NRO/RIR reports
19 October 2018
9 a.m.:

ANTHONY GOLLAN: Welcome to everyone's favourite meeting slot.

I see a lot of Brian well rested faces out there. We are going to mix things up a lit bit, a shortened RIR update session and have unusual NRO and IANA updates, and then my colleague Mirjam will take over for chairing the discussion about the RIPE chair selection process. Also just a reminder if you are registered to vote in the NRO NC elections voting closes at 10:30 so please cast any vote before the end of the session. So if Alun is ready he can kick us off.

ALUN BARRETT: Good morning. I am the CEO of after AFRINIC and I was asked to keep this short and leave out most of the boring statistics we usually present so I am going to try that.

If I read properly here it said "this is not a touch screen". There is a label there.

So, we have recently implemented a new policy around lame delegations. The idea is that if in your Whois records for your Reverse‑DNS you have listed some name servers which aren't working properly, which are lame, then AFRINIC will eventually delete them, but before deletion we go through a process of sending you messages asking you to fix it and this graph you probably can't read the numbers but see it it goes down. On the left‑hand side we have the 29th September, I think, and on the right‑hand side we have today, and the height of the graph shows the number of lame delegations that we have measured in our Reverse‑DNS. Since we started telling people your servers are lame please fix it, they have been fixing it, great. Also, we have had some activity on our Internet routing registry side, some of you guys might be aware that the RIPE routing registry recently went through some changes and you no longer allow out of region IP addresses. And so, big surprise, use of the AFRINIC greeting registry suddenly kicked up a notch the day after RIPE made those changes.

We have also made changes in our routing registry that route and route 6 objects need to be authorised by the IP address holder as always and but no longer are by the ASN holder, although we do check the ASN is valid, we check that it's been issued by one of the RIRs. We have got several policies recently implemented or under discussion. So, there is a policy on, well three different policies on IPv6 which have been ratified by the Board and staff is busy implementing it. It's really around clean‑up, bringing our IPv6 policies into the modern world, taking away some obsolete requirements.

There is another policy on discussion about reworking our policy development process, proposes quite some big changes to the policy development process. There is a policy proposal on number resources review, often called the audit proposal, the idea there is that AFRINIC, an audit that you are using your resources in you way you said you would and eventually could revoke them if there is a problem.

Number 3 there, IPv4 soft landing bis, we are in the middle of our IPv4 soft landing, about 0.4 /8s left, similar level to what is available in the RIPE region, when we get down to a/11 a new level of policy kicks in and we will start burning them probably around the same speed or slower than RIPE is.

Proposal number 3 then is a complete replacement of that soft landing policy to the, I think the idea is to make the rules even stricter.

There is a recent proposal about making abuse contacts mandatory, today it's optional to have an abuse contact, so there is a proposal to make it mandatory and to make AFRINIC verify that email to that abuse contact actually gets a reply.

And there is another IPv6 policy under discussion, it's about sub assignments. Currently, we have the concept of sub allocations and sub assignments, one is to another level of LIR Internet service provider and one is to an end user, but in IPv6 the concept of sub assignments doesn't exist, there is a proposal on the table to add that. So, thank you.


PAUL WILSON: Good morning, hello. Always a pleasure to speak to the hard core of the RIPE community, thanks for your dedication this morning. A quick update from APNIC, again not talking about the usual numbers but in our case, talking about the latest APNIC survey which was conducted and delivered to us just in the last couple of months, so this is hot off the press, like RIPE NCC we do a periodic survey, in our case it happens every two years, and it's the main input into the forward planning processes that we have, obviously.

First half of the survey is about quality of service and value and feedback on our performance and it's always pretty good. I mean, on that sort of, grey being neutral in the middle, it's almost all green on both quality and value, the two bars there, sorry about the line wrap there, but the two bars there are for the last survey in 2016 and the latest survey in 2018, showing just a consistent and slightly increasing regard for APNIC services and value, which is good to see.

The second part of the survey is about, which I will get on to, is about the future priorities that APNIC will undertake to survey our members, a couple of others on our assessment, though:

The question on the upper right is a bit less ROSIE than the other one, it's about how the respondents speak to others about APNIC. On the right‑hand side are those who speak positively about APNIC without being asked, then those who speak positively when asked, then those who are neutral and on the left of those are those who speak critically about us when asked. And as you see, at least that has also improved substantially over the two years as well.

We asked a question about transparency as well is whether app Mick is transparent in its operations and that has improved markedly over the two years between these surveys after some concrete efforts that we made to address that issue after the last survey. That is our performance so far. There is a lot of questions there, this is all available for review on the APNIC website of course.

As I said the second half of the survey is about the challenges of members and what members expect APNIC to do and so this sort of broadest question about that is to ask members to list the one or two main operational challenges that they face and network security has increased as the main operational challenge very markedly over the years and it's way above the rest as a single challenge for our members, followed then, I guess you can read this, but scarcity of IPv4 addresses is the next one, and then hiring and keeping skilled employees is the next one, so these things really do ‑‑ these are very meaningful results for us, they will be taken into account very much directly and explicitly in the work that we are doing at that broad level.

So yeah, there is the top three that I have just listed.

Now, we drill down further and look into the sort of challenges for members and the expectations of APNIC in respect to those challenges as well. So as I said, security is the top challenge, asking within security what are the main challenges, phishing, spam, malware, then DDOS attacks, then intrusion and other breaches. So they are the sort of issues that our community, service providers are dealing with. And the question of how APNIC can help, well, no surprise, given the last slide about challenges is that members are looking for specific security‑related training courses, so that's what we are being asked to do, training is always through the surveys that we have been conducting now for 20 years, training has always been a top priority for APNIC services and so that continues.

Also on security is collaboration with technical security, agencies and organisations and also sharing security insights on the APNIC blog etc.. this is pretty rich sort of feedback for us and it will be reflected in the planning that is currently being done for this year and the following year as well.

IPv4 scarcity is an interesting one, the ‑‑ that was the second main challenge of all challenges for members, drilling down into that the biggest challenge there is deploying IPv6, followed by the cost of buying v4 addresses, followed by the cost and complexity of NATS. And again, what should APNIC do? Now, this is interesting because we are being asked by 57% of respondents to get involved with reclaiming or recovering of unused address space. So, that might sound a bit radical and alarming but of course these things are ‑‑ these responses are subject to some interpretations and so the sorts of things that we might be looking at there on consideration would be identifying unused address space and encouraging those who have it to put it out on to the market because we have certainly got a long waiting list of approved transfer recipients in our system.

We might also be looking at our softer methods for contacting people and asking whether they would return address space that is not being used. We have to define what it is for address ‑‑ address space to be used in the first place, of course. But the very last thing that we would be doing would be any kind of unilateral reclamation of addresses paced on some unilateral definition of address space usage. That is sort of discussion we might have about the ‑‑ about results like this. The other scarcity responses that we have are about monitoring and reporting, usage of address space and, again, another big one is providing incentives for return of unused address space.

So that's a snapshot, really, of where we are receiving our main and most important inputs into the APNIC planning process. What generally what happens with these surveys is that they are commissioned by the APNIC EC so they have got that high level of authority, the EC at the moment is considering a formal kind of response to that survey, public response, that simply says to the community what it is that the EC has read and acknowledged in the survey, and where it is in broad terms it that we should be heading with this, and we should be able to see explicitly in the planning, as I said, in the future plans, the forward plans, for instance, for 2019, some of these very specific issues and topics.

