RIPE main hall.
BoF session
Monday, 20 October 2025
6 p.m.
.
HERVE CLEMENT: Good evening everyone, it's six o'clock now. Okay, one more time. So good evening everybody. I think we can start. So you are here for a BoF with your favourite topic since the last two years, which is the ICP‑2. Even before that period, I was absolutely not aware what an ICP was and I did not know it was an Internet coordinated policy from the ICANN, and now I know. We are here because we are during the ongoing regional and ICANN consultation of ICP‑2, so we are here to listen from you regarding the current version of the document. Just one thing I wanted to emphasise that we are together, it's not because we are part of the NRO NC that we are in one group and you are in the other group. We are all part of the community of the Internet community, and I think that we have the same goal to have an acceptable, more than acceptable document for everybody. We have a specific role as we are part of the NRO NC and we were given the task to update the initial ICP‑2 document that was provided and accepted in 2001 and the other difference that we are connected with the other regions, and so we have to converge in something who takes into account all the comments from the different regions as well.
The idea, but I think some of you know what it is about. So we will have a short presentation but not more than 20 minutes and after the idea is really to discuss, not us to ask a specific question, not to complain about something, but really to discuss together. Thank you.
Updating ICP‑2, that's the title of the BoF, so we are discussing that.
On the second graph of this document, so there have been three consultations, the first it was regarding the 24 principles, if I can remember, about what could be in this document, and two were consultation regarding the regional Internet registry Internet governance. It's not called ICP‑2 anymore because I don't think that the final document which is part of the series of the ICP.
Just to say that in 2023, it's two years ago, only two years ago, or two years is a lot, I don't know, but we have to converge with all the communities, so we need a little time to have something under consensus, and the NRO NC, as I said, is a set of directors of the regional internet registry asked us to update the ICP‑2 to strengthen the regional governance regarding the Internet resource ‑‑ Internet number resources.
Just a piece of timeline on the work we have performed.
We need one year, because there was a lot to read, a lot to discuss, to present the first document, which was the principles document and there was a dedicated consultation between October and December 2024. After that, we reviewed the feedback on this principle document, and in February 2025, so at the beginning of this year, we published the finally the result of the analysis we have performed. And in April 2025, we published a draft of the regional internet registry governance documents, and there was another consultation. And after this consultation, there was another part of analysis of the outputs, and we were able to in 2025 to publish the current document you can read currently and which is under consultation until 7th November.
To see all the documents I talk about, the principle consultation report, you can reach here. The left QR code. The first draft of the current document is the right QR code. And you can read the current document using this QR code. We use QR codes to do this. And to help you to read the current document, we have prepared three supporting documents. The first one is a red line note change, so you can see the differences between this document and the last document. It's something like that with the new wording in blue, the deleted wording in red, so it's something very easy. You have also a rationale document explaining why we did some improvements, and why we didn't take into account some of the comments because it was out of scope of the document. And there is a frequently asked question document as well.
And the rationale is a document something like that, and it explains what we did as a notion of recognition in the document for instance. The idea of course is for you to read the document and to share your input, simply your input can be "I am very happy with the document you share, I support" so, okay, we are happy to read that because so we can see that finally we reach a certain goal acceptable. But you can say also "I don't agree with that" or "I think it could include that" or so and so, and your feedback is vital for that because the document will be legitimate if it agrees and reaches consensus within the community one more time.
So, you have the QR code one more time to read the document, and there is the original document but also at the same time, an ICANN open consultation also.
And with that, I will give the floor to Constanze who will talk about the mentions compared to the last document.
CONSTANZE BUERGER: A small overview about the revised points in the new document, but first of all, we wanted to thank you, the community, for all the inputs we got and all the comments and I can make sure, or we made sure we worked with all comments. We have a really huge Excel sheet and we put all the comments together and we clustered and prioritised all the comments and then we put in the new RIR governance document, the new inputs.
