ARIN XXI Public Policy Meeting
Draft Transcript
Tuesday, 8 April 2008

"This transcript of the meeting may contain errors due to errors in transcription or in formatting it for posting. Therefore, the material is presented only to assist you, and is not an authoritative representation of discussion at the meeting."

Table of Contents

Meeting Called to Order

MR. PLZAK: Okay, good morning. Hello. Okay, a few items. First of all, as I said yesterday at the close of business that we would be reordering the agenda, and that information has been posted on the website. An announcement has been sent to the PPML. In addition, a printed copy of the advised agenda was made available as you entered this morning. It looks like this. I'm sure you can all see that from where you're sitting. If you didn't get a copy, copies are available at the registration desk. So, well, you know, yesterday we were talking about what we were going to do last night and said we're going to paint the town red. Well, we did. And we do have some evidence, and so I will briefly show you that. (Laughter)

MR. PLZAK: There are other pictures available upon request. Okay. So, anyway, reminder, Cyber Café, take a break. It's a good place to get connected. You can watch some of the information (off mike) and to get information. Also, we are going to continue bombarding you with surveys. And we will continue to bribe you to fill out those surveys with prize drawings. Remember, you must be present to win. The help desks are open as shown. And remember, if you are a loser, you're really not a loser because you will still be entered into a consolation prize drawing which is an Apple TV, so ooh.

SPEAKERS: Ooh.

MR. PLZAK: Ahh.

SPEAKERS: Ahh.

MR. PLZAK: Perfect. Okay. So, don't forget. We have another Project X -- Project X Live Demo at 1:00 p.m. It's in here. Reminder of the rules. They're coming up very slowly. They must have been out too late last night, as well. Anyway, we had a quite extensive discussion yesterday, and following these rules makes that discussion easier to contain. And the Chair will enforce the rules. So, at the head table we've got John, Bill, Scott, Matt, Lea, R.S., and myself.

Statement from Sponsor

MR. PLZAK: And so now a word from our sponsor. We'd like to introduce Thad Mazureczyk from WildBlue. And he'll give us a few words.

MR. MAZURECZYK: Thank you, Ray. Members of the ARIN panel. Good morning, everyone.

SPEAKERS: Good morning.

MR. MAZURECZYK: Welcome to the 21st ARIN Conference and to the City of Denver. Hope you're all having a great time at the conference here in the Mile High City. From the slides I just saw, it looks like you are. No one likes a longwinded sponsor, so I think I'll keep it short, especially since I'm in engineering and not marketing. My name is Thad Mazureczyk, and I'm responsible for the engineering group at WildBlue Communications, which is a national space driven spot beam KA Band satellite Internet service. We have a growing business delivering broadband services principally to the rural regions of the Continental U.S. We at WildBlue are grateful for the opportunity to be a sponsor for your conference. The work you all do here is very important to the future of the Internet, and we're glad to be able to play a small role in supporting your conference. We'd also like to thank ARIN for working with us as we launched and continue to grow the Wild Blue business. Since you have a lot of work to do, let me just wrap up by saying we at WildBlue wish you a fun, productive, and successful conference this week. And once again, welcome to Denver. (Applause)

2008-2: IPv4 Transfer Policy Proposal Cont.

MR. PLZAK: Okay. And so now on with the agenda. The first item of business is a continuation of the discussion of Policy Proposal 2008-2. Matt Pounsett will please come to the microphone, and John, I will turn it back over to you.

MR. CURRAN: Okay, folks, we're opening the microphone for discussion of Policy Proposal 2008-2. I will be allowing about 45 minutes to an hour of discussion, at which point we'll do some shows of hands to help the Advisory Council. I'd like people who approach the mic to limit their comments to about two minutes because we really had a lot of people, and we probably will have the same problem this morning. So, microphones are open regarding the Transfer Policy 2008-2. When we broke yesterday, we had nearly 10 people on the queue when I closed them and were endangering the social. This morning it's the same policy proposal, and I just want to point out the microphones remain open. Yes, Mr. Darte.

MR. DARTE: Hi, I'm Bill Darte. I'm on the Advisory Council and represent Washington University in St. Louis. I think first of all that I'm for the general position that a transfer policy change to exist. That's reluctant, however, because I basically don't like the idea of doing it in general. But I think that it's all about a set of signals that we need to send. If we don't do something like this, it seems to me that we open ourselves up to calls from outside our group that we have advocated our responsibility and stewardship for these resources and to provide them as best we can to the community. And so, I think we need to do something to keep -- to dampen any of that sort of criticism. I'm not sure that we need to tweak this too much or to try to rationalize all of these issues, like deaggregation. Not that it's not serious, but we have to recognize that we have a limited amount of control over what happens.

So, while I'm in favor of this, I'm not in favor of it in an unlimited sense because I think -- one of my concerns is that if there are investment dollars to be made towards development, I wouldn't want to send the wrong signal to the development community and say, you know, gee, if you've got limited dollars and you can develop for the v6 world or the v4 world, look, they seem to be saying that v4 is going to be perpetuated. And so I need to make my investments there. Rather, I would like to see people take a queue from our activity here that says v6 really is the world that we need to populate with in the future. And so, while I think the transfer policy ought to go into place, I don't think it ought to be necessarily too critically crafted. And we out to sunset it and basically say five years, seven years down the road it ends. This is a process of transition. This is a way to help folks who haven't embraced this notion, haven't worked it into their business plans or their architectural plans. And so that would be a signal that this -- the v4 world is not the world of the future. And so that's what I think about it.

MR. CURRAN: Thank you. Any response?

MR. POUNSETT: No.

MR. CURRAN: Center microphone.

MR. SCHILLER: Bill, could you stay at the mic for a second? Jason Schiller, Verizon Business, NRO NC. I just wanted to respond to something that Bill said about sunsetting this particular policy. Oh, in general, I think transfer policy is a good thing. And specifically 2008-2, I think I'm warming up to it. With regards to sunsetting, though, I think what I wanted to point out is two things. First, any policy can be revoked by the ARIN membership at any time, obviously. But secondly, if we really do think that v6 is the future and at some point it's going to be widespread adopted with critical mass, won't that just sunset the transfer policy anyway? If the majority of the Internet has migrated to v6, then with the exception of a few people that are really shoehorned into imbedded systems that they can't get a v6 stack in, I would assume that v4 addresses would become nearly valueless.

MR. CURRAN: Bill, do you want to respond?

MR. DARTE: You know, the law of unintended consequences is always in my mind. And while I would hope that your prognostication would be exactly what would happen, I think we can, in fact, accelerate that by being quite explicit about -- I think the sunset clause, you know, just like we can change policies, you could extend the sunset, too, couldn't you? So, to me it's all about sending signals. And the sunset clause sends a very clear signal, I think, and explicitly, that we see this as an interim transition mechanism.

MR. CURRAN: Just to be clear, Bill. As written, without a sunset clause you would still be in favor for a lax transfer policy?

MR. DARTE: I think we have to be from a geopolitical stand.

MR. CURRAN: Okay.

MR. SCHILLER: Bill, just one other question. Do you think it would also -- would it also help for the Board, for example, to make some sort of policy stating that we believe that for the future of the Internet the right thing to do is to adopt IPv6 similar to what they've done in the past?

MR. DARTE: Yes.

MR. SCHILLER: But specifically pointing out that the transfer policy is intended to help with this transition.

MR. DARTE: Absolutely. Applying the principles of stewardship, if ARIN, a nonprofit corporation allocates Internet protocol resources, blah, blah, blah, blah, consensus based policies facilitates the advancement of the Internet through information and education outreach. And I think that is the second and more important phase of our future than this policy.

MR. CURRAN: Okay, thank you.

MR. SCHILLER: Thank you.

MR. CURRAN: Okay, microphones remain open. Far right side.

MR. DURAND: Alain Durand. Commenting on the previous point. In the press articles it was saying that because of transfers there would be (off mike) space in v4 to sustain the next five years. First, questioning the need to move to IPv6. And a lot of people who read the press thought that it was (off mike) so I'd like to be very, very careful in the signals that we are sending through this policy.

MR. CURRAN: Okay, Alain, as the policy is written today, would you be for it, against it, or undecided?

MR. DURAND: As of today I would be against. I may change my mind a few months from now.

MR. CURRAN: Okay, thank you. Microphones remain open. Center rear microphone.

MR. OBERMAN: Kevin Oberman, ESnet. Against the policy as currently written. The statement that was made that the policy can be changed is a valid one. But in a world where markets and network operation activities can happen very fast, ARIN does not pass policies very fast. I think you have to be very conservative in any policy you write in this area. Because once it gets out of hand and the genie is out of the bottle, changing the policy may become utterly meaningless at that point. So this requires, I think, a lot more discretion and care than a lot of the types of policies ARIN deals with.

MR. CURRAN: Okay, Mr. Woodcock, do you want to respond?

MR. WOODCOCK: I think my main issue with this policy is just the one that Tom brought up yesterday, which is that it bundles many different fairly unrelated concepts together into one omnibus kind of policy. I think that it's a complicated enough situation and there's room for reasonable debate on many of these individual points. And I think that we're only going to get the best consensus judgment of the community on all of those points if we have a chance to debate them individually rather than trying to have the AC put together a big package of them and squeeze it through the very narrow bottle neck of a community approval or disapproval on the package as a whole.

MR. CURRAN: Okay, thank you. Front center mic.

MR. LEIBRAND: Scott Leibrand, ARIN AC, Internap. One general point that I think we need to all keep in mind: Human nature is that we consider inaction to have less impact than action. I think in this case it's actually the opposite. So we all need to stop and think what are the impacts of inaction here. And really consider whether that would be better than action. So that's just a general keep that in mind. To Bill's point, I really would like to see some real discussion and proposals about things that we can do other than a complete transfer policy that can stand on their own to the degree that we can actually implement them. As I see it, as soon as we start going down the transfer policy route, unless we have a number of the restrictions that are in this proposal, we open ourselves up to a lot of negative, unintended consequences as I've elucidated in my opinion about unrestricted transfers that I see coming out of APNIC. So, I would love to see specific proposals of either subsets of this transfer policy that we can do that would be good and useful on their own, or some people think that there are things that we could do to provide incentives without a market. And I've seen one proposal on that that I don't think is viable. I'd love to see others that might be more viable. But I just haven't seen a good alternative to the transfer policy yet, and I don't think doing nothing is a good alternative.

MR. CURRAN: Okay, Matt, comment?

MR. POUNSETT: Yeah, I wanted to comment on what Bill had to say as well. I think it would be nice if there was a way that we could take a series of simpler steps. But I'm not sure I see a way to do that. Similar to what Scott said, I think, as soon as we do anything in this area, we have to do more than just the very simplest thing for basically the same reasons he said. If we do anything without some complexity in restrictions, then we open ourselves up to a pretty bad situation, I think.

MR. CURRAN: Okay. Bill, myself, then Lea.

MR. WOODCOCK: Yeah, I would fully agree with the premise that doing nothing is our most dangerous course of action. And I am by no means suggesting that this is a simple situation that is best addressed with a minimalist response. What I am suggesting is that I believe it's a complex situation and I believe it's too complex to be able to give a good yes or no answer to a large bundle of propositions without those propositions being debated individually in addition to, or prior to the whole thing at once. And I agree that there's a lot of work here, and I'm certainly not trying to say it's all work for the AC.