So, thanks very much. There is website. Dot blog for the blog; dot acadamy for the academy; dot foundation and it's all under/social and of course under/survey are all the results of our surveys for the last 20 years for those who would like to follow up. Thank you very much for your time.


JOHN CURRAN: I'm John Curran, president and CEO of ARIN, I am going to give a quick update, we are not doing the laborious everything we have done, just a few slides to give you an idea.

Again, ARIN serves US and Coanda, about 25 Carribean and Latin American economies. We serve 38,000 organisations, about 16 legacy organisations which have their assignment before our formation, about 16,000 end user organisations and about 58100 member organisations, these are generally ISPs what you would call LIRs in this region.

So, what have we been up to, in a middle of major website redesign, the ARIN website is a nice website but it hasn't been refreshed in a while, we did a lot of user interface testing with the community to try to figure out what would make for a more usable website and online system and we actually at the meeting last week, two weeks ago we actually sat down with the community and walked them through a preview of the website so we are happy with that and that will push out between now and end of the year.

We have done a number of community consultations, when we do operational changes that aren't regarding the policy manual, with do that through ARIN's consultation and suggestion process. This year involved the Internet routing registry roadmap for ARIN, ARIN is going to be doing a new IRR, something we haven't put any resources in the last five to eight years and so we will be doing a new one that is integrated with the ARIN database. We have been doing consultation on the ASO future structure, changing how Whois queries are done, securing them, we did a consultation on expanding the ARIN board size which we decided not to do, we had a fee schedule change which we did a consultation, we have had consultations on use of our mailing lists, like should people be able to send attachments on our mailing list when in a discussion of a topic. We have been doing discussions about about removing orphaned records, clean‑up of points of contact organisation records that have either no ‑‑ no relation to resources whatsoever. They have just in the database and has mot been uptated in some time and we will be removing those, that was decided by the community. We look at tightening our facilitator programme, what is known as brokers, looking and seeing whether or not we would make changes to improve that and actually accredit brokers rather than just list them, we decided against that, you can go see online all of the consultations. We have actually also done an ARIN bits newsletter, Thor those people who want to keep up but maybe you are not in the region, we have a quarterly E news update you can sign up for and that will give you the high level changes that are happening to up as an IRR.

Community outreach, we have doing a huge amount, those are meetings outside of our regular meetings, much shorter one day long so the ARIN on the road series is very successful, we have added ARIN in the Caribbean and doing a major push in the Caribbean with several 2000 events for the coming year and we did more than dozen this year, making sure we make ARIN accessible to those economies in the Caribbean that are in our service region. We are also doing outreach and training for government and law enforcement who want to know how the utilize ARIN's database for their work and we have revamped our blog and doing blog entries that are coming directly from the experts, for example on our registration services team, with both online entries and video and that is helped a lot of the community understand our services. So it's been an exciting year with engagement with our community, this is our update and that's it. I would like to turn it over to Oscar now.


OSCAR ROBLES‑GARAY: This is going to be a short update so you can have more time for the regional discussions and chair selection. I am going to talk about two specific updates, one is made by our community, which is PDP process for developing policies and the other one is capacity building that is executed by LACNIC staff.

So, first, we just recently from RIPE 76 in Marseilles to now, we updated the PDP process, basically we clarify most of the things that were already done by the co‑chairs, some clarifications, some modifications of the periods of time so the community may have more time to review and discuss what has been suggested. So, this is what has been changed, and as I said, most of them are clarifications, something that is not in this list of changes is that there were some confusions in the community that once they were asked for, once the co‑chairs were asking to the room to see if they support ‑‑ whether they were supporting a proposal or not, the community felt that they were voting for a specific policy. Of course, the PDP doesn't mention anything about voting. But we decide not to count the raise of hands, just to ‑‑ for the could chairs to see the support to a specific proposal, without counting the number of hands up in the air. So that's an additional change, that it is not part of this because that was not in the PDP.

So that's as I mentioned, it's been already executed by the people in the LACNIC staff that supports this process and it's been already implemented.

The second update, it's about our training efforts in the region, training is the second most evaluated and better qualified service after resources and we have been offering several on‑line courses and in person participants courses, our physical presence courses, and this year we provided ‑‑ we started doing at least one webinar monthly, so‑so far we have done already 12 webinars, most of them focused on the activities that we are responsible for, like IPv6, BGP and you will see a small set of training courses here called entrepreneurship, why are doing things in the region? It has to do with one specific project that we are doing in IT, IT is one of the poorest countries in our region and started this project with IDRC which is the development ‑‑ development bank ‑‑ I mean, Canadian development institution for research, and they selected us to manage these funds. It is a programme to develop some training efforts in Haiti, one is focused on one and the other on IT professionals. So after these we will have all these resources for the rest of the community, but it is like seat programme. If IBRC see that it is working, they will implement it somewhere else in the region. So we are doing this activities where the entrepreneur. Some of these courses are already in our online platform and we will be including some cybersecurity webinars next year and one Internet governance online course next year as well. So that will be our new offer on training courses for our memberships, our members, and also available for the rest of the community.

That is LACNIC update. Thanks.


ANTHONY GOLLAN: Thank you, Oscar. Do we have any questions?

AUDIENCE SPEAKER: Herve Clement. Thank you very much for this, always very interesting presentation, even Friday morning. So just a remark on the question, a remark I found the answer to the APNIC survey very interesting and the question regarding IPv4, what APNIC should do to foster IPv6, very interesting, that ‑‑ some members answer, so the recovering of IPv4 to the APNIC could be done so I think it's very interesting. Just a little question to Alun for the IPv4 calender, I would say do you have even roughly an idea on when this /11 will be reached on then after the IPv4 free pool will be depleted?

ALUN BARRETT: Yeah. It's hard to know when we will reach any particular milestone, but current indications suggest early next year that we will reach the second phase of our soft landing. So when we will have only one/11 available.


ALUN BARRETT: And after that it could take years.

HERVE CLEMENT: Okay, thank you very much.

JIM REID: Just a comment for John Curran from ARIN. As you know, one of your colleagues Cathy Hanley has just retired, and I would like to publicly express my appreciation and thanks to Cathy for the fantastic job she has done for ARIN and the broader RIR community. She has been great and I have personally benefited a lot from her advice and guidance and she is going to be a very hard act to follow.

JOHN CURRAN: Thank you.


Yes, as some of you may know Cathy Hanley has been with ARIN, oh, for now, 12, 14 years, and she is now retiring this year, and she has been a lead element in Internet governance discussions in many forms and it's great to see someone who has gone the distance in this community and made a contribution and has a chance to retire. Her work is retired, and Michele, who many of you are aware of, is now taking over Internet governance team and will be, for example, going to the meetings later on this year involving the UN I touch. And others. (I touch should be I T U)

ANTHONY GOLLAN: We will hand it over to Paul Wilson for the NRO NC update.