So, what we did in the revised RIR governance document, we have some key changes. First of all, the title. You see this, RIR governance document. We changed the preamble. We had an expand of the preamble. We explained the goals of the document and we explained what the document not stands for. It's very important. Implementation procedures. A reference to the separate implement procedures, the document that may provide further implementation details on any obligation.
So, next slide.
So, we have key changes in the recognition and derecognition procedures. The audits and proposals for the audits and emergency continuity. In detail is my colleague acting and providing the solutions and the new states in the following presentation.
The key issue, and this is what I want to figure out is the role of ICANN. Many people from the community are concerned about the role as too strong a role of ICANN, and this I can ‑‑ I can explain now. I cannot unilaterally regular or deregulate an RIR or initiate emergency continuity procedures. So what I can is, what I am able to trigger a proposal a derecognition of an audit or ad hoc audits. So, this is the topic and the role of ICANN and now we go into the detail and this is the part of Andrei.
ANDREI ROBACHEVSKY: Thank you. So, yeah, let's dig a little bit into the detail. I will not go too deep, to come back to you. But you can see all the details. We showed the reference to this rationale summary and rationale document which contains all the details of feedback that, how feedback got into the document and why and which feedback didn't make it to the document and why.
So, the title: It's maybe a minor change but because it's in the title, we highlight it here. So, we thought "Operation" is actually more adequate term for how you sort of operate, not maintain, the registry. So the title was changed in the document.
More important change probably is the rationale. In the previous version it was a very sort of skinny rationale, basically explaining what the document talks about. But it didn't set the context, and, you know, it's a very tricky context. If you look at the RIR system you may be confused and think that this is a top‑down hierarchical system, but it's not. The authority comes from the community, right, an RIRs are acting independently pretty much, and I think that was one of the very strong things we heard at the last BoF in Lisbon. So we expanded this preamble, we talked about the nature of the RIR system. We talked about the goals, what this document is trying to achieve and what it doesn't. So, please read the rationale ‑‑ this preamble and see if it sort of sets the right context under which you can look at the document. Because for a purely hierarchical system, and you may be confused again if you look at the TLD space for instance in ICANN, some of the processes can be much more simplified. So, some of the complexity that we see in this recognition, derecognition process is they are related to the nature of the system so we thought it's very important to explain that it's not complexity for the sake of complexity, but it's a necessity because of the nature of the system.
There is a lot of feedback on well, we need for details and even suggestions on some of the details of be it an audit or be it some of the criteria, and we felt that the document should stay in a high level. The document should provide boundary conditions for future developments, what we call implementation procedures, and in this document we clearly mention those implementation procedures that have to be developed, that have to be sort of flesh out the details how certain things will be done within the boundary conditions set by this document.
So we don't need to, you know, come back and start the whole process again when we change slightly and we realise some things do not work as expected. Again, read the document, see if we did the right judgment on the level of detail that is present in the document.
Okay, now, a recognition and derecognition. A picture says more than 1,000 words. So here is the flow diagram. It's pretty sort of normal, it was in the previous version of the document. Approval is unanimous at this stage. And well rejection as well. Or rejection doesn't need to be unanimous of course, if it's not unanimous approval, then it's rejection. But several people raised concerns about unanimity of that decision, both from the point of view of anonymity is hard to reach because of the RIRs may have more vested interests than maintaining status quo, making the system really closed. But also from the point of view of well what if an RIR is unable to make a decision, or make a recommendation?
.
I have to say, although I don't ‑‑ maybe that doesn't make a material difference, there is no voting here. If you look at the procedures, each RIR, independently of the proposal and submits a recommendation. So the sum of recommendations, it forms the decision. So if all recommendations are in favour, that's what we call unanimous. So there is no like voting at the table in a closed room here.