MR. CURRAN: Let me respond first and then Lea is next in queue. Along the lines of attempting to create policy that would be supported by many, and at the same time get debate on policy elements that might have issues, anyone in the room who isn't for the policy proposal but knows of a chance and modification that would let them support it, that would be a good thing to bring to the microphone. Because that's most helpful to the AC in considering whether small changes would help and what the degree of consensus on those are. So if you are on the fence on this policy proposal but know a change that would let you support it, I really encourage you to find the microphones. Okay. Lea and then the front mic.

MS. ROBERTS: Hi, I'm Lea Roberts from Stanford University and also on the ARIN AC. What Bill said, absolutely. I think personally I believe it's necessary to have a transfer policy, but I think this one, I'm against it as it's currently written because it's just way too nitpicky and complicated. And what Bill said, you know, made me think that the way the policy is written now, it only takes effect when the IANA free pool is depleted. So, you know, you can actually build on or modify the policy all the way up till that time. So it's entirely possible that we could start with a very simple, you know, we are going to allow a relaxed transfer policy and then as Bill says kind of build on the various pieces that can accomplish community consensus. So I think that's a great suggestion. Thanks, Bill.

MR. CURRAN: Okay, front center mic.

MR. SCHILLER: Jason Schiller, Verizon Business, NRO NC. With regard to Bill's point about the omnibus bill before us in the house, I think that, yes, there are a lot of amendments to this bill and that people may have some -- there may be some piece of the policy that they disagree with. In all, they might think that the transfer is good except for this one particular restriction. So I think it is important to look at the restrictions individually. However, I don't think it's as simple as simply looking at the restrictions individually because if you considered a transfer policy that had only any one of the restrictions named, certainly a lot of people who might generally be for it might be against it in that case. So, I think, yes, we have to look at each of the restrictions individually to see what we like about them and what we dislike about them. But then we also have to take a step back and look at the policy overall and make sure that we have enough of the right restrictions of the right type. So, I think in a sense, yes, we need to break it down and talk about them individually, but we also need to talk about the individual restrictions in combination with each other.

MR. CURRAN: Okay.

MR. SCHILLER: In response to John's question about what would you like to see in this policy to make it more desirable, I've been thinking long and hard about this. And from comments of what R.S. has said in the past and what Edelman has said and some other comments about -- that there has to be deaggregation, I think that there are maybe two things in concept that might help me with the transfer policy. And one is if we can make sure that if there is a market, that larger blocks on a per address basis are more expensive than smaller blocks. I'm not sure how we could manufacture that behavior. Certainly people that have a better understanding of economics and free markets, I would encourage to comment. And the second thing would be is if we could encourage people that are vacating space but need to keep some piece of it for their own use, if we could somehow make it beneficial to them to actually remember and give up the entire space as a single aggregate, I think that would go a long way to swaying me more in support of 2008-2.

MR. CURRAN: Okay.

MR. SCHILLER: So two comments. One, can we make sure that large blocks are more expensive than small blocks per unit? And two, can we encourage people to renumber out of space if they need to so they can give up an entire aggregate?

MR. CURRAN: Got it. Okay. Center rear microphone.

MS. ARONSON: I've got to wait for the mic. Cathy Aronson, ARIN AC. I'm not in favor of this proposal as it's written because of the same things that Lea expressed. That I think it's cumbersome and difficult to implement. I think there should be a minimum allocation size, and I think that people should be able to get what they need. And they should be encouraged in some mechanism to get a block -- contiguous block of the size that they need. And I think that I'm a little nervous about the six month requirement to be able to get more space because if the only blocks that are available are /24s and I can only get one every six months, then I'm not going to get what I need. So, I think that we really need to make it so that people get what they can justify.

MR. CURRAN: Okay, thank you. Response or -- microphones remain open. Go ahead.

MR. LEIBRAND: Response to a couple of things, particularly what Bill and Lea said. My own personal opinion is that one of the ways that we ought to go forward with this is the way we did our 2005-1, the IPv6 for provider independent space, which is to put out a survey of here are all the points that we think are up for discussion. Please give us your opinion on each point. And also, please give us your opinion on the entire thing. And that seemed to work really well for 2005-1, which was a complicated policy, and got us towards a consensus that then had quite a bit of support. So, I agree with the comments of Bill and Lea that there are a lot of aspects of this that we need to consider. And I think that there are ways to do that, and I think we will go ahead and do that. So, with regard to C.J.'s comment about there not being enough space and the six-month restriction being a hardship, I can see that as a hypothetical, but remember that we are talking about a market here. So, if there's a shortage, it probably means that's because the price is too low. And so there are a number of possible outcomes that might not be ideal, but remember here that we are talking about a flexible system that will tend to adjust to avoid those outcomes, and by doing so will provide the incentives for people to participate.

MR. CURRAN: Okay, got it. Center rear microphone.

MS. AZINGER: Marla Azinger, Frontier Communications. Right now I have one foot on the fence and one foot on the against side. So I'm still in the decision-making process. I have gone a full 360 of my opinion of this. I actually nine months ago wrote what I thought could possibly be a change to the transfer policy myself to try to grasp really what would take effect, and I've gone from that to thinking, oh, maybe this is a good idea, to, oh, wait a second. If I think about issues my company itself and I would face as a core network engineer, one of the main things is I don't want to get into a battle with marketing trying to sell my address space that I need to run the network. Because if marketing starts telling us that we're going to start using NAT on a large network, I'm sorry, I'm not a NAT lover. I don't want to get into a NAT discussion. But that's just how I feel about how we want to run our network. And a transfer policy change could possibly make that very difficult for us. And the other thing is, you have to put a lot of exemptions. We're all worried about nitpickies, which all of us hate nitpickies in a policy. However, when I went through and wrote it, I realized there needs to be a lot of exemptions written in so that we're not, pardon the word, screwing people when they need to come and get address space to actually run their business. So, we kind of have to come to a grip if we're going to do this that no matter what there are going to have to be a lot of exemptions written into it. That takes a lot of time because if you forget one, yeah, we're going to have to come back and write another exemption in to change the current policy if this were to go into effect. The other thing I'm now starting to more lean towards is, okay, maybe we should work on this, try and make it as good as we can, but not implement it unless we see specific alarms that we're concerned about go off. If we preset what we don't want to see happen and we start to see that happen, then we take this document that we already wrote and we emergency implement it. I kind of more lean towards that now because we're in a 'what if.' This is all a 'what if' scenario. And I would rather see some alarms go off before we implement something this extreme.

MR. CURRAN: Okay. Far left microphone.

MR. LEWIS: Ed Lewis, NeuStar. I want to second what Scott Leibrand had said just a little while ago about doing surveys. I hate surveys, but I think in this case we have so many diametrically opposed principles we're after that I think it's worth identifying what's the most important thing here. Like, I don't like tying any dollar value to an IP address. I don't like that at all. I think it's a very bad road to go down. But I want to incentivize people to become more efficient with what they've already had already on hand. So I think we do need to decide what are the most important things to solve and avoid. As far as the policy itself, I'm very undecided. I've gone back and forth because each principle comes up to me, like, this is important, but this is important. And this is important and that's important. It's hard to come to saying yes or no on this because we have so many things we're trying to fix at one time.

MR. CURRAN: Okay, thank you. Far right microphone. Owen.

MR. DeLONG: Owen DeLong, ARIN AC, speaking strictly on my own behalf. I don't buy into the premise that not changing policy is necessarily the most harmful thing we can do. There are many, many unknowns either way we go in this scenario. And I don't think that there is any way to develop enough data to really know which direction is more harmful. And I will point out that we have a great deal more operational experience with current policy than we do with what would be done by adopting such a radical policy. I'm still not convinced completely one way or another on this policy or on the idea of a relaxed transfer policy in general. But the more we look at the number of devils, and the number of details, and the number of moving targets and variables that have to be factored into any sort of change like this, and the fact that we don't even know of the 12 different competing diametrically opposed targets that we need to try and optimize for, which ones should take precedence, and to what degree over the others. I'm just not sure this can be done in a way that doesn't create more harm than good.

MR. CURRAN: Okay. Center front microphone.

MR. SCHILLER: Jason Schiller, Verizon Business, NRO NC. It's been stated a few times that this policy was the work of the ARIN AC over the course of many weeks, and many discussions, and many arguments, and is not necessarily the opinion of the AC in its entirety. I would like to hear some of the dissenting opinions from that group.

MR. POUNSETT: You've heard several.

MR. SCHILLER: Have we heard all of them?

MR. POUNSETT: Offhand I'm not sure.

MR. SCHILLER: I mean, certainly this output from the AC is great, and I'm glad that you guys got together and talked about these issues and came up with a policy. But if there were other things that were considered in your discussions that we haven't heard up until now I would certainly encourage the other AC members to make sure that those opinions are heard by the membership.

MR. POUNSETT: Okay. Cathy, do you want to respond directly?

MS. ARONSON: Yes, I would like to respond to that. I mean, the reason that the AC as a whole, you know, all of us probably haven't been up here speaking is because we want to know what you think. Okay, because we're here to put together a policy for this community. And it may be that I'm against this proposal, but if you guys convince me it's what you need, then I'm going to vote for it. So, you know, we come here to hear you guys. We want to know what you think. We want to know what you want. We want to know what you need. So, we want you up here, not us.

MR. CURRAN: Okay. Far left microphone.

MR. TITLEY: Nigel Titley, RIPE NCC Board, but in this case speaking in my capacity as one of the authors of the RIPE transfer policy that's currently going through our process. Yesterday, Jean Camp suggested that her predictions of the slow diffusion of IPv6 were largely due to the inability of the market to address long-term risks against short-term costs. The advantage of a transfer policy is that it can provide short-term gains that can be set against short-term costs, and, therefore, could actually bring IPv6 adoption earlier. That's just the point I'd like to make. The only one thing I would say is that the ARIN policy, because it only kicks in when the IANA pool exhausts, to some extent works against this.

MR. CURRAN: Okay, good point. Would you -- I know as a staff member -- as a Board member of another RIR per se is not necessarily a place you want to take a position, but would you advocate that this body consider a different implementation date or a different effective time for this policy?

MR. TITLEY: Yes, I would suggest that that particular clause get struck.

MR. CURRAN: Okay, thank you. Far right mic.

MR. WILSON: Thanks. Paul Wilson from APNIC. One of the premises behind the transfer discussion is that transfers are going to happen. In fact, the transfers are already happening. Personally, I'm fairly convinced by that by seeing the behavior of companies -- members of APNIC in our region. And I've also heard many casual comments from people in this meeting over some years now about how there's a market that exists already. But then occasionally someone will say correctly what's the evidence. They'll suggest that there is little evidence, and I think that's a problem as well. So I'm not sure that we've had too much discussion about this, but what are the mechanisms for actually finding out how much transfer of address space actually is happening outside of the system at the moment either by analysis of fragmentation and routing tables, trying to draw some conclusions from that, or else otherwise by a www.betareporting.com -- for instance, an anonymous survey targeted and very well publicized where we might get people to confess to some out of policy use of address space and so on. I think since this community is going to be -- and all the communities are going to be deliberating over this for some time, there is in that time plenty of opportunity available to start, you know, covering those bases as well.