PAUL WILSON: I am wearing a different hat now just present a different update on the NRO as of today. This is a presentation which similar to one that many of you have seen before, it may be a ‑‑ shortening down to three slides but I have got a few more than that in this case I am sorry to say. The NRO is the umbrella for the regional Internet registries, we are still an unincorporated association but do very various structural components, the mission being to coordinate RIR joint activities, to act as the ICANN ASO as well. We have got so‑called executive committee, I am here on account of being the chair this year, Alun is vice‑chair and secretary, Axel treasurer, Oscar and John as regular members, those positions will rotate, as always, next year, so in order Alun as chair, and Axel and Oscar as pressurer and secretary and /TRAOURBer. (/TRAOURBer) we have got very various communications groups, we also coordinate on public affairs and policy, policy development support on finance and on legals. We have got a number of publications which are available on the website, a regularly updated statistics reports which has been updated as of this year, recently. There is also RPKI stats being added to that in due course as well. There is a matrix style of comparative policy overview which tills for any particular policy issue that you might be interested in, how it's dealt with respectively by each the five RIRs. There is also the address supporting organisation website there for your info.

We split the expenses of the NRO, which are incurred in coordination and so forth, and also largely in a contribution to ICANN which continues at a total of 823,000 dollars, we split that responsibility for those expenses proportionally on a formula that is based on our respective budgets and respective total IPv4 allocations, so the percentage responsibilities are there for each of the RIRs respectively, so in case of RIPE NCC, for the current year it's 29 .53% of total NRO expenses.

A little while ago, we announced a joint stability fund which is really a provision or a pledge that we have all made towards the stability of any RIR that might find themselves in need of financial support, a total of 2.1 million dollars has been, not exactly set aside but it's been provisioned for that purpose.

The IANA Review Committee is something that we also coordinate, it's come out of the IANA review process ‑‑ sorry, the IANA restructure process a little while ago. The purpose of that is to advise us on the services received by us, by the RIRs from the IANA, like the Address Council it's got three members from each region, elected by the community, two of each ‑‑ two of the three are elected by the community, one is a staff representative and their /THREUB rations are recorded online on the website as you see there (deliberations /SPH‑FPLT (the Review Committee members are here from your region, Filiz, Nurani and Felipe.

A feature of the new ICANN by‑laws is the empowered community structures so the ASO is a component of the empowered community and this year we have adopted a whole set of procedures for approval and rejection and for director removal and so forth. These are all powers that are offered to us as stakeholders in ICANN and as members of the empowered community so it's all a little bit complicated but these are powers which are offered to us and which we can choose to undertake or to ignore as we wish.

The ASO review which you may also recall, periodically the RIRs Commission a review of the ASO's effectiveness and the latest review was completed this year so there is reports online, time‑lines and announcements on the NRO website. There was a response that was published to that review, which ‑‑ in which we actually accepted all of the recommendations so the implementation of those is underway. The 18th of them is properly the least trivial in it calls for a regional reviews of the ‑‑ of the ASO's future structure and that is something underway on a time‑line that's not really determined yet but it will in theory produce some sort of a new structural reform or form of the ASO to be taken forward in future.

A few technical projects. The RIRs adopted an all resources Trust Anchor, that was something that was discussed extensively within the IETF and earlier this year. We have got an engineering project on establishing emergency backup operations and we have been working with ICANN on the technical health identifiers project, so identify technology health project and website revamp. That is it from the NRO. Thank you very much.


ANTHONY GOLLAN: Any questions.

SALAM YAMOUT: Speaking on my own behalf. So the question is does the NRO have a coordinated idea about engaging the ITU? Is there a coordinated philosophy for engaging in the ITU especially now the PP 18 is going to be held in two weeks and we hear that the ccTLD dependents and national Internet registries will be on the agenda?

PAUL WILSON: We do have coordination at the public policy level as one of the ‑‑ one of these days informal coordination groups. As for a common philosophy I think it's we to engage with the I touch. And we turn up at those events to add our voice by way of advice, technical input, expertise providing answers to questions that arise and we would be as useful as we can in making sure that the proposals and the discussions get the technical input that they need. I think we have ‑‑ we spend more time helping our supporters so to speak, with launching the arguments that they might need to put in their terms, in the discussions that are going on there, but I think as I say, I think overall the joint philosophy of engagement is to engage to be there and to be supportive and to assist with technical advice. Does that answer what you are looking for, Salam?

SALAM YAMOUT: More not high level. On the ground.

JOHN CURRAN: Yesterday, there was a meeting of the Cooperation Working Group that covered a little bit of this, Chris Buckridge did an excellent job. The RIRs do coordinate, our actual engagement in activities like the ITU in the individual study groups that are taking place, whether that is one or two or like study group 20, and we jointly review proposals, provide updates on information that might be inaccurate in them, we provide, if there is an opportunity to comment, we have actually filed comments on behalf of all the RIRs against specific proposals or specific language that is been suggested. Once you get into an actual event like Plenipotentiary it's a little different because it truly is Member State‑driven, we are in the hauls predominantly, some sessions we are but not all of them obviously and we are there to make sure that there is a sort of accuracy about what the registry system is and isn't but we can't ‑‑ often we cannot shape the agenda; all we can do is be a source of reference ‑‑ a reference source. In terms of do we have, going in, specific positions, as Chris Buckridge covered, there is a lot of documents that are going to be discussed and the documents are usually the starting points of the discussion. So, we haven't formed responses on behalf of the RIRs to every document because, realistically, it will be a week or two into the 21 days of meetings where we will actually know what is the solid matter but what is the key events they are working on. At that time we will work together. All of the RIRs have an overall approach with regard to the engagement with inter governmental organisations that we want to preserve the registry system and its model of self‑governance. So that is our starting point in nearly any response that we provide. Does that help a little more? Okay. Thank you.

ELVIS VELEA: I have got a couple of questions. First one is, any plans to go away with the distinction between the NRO and ASO ‑‑ merge the functions into one body and not have two bodies doing similar things but not exactly the same things? Or do you still plan to continue having two different organisations doing ‑‑

OSCAR ROBLES‑GARAY: Hi. At least in the Latin American region there were no ideas or desires to mix them in a single group because those are two different roles; operational of the registry and the policy definition. So nobody told us that those two activities should be together. There were, of course, comments regarding the confusion of the terms, and also there was ‑‑ there were opinions that we should simplify the terminology but also to have one authority, because currently there is no default authority and sometimes, some responsibilities fall into something undefined and we are very slow to react and to find who is the authority, either the operational part or the policy ‑‑ the global policy body.

JOHN CURRAN: In each of the regions there is a consultation going on and regarding the future structure of the ASO and how we engage with ICANN, in the ARIN region this consultation took place and the output said there is confusion over two names, the fact that when the numbers community works in ICANN, we use the term address supporting organisation, but, really, there is two bodies, a Number Council and an executive and the Number Council is what is referred to as the AS AC and it's very confusing, and so the ARIN region output was let's go to one term or the other. There is a little bit of a challenge in using the term "address supporting organisation" so we came here and said we are the ASO, not the NRO. To some extent supporting organisations are pretty rigorously defined in ICANN's bylaws, so we would be defining ourselves as something that is defined in ICANN's by‑laws and we can do it, but there is no convergence on this point yet; each region is busy discussing, aye know APNIC had a different take on it.

PAUL WILSON: Yeah, there was a poll around taken after the last consultation session which favoured ASO as the name, but I think it's a question of terminology more than one of separate organisations, because, as John said, it's a case of us using a in a different context and in a way, that is obviously very confusing. I mean, the different styles of the answers you are getting here right now might not help with the confusion either but that is exactly the nature of the beast we are dealing with.

