component, prior to IPv4 each network "needed" a network identifier, equivalent to a /8 today. With IPv4 pre-CIDR, this changed to three class of network identifiers; classes A, B, and C, /8s, /16s, and /24s today. You had to meet additional criteria to get a class A or B, otherwise you got a C. This is still based on need, but a little more granularity. By today's standards were most of those class As or Bs needed, maybe not, but there were needed by the standards of the day. RFCs 1366, 1466, 2050, document the criteria used and its evolution, until the RIR policy process take over.

It is, as you state later on, one of those hollow "jesters".

They may have also been hollow gestures, but not a hollow as saying that fairness is necessary without and kind of mechanism providing it. (Sorry, I can't spell for shirt :) , see http://tinyurl.com/l3cop8b )

Furthermore, I believe that now that everyone's operational need can no longer be meet, a state of scarcity, that fairness is doubly important. How does verified operational need provide fairness in a state of scarcity? If someone without verified operational need were to receive Internet number resources, presumably through a transfer, and you have verifiable operational need that can no longer be meet; it would add insult to injury that someone without that verifiable operational need receives Internet number resources when you can't.

This is how Internet number resources have been handled for years; organizations without verified operational needs have received Internet number resources, some in huge quantities.

If you mean address were handed out without need by today's standards, sure, but hind sight is allows 20-20. But, if you mean addresses were handed out without need by the standards of the day then I have to disagree.

One could easily argue that this is one of the root problems with former Internet number resource handling.

Fortunately, IPv6 came to the rescue.

I sure hope IPv6 comes to the rescue, but I'm not sure we have been rescued just yet. I'm only able to get ~40% of my traffic via IPv6, this is moving the the right direction but, we're not rescued yet.

Therefore, verifying operational need for transfers, still provides some minimal amount of fairness to those that are not going to receive Internet number resources.

If the verification is going to be at the level it used to be, that fairness is so minimal that you can't poke a stick at it. It is, at best, an illusion.

While this is a start, just saying fairness is necessary, is a hollow jester, without verified operational need or some other replacement mechanism to provides the fairness you are saying is necessary.

If you want to introduce verification of the operational need, I suggest you write your own proposal to the RIPE community, so that we can discuss and handle the proposed changes accordingly.

I'm not introducing anything new, I'm objecting to the removal of something that has always been there, as are others.

But I cannot, for the life of me, understand what this has to do with 2013-03, sorry.

It is in the very title of the proposal "No Need - Post-Depletion Reality Adjustment".

-- Jan

-- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

On 8/4/13 12:15 , Jan Ingvoldstad wrote:

On Sun, Aug 4, 2013 at 6:48 PM, David Farmer <farmer@umn.edu <mailto:farmer@umn.edu>> wrote:

I'm not introducing anything new, I'm objecting to the removal of something that has always been there, as are others.

As I understand it, there has "always" been a requirement for need. I don't disagree that this is nothing new.

However, you used the phrase "verified operational need", which is something different.

I am a n00b in the RIPE community, so I have certainly not read everything there is, but from what I have seen, read and understand, this verification thing is new.

If you wanted more than the minimum it was always necessary to provide a story that passed a smell test. Also, if you wanted another batch of numbers you always had to explain what did with the old ones you were given, usually providing some data, much of the "bureaucracy" 2013-3 wants to eliminate. Verified: 1. Make sure or demonstrate that (something) is true, accurate, or justified. 2. Swear to or support (a statement) by affidavit. Justified: 1. Having, done for, or marked by a good or legitimate reason. 2. Declared or made righteous in the sight of God. So, please tell me how "verified operational need" is that different. If you would prefer justified or validated operational need. OK, what ever. -- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

Jan Ingvoldstad <frettled@gmail.com> schrieb:

On Sun, Aug 4, 2013 at 7:52 PM, David Farmer <farmer@umn.edu> wrote:

...

If you wanted more than the minimum it was always necessary to

provide a

...

story that passed a smell test.

Yes, and that story was, typically, "I'm a new service provider and need an IP address for my service", or "I want to provide addresses for others by becoming a LIR". It may have been written in more words, but it's not like you had to show that you had contracts with others or whatnot.

Actually this is not true, I had to send copies of contacts and approvals that customers are returning space to other LIRs ... BR Jens

On Mon, Aug 05, 2013 at 01:00:37PM +0000, bmanning@vacation.karoshi.com wrote:

persuant to your comments and David Conrads contribution, I'd suggest that the "smell test" criteria has changed dramatically over the years. Nearly all the IPv4 space I have requested has passed the "verified operational need" using the Warren Buffet test. (he's a man of good character) These days it sometimes goes as far as explaining a business model, showing signed contracts, bank statements, customer lists and credit reports on clients.

Interesting. I've never experienced anything like that, but I've not made applications since "Last /8" came in. I should think that asking for bank statements and credit reports would be a data protection issue to say the least. If this is the case (and again, I've not experienced such requests myself) it is all the more reason to strongly support 2013-03... rgds, Sascha Luck

* David Farmer

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

Do you have a link or reference? (Tried Google, no hits.)

Furthermore, I believe that now that everyone's operational need can no longer be meet, a state of scarcity, that fairness is doubly important. How does verified operational need provide fairness in a state of scarcity? If someone without verified operational need were to receive Internet number resources, presumably through a transfer, and you have verifiable operational need that can no longer be meet; it would add insult to injury that someone without that verifiable operational need receives Internet number resources when you can't. Therefore, verifying operational need for transfers, still provides some minimal amount of fairness to those that are not going to receive Internet number resources.

Let me get this straight, are you saying here that our current goal of fairness in our current state of scarcity is to protect the *feelings* of the LIRs who return home from the second-hand market empty-handed? (Currently 96-97% of them.) In a similar vein, and somewhat tongue in cheek: If you're trying to buy an apartment, but get outbid by someone, do you get upset unless you can see some proof that guy did buy it plans on moving in there? Or do you assume that's probably the case and move on? Tore

* David Farmer

On 8/4/13 09:48 , Tore Anderson wrote:

...

* David Farmer

...

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

Do you have a link or reference? (Tried Google, no hits.)

Try goal #1 in section #1 of RFC 2050. See https://tools.ietf.org/html/rfc2050#section-1

Written in a time of abundance long ago, where the fairness of the RIRs' "all you can eat buffet" worked as there was enough for everyone. That's not the world we live in today, and besides the document is about to be superseded by the 2050-bis:

And in slightly different words, try goal #1 in section #2 of RFC 2050-bis. See http://tools.ietf.org/html/draft-housley-rfc2050bis-02#section-2

...which happens to contain exactly 0 occurrences of the string "fair".

...

Let me get this straight, are you saying here that our current goal of fairness in our current state of scarcity is to protect the *feelings* of the LIRs who return home from the second-hand market empty-handed? (Currently 96-97% of them.)