MR. CURRAN: Okay, thank you. Do you want to respond, Bill?

MR. WOODCOCK: Yes. Paul, I absolutely agree that information would be extraordinarily valuable in the policy input process. But it strikes me that that's of exactly the same nature as, for instance, debate over illicit drugs. Right? People argue about legalization or regulation or whatever. And everyone wants to know in that policy debate just what is the size of the illicit market. And you get widely variant claims about that in support of different positions. And I suspect we would fall into the same trap, rather than arguing about what our goals are and what policy could be developed to support those goals. I don't want to fall into argument about what the nature of the illicit market is and how to control the illicit market. That seems to me like a losing expenditure of energy.

MR. WILSON: You sound like my father who once said that there's nothing like facts to spoil a good argument.

MR. WOODCOCK: I'm not sure that the -- I think it is very likely based upon observation that there exists some amount of illicit market. I haven't the foggiest idea whether it's getting larger or smaller. I believe it's not too terribly big.

MR. WILSON: Just one final comment. We're seeing fragmentation happening in the routing table in the use of addresses. The amount of fragmentation which comes from RIR allocations appears to be, according to Jeff, around 60 percent of the routing table. The long prefixes in the routing table are not any longer dominated by historical space. They're coming from fragmentation of existing allocations. And that gives us some pointers to a community of people following the WHOIS database and all of the reliable information we have there. It gives us some pointers to actually approaching and asking people to respond. And the people who are involved in this sort of thing may be more motivated to respond anonymously about what they're doing and why than the general population in drug use surveys.

MR. WOODCOCK: Is your worry that people are showing up to RIR policy development and making policy which is in explicit disagreement with their actual behavior, or is your worry that it is a different community of people that are showing up to the policy development process than who are crafting routing announcements?

MR. WILSON: Well, I don't know. But the community of people who we're referring to is a lot -- is many sizes larger than the number of people in this room. So there are a lot of people who aren't here.

MR. CURRAN: Our meeting locations probably are not attractive enough yet to get them to participate. Okay, Counsel, did you want to make a comment? No? Okay. Center microphone.

MR. DUL: Andrew Dul, Perkins Coie, speaking for myself. I'm just going to wrap up all my comments from this discussion into this list, I guess. There's been a lot of talk about transfers already happening. I agree transfers are happening. They are illicit, if you want to call them that. Just because they are, I don't necessarily believe that we 100 percent need to condone that with an open market-type arrangement. I think we need to look more strongly at how those effects will affect us long term. There's been talk about forcing pricing on larger blocks versus smaller blocks. That doesn't sound like a market to me, so I don't know how that fits. And to echo some of kc's comments yesterday, I actually think -- I absolutely think we need more scrutiny, more academia research, more people looking at this than people in this room. No offense to Ben or anyone else. I think you guys have helped us as we can, but we need more people from outside this forum. From as many places as we can get, because this is a huge change for us. And as much as we can get. Specifically, I'd like to see what is -- if we can get people to do some sort of modeling based upon assumptions. We realize that we don't have hard data to support what a market would look like, but we can model that based upon assumptions. And we could make educated guesses about what might happen using percentages of people willing to do certain actions. And I'd like to see some work done in that area to see if we can come up with that. And results published in some sort of manner. I'd also like to see other types of research published in terms of liquidity, concerns about hoarding, why we believe hoarding would not occur under certain types of arrangements. How we can make sure that liquidity actually exists, such that we don't have price fixing or those types of issues. I also have concerns about how transfers would delay IPv6 transition. That's probably of paramount concern of how this policy would affect that in the long-term stability of the network, our communities, and our businesses. And lastly, I have concerns about how this policy could lead to more walled gardens by service providers, as well as more NAT throughout the networks. I think that's it. Thank you.

MR. CURRAN: As the policy is written --

MR. DUL: I am opposed to the policy as written and transfer policy in general. I'm not convinced it's the best move for this organization at this time.

MR. CURRAN: Okay, thank you. I was going to suggest something, but we'll let Counsel -- go ahead.

MR. RYAN: I wanted to oppose a particular issue that seems to be -- I haven't heard enough discussion either for the AC or for counsel to evaluate. The key difference between this policy proposal and the APNIC and RIPE proposals is that the ARIN proposal as written contemplates that a need would still have to be proven. That is, the person who -- if I believe I read the policy correctly, indicates that the person who wants to avail themselves of the rule of the transfer market would have to demonstrate a need. And I think Paul characterized it as get a ticket from ARIN. And I think that's correct. From a lawyer's standpoint, the APNIC and RIPE proposal is very easy, because then we don't have to worry about all of the mechanism of doing that. On the other hand, there's a stewardship responsibility that has to be considered here. It seems to me that if the AC is going to be helped by the discussion from the floor, I'm picking this as an example. I'm putting before you to encourage that you discuss as a separate thread. But what we need to do is establish several of these separate threads and actually have people stop reacting globally to the transfer policy and react to these individual slices of it, particularly where the other regions may be contemplating a different policy. Let me make one comment then that's unrelated to that. There's been several lists -- it's like a repeated raising of the issue of will the United States Government as an example of a sovereign government within our region, would it be regulated at the Securities and Exchange Commission if we did something like this? I just want to review briefly that my judgment is in the United States, as one example of the 19 sovereign governments, or 20 in our region, the SEC would have literally nothing to do with a transfer policy related to IP addresses. The SEC regulates publicly held companies. They will not care about this. The consumer -- pardon me, the CFTC, the agency that does look at commodity markets, will likely never look at this either. It will never be the kind of commodity that is addressed in the bundle of rights that we're talking about on a floor of an exchange like they do. So that the government regulatory mechanisms here that will look at this is an antitrust review potentially by the Department of Justice or by the FTC. And I think we're very, very comfortable in knowing, for example, with our Canadian counterparts and our Caribbean counterparts, how we would address those legal issues once you make a decision as an entity.

MR. CURRAN: Thank you. So Counsel raises in particular it would be good to have a discussion on the fact that the difference in this transfer policy compared to those in other regions is the need to go to ARIN and demonstrate the need for resources. Does anyone want to speak about that aspect being a particularly good or bad idea? So, Bill first.

MR. WOODCOCK: I'd like to say that this is the one really critical difference from my point of view. There are many other aspects of the policy that I consider to be necessary find tuning in order to make it work as well as it might possibly work. But this is the core of the policy that we have spent the last 10 years in a public forum having open debate about what the requirements for justification of address consumption are. That is, who gets to have IP addresses and why. And I think that it is vitally important that we continue to build upon that good work going forward rather than using, you know, vague notions about a free market that shouldn't be regulated for some reason as an excuse for throwing out all of that really, really good, constructive work that's been done over the years.

MR. CURRAN: Okay. Far left microphone.

MR. LEWIS: I think getting the ticket from ARIN is the most essentially good part of this proposal. I think that has to be there.

MR. CURRAN: Center mic.

MR. DARTE: Bill Darte, ARIN AC. I also support the fact, the ticketing notion. I think anything other than that says that the course that we've trod all this time we're abandoning. That it is free market, open, and it makes us irrelevant as far as I'm concerned. And so I support it fully and wouldn't support any transfer policy that didn't have that as a component to it.

MR. CURRAN: Okay. Far right microphone.

MR. DURAND: Alain Durand. I'm not sure I'm for a transfer policy in general, but if we need to have one, I think that the need to have basis should be in it. I absolutely support this view because this in essence is why we are all here. This is the bottom up process that actually have proven to work over many, many years. And I don't think that throwing in the towel and essentially letting the baby go with the bathwater is a good answer.

MR. CURRAN: Okay, thank you. Center rear microphone.

MR. DARTE: Bill Darte, ARIN AC again. I'm sorry, I meant to say that IPv6 as it emerges assuming so, we still have a stewardship responsibility for conservancy there as well, and we have spoken to that issue by, you know, addressing, you know, /56s as opposed to /48s as the minimum, et cetera. So I think it's consistent with our future goals as well not to abandon this notion of need based allocations.

MR. CURRAN: Okay. Microphones remain open. One of the items that also stands in difference between this policy proposal and those in the other region, not that that's good or bad but it's something to note, is that this one does specifically take effect upon completion of the IANA free pool. To that extent, there's been comments that that may have positive aspects or not having to worry about this proposal till then, and negative aspects in that it can't be utilized by someone who might want to desire to do so at that point. Does anyone want to speak on the start of the effective date of this policy proposal and the merits of that either way? Okay, I've got Bill and then far mic. Go ahead, Bill.

MR. DARTE: Bill Darte, ARIN AC. I think at the point of exhaustion of the IANA free pool is the appropriate, or in that range, is the appropriate time. Because, again, I think this policy -- the signal that should be sent is that this is an emergency transitional mechanism. It's not the future mechanism by which we see allocations and transfers and things working in the future. I see this as a signal that this is a transitional policy and that's all it is.

MR. CURRAN: Okay. Far left mic.

MR. LEWIS: Ed Lewis. I actually don't like that. I'd like to have this policy take effect immediately to more quickly go to make more efficient use of what we've already given out and maybe delay depletion of the IANA free pool.

MR. CURRAN: Okay. Far right mic.

MR. WILSON: Yeah, that was -- Paul Wilson from APNIC. What Ed just said was one of the points I was going to make. I've had numerous discussions about this particular issue. The advantage of bringing this kind of policy in earlier is said to be both the provision of the alternative mechanism or means for those who want addresses to receive them from somewhere other than the RIRs and the IANA free pool. Another point being that the existing well-known pricing structure, if you like, attached to RIR membership will help to tell people what these addresses are actually worth. That will obviously be influential in a market situation. The other one would be a softer landing. So right until the free pool is exhausted and then to unleash this untested opportunity and facility for transfer of addresses is, to me, scarier than to allow it to happen in a more gradual fashion in the meantime. Thanks.

MR. CURRAN: Okay, Matt, do you want to respond?

MR. POUNSETT: Yeah. We actually considered a few different ways to approach the start date. We considered everything from, you know, immediate or, you know, as soon as we can get it running start date all the way to ARIN's free pool exhaustion. And a couple of the competing interests that we took into account there were the earlier we started up, the more of a smooth ramp up we get in availability, price, you know, all those sorts of things. And -- but on the other hand, the earlier we start up, the more likely we are to either give the impression that we're trying to extend before a great deal, or actually actively extend it, neither of which we felt were really good things to do. So we decided eventually to go with IANA exhaustion because it was, we felt, a reasonable balance between those interests. But, you know, we really want to hear other people's opinions on that.

MR. CURRAN: Okay. Center front microphone.

MR. SCHILLER: Jason Schiller, Verizon Business, UUNET, NRO NC. So far we've heard two opinions, which is we should start it immediately or we should wait until depletion occurs. I don't want to debate those two points, but I want to bring up a third point that was mentioned by Marla. I think that there are probably a number of people in the room, and I see Alain is at the mic over there, that feel that we don't want to necessarily decide on, and support, and enact a policy based on a what-if scenario. I think that there are many people that probably would be in favor of a transfer policy rather than having transfers occur on the black market, but at this moment they may believe that there is no black market for address space, and maybe there never will be. There are other people that believe there already is a black market. The point I'm trying to make here is Marla brought up an interesting point, which is rather than get into a critical situation where emergency policy without the community has to be enacted, it would certainly be more beneficial that the community could agree on an emergency policy that could be enacted at a certain time given a certain event. For example, if someone demonstrated that there was a severe black market in addresses, let's immediately enact this policy as soon as the AC and the Board can mobilize it. I'm not suggesting that that should be the event that causes the policy to be enacted. I'm not suggesting that there should only be one event. But I think that certainly that would help remove doubts from some people that are concerned about voting on a policy to be enacted based on a what-if scenario. And I think that they would be more comfortable if that what-if scenario actually started to play out and we enacted the policy. So I hope people would consider that.