ALUN BARRETT: Let me add to that as well. I think that the NRO and the ASO are two different names for the same organisation, it's not two bodies, it's two names. And it's definitely confusing, no argument about that. Thanks.

ELVIS VELEA: The stability fund, I know you guys planned when this was initiated that the RIRs, if they would be in trouble at some point in the future, would have a few years of funds in the stability fund that would help them continue operations. I don't remember if there was a target but I do vaguely remember that there were discussions about RIRs being to operate for at least five years, even if no other money is coming into the system. I see you guys have raised 2.1 million dollars or something like that, is there any way you could explain a bit what is the target and maybe also how did you get to this amount, which RIR contributed and with how much, approximately?

OSCAR ROBLES‑GARAY: Yes, there were no target and this is separate from the individual provisions that every registry may have for the ‑‑ for their operation, some registry may have 18 months of operation or some others may have 24 or some others may find different, whether it's an operational fund or long‑term funds in order to survive, I don't know how many years. In this case, there was no specific goal and there were only pledges just in case money could be a problem for a specific registry to help it, and besides the amount of money pledged, we work a little bit on the rules that will be applied for the recipient registry and so that is the main work.

I don't recall the figures, but you may see it in full in the NRO website as well, NRO .net.

ANTHONY GOLLAN: Before we take ‑‑

PAUL WILSON: Just for the record there, there is no RIR that has got a five‑year reserve fund or a target to have one. I think the targets are more like 12, 18 or maybe 24 months but it's well short of five years and I thought I should put that on the record. Thanks.

ANTHONY GOLLAN: Can I just add we are trying to save time for the RIPE Chair selection discussion, so if we could just keep the next two questions as short as possible.

JIM REID: Just an observation, Jim Reid. Elvis was talking about this distinction between the NRO and ASO, in the context of another ICANN review that I was involved in this was mentioned in passing and I think this may help to crystallise things: In the ICANN structures, as John mentioned, there is this thing called the ASO, so that is a socket and the NRO is the plug that fits that socket, for the RIR community and the RIRs do other stuff through the NRO and other fora but that plug into the ICANN machinery is the NRO being expressed as the ASO, maybe that might help to clarify that distinction for people approximate. Curran very apt analogy. Now we need to talk about impedance and voltage.

AUDIENCE SPEAKER: Speaking as one of the members of the Working Group, co‑chair of the Working Group which conducted the consultation process back in APNIC region. On the same point that Jim and he will advice suggested, we conducted a survey and sent it out to the community asking their opinion about this thing and almost 70% of them said there is a confusion and you have to have one name, when representing the community in front of the ICANN. Half of them said we don't care about the name. So, whether it is NRO, ASO or A, B, C, but the problems during the discussion came out, if you change the name today for the remaining ‑‑ for the next ten years it will be something else formerly known as ASO in ICANN in every document, in every discussion, so it's better to keep it ASO, otherwise it will be more confusion for the community and people coming in on board later on. That was the discussion. Thank you.

ANTHONY GOLLAN: Okay. Thank you. So now we have Nikolas Pediaditis from our registration services department to give the NRO statistics. And you guys are released.


NIKOLAS PEDIADITIS: Good morning, brave people. This is the NRO statistics report. It's a report that has been created collectively by all LIRs and in our region, at least we do this once every second RIPE meeting, and the idea is to show you an overview of what is happening in our pools, what are the current trends and hopefully maybe you can draw some useful conclusions from that.

So we will start with IPv4. In this pie chart, you can see the entire IPv4 address space, and the way it has been distributed in terms of /8s. This is not something that is changing, because as you most probably already know, since 2011 IANA has handed out all /8s to RIRs, 2012 and 11 was the year that the last /8s were handed out. And as you can see from the numbers, ARIN and the RIPE NCC had a pretty equal distribution with 35 and 36 /8s, APNIC with 45 /8s. AFRINIC joined quite late, so got 4 /8s. And the biggest number central reg with 92 /8s.

When we speak about the central registry we speak about the time before the existence of the RIRs, and that's also known as legacy space.

And if we take a look on how much do we have available in our pools, I am pretty sure you all know this, we are really scratching the bottom of the barrel right now, ARIN is completely out without any IPv4 addresses left; however, ARIN is still handing out some IPv4 blocks, the recycled ones, the ones that being returned and reissued. The RIPE NCC, we have almost half of a /8. What we called a/ ‑‑ last /8 in 2012, so that means 185 /8 is gone now, a few months ago we handed all of that out, and now we are also distributing IP space that is this in our pools either from bits we got back from IANA or mainly from IPv4 space that has been returned to us, that has been deregistered.

AFRINIC, same place, and APNIC with 0.3 of a /8 left.

And if you take a look at, in terms of /8 and how much it has been issued per year and if you take a look in the last ten years, I think it's pretty clear of what is going on. If you look at 2018 there is little bumps left which means basically we are almost out. Very interesting to notice are the two years in time when you see sudden drops in those graphs, say, for example, you take a look at APNIC in 2011 and the RIPE NCC in 2012, you realise that it's quite a steep drop and this is due to the fact that this is the moment when these two RIRs reached the exhaustion point. So starting distributing from the last /8, only giving out /22s and that is what caused this sudden drop.

When IPv4 space is being exhausted then transfers go up, it's kind of a natural course of events. And if we take a look at the number of transfers per year, so this is the number of transfers and every transfer can be multiple IP blocks, of course there is a rising trend but it is stabilising a bit. We don't see any crazy changes there. It's still increasing a little bit but in terms of number it's pretty stable, to pretty much all RIRs that have policies for transfers.

If we take a look in the number of IPv4 addresses being transferred, then it's a kind of a slightly different picture. We don't have are a stable trend, some years we see spikes in some RIRs, some years we see drops. I assume this has to do with pricing, because what we observed this year is a lower number of IPv4 addresses being transferred, so maybe it's the pricing, maybe it's IPv6 picking up, lets hope for the latter.

We also have Inter‑RIR transfers, these are for the newcomers in the RIPE meeting, these are transfers that take place between organisations residing in different RIRs. We have three RIRs participating in these type of transfers because they have compatible Inter‑RIR transfer policies. The vast majority of the transfers are taking place, in terms of number of transfers, are taking place between ARIN and the RIPE NCC and ARIN and APNIC, and also we have a growing number of transfers between the RIPE NCC and APNIC. And if we take a look on the number of addresses being transferred, the majority is between ARIN and APNIC, with 16 .5 million IPv4 addresses towards that direction, and 5.6 million between ‑‑ from ARIN to the RIPE NCC, and we have another 1.2 million IPv4 addresses transferred from the RIPE NCC to APNIC, the opposite directions you see a huge difference in numbers. So it's a very clear trend of the flow of IPv4 addresses between the RIRs.

So, moving from IPv4 to IPv6, most probably you have seen this before, this is all IPv6 varies, we have a/0 and one‑eighth of that has been handed out to IANA, to hand it out further to the RIRs. And as of 2006, as of 2006 every RIR got a /12. Prior to 2006 there were some smaller IP blocks that were handed out to RIRs so if you deployed IPv6 very early in time then most probably you have a prefix that is different to a 00 from the RIPE NCC.