You always have to look at fairness from the perspective of those who don't get what they want. Fairness isn't usually an issue raised by those that get what they want.

Yes, I'm sure that the LIRs that didn't get the resources they wanted (and "needed") feel it's a great consolation to know that their competitors with deeper pockets also "needed" the resources in order to absorb the customer growth they now are unable to take on themselves. Or how starving folks around the globe surely must feel that their situation is "fair", as Tore and David managed to finish all the food on their dinner plates today. (We "needed" it, after all.) Sorry. I just don't at all buy the argument that the unprivileged are treated "fair" just because the privileged happened to "need". Everyone "needs", yet not everyone gets. The way I see it, the scarcity situation is going to result in unfairness no matter what and there's nothing our current policy nor 2013-03 can do about it.

Personally, I'm not fundamentally opposed to operational need going away, especially the current overly bureaucratic way it is determined, if you can find something that replaces it that provides some sense of fairness that you seem to agree is necessary.

Personally I don't think it is at all necessary to keep the philosophical "fairness" goal in the policy, as we have no working mechanism in the policy that actually ensure any form of fair distribution in this time of scarcity. That's my pragmatic view. However there were a couple of other WG participants who felt that it should be left in there for high-level philosophical and political reasons, and to ensure the door is kept open for a potential future discussion of "what is fair nowadays and how do we go about enforcing it exactly"? I have no problem with that per se, hence my offer to leave it in there. Tore

* David Farmer

you asked about "operational need" not "fairness";

[citation needed] I seem to remember my question being: «In a state of scarcity, what is "fair"?» While discussing the amendment: «Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.» And finally asking you to provide a reference to the «primary definition of fairness the RIR communities have been using». So the discussion was never about "fairness"? Hmm. English is only a secondary language of mine, so I guess I must have misunderstood you. Apologies.

1) Allocation Pool Management: Due to the fixed lengths of IP addresses and AS numbers, the pools from which these resources are allocated are finite. As such, allocations must be made in accordance with the *operational needs* of those running the networks that make use of these number resources and by taking into consideration pool limitations at the time of allocation.

So depending on how you look at it, either: A (if ignoring the "last /8 pool"): The RIPE NCC's allocation pool is empty. No point in talking about the Management of an Allocation Pool that does not exist. B (if including the "last /8 pool"): I've proposed to retain the current requirement that LIRs must use the last /22 allocation for making assignments to its End Users. Either way, case closed? Tore

* David Farmer

By asking me for a quotation, I interpreted that as to not only being about "fairness", but also about my tie of "fairness" to "operational need". The concept of "operational need" still remains in RFC 2050-bis even though the "fairness" was dropped. But I think RFC 2050-bis and "operational need" remains relevant.

I wanted to know where you'd seen this "primary definition of fairness". I don't dispute that 2050-bis mentions "operational need", but I do dispute that it says it is the "definition of fairness". If you didn't mean to make that claim I suppose we were talking past each other.

I believe you are incorrectly equating "Allocation Pool" with "Free Pool", there is nothing that says the "Allocation Pool" doesn't include resources that are available for "Re-Allocation".

If already allocated resources are to be considered as part of the allocation pool, then the pool would essentially be infinite and limitless as you can re-allocate the same addresses over and over and over and over again. So IMHO this is quite a stretch.

Furthermore, the statements in 2050-bis apply are intended to apply to the whole Internet Registry System, IANA, the RIRs, and LIRs, so even if you consider RIPE's Allocation Pool empty, the LIR's Allocation Pool isn't empty and operational need should still apply to the LIR making the Re-allocation of resources.

In the RIPE region LIRs don't allocate, they assign. (Sub-allocations being the extremely rarely used exception.) Tore

Hi, On Aug 4, 2013, at 3:17 PM, Tore Anderson <tore@fud.no> wrote:

I don't dispute that 2050-bis mentions "operational need", but I do dispute that it says it is the "definition of fairness".

Just to be clear: "operational need" has absolutely nothing to do with "fairness". The former is, at least in theory, objectively verifiable by, e.g., measuring requester utilization efficiency, customer growth patterns, etc. (ignoring this has never been done and the IANA/RIR system is not capable of doing this -- the entire system relies on trust and passing the giggle test). The latter is a purely subjective value judgement. The intent of goal #1 in 2050bis is to encourage people to look at the free pool availability when considering allocation policies. In cases where the allocation pool is tightly constrained, I believe it safe to say it was the consensus opinion of the authors that "the operational needs of those running the networks that make use of the number resources" should be a key consideration. In my personal view, this should true even if it is considered "unfair" by some.

...

I believe you are incorrectly equating "Allocation Pool" with "Free Pool", there is nothing that says the "Allocation Pool" doesn't include resources that are available for "Re-Allocation".

If already allocated resources are to be considered as part of the allocation pool, then the pool would essentially be infinite and limitless as you can re-allocate the same addresses over and over and over and over again. So IMHO this is quite a stretch.

An "allocation pool" is the source from which resources are taken. Once a resource is allocated, it is removed from the allocation pool. As mentioned in 2050bis, "the pools from which these resources are allocated are finite." It is, of course, true that the allocation pool can be replenished, e.g., when someone returns a block of addresses to some part of the Internet registry system, however that is a relatively rare occurrence. AFAICT, the question 2013-03 revolves around is whether or not the RIPE community considers the free ("allocation") pool exhausted. If it is exhausted, questions of "fairness" or "operational need" are irrelevant in allocation pool management -- the allocation pool size is zero so there is nothing to consider.

...

Furthermore, the statements in 2050-bis apply are intended to apply to the whole Internet Registry System, IANA, the RIRs, and LIRs, so even if you consider RIPE's Allocation Pool empty, the LIR's Allocation Pool isn't empty and operational need should still apply to the LIR making the Re-allocation of resources.

The statements in 2050bis are goals and recommendations of a small number of people interested in address allocation policy, not laws or axioms. One of my reasons for helping with 2050bis was to get away from the "quotations from holy text" approach of address policy making that has infected the community as a result of 2050 and instead, encourage people to evolve policies through "an open, transparent, and broad multi-stakeholder manner". I would personally be distressed to see the Old Testament replaced by the New Testament. Regards, -drc

Hi, Since I did not mention it clearly so far, please let me state *i do not support 2013-3*. Unfortunately the late amendments proposed here to reach a consensus look like cosmetics to me, and do not address many of the objections that were raised here. On 05/08/2013 07:56, Tore Anderson wrote:

...

An "allocation pool" is the source from which resources are taken. Once a resource is allocated, it is removed from the allocation pool. As mentioned in 2050bis, "the pools from which these resources are allocated are finite."