MR. CURRAN: Would you -- I want to get an idea of sort of which idea you're advocating here. Would you like to see something in NRPM with some specific triggers attached to it? Or would you like to see discussion of policy that we never quite put in NRPM that's just there and available for the Board to enact when they feel, or when the triggers that we've talked about come up?

MR. SCHILLER: I think either way is acceptable as long as it's clear that the membership has decided that this is what the text of the policy should be, and these are the triggers, and this is how you measure the triggers. Whether that's actually recorded in the NRPM or not, I don't think is really that important, except that I think that the membership may want to revisit the issue in six months and maybe revise the triggers or revise the policy. So, I think it needs to exist somewhere where everyone can see it and maybe in the future revise it. But it doesn't necessarily have to be in the NRPM. It can either be in the NRPM going forward or it could be an emergency policy that has support by the community that can be enacted at such time.

MR. CURRAN: Okay. I'm going to be closing the microphones. Approach the mics if you need to comment on any aspect of this policy proposal. Center front.

MR. FARMER: David Farmer, University of Minnesota. I would support a multi-trigger kind of idea because there probably are more than one reason we might want to do something like this. The one thing I would say, though, is we know that there's a trigger that's coming, so we can probably plan for that one. And this is going to be a big change, so we need to make sure that we ramp into this and make sure that we have enough time. Let's say we guessed wrong on the mechanisms. We need to tweak it. It would be good that we have enough time to tweak it before we actually have to rely on this policy to keep things going. I'm not sure when the IANA free pool is exhausted is enough of a cushion for that because everybody is going to panic then. And so it would be good to have this enacted a little bit before panic sets in. Hopefully, enacting this doesn't cause the panic, but, you know, that's just an opinion.

MR. CURRAN: Okay. Thank you. Microphones are now closed. Only people remaining in line. Far right microphone.

MR. DURAND: Alain Durand. I would like to remind us that about two years ago most of the industry was in denial about this incoming IPv4 completion. And things have changed. At the same time we talk about this transfer policies, but we are very early in this process. There are a lot of unknowns that we don't have good answers for today. So I think that it would be wise to not implement it now, to keep the discussion going, and try to understand better what will be the consequences of implementing such a thing.

MR. CURRAN: Thank you. Far left mic.

MR. HAIN: Tony Hain, Cisco. I have a couple of different comments. One, I think that you should actually enact it right now if you're going to enact it to work through the issues because it will take a while. And while the IANA cushion is not enough, I absolutely agree you probably don't have enough time and enough cushion anyway. You should have started two years ago. The other point is the whole needs based justification, I think, needs some serious deep thought, because one of the things that this community has been very clear about is not making any statement about the routability of any prefix that may be assigned. But the entire point of need in a market driven address space transfer is to get it routed. So, you're stepping across that line whether you intend to or not, so that's why I think you need some serious deep thought as to exactly how that plays out. And that's exactly why you need to move quickly as opposed to wait until later because if you wait and then realize you're stepping across that boundary, you may not find a good, clean foothold to get across.

MR. CURRAN: Okay. Good point. Center rear microphone.

MR. DARTE: Yeah, to reiterate -- Bill Darte, ARIN AC. To reiterate, I think it's all about sending signals, and I think the signal that should be sent now is that this is a problem that we should educate in our role to let people know that this is a problem, that we are not abandoning our stewardship or our role in supporting people's acquisition of IP address resources. We should let people know clearly that we have a response plan in place and to keep that dialogue going as long as possible. But to clearly state that this is a transitional process.

MR. CURRAN: Okay. Thank you. Far right microphone.

MR. HAZEN: John Hazen from Tellme/Microsoft. I want to say I'm in favor of having some need-based clause in the policy. And that if the criteria for the need-based clause is the same as the existing criteria for getting allocations from ARIN, then I see no detrimental effects other than possibly the philosophical thing of we plan to keep supporting v4 of implementing immediately rather than waiting for exhaustion. And I think the benefit of that is that, you know, there can be a gradual price setting or whatever, that while we have the ability to have allocations directly from ARIN and transfers, then I think there will be a smoother transition than trying to implement it when you're already done.

MR. CURRAN: Okay. Thank you. Center front microphone.

MR. LEIBRAND: Scott Leibrand, ARIN AC. I just want to caution about triggers, that we don't put triggers that get triggered too late. Effectively the policy has a trigger right now, which is IANA exhaustion. I think that's a good and valid trigger that gives us just barely enough buffer time to get this thing ramped up. I'm sympathetic to the argument that we should do it earlier. I think it's dangerous to do it any later. So, I would caution against trying to put in triggers around black market formation or such like because if you wait until something like a black market is up and running, you're behind the curve and you're going to have a lot harder time getting people back out of the black market into the transfer policy than if you have that thing available from the beginning and they haven't already committed themselves to using the black market.

MR. CURRAN: Understood. Microphones are closed unless you want to directly respond to that.

MS. AZINGER: Can I?

MR. CURRAN: We have two people who have stepped up to the mics. Mr. Hannigan, are you responding directly to this point?

MR. HANNIGAN: Martin Hannigan, Verne Global, undecided. And wake up, the black market is here.

MR. CURRAN: All right. Front.

MS. AZINGER: I do agree. The black market does exist. It's not gigantic, though.

MR. CURRAN: Okay.

MS. AZINGER: And for markers, the triggers that Scott says -- by the way, I am Marla Azinger again, Frontier Communications. I am on Advisory Council as well. Those are not the triggers that I agree with. Mine are more later. Not that IANA run out. Mine is I want to make sure that we're not putting something in place because of a 'what if.' The black market may not become a gigantic monster. We're assuming it will, or some are.

MR. CURRAN: Thank you. Okay, microphones remain closed. Final speaker far side.

MR. GRUNDEMANN: Chris Grundemann, Colorado Chapter of the Internet Society. I just want to say I'm against the transfer policy in general. I think from a stewardship standpoint the idea would be to promote IPv6 and not to promote IPv4. To do everything we can to push people to IPv6, and I think this is counter to that. I also think that privatizing or giving the impression of privatizing a private resource such as IP space, it sets a bad precedent and invalidating that with a dollar amount on IP addresses, I just don't like the idea of it in general. I think that there is a black market, and I think that the only compelling strong reason I have heard for this policy is that there is going to be or that there is a black market. And I actually feel the opposite. I think the fact that there is a black market today is evidence that this policy is not necessary because there is IPv4 space available right now. You can go to ARIN and get space, but people are trading for it already. And so when it's available and they can get it and they're still trading for it, I don't know that a transfer policy is going to stop that in the future. I think that the current system that's in place of allocating numbers and of returning numbers that you're not using should stay the same.

MR. CURRAN: Okay, thank you very much. Okay, that concludes the discussion. I'd like to thank Matt for your ample guidance. (Applause)

MR. CURRAN: So, the AC has an interesting job to do. And with some policy proposals it's more interesting than others. And as several members actually said at the microphone, while they're elected experts from the community, they also want the community's input to make sure they're following the guidance of you to get this job done. So, I'm going to ask for a few shows of hands. I'm going to ask them in a very particular order based on the discussion that we've had. And the first one will be on whether or not ARIN should continue work on a relaxed transfer policy. That's the only question. Just should ARIN continue work on a transfer policy? And to the extent that someone says yes, then the AC will continue. And the extent that the consensus says no, the AC will have to take that into consideration. Again, this item on the AC work list was sort of prompted by the Board, so it's not inevitable that ARIN needs to go this direction, but clearly there are other RIRs doing it.

So, I'm going to make sure my pollers are in position. I'm going to ask for a show of hands. It's going to be everyone in favor of ARIN continuing work on a relaxed transfer policy. A vote of aye or raising your hand says ARIN will go that direction. So, on the issue of Policy Proposal 2008-2, in general, all those in favor of ARIN continuing work on a relaxed transfer policy, please raise your hand now. Nice and high. Okay. All those opposed to ARIN continuing work on a relaxed transfer policy, raise your hand now. Very high. Nice and high. Thank you. Waiting for the poll results. Okay, on the matter of continuing work on a relaxed policy, the transfer policy: Number of people in the room, 123; number in favor, 49; number against, 7. This will be provided to the AC for their use. We've discussed various aspects of the Policy Proposal 2008-2. Presuming that work continues on such a policy proposal, I'd like to know of the people in the room who was in favor of us retaining a need based qualification for recipients. So I'm going to ask for a show of hands on that topic. A vote of aye or raising your hand says that in a transfer policy proposal adopted by ARIN, you would like to see the need-based qualification retained in such a policy. A vote of no says you do not wish to see that retained. So, I'm going to ask for a show of hands on retaining a need-based qualification in our relaxed transfer policy. All those in favor of retaining a need-based qualification, raise your hand. Nice and high. This is actually the exercise benefit of your ARIN membership. (Laughter)

MR. CURRAN: Okay. All those opposed to retaining a need-based qualification for recipients, raise your hand. Okay. Thank you. Okay, 123 people in the room. No one managed to sneak out in the last three minutes. On retaining a need-based qualification, in favor, 54; opposed, 3. This will be provided to the AC for their consideration. Presuming ARIN should continue with work on a relaxed transfer policy and continue with a need-based qualification of recipients as currently stated, one of the issues that's come up is whether or not this policy should begin sooner than the depletion of the IANA free pool, whether that should be considered by the Advisory Council. As stated, it begins upon depletion of the IANA IPv4 free pool. So, under the assumptions that ARIN continues work on this proposal and that a need- based qualification is retained, I'm going to ask for a show of hands for all those who believe that this policy should take effect sooner than the depletion of the IANA free pool. A vote of aye is a vote saying the AC should consider an earlier effective date. A vote of no is that it should be retained as is in 2008-2. So, let me make sure my pollers are ready.

On the assumption that ARIN continues work on the policy proposal and maintains a need-based qualification of recipients, all those in favor of an effective date sooner than the IANA IPv4 free pool depletion, raise your hand. Nice and high. If you're behind a tall person, raise it extra high. Thank you. And all those opposed to an effective date for the policy proposal sooner than the IPv4 IANA free pool depletion, raise your hand. Thank you. On the matter of an earlier effective date than the IPv4 depletion date, in favor, 35; opposed, 13. This information will be provided to the AC for their consideration. Thank you very much. (Applause)

MR. CURRAN: I'm sorry, is there someone at the microphone?

MR. DURAND: Alain Durand. I wanted to point out that not wanting to have this policy enacted now is not necessarily incompatible with wanting to have it sooner than the IP for completion. There's a gray zone in between.

MR. CURRAN: There is a very gray zone.

MR. DURAND: My question was, it could be a little bit in a weird situation.

MR. CURRAN: It is very truly a very large zone. Unfortunately, we can only make incremental progress towards consensus, and I can only poll for that which I've seen adequate discussion on both sides. Okay? Thank you.