If we take a look on the number of IPv6 allocations issued by RIRs and this chart shows the number of prefixes, we see that mostly the the RIPE NCC and LACNIC that are handing out the most prefixes, but bear in mind this is number of prefixes. This could be allocations and assignments, but to make the picture more clear, we need to separate those. So if we look at the number of allocations in terms of /32s, we see that the RIPE NCC has by far to the second issued out the most IPv6 /32 allocations. There are multiple reasons for this; the main reason is the difference in policies between the RIRs. For example, in our region, the IPv6 policy says that we can give up to /29 to every LIR without asking any questions. In APNIC, for example, that number is a /32. What is interesting to observe is the difference, for example, in LACNIC; if we take a look at the IPv6 allocations and compare that with the IPv6 assignments in the same region, and the same goes for our region, the RIPE NCC. So ‑‑ and this is the slide of IPv6 assignments, those are what we call in our region PI assignments. In terms of /48s, and here you see the exact opposite, the RIPE NCC is well below everybody else and LACNIC is at the top and well above everybody else. This has to do with the different way of distributing IPv6 and the difference in the policies in every RIR.

If we take a look at the percentage of members with IPv6 in each RIR, we have LACNIC, that seems like something is working there. I don't know if this is one to one relation with IPv6 deployment. We also have an interesting number from ARIN, because ARIN has quite a big number of members with IPv6 only. Now, this could be that something amazing is happening, we know that the ARIN region has a got rate in IPv6 deployment, but also it could be due to the fact that ARIN's IPv4 pools are exhausted so people need to get IPv4 maybe from their up‑streams and only get IPv6 from ARIN.

And if we take a look at the AS numbers, this is a holistic picture of all the AS numbers we have. Of course, the biggest portion is the 32‑bit AS numbers. And if we take a look in time and how these AS numbers have been distributed during time, we see all RIRs, if we got ten years back, started issuing primarily of course 16‑bit AS numbers and those ‑‑ as those are getting low then the 32‑bit AS numbers pick up. Nowadays, by default, for example, the RIPE NCC by default distributes 32‑bit AS numbers and only if there is a specific need if an organisation specifically asks I want a 16‑bit AS number we give it to them but it makes absolute sense that the 32‑bit nowadays are the biggest part of what we distribute.

And the total assigned by each RIR by year, then it's basically justifies the same thing. Historically, 16‑bit are more than 32‑bit AS numbers they wants to have distributed but this trend is changing so eventually this will change.

That's it. The NRO statistics are available on the NRO website. And you can also visit the IANA page to see the numbers, especially when it comes to the entire IPv4, IPv6 and how it has been handed out to the RIRs. Any questions?

ANTHONY GOLLAN: No. Elvis, we don't have time for questions.


Thanks. So now we have got the IANA PTI update from Kim Davies.

KIM DAVIES: Good morning everyone. I am here to give you the numbers update on behalf of the IANA team, I am Kim Davies. For those that aren't familiar with IANA, this is one of three things that we specialize in, the others being managing the root zone and other key Internet domains and managing around 3,000 protocol parameter registries on behalf of the IETF.

The team delivers the IANA services to you is on the screen, there is 14 of us at the moment. And obviously we all have a hand in some respect to delivering services to you.

Currently we are conducting our 2018 annual customer survey. The way we do this is, we segment all of our different customers in areas of interests into different groups. With respect to performance of the numbers function, we send the survey to the CEOs of the RIRs, as well as the registration services managers of the RIRs. It's currently open for responses at the moment and we expect to have the results available for you at the end of the year. That being said, for last year we have the compiled results and we had a 100 percent satisfaction rate from the numbers community and our average over the past four years was 99%. We ask the respondents to rate areas of focus for us, whether it should be focusing on different aspects of our operations and time and time again we keep getting the response that accuracy and timeliness are the key factors that we should be monitoring when it comes to performing the numbering function.

One thing we are changing at the conclusion of the survey period this year is, we are going to start sending out surveys immediately after we have done customer interaction, right now we are doing solely an annual survey process. The problem with this of course is that a lot of our interactions with our customers may have happened six, twelve months prior, and the customer is scratching their head not actually recalling how well that interaction went. So, we are sort of by if Iicating our approach, we will survey them on how that interaction went and that will represent the assessment of the quality of the services that we are delivering in a direct way and we will still continue to do the annual survey but it will be more strategic, we will ask strategic questions about the direction we are heading in.

We publish performance of the numbering functions on our website. This is something that happened post transition in 2016. There is quite a detailed SLA that we adhere to and you can drill down into all the details of all the allocations we have made in the last two years to see how we adhere to the SLA.

Really big part of the culture of our team is conducting audits and we have third party auditors come in and review our processes, our adherence to the procedures and policies that we implement as well as our security procedures as well. We, for the first time the previous audit period including the numbering functions in the audit. That report was issued without exception, which means we met all of our controls. We also separately have an audit of how we have managed the root zone KSK and that was issued without exception in 2017. One change that we are making this year is, we have switched auditors, we have been audited by 2010 by PricewaterhouseCoopers and we are now switching to RSM for our 2018 audit programme.

Recovered pool allocation, we do have a very small number of IPv4 addresses, those that have been recovered by some means or another, been returned to IANA in some fashion. And there is a policy, a global policy for post‑exhaustion IPv4 allocation that we have been implementing now for several years. We are really scraping the bottom of the barrel at the moment, we have made an allocation in September to give four /24s to each RIR, based on the current number of addresses we have in the pool, we will make one final allocation of two /24s to the RIRs. And that will be it. So, unless someone wants to volunteer anything back to the RIR, I think that will be the end of that mechanism.

This was covered in a previous presentation, so I won't belabour it but we have reviews by the NRO/ASO community, and this is one of the accountability mechanisms that we have to the community. So on an annual basis they review our performance and produce a report on that.

We are currently ‑‑ we recently reached a milestone for a multi‑year project that we have been conduct the KSK roll‑over project. This has been sensitive because the first time we have ever replaced the Trust Anchor for the DNS and how software would cope with such an eventuality was quite an unknown. It hadn't really been tested, so there was a lot of preplanning, outreach and so forth leading up to this but on 11 October we switched the key for the root zone and the preliminary response is there was very little impact which is very positive for us, we continued to execute the project, it's not quite done yet, we still need to revoke the old key and then destroy the old key and we are continuing to receive feedback on how that transition went. But once that's all said and done, we expect to take lessons learned from the process and then study a schedule for future roll‑overs.

Lastly, I just want to mention that we are currently going through our annual budgeting process and this is for fiscal year 2020. Due to the amount of consultation that is involved, the lead time for the IANA budget is quite long, so this is money that will not be used until the 2019 to 2020 period. The PTI and IANA budgets are devised accept federal the ICANN budget and but once they are decided they are then rolled up into the broader ICANN budget. So we did preliminary discussions with various community groups earlier this year, that has provided the initial basis for our draft and that draft is now out there for you to review. Please take a look and comments are due by 12th November. That is all I had.

ANTHONY GOLLAN: Thank you, Kim.


I think we have got room for one question if anyone has one? No. Thank you.

MIRJAM KUEHNE: Thanks, Anthony. That was brilliant to squeeze all this in in the first hour of the session. Wear few minutes late but without further aado, I would like to introduce the next session. I am Mirjam Kuhn that, thanks for putting this in, that's true. I am chairing this next year as we agreed on the mailing list we want to split the RIPE chair role description discussion from the RIPE chair function ‑‑ function process discussion and so Hans Petter is going to go quickly over the first part and hopefully we can find consensus here.