Now, for this the procedure was amended and that's what we call reconsideration review. So there is a provision where an RIR can be recognised with less than unanimous support and less than unanimous support there might be one dissenting view from one RIR, and we are considering a scenario where, for instance, you know, a significant part of the membership of the existing RIR goes to another RIR, so there might be a vested interest. And for this there might be a situation where maybe judgement may be affected by this thing.
There was also ‑‑ and let me show another picture with this consideration recognition review phase ‑‑ that maybe employing a third party, independent third party is a good thing to review sort of this assessment of the RIR. So, if it there is one dissenting RIR, there will be an independent commission that will review the judgment and if there is a material error in fair judgment then there will be reconsideration phase. And after that there might be a rejection. So, there might be a reversal when the opinion of dissenting RIR, if they are material RIR or unfair judgment can be overridden and therefore recognition can be made with minus one RIR. So that's that the proposal for the new more complicated process that will allow some dissent.
I have to say before I go to this derecognition thing, that you might think yeah well it's not that easy to become an RIR, right? But I don't think we want to have make it easy to become an RIR because there is a huge cost of this coordination of cooperation, keeping this global system collaborative and cooperative and acting as one single. So we need to think of this cost and really sort of set the threshold pretty high. So, not totally closing this up but setting this high so really, then there is a real necessity, then the RIR can be created and the club can be extended.
I hope you saw the picture, I can't go back for some reason. The recognition proposal there, there is not much change there, we left the unanimous decision because it's a big decision of course, well a part from the RIR that is being derecognised, how all other RIRs need to make a unanimous decision on derecognition.
What was added here is that, well minimum requirements for derecognition proposal. That the, there should be rationale, it can't be just well I would like to derecognise. They should be rational and this should be public. Response: There should be an opportunity for the RIR to publicly response to that proposal and transparency. So when a decision is made, all rationale should be public.
So we want to make this process as much transparent as possible so community where the authority comes from, can actually independently review whether those decisions were made and how.
Also it was added that there is a new threshold for members, for initiating derecognition proposal. Now membership of an RIR can initiate derecognition proposal against this RIR, and there is two holds, 2000 to 25%, why is that? Because different RIRs have different membership structures and member, a definition of a member is different in different areas, so to sort of make a more, a fair comparison, we made these two will thresholds for that process, for this trigger.
It also allows ICANN to initiate derecognition process. That was also a request from the community that in some situations there may be some impasse or an ability for areas to trigger that proposal for their reasons so maybe ICANN shouldn't do this. Now I have to be very clear that here we're not talking about the derecognition by ICANN, we can talking about the triggering of the whole process, so this out left of the diagram that I very briefly showed you.
.
Audits: That's another thing that was added. And there are two types of audit. There are regular audits and there are ad hoc audits and I think they are a very important component for the whole thing because we don't want to get to the derecognition. We want to capture problems as early as possible and be able to rectify them.
So, previous audits. There is a suddenly cadence of audits that areas should do, I think it's three years now. At least one every three years, so pretty frequently. And are so we can maybe capture this earlier, the problems I identified early on.
There is a possibility for ad hoc audits. They are not built in in derecognition process but it's pretty clear that maybe ad hoc audit will be requested for that if necessary. So,ing there might be other RIRs that can request an ad hoc audit, again the same thresholds, 2 thousand to 25% members of the audit can request this RIR audit and ICANN can request an ad hoc audit. Why we did this because really, well ad hoc audit you it's not cheap maybe, but we don't want to raise the threshold of triggering this audit too high because it's counterproductive. We want to detect errors as early as possible, so if there is a reason to believe that yeah, there's something wrong going with this RIR, I think we want to have this ad hoc audit so there is a possibility, there is ‑‑ we haven't passed the point of no return, so to speak.
Emergency continuity. That's a completely new thing that was also recommended by some feedback from the community. And this is not to be confused with any derecognition. This is more of a force majeure situation when we talk about business continuity, operational continuity of the global registry system, and so there are mechanisms in place to quickly transfer operation of the, an RIR, to another designated emergency to operate in urgent situations and of course in urgent situations we have to prepare beforehand not when the urgent situation happens. There is something in the document to say well who is the emergency operator and what it needs to perform.