NRO Activities Report

MR. PLZAK: Thanks, John. Now Paul Wilson, who is the current Chair of the Executive Council of the Number Resource Organization will give an activities report. Paul?

MR. WILSON: Good morning, everyone. The NRO in 2008 has had its regular rotation of office holders, so in addition to myself as Chair, Adiel Akplogan from AfriNIC is the Secretary. And that makes AfriNIC the secretariat of the NRO. And the Treasurer is Axel Pawlik, so off the hook for the time being are Ray and Raul. But thanks very much to Ray for his services as Chair last year.

As some may know, we did finally make steps towards the finalization of the relationship between the RIRs collectively under the NRO and ICANN last year in December, with an exchange of letters which was less formal and less binding mechanism than the sorts of contractual relationships and arrangements which had been discussed for quite some time. It was still some reasonable step forward, I think. What the exchange of letters served to provide is a statement of mutual recognition and support between the NRO, RIRs, and ICANN. They clarified that the NRO will continue to make voluntarily contributions to ICANN, which is something of a change from the more formal service- oriented fee and fee-oriented relationship which had been discussed for some time. The letters also say that we will pursue further contractual arrangements of a more -- as it says, a more appropriate formal relationship between the two parties within a year. So if you'd like to have a look at those letters, they're on the NRO website. A long URL there, but you can also find it fairly easily from the site itself.

We all attended an ICANN meeting recently. The latest one being in February in New Delhi. We spoke -- appeared and spoke at a session of the Government Advisory Committee of ICANN on IP addressing on the now-common themes of v4 consumption and v6 transition. We have seen increasing interest in the ICANN community in this topic, and at the next meeting we'll be spending at least, I think, a couple of hours with the GIC on a more extended discussion about this stuff. And that's a very substantial allocation of time for the GIC. You're lucky to get 10 or 15 minutes with them normally, so that's an indication of how important stuff is being seen within ICANN. We also had, thanks to IANA, a workshop, or appeared at an IANA workshop on IPv6 during that ICANN session as well. And I guess that's sort of an informative or educational-type of session that's going to, I expect, be continued as well. We met -- in accordance with the exchange of letters, we met for a review of the contributions that the NRO makes to ICANN, the voluntary contributions. And we agreed to some total of $823,000 U.S. as the contribution for this financial year, that is for 2008. And we agreed that we'd be reviewing that on an annual basis with no particular discussion about the basis for this fee or basis for any increases. It's just, as I mentioned, a voluntary contribution that we make in support of ICANN.

The NRO also, in an associated move, we wrote a response to the U.S. Government's call for comments on the JPA. So we said a few things there, but basically the message is that the NRO continues to be supportive of the ICANN model. We support the ending of the JPA in due course. We didn't go into the -- and quite deliberately didn't go into the details. There were 10 points that were under discussion about ICANN's performance, but the message behind our response was really where we're now in a position -- we feel ICANN is in a position for us to deal with ICANN directly on those issues of their performance and all of those details. And don't really see it as necessary to contribute that through this formal U.S. Government-sponsored process.

The Internet Government Forum is the aftermath of the WSIS, the World Summit on Information Society, in the working group in Internet governments. It's a process we heard about from Lynn St. Amour yesterday. During the last year we were quite active in that. We had, fortunately, I think, two of our NRO representatives, members of the AC, both Adiel Akplogan and Raúl Echeberría, on the multi-stakeholder advisory group. And that's a group whose seats are very hard to get and under a lot of contention. So to have the two NRO reps there, I think, was a good evidence of the success of the work that we've done in promoting the importance of having our expertise available to this forum. And so in its typically opaque style, we're not sure what's happening to the MAG of the IGF this year, but it's under some discussion. At this last meeting of the IGF that was in Rio, in December last year, we had a NRO display and new had materials. We had produced a report called "The Continuing Cooperation Report of the NRO." And that was circulated there as well as there were numerous presentations and sessions going on there. And it was just part of the ongoing road show that we're finding ourselves involved with. The next meeting we'll be doing it all again. It will be at the end of this year in Hyderabad in India. Now, you can't see the beautiful screen grab cover that I prepared for -- which is just a copy of the cover of the NRO continuing cooperation report, but you might be able to see the URL there or as I said before, you can find that easily linked from the NRO's website.

So activities coming up this year, again, the road shows. We're doing the addressing stuff at ICANN meetings, of course. Representing ourselves in the IGF in the run up to the IGF and in the event at the end of the year. We've got two of the ITU telecom events, regional events coming up in Africa and in Asia this year, Cairo and Bangkok. And the NRO is doing a small activity. They're represented by the corresponding regional Internet registry, who is sort of coordinating that show. That'll be AfriNIC, of course, and APNIC for those two. We've been discussing a more visible statement of support for IPv6 or the position of IPv6 among the NRO. Looking at the possibility of setting a timeline for some sort of full support as defined by all of the RIRs for IPv6. I'll say a little more later about what that means -- what that's meant in APNIC's case.

Concretely for the NRO, one issue there that we do need to discuss, and that's happening through the engineering group, is how we're going to go about the ipv6.arpa, the management of the reverse delegations for IPv6. And the other thing that is on the engineering front is the resource -- the issue of resource certification across the NRO and what the NRO is doing there is nothing operational, of course. But within that engineering group looking at to what extent we should be defining a common architecture and what that common architectural should look like in the event that resource certification is undertaken on a global basis or amongst a number of the RIRs. And the old chestnuts, IPv4 and IPv6 issues, are under active discussion within -- obviously, within each of the Board governing bodies of the RIRs, and also there's some coordinating discussions going on among the RIRs. Boards -- we had one of those in formal meetings this week. And we'll probably be having at least a venue provided for that kind of discussion to be happening at all of the future or hopefully all of the future RIR meetings. It's unlikely, unfortunately, that all of the RIR Boards can ever be in one place at one time. So those meetings are informal and they really get together for an exchange -- for an ongoing exchange of view about these current topical issues. So thanks. That's all from the NRO.

MR. CURRAN: Thank you. Any questions to Paul as head of the NRO? Rotating head.

MR. WILSON: Rotating head, exactly.

MR. CURRAN: No? Thank you very much, Paul.

MR. PLZAK: Okay, we are at time for a break. We're a little early, but that's okay. So, let's take a break. Get out to the Cyber Café. Remember, fill out those surveys. Help desks are open. Please take advantage of them. And be back here at 10:45 by this clock. (Recess)

RIR Update - APNIC

MR. PLZAK: Okay. Welcome back. We are now going to have another one of our regular reports. This is the report from APNIC. It'll be presented by the Director General Paul Wilson. Paul?

MR. WILSON: Thank you, Ray. Okay, you get to hear from me again. And this is the -- an update of a few things, items of news from APNIC. I try to vary this report from meeting to meeting so that, well, it gives me something to do on the plane on the way here. And also, maybe, if you're interested, you might just see some different aspects of what we do at APNIC.

The structure, these days, has got four -- three different areas under the DG. They are the services, communications, and business area. We've also got the mad chief scientist out there as a separate member of the senior executive team at APNIC. So that's -- that team of people is now complete. We had some vacancies over the last year, but we've been recently joined by German Valdez, who you might remember from previous ARIN meetings having reported on behalf of LACNIC, which is where he earned his stripes. But he's moved to Brisbane recently and joined us. I don't normally show you what our staffing looks like, but, so, that's the -- I thought that might be of some interest. We have got just over 50 staff at the moment at APNIC. The rate of growth is tailing off, although -- and in the last year, we had no net growth at all. That was more for financial reasons than operational needs. But the pattern there is definitely of a decreasing rate of growth, and a fairly low staff turnover, as well.

The planning process at APNIC, I thought I'd say a few words about this. It's something that we're developing now with this new senior management team that I mentioned. The process is based around, or seeded by, the regular membership surveys that we have, now, every two years. We conduct or contract KPMG to anonymously survey the community -- not limited to members, in fact, but members and stakeholders. And so we use the results of that anonymous survey, which are quite detailed, to base our future planning around. So using those results, and also the results of our open policy meetings, both formal policy decisions and the operational outcomes of those, but also less formal inputs from our other special interest groups, we plan activities in an internal secretariat process. We base our budget on that, and that goes out through the Executive Council of APNIC, the Board, for monitoring during the year.

Now, one of the challenges that we've had in that planning process, in recent years, has been related to this graph here. I guess it would be this. What you're looking at here would be familiar in form to all of you, because this is your currency. Against the Australian dollar, the U.S. Dollar has really diminished lately. So, back in January 2001, we were getting 2 Aussie dollars for -- just over 2 Aussie dollars for each U.S. dollar that we received. And that crashed to just around -- just over a dollar over the previous six years. So that's both currencies moving in different directions. But you can see that that would cause a fair amount of problem for an organization that's spending Australian dollars in trying to provide services. Youch. The other problem here is the unpredictability of it. So, we started the 2007 calendar year with an official forecast that we'd be getting $1.27 for each -- local -- for each U.S. Dollar. But from that point on, from the beginning of the year on, it still went down and down and down. And it was looking, at one point, of going down to a dollar or lower. So, we really needed to do something about that. You can't run a business with those sorts of variations happening even -- happening live, with no ability to predict. So what we finally managed to do, after numerous discussions within the community, which were not only about our fee levels, but, in fact, about the overall structure of APNIC fees. We did, at least, make the small change of converting our existing U.S. dollar fee structure into -- from U.S. Dollars to Australian dollars. So, that really solved what could have been a -- or addressed what could have been a very bad year this year for APNIC, where you set that fee structure at that rate that you see there, which was the predicted rate for 2007.

The interesting thing about this is that we've got a lot of currencies in the region. I don't expect you to read this, but it's available online. We've got a lot of currencies in the region and the actual impact of changing our fees to Australian dollars on each of those currencies was completely different, widely varied across the regions. So, the latest APNIC newsletter, APSTA, has got this analysis which I've written up, which shows the impact for a whole selection of currencies in terms of what a small membership costs in that local currency at -- over the last, quite a number of years, since year 2000. So, that shows that in most cases, for most currencies, the amount that APNIC costs now is still lower than it has been in previous years during 2000 -- since 2000.

Okay. Just a few graphs here, showing what we're doing in terms of services. The top left is IPv4 allocations measured in total /8s. And so we've gone in 2006, having allocated just over 3 /8s, to in 2007, having allocated about 4.25 /8s. That's a fairly large increase in rate. The next one across at the top right is IPv6. This is something I mentioned yesterday during the discussion about where IPv6 deployment is happening, if anywhere. And it's certainly not in the Asia-Pacific, as you can see here, in any -- it is happening, but it's not in the sorts of numbers that some people sometimes assume. So, 2007 seems to have been our biggest year in terms of total prefixes allocated. But it's still just over 50, 55 prefixes in total. We're reporting prefixes here because, well, the size, the variation in size of v6 allocations which are made is pretty dramatic. And the actual numbers in terms of addresses don't seem to make too much sense just yet. ASNs, bottom left, are moving up, and members are also moving up at a slow, but exponential accelerating growth of -- rate of growth, which is good to see. And you could contrast that, those growth curves, with the staff growth curve, and it's looking fairly good in terms of the overall efficiency, I think, of the operation. IPv6 is on everyone's lips at the moment.