HANS PETTER HOLEN: So, I am the RIPE chair, for those who haven't noticed yet. There was an updated version of the document circulated on the chair discuss list yesterday, I believe, and Daniel has been very kind in putting this on paper. The intent with this document is to describe the role of the RIPE chair as it has been and developed up until today. So, the idea is that we document what it is right now and then if the community wants to change that in the future at least we know what we are changing from into something else. So, I saw only positive comments with the last version on the list yesterday, so I will make this really short, unless there are anybody in this room that are unhappy with the current RIPE chair function description, I will call for consensus in the plenary today and on the mailing list after this meeting and then hopefully we will have a document that describes the function of the RIPE chair. Assuming that we have that, I will hand over the microphone to Anna, who will ask some questions to you regarding the process to select the RIPE chair which I think is the most important one that we have agreed on the function description. Filiz.

FILIZ YILMAZ: I agree with the process defined and it serves the process. It could have been better but it will take us to the next step but I personally don't have any objections but I have remarks to make here, Hans Petter. This is almost ten past ten on the Friday, this process has been on the list like, reaches maturity for the last four or five years. A lot of people have already left and this is an important discussion. This should have been in the agenda, it should have been more announced better and what would have happened, because we knew this was coming, all right, from Marseille, we discussed that was the initial input coming out. I just want to draw attention there, let's not do this. We pride ourselves for openness and transparency and that comes and stems from good engagement and for people to engage, first need to understand, understand the discussion well. Remark for this, and you know where this is coming from.

HANS PETTER HOLEN: Thank you, Filiz. That is much appreciated. And yes, I have to take it on board myself. I thought we had consensus after the last meeting and that document that was published ahead of this meeting was something that we had consensus on. Therefore, we did not set aside for instance, evening session to discuss this. It has been on the agenda that it will be discussed in this time slot but I fully appreciate that yes we should have set even more time and advertised it even more. So I think the important thing here now is the question that Anna has identified from ‑‑ oh, there is another gentleman that wants to discuss this as well.

JIM REID: Filiz's points are well made but I think they are a little bit unfair. If we were actually on the verge of having a selection process document that was reaching consensus or close to consensus, then by all means we must have a very prominent profile for that and which scheduled well in advance in the meeting planning plan and in actual fact because of the discussion on the list it was very clear we do not have consensus yet on what that selection process is, so I think it's not unreasonable that we have a discussion here at the meeting right now, with what we have in front of us, given that we still have some more work ahead of us and what Filiz is saying is when we are closer to having a bigger version of that document we must have that advertised in the meeting plan well in advance, and I am sure that will take place.

HANS PETTER HOLEN: We will make sure that that happens at the next meeting, definitely. Thank you. So any more questions before we get to the meat of the discussion, the key elements in the selection process? Anna, the floor is yours.

ANNA WILSON: Thanks very much, Hans Petter. Rob was our chair for 25 years, and when he stepped down his only instruction to his successor was that he established a process for selecting the chair. So, we have 19 minutes between now and the coffee break and the coffee break right now is, I can't extend into it very far because there is stuff that has to happen in the coffee break and I have to clear the stage in time for that so they can set up. So let me very quickly summarise and we will work out what we are going to do from here on. Where we were last time we spoke about this in Marseilles, well, we ‑‑ it was clear that any change to how the RIPE community operates needs a consensus of the community. So, that's us. We had a very productive discussion in the last RIPE meeting about this, it seemed to work pretty well for getting views aired and for reaching some sense of what the community wanted. At the time we had the idea that the RIPE NCC would take that discussion, draft a proposal and circulate it with the aim to get consensus at this meeting. NCC did so. There was some discussion. They took that and made a revised proposal at the beginning of October, which, in summary, looks something like this.

The big deal is that there is a Nomcom, the Working Group chairs in that proposal are responsible for serving it and also one other thing that is not on the slide is that there is a vice‑chair positioned which is selected the name way. No assumption that the vice‑chair will become chair but they are allowed to be nominated.

Now, three things came up with that that I want to mention.

One is I personally, I am not driving this, the ‑‑ I don't own the proposal, RIPE NCC have been doing the hard work here and I think they are doing a wonderful job on this. I have just been brought in because someone from the community was to be asked to help moderate the discussion rather than the chair moderating this particular same discussion.

The second thing is, it's clear some people were surprised by, at the last minute, by the wish to get consensus at this meeting. I used the phrase 'last call' and that wasn't really a formal thing that we had used. But what we did get was some really excellent discussions in our community, on the mailing list. The mailing list by the way, the formal link are here. It's been very constructive, particularly over the last couple of weeks, some really great stuff in there and I think wave clear idea now of which parts in the proposal need to change and some good ideas on how to change them.

But this does mean the third thing: I think it's clear that the proposal as it stands as of the beginning of October feels like the right shape but doesn't have consensus. So, with that, I propose now for the remaining session today is we take the time to discuss the outstanding issue, I think there are three, maybe now down to two, and then we go through the same iteration again to see with the feedback we get today, produce another draft, I will probably ask for a shows of hands hands to get a sense. We have further discussion on the mailing list and further drafts if needed, so please do check the mailing list and get your comments in early so we revise the draft as quickly as possible and come back and do attempt to get consensus at the next meeting in Reykjavik. Does anyone object to that at this time? Okay. If not, the outstanding issues that I see reading the list, talking to people at the meeting, I think there are three that would stop consensus, one was the request that we complete the RIPE chair function description before proceeding with this process; one was the question of who selects the final candidate in the proposal that was a Working Group chairs; and the third is, how do we choose the make‑up of the nominations committee, which in the proposal was also left to the Working Group chairs.

There was a suggestion, first of all ‑‑ oh, what I propose to do, we have 15 minutes left. So let's go lieu this one and see what time we have. There was a suggestion that it's difficult to discuss the chair selection when the job description or the function description isn't final. So I will put this one first because, well, maybe the answer to this someone we stop this discussion right now. Hearing what Hans Petter has just said and there seems to be consensus in the community for the description that is on the list, it sounds like we are going to get proper consensus on that quite soon and we pretty much have something we are happy with there. Are there any objections if we continue to work on this chair procedure? No objections to that. Okay. Cool.

So we don't have consensus to stop this process, so I will continue.

That leaves two substantive questions:

What I propose to do for these ‑‑ oh, I am sorry, Daniel does want to respond, please.

DANIEL KARRENBERG: I think the point on the mailing list was not to stop the discussion but to delay the call for consensus, so I don't think you put this right.

ANNA WILSON: Very well. It sounds like we are aligned then, we will not be calling for consensus on the chair selection process today.

Are you happy with that?


ANNA WILSON: Thank you. Okay. Then what I propose is we take these two other items, and we spend five minutes on discussion on each of them. I will have to make a hard stop because we do have to stop by 10:30 and I make ask for some shows of hands based on the feedback and ideas that come out in the discussion.

The first one is, who selects the final candidate? Both these remaining questions are wrapped up in Working Group chairs, but to my eyes there are two separate questions here. There was a strong feeling on the list that the process, the draft process means a lot of the Working Group chairs, they do excellent work but they are selected for a particular role and it's very attractive when you have a group like that to load other duties on to them and that is not necessarily the strength. On the other hand, it is a group that clearly has the backing of the community and the RIPE chair is the chair of the Working Group chairs, so there is an argument that they should have some say in the selection.