Aren't allocation transfers to be considered as part of this pool, since they are (so far) supposed to be made of empty allocs, and subject to the same conditions than direct allocations from the Ripe NCC ?

The last /8 pool has a very blunt definition of "operational need"; it equates it to 1024 addresses. The "giggle test" applied to this pool is essentially «do you need anything at all? yes/no». The current version of 2013-03 proposes to remove this giggle test (but not the one-size-fits-all definition of "operational need").

That "one-size-fits-all" is a myth because of allocation transfer mechanism in my opinion. And also because of the PI to PA migration projet, BTW. But also 2013-3 does not apply to such a small space, since, as Tore said, in another mail in this thread the "few" returns to the pool since "depletion" amount to about 4,600 /22s (to be compared to the 16,000 remaining ones in the Ripe's pool). Last /8 policy being set in order to deal with scarcity of the public ressource I guess the importance of the ressource is a point everyone admits (and no one knows how long it will remain vital). However when I read through discussions here I feel like a "come on, IPv4 is over, let's just stop the administrative burden", which reminds me the last days at school before holidays when nobody was minding about being still nor teaching or learning since it was the end anyway... Which makes me ask two questions 1. why should Ripe NCC consider *now* that these rare ressources need less attention and protection than previous ones, and not to be treated in respect to previous conservation principles, since these quite free and subject-to-be-returned addresses do not represent a ridiculous pool ? Turned another way : when did reducing charges become more important than the conservation goal ? 2. why, if IPv4 free pools were completely depleted, would Ripe NCC be discharged of the responsability to supervise assignments for what still is a vital and public ressource (yes they are, despite AW made it much easier for everyone) ? It might be a wrong impression really, and sorry for that but I feel like if Ripe NCC would not really want to mind with IPv4 anymore, and just wanted to turn away, being OK to just drop responsability to the market, whatever happens. Hope I'm wrong. Best regards, Sylvain

On 12/08/2013 23:03, Sander Steffann wrote:

Hi,

...

It might be a wrong impression really, and sorry for that but I feel like if Ripe NCC would not really want to mind with IPv4 anymore

Just making one important point here: the RIPE NCC doesn't want anything here. The community / this working group sets the policies, not the RIPE NCC. They only implement them once this working group declares consensus on the proposed policy.

Agreed. But since I do not want to project my perceptions as intents on someone in particular either... Please replace subject "Ripe NCC" by a subject "the proposal".

On Sunday, August 4, 2013, David Farmer wrote:

On 8/4/13 09:48 , Tore Anderson wrote:

...

* David Farmer

I believe the primary definition of fairness the RIR communities have

...

been using is, "only those that have *verified operational need* get Internet number resources".

Do you have a link or reference? (Tried Google, no hits.)

Try goal #1 in section #1 of RFC 2050. See https://tools.ietf.org/html/**rfc2050#section-1<https://tools.ietf.org/html/rfc2050#section-1>

since this is going to be replaced we should probably make sure we are in line with:

And in slightly different words, try goal #1 in section #2 of RFC 2050-bis. See http://tools.ietf.org/html/**draft-housley-rfc2050bis-02#**section-2<http://tools.ietf.org/html/draft-housley-rfc2050bis-02#section-2>

another question is if rfc2050 is "binding" and "limiting" for the RIRs and if draft-housley is going to be it may have go go trough some RIR processes too? -hph -- Hans Petter Holen Mobile +47 45 06 60 54 | hph@oslo.net | http://hph.oslo.net

On 8/6/13 10:20 , Hans Petter Holen wrote:

On Sunday, August 4, 2013, David Farmer wrote:

On 8/4/13 09:48 , Tore Anderson wrote:

* David Farmer

I believe the primary definition of fairness the RIR communities have been using is, "only those that have *verified operational need* get Internet number resources".

Do you have a link or reference? (Tried Google, no hits.)

Try goal #1 in section #1 of RFC 2050. See https://tools.ietf.org/html/__rfc2050#section-1 <https://tools.ietf.org/html/rfc2050#section-1>

since this is going to be replaced we should probably make sure we are in line with:

And in slightly different words, try goal #1 in section #2 of RFC 2050-bis. See http://tools.ietf.org/html/__draft-housley-rfc2050bis-02#__section-2 <http://tools.ietf.org/html/draft-housley-rfc2050bis-02#section-2>

another question is if rfc2050 is "binding" and "limiting" for the RIRs and if draft-housley is going to be it may have go go trough some RIR processes too?

I never said RFC 2050 is binding on the RIR's. I've been saying it articulates fairness and why operational need is important. RIPE and any other RIRs can and have ignored many parts of it, and are fee to continue to do so. However, RFC 2050 and the upcoming 2050-bis are the only common statement of goals or principles other than ICP-2 that there is for the Internet Registry System and the RIRs as a whole. If we ignore those goals and principles, without careful consideration, bad things are likely to happen. If you are suggesting it would be a good thing for there to be a common set of goals and/or principles for all the RIRs that have gone through the RIR's policy processes or maybe the global policy process, that may not be a bad idea. Note: Within the ARIN Region there is such a discussion going one see ARIN-2013-4, right now the scope is intended to be within the ARIN region only. https://www.arin.net/policy/proposals/2013_4.html I believe that operational need is one of the few useful concept of fairness we have from a Internet number resources policy perspective. Impartiality and community consensus supporting policy being the only others I can think of as useful concepts of fairness. -- ================================================ David Farmer Email: farmer@umn.edu Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 1-612-626-0815 Minneapolis, MN 55414-3029 Cell: 1-612-812-9952 ================================================

On 04/08/2013 16:48, Tore Anderson wrote:

In a similar vein, and somewhat tongue in cheek: If you're trying to buy an apartment, but get outbid by someone, do you get upset unless you can see some proof that guy did buy it plans on moving in there? Or do you assume that's probably the case and move on?

We got something in France that is called the right to housing. The question is not, of course, but is not fully disconnected of knowing if does Ripe wants to tend to protect the public right of weaks to get a place. You asked repetedly here for precise statements to explain how 2013-03 would, on its own, make things worse. And you explained many times that wrong things could already be done independantly from 2013-03 being accepted or not. Things cannot be split that way in my mind. My point is, that 2013-03 is part of a much larger evolution that is progressively making the deal completely different. I cannot help but consider thinks together and observe a general move that breaches yesterday's principles : - the ability to resell IPv4 allocations - the proposal to allow v4 PI changed to PA (aggregation breach) - the suppression of the conservation goal - the documentation need removal To me, the general picture looks like is that considerable ground is given to a fully commercial hand over the public ressource management. This seems to be motivated by kind of a hurry to get rid of the burden of the IPv4 administrative work. Which I feel hard to understand since lots of operators still need a proper and fair ressource attribution to continue or to start their business today (which is being more critical because of scarcity) and also because Ripe members still pay membership fees to finance it. Sorry I went quite far from simple 2013-03 discussion here. However reducing the discussion field does not allow to build a global policy that is consistent and has a coherent philosophy, I think. Best regards, Sylvain

Tore

-- http://www.opdop.fr - mutualiser et interconnecter en coopérative -- ManyOnes.COM SARL - RCS Paris B 449 031 574 - TVA n°FR56449031574 en cours de mutation comme Société Coopérative d'Interêt Collectif -- ManyOnes, c/o Sylvain Vallerot, 3 rue Erckmann Chatrian, 75018 Paris

On 01/08/2013 07:38, Tore Anderson wrote:

* Malcolm Hutty

...