So, as I mentioned earlier in connection with the NRO, APNIC has a goal for 2008 of having -- providing full support for IPv6 across all of our services. Mostly it's done, and there're a few things underway, but these are things like making sure that our IPv6 request process is as fully automated as it is for IPv4. Our request management process, our internal resource management system, support for IPv6 is improving. Our DNS management, also. We've got, I think, all of our online services now speaking v6. So, DNS services websites, MyAPNIC, all dual-stacked. And also, importantly, that when we're doing that, we need to actually have reachability, and so we've quite some substantially improved the v6 connectivity into our sites. And there's a lot of communications involved. So, in the training area, there's IPv6 training under continual development and being deployed quite often. A lot of information services and, also, the notion of IPv6. And I put that in brackets because we're not doing, sort of, IPv6 evangelical sessions. I don't think any of the RIRs are, and we're not being asked to. But we're being asked to promote awareness of the need, and the benefits and realities of transition to IPv6. And that does seem to be quite a dire need in our part of the world. And I won't click on this, but we do have a little -- a happy little IPv6 man on our website. So if you connect with v6, you get a little treat there. And that's just following the Kame turtle and others who've done similar stuff.

SPEAKER: Treats?

MR. WILSON: Hmm? Treats. Treats. Okay. This is the policy meeting and so I thought I'd mention the policy outcomes of our latest meeting. That was APNIC 25, which was held in Taiwan, alongside APRICOT, earlier this year.

We had three proposals accepted at that meeting. One was to change the minimum IPv4 allocations size to a /22. The next is some operational changes in relation to our new automated DNS shared zone management system, between APNIC and the NIRs that we support. A proposal to change the IPv6 initial allocation criteria. So in addition to the requirement to plan for 200 allocations, or assignments within the first two years, there's now an alternative, which is that you don't have to have precisely that plan. But if you have v4 addresses and you're basically acting as an ISP, providing assignments and so forth, then you can qualify on that basis rather than the 200 assignments. So we had, also, a policy proposal for the allocation of the remaining IPv4 address space. That's the proposal that's often referred to as the last /8, or the last five /8s. The proposal that the last IANA blocks be distributed to the RIRs. That one's gone back to the policy special interest group. We abandoned a policy on cooperative distribution of the end of the IPv4 free pool. And, also, a proposal to create an IPv4 shared use address space among LIRs, which was, in effect, an addition to the RFC 1918 address space. And further work by the authors is being done on the IPv4 Resource Transfer Policy, and on another policy from IPv4 Soft Landing. Now, coming up at APNIC 26, which is our next meeting, we've got a proposal already submitted which has got to do with generating valid routing registry data from the resource public key infrastructure, so the resource certification system is able to be used to construct the data for input into the routing registries.

Okay. Meetings. I don't think I've ever been able to announce four upcoming meetings at one time, but here are the next four. We're meeting in Christchurch in New Zealand this coming August. In Manila, next year with APRICOT 2009. That's the regional operators' conference. We're meeting at a city to be announced in China in 2009, and in Kuala Lumpur in 2010. And all of you would be very welcome to come to any or, indeed, all of those meetings, if you can make it. Thanks very much. (Applause)

MR. PLZAK: Anybody have any questions of Paul?

2007-23: End Policy for IANA IPv4 allocations to RIRs

MR. PLZAK: If not, Chairman John Curran had to leave the meeting, so Scott Bradner, the Vice Chairman of the Board, will now assume the duties of Chair of this meeting. And we are now on to Policy Proposal 2007-23. This is a global policy proposal in that all of the RIRs have to agree to a common text of the proposal. And for it to be enacted globally, it would have to be approved, or ratified, I should say, by the ICANN Board. The action by this group would be whether or not to accept this proposal. And it would be, then -- and when the Board adopted it, it would be contingent upon implementation upon the exact same text being ratified by the ICANN Board. So, having gone through that convoluted explanation, this proposal's been around since August of last year. It's -- at the last meeting, it was -- community consensus was revise it. And you can see that it is a global proposal. It has been under -- it is -- at the last APNIC meeting, it was returned back to their list. And it will be discussed at the upcoming meetings of AfriNIC, LACNIC, and the RIPE NCC. The text is in your meeting packet. Basically, the understanding is, is that IANA will take the last five /8s of IPv4 space in its free pool and allocate one to each RIR. The AC shepherds are Dan and R.S. No liability risk is seen by the council. We do have some concerns from staff, in that it conflicts with the spirit of RFC 2050, Fairness and Efficiency. But staff, on the other hand, sees a minimal impact, saying that 30 to 90 days after ratification by the ICANN Board, we could implement this policy. Here's the PPML discussion. Not much. There was a lot earlier, but there hasn't been much since it's been revised. And so, I would like to call on, I believe, Izumi? Or who's going to present this? Oh, there she is. I was looking for you over there. How are you?

MS. OKUTANI: Hello, everyone. My name is Izumi Okutani, and I am here to present a proposal on how we're going to distribute the last piece of IANA IPv4 address blocks to RIRs. And this is actually a joint proposal made by AfriNIC, LACNIC, and JPNIC team. So I'm just here to present on behalf of the whole team.

The brief introduction. This is intending to address a very short-term issue, when we face the time of IPv4 address exhaustion, to clearly define how we're going to distribute the last piece of IANA blocks, so that there will be a minimum of confusion when we try to, you know, finish up the IANA blocks. And there were actually two proposals made at the last ARIN meeting. One was the same proposal number as this one, and, basically, the same concept as this proposal submitted this time, which is to allocate a single /8 to each RIR. And the second proposal was made by AfriNIC and LACNIC team, which was to allocate two /8, instead of one. And the discussions at the ARIN meeting was that, generally, there were favorable comments, generally supportive comments, for the idea itself. And the show of hands expressed was roughly 70 for the support and 10 against. And there were more support confirmed for one /8, instead of two /8s, for the allocation size. And the council also confirmed that there were no negative legal implications for ARIN if we implement this policy.

So, what we've done after the last ARIN meeting was that we got advice from the AC for the two authors of the two proposals to get together and draft a single proposal, since it's trying to address the same issue. So, we tried to maintain the same concept as what was proposed in the last meeting, and then stick with the size of one /8 to each RIR, instead of two, as there were more support for the size of one for --- within the ARIN community, as well as in other RIRs' communities. And we have also taken in the concept of reserving this particular block for our standard IANA pool, which was an idea introduced in Proposal 2007-18. And our proposal itself is very, very simple. As you can see in the diagram over here, we proposed to separately reserve five /8 from the standard allocation blocks that IANA makes to RIRs, as you can see in the diagram below. And five because we currently have five RIRs. And then, we expected IANA to carryout allocations as it is under the current policy for the standard IANA pool, under -- as it is under the current practice, until the last RIR makes an allocation request and standard IANA block is gone. At that point, IANA will allocate from the reserved IANA block to each RIR and, as a result, each RIR will get a single /8 each. That's what it is. Very simple. And then we tried to consider what would be the implications of implementing this policy. Firstly, possible negative implications. Some people may feel concerned that for RIRs with high consumption rates, such as ARIN, APNIC, or RIPE, they would lose out the chance of receiving more than one /8 allocations by restricting the size as one. But when we actually think about the possibilities, depending on the timing of the request, those RIRs, for example, ARIN, equally would have had the chance of receiving no allocations, whatsoever, if the timing of the request is too late and IANA block is gone. So, yes, it's true that ARIN will lose out the chance of receiving more than one /8 blocks, but it also ensures that ARIN will, at least, receive one /8 at the very end, instead of zero.

So it's very difficult to say that ARIN will or RIRs with high consumption rate will lose out as a result of this policy. And also, as introduced by Ray, since the consumption of one /8 is an equivalent of one to a few months' worth of consumption in RIRs with high consumption rate, so, I think it is also fair to say that the impact is minimum. And also, in terms of positive implications, it ensures and makes sure that RIR will know what size they will get at the very end. So, it would reduce confusion as much as possible, and makes the size of allocation predicable. That's all. Thank you.

MR. BRADNER: Okay. Microphones are open. Any discussion? Don't all leap up at once. Ah-ha, an instigator. (Laughter)

MS. TAYLOR: Stacy Taylor, Time Warner Telecom. I support this proposal. And ARIN AC.

MS. OKUTANI: Thank you.

MR. BRADNER: Front mic.

MR. FARMER: David Farmer, University of Minnesota. In general, I support the idea. The only question I would have is since this is sort of figuring out the timing of the trigger for the previous proposal, some thought should be given to that. In other words, is this, you know, is this -- does this pull the trigger at the right time? Would allocating two at the end give enough cushion for that trigger, or something along those lines, would be my only consideration. Otherwise, I fully support it.

MR. BRADNER: Okay. Don't go away from the mic yet. One of the things Ray pointed out in the initial -- when he presented it originally, is the text for all of the RIRs needs to be the same.

MR. FARMER: Yes.

MR. BRADNER: So, is that -- is your concern enough of one to push it back into the reset button to start thinking about that in all the RIRs again?

MR. FARMER: I don't know. I think the whole end -- I don't know that -- this is sort of one of many proposals around the endgame. And so, does -- should this one be the first one or should this one come in at the same time as the other endgame proposals get finished up? I don't know. It's a valid concern. I honestly don't know.

MR. BRADNER: Would you support the proposal as presented?

MR. FARMER: I would support the proposal as written.

MR. BRADNER: Okay.

MR. FARMER: Just saying, hey, you know, it's the trigger for the endgame. Is this when we want the trigger?

MR. BRADNER: It's a good consideration.

MR. LEWIS: Lewis. We've spoken about this proposal last time around, and I think that the first thing I want to say, this is a good end signal saying that v4 is about to run out. This way, no one's really caught by surprise. You have one /8 left to play with. The other -- the question that was just brought up about one or two, we've debated this in Albuquerque, also. And I -- it's not like I say we shouldn't debate it again, but I'm sure people who weren't there might want to hear more about why -- I sort of prefer one because it sticks us closer to the needs-based allocation until the last minute, and then still gives the signal that, okay, we're almost done. Also, I believe there're some other considerations that people might want to bring up about that. But we have discussed that and, for those who haven't heard it, it's probably good to go over it, if they really want to hear it again.

MR. BRADNER: Far microphone.

MR. DeLONG: Owen DeLong, ARIN AC. I support the proposal as is. As my predecessor said, we debated one versus two pretty thoroughly in Albuquerque. The conclusion reached there was one. I don't see any reason to change that at this point. And I think this allows us to better get on with some of the other endgame proposals that are needed.

MR. BRADNER: Rear center mic.

MR. ECHEBERRÍA: Thank you. Raúl Echeberría from LACNIC. A chance to make a minor correction about the status of this policy in LACNIC region. The policy has been approved in LACNIC region, but with a larger number of /8s to be distributed. Due to the fact that it needed to be harmonized with other proposals, as Izumi has explained, it has to be -- it needs to be voted -- discussion again in the next meeting. But the concept has already been approved. So, I think that it will be approved again with the minor number of /8s to be distributed. Thank you.

MR. BRADNER: Thank you. Thanks for that update. Center front mic.