What I am seeing on the list so far it's generally agreed there should be a Nomcom and that should select the final candidate. So we will come to the membership of that shortly.

Can I ask what the feeling in the room, what are the comments on the role of the Working Group chairs in the process of selecting the final candidate, whether it should be left to the nominations committee or something else? So back microphone first, I think.

JIM REID: Sorry to be monopolising the mouthpiece ‑‑

ANNA WILSON: Sorry to interrupt, Jim. I will say, because time is short, what I will be asking for right now is we want to get a sense of the feeling in the room, I would suggest that detailed proposals go on to the list but we want to get some ideas so we can get a hunch right now, what we are looking for right now is broad rather than deep.

JIM REID: Short high level ideas, perhaps have a member or two of the Working Group chairs serve on the Nomcom, I would perhaps agree with the comments being made recently on the list that having the Nomcom having more of a role in that, in the process is maybe a bad idea, fewer moving parts, less suspicions of things being done in a smoke filled room, more openness, and I think we should develop on the suggestion that Bijal put forward before. When it comes to the Nomcom having final decision, I would hope they come up with one name so it's easier for the community to endorse or reject that decision. If they come up with a list of candidates, it's going to be difficult to find a way of having the community endorse or select one candidate from that list, and after all, the Nomcom's job should be to nominate the candidate. Thank you.

ANNA WILSON: Thank you. Because we have five minutes for this and about eight people at the mics, that is 30 seconds each.

FILIZ YILMAZ: Working Group chairs are foremost RIPE community members, first of all. That's why they are Working Group chairs. So, if you can come up with a Nomcom formation process which will be open and inclusive to all community members and I ‑‑ nobody makes a distinction from the Working Group chairs collective and the RIPE community, I don't see why some Working Group chairs will be part ‑‑ will not be part of that Nomcom. Thank you.

DANIEL KARRENBERG: I think the original proposal that was on the table is a good one. The ‑‑ a Nomcom is a Nomcom so it's best used to select a slate of candidates or fill ‑‑ yes, to select a slate of candidates but it's not good to select, make a final selection. I think the Working Group chairs are actually a good body to make the final selection from a slate of candidates that has been published before and has it it had discussion and feedback. I think there is a danger that if the Working Group chairs are not involved here, that there could be friction between the candidate finally selected and the Working Group chairs, which would not be a good thing for RIPE.

ANNA WILSON: Thank you.

AUDIENCE SPEAKER: Elvis: So I don't agree with the idea of a Nomcom. I think that could be seen as a gentleman's group, could be seen as a closed meeting, let's see what our best friend is kind of a thing. I really don't agree with the idea of a Nomcom. I think that anyone should be able to nominate themselves or nominate someone for the chair position and maybe the selection could be done somehow into the RIPE meeting ‑‑ in the RIPE community selecting maybe only three or five in the end and leave it up to the Working Group chairs to decide who their chair should be. But the idea of a Nomcom making decisions behind closed doors is not really open and it's not really a good idea.

ANNA WILSON: Let me ask you one question, I am not making a proposal here but just trying to feel the shape of the objection. If the Nomcom was randomly selected from the community, would that change your feeling about a Nomcom?


AUDIENCE SPEAKER: Currently Working Group chairs are selected not to represent community on such high level, speaking of me I can trust some of Working Group, current Working Group chairs on topics of the Working Group but can't trust them on making high level and more community‑related decisions so I think that is too much of Working Group chairs in these proposals.

AUDIENCE SPEAKER: Hi. Bijal. I made a proposal on the mailing list and that doesn't ‑‑

ANNA WILSON: Thank you very much for that, by the way.

Bijal: That doesn't include the Working Group chairs. I am not sure I understand Daniel's point on why there would be friction if the Working Group chairs were not involved with the selection. Secondly, I still think that the Working Group chairs if they want to, there is nothing stopping them putting themselves forward, but I do think that the ‑‑ you know, having the Nomcom selected randomly is the way, like I said, we should go.

RANDY BUSH: As I look around the different organisations in our culture and how they select things; beauty contests suck. Working Group chairs are Working Group chairs, they are not selected as representatives of the community and they are having enough trouble producing work and Bijal and a number of others have said essentially steal the IETF method, which is a randomly selected Nomcom from qualified individuals who are qualified in this culture, who attended recent meetings.
And to Daniel's objection that the result might not be compatible with the Working Group chairs, the IETF Nomcom specifically goes to the different components of the community and says, can you work with Mary Anne? You know. And that's ‑‑ the Nomcom is not ‑‑ their purpose is to see what works and what fits.

ANNA WILSON: Thank you. And last comment.

SALAM YAMOUT: How about if the exiting chair would make the final selection? I don't want to make sound like a monarchy, but because we have trusted the chair for so many years. Unless the exiting Chair he has been impeached I think he is the one who could find the consensus needed and could find the best candidate from the Nomcom, that he didn't interfere with the process so far.

ANNA WILSON: Thank you very much. I have some notes written down that I will ask the room to give a show of hands‑on shortly, so that a new draft can be made.

But let's, before we do that, let's keep talking about the Nomcom because we are really getting into that discussion now. The original proposal made the Working Group chairs responsible for more ‑‑ form willing the Nomcom and didn't specify any method, they could do anything they want. It looks like there wasn't consensus for that on the list. And there was Randy and Bijal have mentioned the suggestion of using some sort of random selection from the community with some qualification there for membership, for example having attended meetings from the last ‑‑ or being active on mailing lists for some number of years.

Can I have comments then on, if there is to be a Nomcom, bearing in mind we have had comments that maybe there shouldn't be, if there is to be one, how should that be composed? Does anyone want to comment on that? One person does, two, three, excellent.

AUDIENCE SPEAKER: I think it should be formed randomly with qualifications. I think we need to ‑‑ there needs to be some qualification, it can't be a random person on the street who is putting their name in the hat. I think there needs to be either ‑‑ well, either the attendance to the forum or some sort of activity on the ‑‑ on the RIPE mailing list.

ANNA WILSON: Thank you.

DANIEL KARRENBERG: Just for clarity. It's a random selection with qualifications from volunteers, right? It's not draft people.

ANNA WILSON: It's not jury duty.

DANIEL KARRENBERG: The other thing is, the key thing here is what is the qualification? And if people ‑‑ I would encourage those who propose a random selection with qualification to specify the qualifications very clearly.

ANNA WILSON: Thank you.

AUDIENCE SPEAKER: Elvis: What Daniel said, I want to say both things as well. So if there will be a Nomcom, I think it should be from randomly selected people; a list of volunteers. I don't think those people have to have specific qualifications. It would be good to have, but then you would randomly ‑‑ well, not randomly, but you would make that list thinner if you put all kinds of barriers into getting on the list. But yeah, if there are to be ‑‑ if there is to be a Nomcom, then it's a very good idea to have it selected from randomly volunteered people ‑‑ randomly selected from volunteer people.

ANNA WILSON: We have three minutes left, while I am loath at this minute to ask people to come to the microphone, I would love to hear what those qualifications should be.