Now, there is some safeguards. It's not a capture or not taken by other RIRs, for instance they request unanimous agreement of RIRs in ICANN. That has to happen. It must be time limited. It's 90 days, which can be extended but again there is a discussion around that. And should be public. And publicly justify and followed by community consultation.
Right. So those safeguards, it's a continuation of the previous slide. To ensure that it is not, it cannot become a substitute for derecognition or take over of other RIRs. 90 days and postmortem, lessons learned of what happened and how we should avoid that.
There are also other things, not less important but we couldn't group them in one particular category. Rectification, again, different feedback people were suggesting certain grace periods when RIRs have to bring their back yard in order, or their potential amendments. Or with this document, when this document is rectified as well. So, what is specified here is that it depends, it depends on the amendment really. Some amendments can be rectified in a month, some of them take longer.
So, that's why when an amendment is proposed, part of the amendment is actually the grace period, what is ‑‑ how much time is given for the RIRs to become compliant with this new amended document.
For the migration from ICP‑2 to this document right, that's right the same thing, and that's a change and that's case in point, a three year grace period is specified. So RIRs are given three years to be fully compliant with this new RIR governance document.
Dispute resolution. That's one thing that I mean the RIR should have a dispute resolution process that allows their members not to, you know, go ballistic but maybe resolve the conflict with more peaceful ‑‑ by more peaceful means.
Good faith: Well that's very important. The whole proposal assumes good faith, that RIRs are acting in good faith and it is describing the system where RIRs are cooperative, right, where the system is actually operating on collaboration, it's not hostile, it's not competitive. My personal opinion is that if those principles, they change and there is no good faith, we'll need another document, this document may not work.
Right. I think that's all the sort of highlights of the changes. Again, please read the document itself, or the summary rationale, there is a few reminders, so the consultation period is open till 7th November. So please submit your feedback on the mailing list. Now, here, soon, as well, and with that, I think I'm done with my part.
HERVE CLEMENT: Thank you Constanze and Ondrej, two things. Before coming to the discussion, you can go behind the microphone if you want to ask your question or to say something regarding what we presented. There will be online questions as well. Just to define what the culled the CCG, so the communication team from the regions who help us a lot about drafting the document, helping, gathering all the comments. So you have Ulka, for instance, thank you. Also, thank you because we have a lot of legal questions, something we are not ‑‑ some of the members of the Council are legal but personally I'm not, so we have the support of the legal team, so you have Athina and Michael for instance, Michael from the ARIN region who helped us. They don't decide, be sure, they just support regarding that. And the other thing that has been recently a change, you know that there have been an election within the AFRINIC region, the AFRINIC has no Executive Board with quorum and we hope very soon to have one, two or three people from the AFRINIC coming to the NRO NC, so they won't have this issue of not having completely the absence of the AFRINIC there which will be a very, very good thing.
I see Jim behind the mic and after Jim perhaps we can go to online.
JIM REID: Thank you Herve, I am speaking purely for myself, I have no organisation behind me. I have two points or observations. An easy one and one that's maybe not so easy. So the easy one is I understand you and your colleagues have done a fantastic job, have had some legal advice about the proposals you are putting forward. Would it be possible for that information to be made public? One of my concerns around this stuff is that parts of these adopts smells a little bit too much of a cartel anti‑competitive in some respects, even though we are all supposed to be collaborative and working together. From the outside it might look that is existing RIRs are acting as one where we decide you can join or club or not and that might not stand up to a legal challenge. So if there is some legal advice around that I'd appreciate if it could be made public. I realise there might be some discussions with lawyers around that but hopefully that can be solved out.