However this argument is fallacious. The Conservation policy, even as stated, expresses *two separate* policy objectives: 'fair distribution', and maximising the lifetime of the public address pool. Depletion means that reality has superseded the second objective, but not necessarily the first.

Hi again Malcom,

Having slept on it, I have a suggestion that hopefully will make us reach common ground. How about, instead of removing the old Conservation goal completely, we rewrite it as follows:

«Fair use: Public IPv4 address space must be fairly distributed to the End Users operating networks.»

This would only remove the second "reality-superseded" objective, but not the first. It would also continue to provide the philosophical foundation on which we may build "enforcement" policies in the future, so we certainly would not be "closing the door" on any such discussion.

Furthermore, it might even help the NCC in their political/government interactions, as they can truthfully maintain that their goal with regards to IPv4 is still to provide for fair distribution and fair use.

Would such an amendment make the proposal more appealing to you, at least enough to make reach the "I can live with this" point? (This question goes for the others who have noted their reservation towards the proposal too, feel free to chime in!)

Tore, Firstly, thank you for being so willing to engage with a view to meeting objections, rather than just defending your proposal. I hope my answer meets you in the same spirit. That amendment would *definitely* make the proposal more appealing to me. To my mind, this moves us from "2013-03 looks bad" to "2013-03 seems broadly OK but needs thought to see whether a little further wordsmithing is needed". Which is, I think you'll agree, major progress. How long do we reasonably have to think about that wording? I realise we're at a late stage, but I would welcome introducing more eyeballs to this problem. In particular, I'm not sure I like limiting fairness to End Users, as you have done. Referring back to your other comments on fairness, to the effect that "fairness" has already been abandoned in the last /8 for NCC->LIR, I don't necessarily agree or think it should be presented like that. On the contrary, we could assert the fairness of the current policy, something like: "In order to ensure fair access to IPv4 address space, and in particular that as many parties have access to IPv4 address space as possible, no LIR shall be assigned more than 1024 addresses from the last /8. For the purpose of efficiency, any LIR [with a legitimate need for IPv4 address space][for their own use] shall be deemed to need 1024 addresses." The text in the first set of square brackets represents that bit of the current policy that an unaltered 2013-03 would remove. The text in the second set set of square brackets represents something that isn't currently policy, but perhaps should be (or perhaps not), and which couldn't be policy unless we retain the principle. I suspect if we adopted the text above without alteration the effect would be to still require a documented need for at least a single IP address. We could further refine it to remove even that - but it has already been mentioned that this is a negligible requirement; how would you feel about leaving it in, given that it's so little? Taken together with the bit in the second set of square brackets, this policy would exclude those that apply simply in order to re-sell, but nobody else. I would think that a simple checkbox "Do you intend to use at least one of these IP addresses on a network you operate?" should satisfy the documentation requirement. Now, you're fully entitled to think that the above policy isn't really all that fair. But it asserts a community view of what is fair, and provides a suitable hook for the community to debate and amend its view by changing the policy in the future. Meanwhile, the NCC will be able to continue to claim that the community aims at fair and efficient distribution, maximising the availability of resources where needed, despite the challenging circumstances - and to carry on saying the same thing if the definition of what is fair changes again. Malcolm. -- Malcolm Hutty | tel: +44 20 7645 3523 Head of Public Affairs | Read the LINX Public Affairs blog London Internet Exchange | http://publicaffairs.linx.net/ London Internet Exchange Ltd 21-27 St Thomas Street, London SE1 9RY Company Registered in England No. 3137929 Trinity Court, Trinity Street, Peterborough PE1 1DA

On 03/08/2013 11:47, Tore Anderson wrote:

So here's my suggested text. In the proposed section 5.1 (http://www.ripe.net/ripe/docs/other-documents/draft-v2-ipv4-address-allocati...) we add a new point:

«4. In order to qualify for receiving the allocation, the LIR must state an intent to use it for making assignments to one or more End Users.»

I may be wrong, but this seems to be quite a substantial shift away from v1 and v2. The previous two versions completely removed any requirement for stated intent, but this change would partially bring it back. This isn't a complaint, btw - just an observation. As an aside, I don't think it will make too much practical difference whether this statement is included or not, at least from the point of view of speculation. The language is vague enough that any respectable lawyer could easily side-step it. If the intent is to ensure that the LIR transferee makes the assignments, then the language could be tightened up. Nick

Hi, On Wed, Jul 31, 2013 at 02:26:31PM +0100, Malcolm Hutty wrote:

I don't support the approval of 2013-03 "No need" at this time.

While you're free to do so, I'm just so slightly annoyed that you're doing this *again* - wait until the very last second, and then state fundamental objections. Generally speaking, this sort of blunt "the fundamental principle is wrong" statement of opposition at the very end of the review phase is wasting lots of effort that has gone into making the policy text, creating the impact analysis (lots of time spent at the NCC), and in the preceding discussions. Fundamental opposition really should be voiced in the intitial phase of a proposal, where we try to find common ground to then build the specific text that is then evaluated for the impact analysis. (Technically speaking, your opposition was actually voiced *after the end* of the review phase - which was on July 30. But since we intended to extend the review phase anyway, based on the ongoing discussion, this point is somewhat moot) Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (89) 32356-444 USt-IdNr.: DE813185279

Malcolm: I find your arguments against 2013-3 to be unusually off target. Let's begin with this: -----Original Message-----

I am not convinced by the the principle argument being made in favour of abolishing the concept of "need"

The policy change does not "abolish the concept of need" it only removes it for the remaining scraps of the last /8 in IPv4, and for transfers of that space. My understanding is that some kind of demonstrated need would still be in place for IPv6.

Finally, as noted by the RIPE NCC, this takes away a key message the RIRs have used to justify their independence from government. That may not be sufficient reason to oppose the policy if it is otherwise necessary, but it does seem to me to indicate caution and counsel against acting precipitously.