MR. LEIBRAND: Scott Leibrand, ARIN AC, Internap. I support the proposal. I think it's important that we, like we've said, get on with the business of figuring out what to do with the last /8. And this kind of sets the stage for that. With regards to the question about triggering, since this is a global policy, and this seems to be the consensus among everyone else that we should do one /8, I think we should go ahead and approve this. And then we can set our transfer policy, or whatever other endgame proposals we want, to have language that deals with that, and sets the date to the date we want, based on the fact that we know that there's going to be this one /8 given out to everybody at the end. So, I think this is the prerequisite to doing anything else endgame, and we should advance it.

MR. BRADNER: I think Ray was next. (Applause)

MR. PLZAK: Yes, Scott. Just as a manner of mechanics regarding the common language between the RIRs, the ASO MOU provides for the staffs of the RIRs to get together to perform any minor editing and grammar corrections they maybe need to make to get a common language.

MR. BRADNER: But it wouldn't extend to changing the number of --

MR. PLZAK: No, no, no, no. It's strictly grammar and that kind of thing.

MR. BRADNER: As long as staff knows their position. Far mic, and then the table.

MR. DURAND: Alain Durand. I would like to say that I support this proposal for (off mike) region. I think that (off mike) right set of compromise and wish that we could pass it today.

SPEAKER: Thank you.

MR. BRADNER: And on the table? R.S.

SPEAKER: (off mike)

MR. PLZAK: I'm sorry? No, I already spoke. (Laughter)

MR. BRADNER: Having too much fun with your e-mail?

MR. SEASTROM: I'm having too much fun trying to shuffle through the -- collect my thoughts. I'm just -- I've heard this sentiment several times that this provides ample warning that we're running out of space, and that, you know, that time has finally come. My theory on this is that when that time finally comes that the IANA free pool is actually exhausted, that it's going to be showing up in places from The Economist to the Wall Street Journal. And that there will be -- nobody will be able to say that they haven't heard this. So I think that's sort of a red herring of an argument. I am opposed to this as it is currently written.

MR. BRADNER: Center front mic.

MR. SCHILLER: Jason Schiller, Verizon Business, UUNET, and the NRO NC. I just wanted to point out a little bit of some of the discussions that have occurred in other regions regarding this policy. The intention of this policy, basically, is when we are in the endgame and there's only one /8 left, we may want to do something different with it. We may want to give out space from that last /8 in a way that's different than our current policies support. And what I've in some of the other regions is that they feel comfortable with that approach, however, they don't want to ratify this policy until they know what the rules are going to be about how the space is dealt with. So, I just want to caution the ARIN membership --

MR. BRADNER: Just a clarification. Do the other RIRs believe that there'll be a global policy on how that space is -- the last /8's dealt with? Or would that be a local decision?

MR. SCHILLER: No, specifically, that would be a local decision because each region may have differing needs based on how they want to consume that last block. The point I'm trying to make is that some of the other regions have concerns about ratifying this policy without knowing how the block is going to be used. And I just want to urge the ARIN membership to think about whether or not they are comfortable ratifying this policy without knowing how that last block will be used. And also, to encourage people to start writing policies about how that last block should be used. Thank you.

MR. BRADNER: Back center mic.

MR. DARTE: It's Bill Darte, ARIN AC.

MR. BRADNER: Are you trying to speak to this point?

MR. GAGLIANO: Hello? Yeah, I'm Roque Gagliano. I'm LACNIC staff, but I take that hat off because I'm also one of the coauthors of the policy. I just want to mention that it's true, what Jason said, it's just one region that came out with that issue. But you have to understand that this is a global policy. And it's a policy that takes, you know, two years, perhaps, in the whole circle, to get approved. And those -- and actually, that's something in favor of this policy, is that it allows the regional RIRs to have those discussions. And that's something that, I think, is going to be encouraged by the approval of this policy, in the sense that there's also been concern on people saying, we don't want to start discussion whether it's going to happen with that last /8, we don't even know we can have it.

MR. BRADNER: Okay. Thank you. Center back mic.

MR. DARTE: Yeah, this is Bill Darte. I'm on the ARIN AC. I support this policy as written. And I think that, you know, to Rob's point, I really think that notification about these issues shouldn't come as a surprise. One of things that I think the ARIN community needs to do is get on the PR bandwagon and make these issues much more widely accessible to people in the broader community, and the impacts that it might have on them.

MR. BRADNER: Any other discussion? Do you want to say anything else?

MS. OKUTANI: No.

MR. BRADNER: Okay. Thank you very much. (Applause)

MR. BRADNER: Okay. Got some input for the AC? There doesn't seem to be a whole lot of variant in the discussion, so I'll just ask the simple question. How many support this proposal and how many oppose it. And, so, first, how many -- are the counters ready? No. We've got to wait for the counters to be ready. Don't have enough practice at this. Okay. Counters are ready? All right. So, how many support this proposal? Keep them up. You're not -- your hand's not up. Put it up. There you go. There you go. There you are. Got to really participate here. Do we have a count? We have a count. Okay. So you can relax. And how many oppose this proposal? The count should be quicker. Okay, you can put your hands down now. Okay. On the matter of Policy Proposal 2007-23, the number of people in the room is 123. The number who's indicated in favor is 48. The number of opposed, five. This information will be provided to the AC.

2007-14: Resource Review Process

MR. PLZAK: Okay, the next proposal is Policy Proposal 2007-14, Resource Review Process. First introduced in May of last year and was discussed at the last ARIN meeting. The proposal has been revised. This is an ARIN-only proposal. Text is in your packet. And understanding of the policy is that it provides clear policy authority to audit, or reclaim resources, guidelines for how it shall be done, and a six-month minimum grace period for the user to stop using the space. The shepherds are R.S. and Leo. And the legal liability risk is none. However, counsel did make a comment about this, supporting this version of the policy being an active (off mike) adoption policy could reduce future legal fees. The only concern the staff has is about sharing results of audits regarding fraud or suspicious activity investigations and see a minimal impact if we did put this into place 30 to 90 days after ratification. Not much discussion on the list. And the proposal will be presented by R.S., and Scott will lead the discussion.

MR. BRADNER: While R.S. is trying to outsmart his computer, I want to mention that in the last two shows of hands, we've had remote votes in both of them, one each. I'm not sure whether it's the same person each one, but there's one each. I think that's as good, but we didn't really prepare for that. So we don't have any process in place for dealing with remote votes or of ensuring that remote votes are adequately represented in the discussion. (off mike) tells me that the staff is looking into ways to do that in the future. So while we appreciate the remote input, we will note it in the record, but it won't be counted as -- particularly in the vote count, per se. But certainly the AC will understand that that's part of the record. But in the future, we'll try and figure a way to empower the people that can't actually get here. Thank you.

MR. SEASTROM: Thanks, Scott. So, I'm Rob. I said I would present this for Owen, who's a little under the weather. 2007-14 has already been presented here once. There have been some changes made to it. Wow, look at those transition effects. (Laughter)

MR. SEASTROM: The existing policy does not address audit or reclamation. The NRPM is silent on it. The RSA uses terms like "cooperation." It goes into some detail about the kind of information you might be asked for. The authors were concerned that a little bit more framework ought to be defined. So, the intent is to provide a framework for audit and reclamation, and provide protection for registrants from unreasonable or abuse of audit processes, and provide a defined appeal process, and a mechanism for cooperating with ARIN to return stuff that you may not be able to justify having. So, ARIN can -- this policy preserves ARIN's ability to go after fraud at any time, while offering some reassurance to registrants that there won't be audits happening so often as to represent an unreasonable burden to their business process. It requires that ARIN disclose summary results to the holder of the resource, and provides the processes that we were talking about a moment ago. More framework is desirable, but we wanted to be able to have ARIN be able to -- whoops, I'm off a slide. That doesn't make any sense. So -- actually, I'm just going to skip forward to that. Never mind. Okay. So, ARIN was concerned that release of too much detail about the audit process might enable workarounds and compromise the integrity of future audits.

MR. BRADNER: I'm not one to criticize what you can do with a Mac, but. (Laughter)

MR. SEASTROM: I was not the person who did this and I flipped through it in composition mode, so I -- frankly, this is as much of a surprise to me as it is to you. (Laughter)

SPEAKER: Okay. I would call this --

MR. SEASTROM: And I have to say, I'm pretty impressed, too. (Laughter)

MR. BRADNER: I would call this decomposition mode. (Laughter)

MR. SEASTROM: So, so --

MR. BRADNER: Just the facts next time. (Laughter)

MR. SEASTROM: So here are the revisions that have been made since the last time you saw this policy. Owen, where are you? This is the last time I'm doing this for you. (Laughter)

SPEAKER: And that's how you've got --

SPEAKER: (off mike) and change to the disco sparkle transition, and you'll be set. (Laughter)

SPEAKER: And now you've got the --

MR. SEASTROM: You know, this is just too difficult. We're going to do that.

SPEAKER: (off mike)

MR. SEASTROM: Yeah, I see that. So, you should have the text in your folder. Are there any comments or questions? Any questions about the presentation itself will not be entertained.

MR. BRADNER: Yeah, we'll ignore the presentation, if we can. The microphones are open. Any discussion?

MR. DeLONG: Okay, so, Owen DeLong, ARIN AC. I'll apologize for the excessive transitions. The staff concern, actually, was, I think, pretty much addressed in discussions with the staff offline, in that what is required is only the disclosure of, you know, we've taken a look at your resource utilization and no further action is required at this time. Unless, of course, there is some other action that they feel needs to be taken. In which case, I don't see how they could get that action taken without contacting the resource holder. So, the policy really doesn't require the staff to disclose the mechanism by which they audit things, or the fact that they are taking a look at something, until they've come to some conclusion on it. And so, we don't believe -- the authors don't believe that that will actually negatively impact the staff's ability to look for fraud.

MR. BRADNER: Any other comments? Yes.

MR. LEWIS: Ed Lewis. I have a question about the staff concern. Was the concern that they were afraid, dealing with a suspected case of fraud, that they didn't want to accuse the person of fraud? Was that the concern, or?

MR. BRADNER: Could we have whoever from the staff who wants to speak for the staff?

MS. NOBILE: Yeah, that was one of the concerns. We don't want to alert people when we're suspicious of them and, yeah, what you said.

MR. LEWIS: Do you think that the proposal now, as it's being presented, addresses that or are you still concerned with that?

MS. NOBILE: I don't think it addresses it, no. But I do -- well, I mean, we had a suggested rewrite of that, but it wasn't taken. We thought that it should say we'd only notify the customer that we did an audit if an action was required by one of us. If there was no action and we were just looking, we didn't think we should be notifying them. Because, like I said, we don't really want to alert them to that fact, if we're actually suspicious. And, we don't want to accuse them. So, it is -- it's -- you know, double.

MR. BRADNER: Any more? Yes.

MR. LEIBRAND: Scott Leibrand, ARIN AC. I have a question then. I got the impression, from the discussion, that telling people the results of their resource review will only happen if you actually, like, ask them questions and initiated something. If you just did behind the scenes investigation of your own, that that wasn't properly an audit that had to go through this process. So, I'm wondering if maybe the authors could speak to the --

MR. BRADNER: The text in the proposal is, "ARIN shall communicate the results of the review to the organization."

MR. LEIBRAND: I guess, what constitutes a review? Is it a review that involves the --

MR. SEASTROM: I'm going to defer to the authors on that. But having discussed that with, you know, as a shepherd, I've never been of the -- under the impression that anything informal really counted as an audit.