The more difficult issue I have is on this question of unanimity. And as I have made comments on the list and also privately to yourselves, is we could have a situation where an RIR is unable to express an opinion or make a recommendation on the formation, the creation or removal of an RIR for whatever reason, not necessarily because of a situation where an RIR hasn't got a functional board, it could be joined some kind of legal action or whatever, but if we're in a situation where one the existing RIRs cannot express an opinion, then you don't have unanimity, therefore you cannot make a decision about anything, you are effectively stalemated at that point. I think this problem really needs to be fixed.
Now, from what Ondrej just said a moment or two ago, he was talking about, I may have misunderstood him as suggesting that unanimity can just be the RIRs who expressed an opinion. Well that's not the legal definition or the dictionary definition of unanimity. It's essentially those who voted although we're not talking about voting here. That's not the same thing. I think we need to solve this in another way. My preference would be to do this through consensus, because that's what we do in the Internet everywhere, we make decisions by consensus. And I understand there is a problem there because some RIRs don't have a definition of consensus or they don't have a common definition of consensus. So one might be to define consensus in a very simple way which is lack of sustained reasonable opposition.
Now, if we can't make that fly for whatever reason, I think the other possibility then is to have some kind of super majority like say two thirds of the existing RIRs rounded up to the nearest whole number expressing a recommendation would be enough to give the strong sense that yes, there is a recognition from the existing community that an RIR could be created or removed but not unanimity in the sense where all of the existing RIRs have to make that decision. I think that needs to be tackled.
Now that I am finished ranting and saying my piece, I just want to repeat once again I am grateful to you and your colleagues for all your hard work, it's an important job and it's important to get it right and I am glad you are doing it because in some ways it's almost a thankless task so keep at it and I very much appreciate all your persistance in all of this. Keep it up.
(Applause)
HERVE CLEMENT: Just a few words perhaps, I don't know, perhaps my colleagues want to answer, because we know this discussion, I don't know if you are behind the mic, perhaps for other things or the same thing, Athina?
ATHINA FRAGKOULI: I would just like to address the first question you posed if I may? Like, indeed, these concerns came in our minds like very early in the process, and that's why we had external actually legal advice on the matter.
The preliminary outcome of this advice actually is very positive. According to that, the RIR activities fell out of the scope of competition at least from the EU perspective, EU competition law perspective, and if one suggests that no actually they do fall within the scope, we fall also under an exemption. So we feel that the concerns are addressed and we're looking into having like a solid final opinion on the matter that we can also publish for the rest of the community.
JIM REID: That would be fantastic, thanks Athina.
HERVE CLEMENT: Also, regarding consensus, so your comments are very rich and so we can spend a lot of time about that information. So we will meet in November which is a LACNIC legal place, to discuss all that. So that there was a change regarding the consensus process compared to the previous document but there is a lot you say that if we go to a consensus something could be difficult also because of the definition, and it's not the same. So that's my preamble. I don't know if Andrei or Constanze want to complete what I say because there is a lot one more time to do and we really appreciate the comment.
CONSTANZE BUERGER: Just, I can add your concerns about unanimity. We take this with us and we discuss it in the team, and this is a very good question, yes.
HERVE CLEMENT: Next.
AUDIENCE SPEAKER: Desiree, speaking as a member of the community. First of all, thank you so much for this work for trying to tighten up the governance model and adding the auditing for every three years, I think it's a good improvement but I am glad also that Jim spoke before me because of what he raised was that aiming for unanimity in the vote is especially difficult especially if one RIR is in question, therefore you wouldn't be able to achieve much. In favour of this, two thirds or removing the body that is being, you know, questioned or triggered or...
So, otherwise the document reads completely differently, that the run that has been challenged could also participate in the vote. So thank you for looking into that. Thank you.
HERVE CLEMENT: Thank you Desiree, I think we'll spare half a day to discuss this question.
NIALL O'REILLY: It's Niall O'Reilly and I see I am going to talk about something I don't really have any qualification to talk about but ignorance has never stopped me before.