FILIZ YILMAZ: This is the core of the process, I think, Anna. I want to have a Nomcom personally and I am happy to repeat it again and again and again as whenever it is asked. It is the right way of going forward, so without any, you know, giving super powers to a pre‑defined collective or a group, without any prejudice, it's just good governance. For Nomcom specifics, a random, I think people when they hear that word, it's connotates with pulling a name out of ‑‑ it's like a lottery. It's not. I think Bijal's mail can be digged into as we are going through the consensus process as you described in the coming weeks, until Reykjavik and we can come up ‑‑ I have trust this community can come up with some criteria that will not make it, you know, look like oh we just got blindly appoint a person but come with a list that will consist of people who will show some reflection of trust from the rest of the community.

JIM REID: I think we should try to follow the IETF model to some extent along those lines and I think the criteria for the Nomcom should predominantly be N of N meetings, so you have to have attended N number of meetings for the last N meetings for some definitions of N and N, I don't care what those numbers are, that some implementation detail. And the reason for that is we have got people at the Nomcom who really understand how the RIPE meetings work because we are not talking about somebody that is come in from the outside, we immediate to understanding of how the RIPE meetings work and how they are put together and what the RIPE chair does and how it interacts with the rest of the RIPE meetings and Working Groups and NCC staff and having familiar with that familiarity I think is very important criteria for the Nomcom membership.

To what Daniel was talking about, about consultations with the Working Group chairs or whatever, I don't think that is too much of a concern but the way we should probably handle that having some guidance for the Nomcom, when it gets down to do its work it's expected to consult with various, I hate to use the word stakeholders, which include, for example, senior members of the NCC members of the ‑‑ long‑standing members of the community, Working Group selectives and other people from other organisations outside like, say, for example, the other RIRs. Thank you.

ANNA WILSON: Thank you very much.

RANDY BUSH: Pretty much what Jim said, the purpose is to see that they are members of the community, and how do we define the community? Sort of mailing list but I don't think we really want to go there because the most frequent posters tend to be idiots like I. And so meeting attendance seem to be the only other criterion we have that is reasonable and measurable. So beyond that, Jim said it.

ANNA WILSON: Can I ask before you step down, Randy, what about the idea that because there is some nervousness about certain constituencies, what about the idea that we ‑‑ that there is a minimum expectation or the algorithm would have a minimum number of people from certain constituencies which might, for example, be current and former Working Group chairs, might be people who have never been Working Group chairs or might be people who have been in the community for a certain amount of time or not for a certain amount of time?

RANDY BUSH: At first blush, I have sympathy with that but I have a fear of deciding we have constituencies, because I think that is the poison that makes IANA so insane, creates opposition and cultures of importance.

ANNA WILSON: Thank you.

RANDY BUSH: I think we are a loosely coordinated group and I think that is at the ‑‑ close to the core of this culture that we want to preserve, and not formalise divisions and representation and so on and so forth, which is part of the reason that I like the qualified selection process is, it's random. That is a feature, not a bug.

ANNA WILSON: Thank you very much. We have two more people in the queue and I will close the queues at that point.

BRIAN NISBET: Just whatever we do, the Nomcom, whatever the criteria are, can we please ensure that whether it's a ‑‑ there is a couple of specific spaces on it or something to guarantee it's not the same people, it's not the people who are at the mics at every meeting, it's not a bunch of Working Group chairs. That we get some new voices in, honestly whether they have been to meetings, whether there is two spots for someone who has never been or been to one meeting or something, but we make sure that there is people there who can give, can give new voices and make it clear that we want that loose band of volunteers from whom the Nomcom will be chosen, we make sure new blood comes into that particular thing. And can we just make sure also, please, that the outgoing chair does not have the final say on who the new chair will be, much and all as I love them all, this is not a monarchy, we do not ‑‑ we don't want to keep handing over the sceptor, we have done it once, I think that was once too many, much and all as I love our current chair, let's not do it again.

AUDIENCE SPEAKER: I am on the chat monitor and have a comment.
For the record, I support the concept of Nomcom, I support Bijal's proposal and it makes sense to separate the role of the Working Group chairs from the RIPE chair selection process.

ANNA WILSON: Super. Thank you very much indeed. Thank you, that is a lot of good stuff. Let me ask for some show of hands so we can get a feel for what might work in the next draft. For who selects the final candidates ‑‑ the final candidate, singular, can I have a show of hands for the final candidate should be ‑‑ the final candidate will always in the proposal be ratified by the community at a plenary, but for the selection of that final candidate can I have hands up for that should be done by the nominations committee? By the Working Group chairs?
And by the outgoing chair?

Okay. Nomination committee has some support and there is less support for the two other.

The nominations committee should come up with one name or a slate?
Nominations should come up with the final candidate?
Probably similar to what we saw before.

And the nominations committee should come up with a slate of candidates? That is quite strong, but that means no one has told us who should choose a final candidate, that is also left to the Nomcom.

Okay. And we have talked about the existing chair already.

I don't know how to solve that one unless anyone has any magic ideas how to get consensus there, otherwise we will have to leave it hanging.

MIRJAM KUEHNE: I have the feeling that the last question was not clear because it kind of, got the opposite show of hands than before. So I think the distinction is what is the role of the Nomcom, are they actually selecting the final candidate, which is then endorsed by the community, or are they coming up with a number of candidates like a slate which will then be selected by an undefined process if we haven't actually discussed yet, or the Working Groups chairs collective that is currently the one? So try that again.

ANNA WILSON: If the chair will allow me one more minute then.

AUDIENCE SPEAKER: Maybe try both questions because having the two questions in mind, maybe someone wants to say the reply in the first question as well.

ANNA WILSON: What I propose then, if the Chair will allow me another minute, is to ask those two questions again in the reverse order, and let's try that again.

It turns out that scribbling things as people are talking isn't a great idea idea for sequencing.

Okay. So we had the idea that the Nomcom should come up with a slate of candidates which should be chosen by that shortlist by someone else or the nominations committee should provide the final candidate which will ‑‑ which will be ratified only by the community. The community would ratify in either case. Can I see a show of hands for the nominations committee should provide a slate of candidates and that ends their work?

DANIEL KARRENBERG: No, because it can do both.

ANNA WILSON: Sorry, the ‑‑ okay. My understanding then is nominations committee will always make shortlist and then may provide a final candidate? Is that the picture?

So, the question I will now ask is, the nominations committee makes a shortlist and then stops, yes or no? Hands up for yes? Okay.

DANIEL KARRENBERG: It's totally unclear, I think this is not the method to do this.

ANNA WILSON: Okay, what method shall we have?

HANS PETTER HOLEN: From listening with the notes that we have already, I think that we have enough to make a new draft. I do not know whether we will have consensus on that new draft, I think that's left for the mailing list. Let's make it as simple as that.


Thank you. Just to be clear, I will not accept the responsibility of appointing my successor. So nice try, folks, but I won't accept that responsibility.


ANNA WILSON: Do we have new information on the other items then, can I ask over there, you are happy with that? If you have the information you need my job is done. Thank you all very much, very grateful.


MIRJAM KUEHNE: Thank you very much, and I know that was an impossible job in such a short time. Sorry to put you on the spot there. I think there is some RIPE NCC board announcement going on during this coffee break so I hope we left enough time for that. With that, this session is over.

HANS PETTER HOLEN: I want top thank Mirjam and Daniel for facilitating this work and producing the documents and pushing me into progress here, and in order to make even more progress, Mirjam has agreed to keep track on me every second week moving forward so we will be more timely in the future, we hope. Thank you.