This argument I simply don't understand. The presence or absence of needs assessment has absolutely nothing to do with the status of RIRs as independent membership nonprofit who administer a shared private resource space. Waving the bloody flag of the ITU really doesn't cut it here. As a private organization, RIRs have to be very careful about competition policy considerations. In this respect, needs assessment, which is completely nontransparent and could in some respects be considered more easily manipulated than a market, puts RIRs in much more danger of external scrutiny than an open and uncontrolled market for IPv4 transfers. When you say there are governments out there who want to take over number allocation, you are correct. But as someone who is as familiar with that environment as you are, I can tell you with complete confidence that these governments don't care a whit whether RIRs eliminate needs assessment or not: they want power in the hands of states, and neither passing nor not passing this policy will have any impact on the course of that debate. Indeed, your view could backfire. It would not be difficult to make a case that continued controls on allocations keep the price of number blocks artificially low for buyers, and by the same token suppress the market for sellers by eliminating many potential buyers from the market. Since most of the dues-paying members of RIRs are ISPs/buyers, it would not be too hard to paint a picture of this situation as a buyers' cartel, which would be actionable under antitrust law. If you think needs assessments are protecting RIRs from external interest and possible intervention, you are 180 degrees off the mark.

The Conservation policy, even as stated, expresses *two separate* policy objectives: 'fair distribution', and maximising the lifetime of the public address pool. Depletion means that reality has superseded the second objective, but not necessarily the first. So my first question to Tore is: "Why should 'fair distribution' of addresses between users no longer be considered an overarching objective of IPv4 address management?"

The simple answer to your question is that needs assessment are not "fairness assessments." If there are 50 ISPs contending for the same number block, needs assessment only tells you whether they cross some threshold (largely theoretical and unenforced) of operational justification. It may be the case that 2 or 3 of those ISPs have far greater merit in their claim than the other 47 or 48. Current practice doesn't care about that. Your argument falls on this simple fact alone.

2. "Fair distribution" establishes a basic goal for IPv4 address management policy. Other policies exist to pursue that goal.

Your second statement here is an unjustified logical leap. There is nothing in the current policy that either defines what "fairness" is or that makes it the overriding concern, more important than efficiency and conservation. One could argue that when there is scarcity it is more fair for people willing to pay the highest price to get the resource. I know that that logic doesn't always work, and doesn't always seem fair, but it does work for about 90% of the economy as a whole. My view is that when we are dealing with the final scraps of the v4 space and thousands of claimants can all claim that they "need" numbers, it is indeed the most fair option to give it to the person who bids the highest. Not that this is a wonderful option, only that it is better than the alternatives, including arbitrary, nontransparent, bureaucratically demanding needs assessments.

limiting the maximum allocation size directly relies upon the notion of fairness (we're only giving you X, even though you want Y, so that the next guy can have some at all).

Yes, we still have the one /22 restriction. Indeed, the last /8 policy giving one small block to a customer is based on a fairness principle. That policy is not fundamentally changed by the elimination of needs assessment; the most important aspect of the policy is the limitation of one block per customer. Absence of needs assessment will not have much impact on that basic operation of that policy. And so you have completely refuted your own argument that there is no "fairness" left in the system if 2013-3 passes.

With the introduction of a transfer mechanism, it is possible that market pressures may prove adequate to ensure efficient and fair allocation of IPv4 addresses; in this transition period, the consequences of run-out are uncertain. RIPE has chosen not to act pre-emptively, to give market mechanisms a chance to succeed. However if market mechanisms prove inadequate to safeguard the fair and efficient use of IPv4 address space, RIPE stands ready to introduce new measures as necessary."

So, you are suggesting that we pass 2013-3 with this language added? The idea is not an entirely bad one, but as currently drafted the statement is unclear. It should read something more like this:

With the introduction of a transfer mechanism and the elimination of needs assessments for IPv4, we are assuming that market pressures will prove adequate to ensure efficient and fair allocation of IPv4 addresses. However if market mechanisms prove inadequate to safeguard the fair and efficient use of IPv4 address space, RIPE stands ready to introduce new measures as necessary."

Such a disclaimer may not be necessary, because RIPE can always change its policies. But if it gains more support to put it that way, go ahead! --MM

Yes, it would be acceptable to me if acceptable to Malcolm and Filiz -----Original Message----- From: Tore Anderson [mailto:tore@fud.no] Sent: Saturday, August 3, 2013 11:59 AM To: Milton L Mueller Cc: Malcolm Hutty; Sander Steffann; address-policy-wg@ripe.net Subject: Re: [address-policy-wg] 2013-03 New Draft Document and Impact Analysis Published (No Need - Post-Depletion Reality Adjustment and Clean up) * Milton L Mueller

[...] But if it gains more support to put it that way, go ahead!

Exactly. I'm sure we all could discuss these topics for decades without finding full agreement. Fortunately we don't have to, all we have to do is to converge on a text that has consensus as acceptable (or better). It's to that end I proposed the two amendments, which I hope will bring Malcom and Filiz on board with the proposal: 1) retain the philosophical "fairness" goal from our current policy, and 2) make LIRs that want to obtain their "last /8" /22 pledge to use it for their End Users (a continuation of current practice). I understand that you're mostly interested in the second-hand transfer market, so I'd like to point out that 2013-03's impact on the transfer policy will not change due to these two amendments. Even though you may find them unnecessary, I hope you'll also find them acceptable, and not something that would prevent you from continuing to support the overall proposal. Best regards, Tore Anderson

Nigel, Thanks, interesting. I had not read the full analysis before With due respect, I think the legal analysis in the impact statement is confused or at least unconvincing. The analysis says: "From an EU competition point of view, it is important that IPv4 allocations and transfers are conducted in a way that cannot be seen as discriminatory or exclusionary." This I agree with. And a no-need policy, which eliminates the RIPE-NCC's judgment from the picture, also eliminates the possibility that RIPE-NCC's judgment could be construed as discriminatory or exclusionary. But the statement goes on to say: "The fact that the allocation of IP addresses by the RIPE NCC occurred on the basis of the proof of objective need, rather than purely economic considerations, has protected the RIPE NCC from competition law claims." First, I would ask for evidence. Has anyone ever threatened NCC with competition law claims before and had their case thrown out based on th existence of needs assessment? I strongly suspect not, but if so, let us know. It seems obvious to me that RIPE-NCC is more likely to be challenged legally if it is directly responsible for making the decision as to who gets and does not get addresses. A court might uphold their methods of assessing need, but this does not tell us anything about whether they would be more or less subject to such challenges if needs assessment goes away. Second, the statement says "Removing the needs-based requirement would create an exposure to competition law claims based on discriminatory treatment or refusal to supply." This is just wrong. If RIPE is not involved in a transfer of number blocks, it simply cannot be accused of discriminatory treatment. It offers no "treatment" whatsoever. Indeed, the legal analysis suggests the correct conclusion: "Competition law claims would be mainly addressed to LIRs that could be seen as coordinating in stockpiling and applying discriminatory practices." In other words, a more market-based allocation regime might involve competition policy claims against individual buyers of IP numbers who are stockpiling addresses for anti-competitive purposes. But RIPE would not be liable. The statement adds, "Such claims may also be addressed to the RIPE NCC, if only in order to make more publicity for the claim." In other words, these claims would have no legal merit and might be filed only for publicity purposes. But if publicity is the object, then the existence of needs assessment will not protect RIPE either - a litigant could sue RIPE-NCC to gain publicity either way. QED -----Original Message----- From: address-policy-wg-bounces@ripe.net [mailto:address-policy-wg-bounces@ripe.net] On Behalf Of Nigel Titley Sent: Sunday, August 4, 2013 9:28 AM To: address-policy-wg@ripe.net Subject: Re: [address-policy-wg] 2013-03 New Draft Document and Impact Analysis Published (No Need - Post-Depletion Reality Adjustment and Clean up) On 03/08/2013 14:11, Milton L Mueller wrote:

This argument I simply don't understand. The presence or absence of needs assessment has absolutely nothing to do with the status of RIRs as independent membership nonprofit who administer a shared private resource space. Waving the bloody flag of the ITU really doesn't cut it here.

With due respect, our lawyers (specifically expert in EU competition law) disagree with you. I'd prefer to give *some* credence to their opinions, which is why the impact analysis specifically passed on their comments. It is, of course, up to the RIPE community to take this opinion into account, or not, as they wish. Nigel

* Milton L Mueller

"Competition law claims would be mainly addressed to LIRs that could be seen as coordinating in stockpiling and applying discriminatory practices."

That's a feature, not a bug. :-) I'd much rather have LIRs that would engage in stockpiling and applying discriminatory practices in breach of competition laws *not* be able to defend themselves with «don't look at us, the RIPE NCC said it was OK!» Tore

On 5 Aug 2013, at 12:24, Milton L Mueller <mueller@syr.edu> wrote:

They [NCC's lawyers? - JR] need to explain a) whether there is any REAL litigation in which the existence of needs assessment has actually shielded RIPE-NCC from a claim; b) how the use of needs assessments make any less likely claims made purely for "publicity" purposes; c) why RIPE-NCC's involvement in allocations of a highly scarce and increasingly contested resource via needs assessment wouldn't make it _more_ likely to be entangled in litigation

Not quite. None of us here are practising lawyers AFAIK, far less lawyers who understand Dutch/EU law on this particular topic. It would be wise to defer to the experts -- presumably the NCC engaged a law firm who know this field -- instead of continuing our amateur speculations. If further clarifications and explanations are needed, I suggest we start at the beginning by asking the NCC to publish the brief given to its legal advisers and their response. This way we will know if the right questions were asked at the outset. Everyone can then choose to hire their own lawyers to advise on the correctness and completeness of the reply. Or do some research of their own in a law library, assuming people do that sort of thing these days. Or we could just take the advice of the NCC's lawyers on trust: that's what they get paid to do after all. I have no idea if the things on your wish list above are necessary or sufficient to address the issues (excuse the pun). For this non-lawyer, they do seem to be putting the cart before the horse.

On Mon, Aug 05, 2013 at 01:13:14PM +0200, Tore Anderson wrote:

That's a feature, not a bug. :-) I'd much rather have LIRs that would engage in stockpiling and applying discriminatory practices in breach of competition laws *not* be able to defend themselves with �don't look at us, the RIPE NCC said it was OK!�

I think this discussion has come off the rails a wee bit. As I read the proposal, it is (correct me if I'm wrong) simply about removing a phrase -and associated bureaucratic process that causes work and delay for operators- that has become meaningless in practice even before IPv4 depletion. All the "needs assessment" consists of is a check that the application conforms somewhat to the letter of the policy (the aforementioned "giggle test"). There is even a specialised skill in creating assignment/allocation requests that will pass the "needs assessment" with minimum hassle. (Full disclosure: This skill has paid my bills a few times) Since there is no point in keeping around a meaningless procedure, there are, in my opinion, two options: 1) implement 2013-03 and give policy standing to a situation that obtains anyway. 2) actually assess the needs of each application on its merits, as some participants in this debate seem to be hinting at. This is something that not many network architects are qualified to do, let alone RAs (with no offence to RAs intended). Even assuming the NCC hired highly qualified networkers to perform these assessments, the process would still be largely arbitrary (ask any two networkers about the "right" way to build a network and you will get three opinions), expensive and totally disproportional in effort, especially considering that there is not that much IPv4 space left to distribute in whichever way. As far as hoarding and speculation is concerned, I'm not sure this is such a massive problem. IPv4 space is a risky speculative investment, as soon as IPv6 usage gains significant traction, it is likely to become progressively worthless and the effort and cost to obtain large amounts for speculation (remember, 1024 IPs per LIR) should make it unattractive. [Caveat: IANAEconomist] For the avoidance of doubt, in case this wasn't clear from the above, I hereby restate my support for 2013-03. rgds, Sascha Luck

Hello, On 25 Jul 2013, at 21:55, Jan Ingvoldstad <frettled@gmail.com> wrote:

I'm bewildered, confuzzled, and wide-eyed, yes, nearly astonished, that Tore's points don't seem to penetrate the fog.

I can say the same thing. I did not really think what I was saying was so confusing…

I've read this proposal a few times, and as far as I can tell, most if not all of the alleged counterpoints to 2013-03 are _not_ counterpoints to 2013-03, but to something else.

I may have misread the proposal, and therefore I request that Filiz, Michele and Nick please point out exactly which points _introduced_ or _altered_ with 2013-03 they disagree with, so that I, and other confused people, can have a better chance of understanding your arguments.

Please?

Just because you asked nicely :).. I am not going to go through a very detailed word smithing or referencing, sorry, in my case I do not believe this will be helpful. But I will try to put my point of view in a different way again, using some different words and then I will stop because I feel I am repeating myself and some of the members of this list may already feel overwhelmed about the discussion. Here it is my main point: "Justification for need" and "evaluation of justification for need" are two different things. First one, "Justification for need", is perfectly a policy matter and I believe IPv4 policy should still mention this, as long as RIPE NCC continues allocating space to its members and the last /8 is totally exhausted. Say something along the lines, "LIRs requesting address space from the RIPE NCC should have a need for the requested space for a network of their own or their customer". So that we at least put a barrier in front of those who would just ask for an allocation to immediately turn it into an asset. But those who really are in need are primarily highlighted by the policy. Current policy has the following text: "Members can receive an initial IPv4 allocation when they have demonstrated a need for IPv4 address space." Tore's proposal is removing this totally and I do not agree with it. The latter, "evaluation of justification for need" is totally an operational matter that is performed by the RIPE NCC. Neither the current policy nor Tore's proposal has any significant text on this but this is one of his arguments for his proposal. In my opinion, real solution to this procedural problem of evaluation is on procedural level, not on the policy level. RIPE NCC may be asked to change their evaluation tools/systems/mechanisms. This does not require to remove the entire "need" notion from the policy text. Kind regards Filiz Yilmaz

-- Jan

On Thursday, July 25, 2013, Jan Ingvoldstad wrote:

Here it is my main point:

...