SPEAKER: So, why don't we --

MR. DeLONG: As one of the authors on the (off mike) ARIN AC, the intent in stating that staff would communicate the results was the authors didn't believe you had results until staff had come to a conclusion of any audit. Therefore, communicating the results at the end of such process, it is our opinion, shouldn't really interfere with the process itself.

MR. BRADNER: I think the question is does the staff, if they're just taking a look, but not doing a formal review, have to report that -- the results of the informal look -- to the organization?

MR. DeLONG: I would say that an informal look would be part of an ongoing review, and there's nothing in the policy that limits the amount of time ARIN takes to conduct a review. So, I would think that the correct action for staff in that case would simply be to keep the review open, until such time as they come to a conclusion.

MR. BRADNER: It does sound like there is a disagreement over when something triggers the requirements for disclosure, et cetera. Just keeping something open as an escape mechanism to not report, doesn't sound quite ideal. But other comments? I think you were -- far mic.

MR. LEWIS: I just want to finish off my comment that I think the proposal as it is, is a good idea, but I have a big concern about the issue that the staff has had. And that makes me say I probably don't want this until the staff is happy with what they would have to reveal. I have a hard time going with the proposal as written.

MR. BRADNER: I think the front center was first. Okay. Back center.

MR. HANNIGAN: Martin Hannigan, Verne Global. What's the check and balance to make sure that this process isn't abused? Is there something in the back office that takes place that this just can't be implemented for any reason whatsoever? Does there actually have to be a reason?

MR. SEASTROM: Well, the whole idea was that it could be done for any reason, or no reason, whatsoever, but that it was limited to happening every 24 months unless there was extenuating circumstances.

MR. HANNIGAN: Complies with the Privacy Policy as stated on the website?

MR. SEASTROM: Pardon?

MR. HANNIGAN: Complies with the Privacy Policy as stated on the website? Privacy concerns, I mean.

MR. SEASTROM: We were more concerned -- I think that the authors were more concerned about having it happen too often, like getting audited every three months.

MR. HANNIGAN: Okay. I'm for this.

MR. BRADNER: Owen, you wanted to speak to that?

MR. DeLONG: I don't see any way it would affect the Privacy Policy on the website, because the disclosures required under this are strictly between ARIN and the organization that has received resources. There's no external disclosure.

MR. BRADNER: Okay. Center front mic.

MS. SCHILLER: I had a question for Registration Services.

MR. BRADNER: And you are?

MS. SCHILLER: Heather Schiller, Verizon Business. My question is, when you perform audits today, when, for whatever reason, but I imagine it's largely when folks come back for additional resources, what is -- how often can you perform an entire audit without input from the resource holder?

SPEAKER: Do you want to answer --

MS. SCHILLER: It seems like you would, in order to really perform an audit, you would actually need input from the resource holder. So, that's how I see it. And so I'm wondering, like, how you could even do an audit without getting their input? So, how would you even be able to an audit without them knowing.

MS. NOBILE: I'm trying to figure out how to answer that. We do audit them when they're coming back for additional resources. That is generally when we're interacting. And that is true, we will ask questions. But there's other times when we have other red flags, that I don't want to go into, that, you know, that will cause us to take a second look at someone who's not in for resources. Just, you know, searching around the web, checking out, you know, customer lists, or, you know, we just do a general look. And that's sort of a semi-audit, you know. And we wouldn't necessarily want to inform someone of that, if we're just taking a look.

MS. SCHILLER: So what he's saying to me off to the side is the difference between an informal audit and a formal audit. And so if you're formally auditing someone, I would imagine you'd have to get to the point where you do talk to them about the results, and that, in this case, you would be notifying them. Because you're asking them for information, you are notifying them that you're doing this audit. And so, I don't know if that needs clarification or not, but. Thanks.

MR. BRADNER: Thank you. Center back mic. Wait until the mic turns on. Okay. Try that. No? Doesn't work. Why don't you move to the front mic if that one's not working?

MS. WAGNER: Hi, I'm Jennifer Wagner with San Isabel Telecom. I'm just asking about item 7. It says, "ARIN shall continue to maintain the resource(s) while their return or revocation is pending, except no any maintenance fees assessed during that period shall be calculated as if the return or revocation was complete." Just --

MR. BRADNER: That is accurately read and inaccurately written. (Laughter)

MS. WAGNER: Okay, but how should it read?

MR. BRADNER: How should it read? Owen?

SPEAKER: I'm going to defer to the author on that one because we didn't pick that one up either.

SPEAKER: (off mike)

MS. WAGNER: Okay. Thank you.

MR. BRADNER: I assume that the word "any" is extraneous.

SPEAKER: And you'll perhaps --

MR. BRADNER: The text currently reads, "While their return or revocation is pending, except no any maintenance fees assessed during that period shall be calculated as if the return or revocation was complete." So I think the word "any" is just extraneous.

SPEAKER: That's correct.

SPEAKER: Other way around.

MR. WEILER: Sam Weiler. I think it's the "no" that's extraneous. But I'm sure you all can wordsmith that.

MR. BRADNER: I think either one could be extraneous, but. A good catch, in any case, that the AC will be cognizant of the next pass.

SPEAKER: Strike the word "no."

MR. BRADNER: Center front mic.

MR. BARGER: Yeah, Dave Barger, AT&T. Item 1, "ARIN may review the current usage of any resources issued by ARIN to an organization." Does this mean to imply that ARIN will not review resources, legacy resources, prior to ARIN's existence?

MR. BRADNER: Thank you. I'd hoped somebody would ask that question, since I'm related to an organization that has some legacy resources and I couldn't. Thank you. So, would somebody answer that?

MR. DeLONG: It's the belief of the authors that doing so is somewhat outside the purview of ARIN or, at least, nebulous at best. So, we felt it best strictly to address the resources that were clearly under ARIN control.

MR. BRADNER: So you meant it as the text is in the proposal?

SPEAKER: Okay. Thank you.

MR. BRADNER: Leslie wants to speak?

MS. NOBILE: Yeah, I'm actually glad someone caught that, as well. We did not. Currently, ARIN, under ARIN policy, we can audit all resources issued to a resource holder, when we're reviewing requests for additional space. This seems like it may limit our ability to do that, so -- and I've used this example before and I hope no one minds, but if you're a Class A holder and you come to ARIN for additional space, and we have -- if we have no right to ask about that Class A, it seems to me that it's a waste of resources and it's just not efficient. So.

MR. BRADNER: So, I think the fundamental, underlying question is does this replace or usurp the normal review done when resources are requested? And that's a question to the authors of the proposal.

SPEAKER: Can I comment on that one?

MR. DeLONG: Can you look at section 8 of the proposal, please. I believe it addresses that concern.

SPEAKER: Yes, it does.

MR. WEILER: On the paper version, it's the last line.

MR. BRADNER: That says, however, "The legacy resources in active use, regardless of utilization, are not subject to revocation by ARIN pursuant to this subsection. However, the utilization of legacy resources shall be considered during a review to assess overall compliance." But that conflicts with the strict wording of the first one -- bullet.

MR. LEIBRAND: I'd like to comment on that point. Scott Leibrand, Internap/ARIN AC. I've always thought that the word audit is something you go back and do after the fact, and is not part of the application for new resources. This issue is actually part of the safe harbor concerns with 2008-2, as well. We're making a safe harbor against audit. We're not intending to make a safe harbor against getting reviewed when you want new resources. So, maybe we need legal input here or whatever. I think there's an intent here of both this -- these authors and the authors of 2008-2, that the word "audit" generally refers to after the fact review of resources already allocated. And there's a separate process of resource review that is put in place when you go for new resources, and we're not talking about that. And Owen has specifically put in text to try and confirm that he's not talking about that. So I don't think the two conflict in his policy, but I think there's a more general question about what's an audit?

MR. BRADNER: Just to be clear, the term "audit" is used post -- ex post facto, post an event. But it's also used to find out how many of something you have, at any time, independent of anything else. To audit, audit your returns or audit the number of books you have on the shelf, they're both good uses of the term and so that may be confusing.

MR. LEIBRAND: Yeah, so I think that we need to define, for purposes of ARIN policy, whether we call the resource review, at time of request for more resources, an audit or not. And make sure our language is consistent with our intent.

MR. BRADNER: Bill.

MR. WOODCOCK: This is kind of a minor nit, but resources assigned by ARIN, while it may sound unambiguous, is a different criterion than is used in other places. And it brings up the ambiguous case of things that have been traded under the amnesty policies. So, if someone has legacy space and they've handed in all or a portion of the legacy space to ARIN and received another assignment in return, those amnesty policies specifically say that that space is to be treated as legacy space for billing purposes. But this might, I mean, if taken literally, that's ARIN-assigned space, even though it's ARIN-assigned space in exchange for legacy space, you know, and intended to otherwise be treated the same as legacy space. So, you might, instead, just look at saying that the supplies to people under RSA, under the terms of the RSA that they've signed.

MR. BRADNER: Far mic.

MR. WEILER: Thanks, Scott. Sam Weiler. I generally support this. Addressing the concern about when does an audit start, I would propose that the metric that be used -- I don't think this necessarily requires a textual change -- is at the moment that ARIN asks for data from the resource holder. I would strongly support a hard cap on the duration of an audit. Someone mentioned the risk that ARIN would keep one of these open for a long time. I strongly support a hard time cap. And lastly, section 7 talks about fees. We try not talk about fees in policies. And I'm just a little bit wary of that. That's all.

MR. BRADNER: Okay. The other two people at that mic, and then the center mic afterwards.

MR. WEST: Ken West, Sprint. I have a question about the informal versus formal audits. Specifically, on section 2b, "Any time without having to establish cause unless a prior review has been completed." Now is that a prior formal review or informal? And I'd suggest that "formal" and "informal" be used in front of "review" or "audit" in this document across the Board.

MR. BRADNER: We could also ask the authors what the intent of this was. Is this to minimize the disruption on the part of the auditee? Or is this to keep the auditee from feeling picked upon or what?

MR. DeLONG: So, a few clarifications. Number one, there is no intent or, I believe, language in this policy that states that it is intended to affect the resource application process. It is assumed that the reviews of resources under that circumstance would remain unaltered by this policy, and ARIN would continue to behave as it does in that regard. The intent is to create a framework whereby ARIN clearly has the authority to audit utilization of resources at other times than applications and to prevent abuse of that process by, say, conducting an audit every six months and asking the resource holder for a bunch of hard to produce information or what have you. By putting a 24-month limit on that, the intent is to, you know, prevent ARIN from coming back to you and picking on you too often. Because for certain large organizations, the process of complying with one of these audits could actually be somewhat expensive. Having said that, there's clearly, you know, an escape clause for ARIN in that if they think you've committed fraud, they can come back at you more often than that. And that was intended to address several concerns that were expressed by staff and by legal. And, basically, the intent here is to affect, primarily, resources that are covered under RSA. But in the case of someone who has both RSA and non-RSA resources, we would not want to exclude their utilization percentage of their non-RSA resources from being factored into the overall utilization percentage of their total resources. Resources issued by ARIN was used specifically to include pretty much everything that is affected by an RSA. I believe, although I could be wrong about this, that resources issued under the Amnesty Policy are, actually, covered under RSA, even though they are not subject to fees because they have a specific fe