"Justification for need" and "evaluation of justification for need" are two different things.

First one, "Justification for need", is perfectly a policy matter and I believe IPv4 policy should still mention this, as long as RIPE NCC continues allocating space to its members and the last /8 is totally exhausted. Say something along the lines, "LIRs requesting address space from the RIPE NCC should have a need for the requested space for a network of their own or their customer".

So that we at least put a barrier in front of those who would just ask for an allocation to immediately turn it into an asset.

That barrier is a paper tiger, unfortunately.

and has been so since the beginning.The amount of paper have changed over times. If the policy states "need" then NCC feels obliged to figure out how to determine "need"

...

But those who really are in need are primarily highlighted by the policy.

Current policy has the following text: "Members can receive an initial IPv4 allocation when they have demonstrated a need for IPv4 address space."

Tore's proposal is removing this totally and I do not agree with it. (...)

So, in essence, what you state is that:

a) There is no need to change or remove the "need" statement in the policy. b) The RIPE NCC should decide how "need" is determined as a matter of procedure.

Given what appears to be the majority view here, the NCC may just as well decide to interpret the community's view on "need" as something that does not need to be documented in itself, other than placing a request for a network block.

Be careful here - we are not operating by majority but by consensus - so we need to get everybody - or at least most - to understand and not object.

Altering this particular part of the policy document as Tore suggests will change very little in practice and procedure.

That is still not clear to me. Hans Petter -- Hans Petter Holen Mobile +47 45 06 60 54 | hph@oslo.net | http://hph.oslo.net

http://www.ietf.org/id/draft-resnick-on-consensus-02.txt may be of help

Hi, On Sat, Jul 27, 2013 at 11:32:29AM +0200, Randy Bush wrote:

http://www.ietf.org/id/draft-resnick-on-consensus-02.txt

Thanks for the pointer. While this is an IETF draft, the principles by which Sander and I try to assert consensus are similar - so while we certainly prefer to reach unanimous agreement (or at least, abstention instead of opposition from those that do not support a proposal), we've had proposals in the past where single individuals very strongly opposed the proposal, but nobody else agreed to their reasoning, and it wasn't possible to convince them that their argument does not hold - so we went ahead nevertheless. For this particular proposal here, Sander and I are reading and listening carefully, and will write a very detailed summary at the end of the review phase... Of course it helps us enormously if those that post their concerns about things, for example, "abuse of the last /8 policy" or "unrestricted trading of addresses", clearly state whether they see this as a consequence of 2013-03 and actually object to 2013-03 because of that (and if yes, which parts of 2013-03, so Tore can answer to that). If we can't see specific opposition to changes brought by 2013-03, we tend to evaluate such statements as "neutral as far as the proposal at hand goes". Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (89) 32356-444 USt-IdNr.: DE813185279

On Sun, Jul 28, 2013 at 1:34 PM, Randy Bush <randy@psg.com> wrote:

it would be *really* nice is the few in disagreement could live with the general consensus. and we certainly need to pay good attention to their points. but sometimes we need to move ahead with there still being a (very few) folk in disagreement.

I think this is where the point about having a mutual understanding is important: - what the proposition changes and does not change - what the opposing arguments are, to which points of the proposition … because, as someone aptly noted earlier, the number of ayes and nays do not necessarily reflect the opinions of a majority or minority of the members. It seems to me that we are just about there, now, and that any unclear statements, suggestions and criticisms can probably be considered "neutral". Then again, this is merely my personal opinion. :) -- Jan

On 26 July 2013 07:32, Tore Anderson <tore@fud.no> wrote:

Second, on a more pragmatic level, when we hit IPv4 depletion last year, the requirement to demonstrate need in order to receive initial allocations changed. It is no longer "specify how many addresses do you need?", but a boolean "do you need more than 0 addresses? yes/no". Demonstrating the need for 1 IPv4 address is, in my opinion, merely a formality - anyone could do that with almost no effort.

<specific bit> However, I know of several organisations who refer to the requirement to require IP address when asking for them from customers using the line: "We have to be able to justify the needs basis to get address space from RIPE and they require us to make sure that you have a need for the address space too" By removing a need at the LIR/RIR boundary you will create a problem at the Customer/LIR boundary. So keeping the text about need but removing the conservation goal bit is actually a desirable outcome. <general bit> I think the goal of making life easier is nice but I fear we are rearranging deck chairs again :) J -- James Blessing 07989 039 476

Hi, On Fri, Jul 26, 2013 at 11:40:43AM +0100, James Blessing wrote:

I think the goal of making life easier is nice but I fear we are rearranging deck chairs again :)

IPv4 is going to be with us for a long time, and the specific bits that Tore is attacking with 2013-03 are going to cause work for LIR network admins for the next 10 years, or so. So there is more benefit to this than mere deck chairs... Gert Doering -- APWG chair -- have you enabled IPv6 on something today...? SpaceNet AG Vorstand: Sebastian v. Bomhard Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann D-80807 Muenchen HRB: 136055 (AG Muenchen) Tel: +49 (89) 32356-444 USt-IdNr.: DE813185279

On Thursday, July 25, 2013, Jan Ingvoldstad wrote:

I'm bewildered, confuzzled, and wide-eyed, yes, nearly astonished, that Tore's points don't seem to penetrate the fog.

I've read this proposal a few times, and as far as I can tell, most if not all of the alleged counterpoints to 2013-03 are _not_ counterpoints to 2013-03, but to something else.

Well, it could be the title: "No need - Post-Depletion" -hph

-- Hans Petter Holen Mobile +47 45 06 60 54 | hph@oslo.net | http://hph.oslo.net

On Jul 25, 2013, at 10:20 AM, Nick Hilliard <nick@inex.ie> wrote:

that is true, but it might be useful to make some sort of estimate particularly with regard to the class A+B legacy allocations. Eyeballing, I figure there's probably quite a good chunk of unrouted legacy space or legacy space which ended up in commercial organisations who probably don't need all that space, and which could eventually come into the market - maybe 30-40 /8s as a rough estimate. Given that the RIRs ploughed through 94 /8s since the early 1990s, this represents a reasonably large amount of integers from the point of view of potential future liquidity.

Nick - It's probably best to look at the last 2 or 3 years of IANA->RIR allocations as an estimator of demand, rather than averaging across the entire period, e.g. reference Geoff Huston's recent blog entry for some very interesting pre-runout statistics (including APNIC numbers in excess of 12 /8's per year) - http://www.potaroo.net/ispcol/2013-08/when.html In terms of available space that may come to market, I'll note that many of these address blocks are actually in use by organizations, but in terms of routing hidden behind various firewall/nat combinations. Taking both the recent burn rate and in-use-but-not-routed factors into consideration, it's likely that (at a high enough price) that there are several years of additional IPv4 liquidity that could be obtained, but rather uncertain as to whether it could provide, say, 10 more years of continued IPv4 Internet growth... FYI, /John

* Gert Doering

(The /22 in the "last /8 policy" was chosen to ensure that every single LIR in existence today can get their /22, and we'll still have some left - the /8 will last for 16.000 /22s, and the number of active LIR is still below 9000 today)

It's last much longer, actually. Since depletion, the NCC has recovered 922,824 IPv4 addresses (outside of 185/8), and the NCC's one-fifth share of the IANA Recovered IPv4 Pool currently numbers 3,820,953 addresses. Put together, this gives us another 4,632 /22s to give to new LIRs. When taking these not-185/8 (yet still covered by the last /8 policy) pools into account, it is actually the case that the "last /8" pool has seen a net *growth* since the NCC hit IPv4 depletion. :-) Tore

Filiz, Michele When number blocks are traded by companies in a market, the price of number blocks reflects their scarcity - i.e., the relationship between supply and demand - and thus companies have very strong incentives to allocate and conserve. These incentives are at once both more effectively enforced and more flexible than a regime in which a central bureaucratic authority looks at some technical indicators submitted to them on pieces of paper (or digital forms) and tries to assess "need." So there are still strong incentives for responsible management without needs assessment. And since v4 is depleting, the idea that the price would rise and people would over time be discouraged from using it, ought to be seen as a good thing. Administrative needs assessment is not the ONLY method that could EVER be used to allocate resources. There are a variety of tools and mechanisms and the authors of 2013-3 are correct that the conditions that led to traditional needs assessments for numbers are gone, and not applicable to the current situation. I've always asked the religious believers in needs assessment whether they think their rental of office space should be subject to a needs assessment. It's not meant to be a challenge, just something to make you think. The economic characteristics of the resources are not that different (space, like number blocks, are occupied but not consumed, and there is a short-term fixity in supply). So let me volunteer to do a needs assessment on Michele's offices, and his home, and perhaps a few other belongings. I just want to see if he is wasting any resources. ;-) Or would Michele prefer to just pay for what he thinks he needs? From: address-policy-wg-bounces@ripe.net [mailto:address-policy-wg-bounces@ripe.net] On Behalf Of Filiz Yilmaz Sent: Wednesday, July 24, 2013 6:53 AM To: Tore Anderson Cc: address-policy-wg@ripe.net Subject: Re: [address-policy-wg] 2013-03 New Draft Document and Impact Analysis Published (No Need - Post-Depletion Reality Adjustment and Clean up) Hello Tore, While I am curious for Michele's response to your mail too, here is my bit, as I feel I agree with him in terms of principles. First of all, I am not totally against your proposal; it cleans up the policy text reflecting today's circumstances to some extent and it makes a long document way more readable. This is also why I am writing back because I've read the draft policy document from the beginning to the end. My observation is that the new document proposed is not exactly a "registration" policy document. It more looked to me like a description of how address space management is done within RIPE NCC region "by the RIPE NCC". If I am not missing anything crucial, the main points described are: - The language is English, - How big an allocation can be, - That there is no PI anymore to be assigned directly from the NCC pools to End users (except for the IXPs) so all resources will only goto LIRs - And these blocks can be transferred between LIRs. The only bit about registration I see in the new text is section 4.0 Registration Requirements and it does not go more than saying details should be recorded in the database. So it does not contain any substantial information for registration or address management on the LIR's side. This is interesting as now with this proposed policy any End user's chance to get any IPv4 address space will be through an LIR and hope that these LIRs are responsible and know what they are doing. I would like to see some guidelines or at least principles mentioned in this document so the LIRs know their responsibility in terms of fair address management as well as the End Users so they know what to expect from these LIRs. This is what I would be expecting from a transparent documentation of a set of policies and principles that are still in place. We may not have too many specific policies to set for the few left-over resources but I would like to believe we still have "principles" towards the responsible management of these resources. In that sense Michele has a point and I argue that LIRs need to be guided for "good address management" even without the "conservation" principle as the top priority in the new IP world. This is missing in the proposed policy text for it to be considered as a helpful "registration" policy in my opinion. In practice, I can set up a new LIR now and ask for a new allocation and I may be someone who does not have any previous RIPE or RIPE NCC experience. If all I have is this document, I am not sure if it tells me enough about my responsibilities, while I will be a critical token in the EU address management and registration system by just becoming an LIR. My other concern is in regards to the transfers. As neatly put by the NCC Board in the Impact Analysis: --- * As mentioned in previous sections, the policy proposal would negatively affect the ability of LIRs to engage in inter-RIR transfers, as the RIPE NCC's service region would be the only one without a needs-based requirement for transfers. * Implementation of the policy could expose LIRs to legal challenges under EU competition law. --- I think singling out the RIPE NCC region in the world of transfers may not be the best idea at this stage. Kind regards Filiz Yilmaz On 24 Jul 2013, at 10:52, Tore Anderson <tore@fud.no<mailto:tore@fud.no>> wrote: * Michele Neylon - Blacknight As previously stated, I do NOT support the "no need" policy and cannot support this document. IP addresses are a finite resource, as we all know, and obliging people to provide some level of justification makes sense. The argument for "conservation" may no longer be valid, but there will always be a compelling argument in favour of good resource management, which I believe the policy covers. RIPE should not remove the requirement to provide justification. Hi Michele, I doubt you'll find anyone in the working group who is against good resource management. I am convinced that the proposed policy is not in conflict with good resource management, otherwise I would never have proposed it. While I can obviously only speak with certainty for myself, I assume that the people who support the proposal feel the same way. While it appears you believe that the proposal will bring about poor resource management, your message neglected to explain why or how. This makes it rather difficult for me to try to alleviate your concerns. As Gert also pointed out recently, the main reason I believe that IPv4 would continue to be consumed responsibly under the proposed policy, is that the LIRs in the region are painfully aware that there is no more IPv4 to be had from the RIPE NCC. Should an LIR anyway decide to go on a "spending spree" with its remaining inventory, it would only end up hurting itself by expediting its own depletion date. The community will not be impacted - without a Common, there can be no Tragedy. Best regards, Tore Anderson