I think the question is important. There are two aspects to the outcome of this. One is the new processes which would be adopted. And that has been made sort of ‑‑ that looks to be very clear can and probably to be converging, and the effort, as Jim already said, of the NRO NC in doing this work and from my point of view, particularly for the delegates from the RIPE region, is something I am very grateful for.
But the second aspect of the outcome is the community narrative, the documentary history of the process, and what's not clear to me is whether this will be a new document in the ICP series with a new number, whether it will be a first document in a new series, or something else, but I think it's important that ICP‑2 will be frozen like an insect in amber as part of the record of this status quoante. It's what I see in the RIPE community in the IETF, and old document is never hidden, it's kept, it's marked obsolete, it's superseded by a new document, and this supports the narrative continuity of the community, and I think that's a very important aspect which shouldn't be lost and I am wondering what the end point of the documentary history will be?
HERVE CLEMENT: So, we totally of course we accept your question, so, it was not the good work but we are very happy you asked the question why, because we discussed that this afternoon and in your specific words, so you are totally in line when specific that. I see Hans Petter behind the mic because I know that you have answers or something to say about that.
HANS PETTER HOLEN: I wish I had the answer. The answer is an idea. Hans Petter Holen, I am Chair of the NRO NC. ICP‑2 was a document series that was started by ICANN in the early days and it got to number 3 or 4 and has since been disbanded so the successive documents will not be in that document series are because it's a nonexistent document series.
Most of the other ICANN SOACs that we like to compare ourselves with, have at some point actually quite a bit back figured out that it would be very good to have a document series that we numbered the numbers in. So SSOAC at an early stage said okay, we have published these documents, let number them and then continue the series. RSAC did the slightly different flavours, they do a version of documents so that's a bit so. Creating a new document series for the ASO NRO space along those principles like your RIPE document series, RFCs in that spirit, is kind of on my idea look to propose to the NRO NC and the NRO NC NC, it's kind of like should it be the NRO series or should it be the ASO series and should we retrospectively number everything that we have here so that we have a record for the future. So yes I am in favour of that but the exact proposal isn't really on the table yet but hopefully it is before we have consensus on this document. If you have ideas, come here and ‑‑
NIALL O'REILLY: Sure, but what I wanted to do today was simply to give input from the community that this was a matter of concern, and if you need sort of documentary input from me, you know where to find me.
HERVE CLEMENT: Thank you Niall. And when speaking about the the ICANN, I realised that there are great guys, I didn't thank because there is ICANN under consultation, there is the ICANN community, is the ICANN which is a great support, we have Carlos and Andres there, I want to thank you also.
AUDIENCE SPEAKER: Hello, thank you, Tom Hill from BT. I am minded to think at the moment about abuse of this policy, and in the way that people may contort the language that we have ultimately write down. I recognise from the summary document that there has been a lot of questions raised in the past about whether or not RIRs, new RIRs overlap with another service area, it's good that that's been clarified. I have been wondering about what the role that the NIRs play in this situation, representing communities, countries, potentially multiple countries bandied together. Have we given thought to the idea that a group of NIRs represents multiple nations could potentially put forward a proposal to cede from an RIR and develop their own Internet registry. Do we want to call that out explicitly? Do we want to deal with that or do we want to say that's normal, that's fine?
HERVE CLEMENT: I would have ‑‑ thank you very much, first. I would have a first answer and then I will let my dear colleagues to complete for that. So regarding the national Internet registries, it was not decided, so the purpose of the document is to say that it's open to the regional internet registries to deal by themselves how they have the management of resources internally. So we are not against, I would say we are not of a strong opinion about the presence or not of regional internet registries, but if ‑‑ so the condition to have national Internet registries is, there won't be an impossibility for the regional internet registry to fulfil finally the principle raised in the document. That's the first thing. After... so I'll let you, if you want.
ANDREI ROBACHEVSKY: I think if you ask me, that would be the most natural way of a new RIR to appear, right. You already demonstrated community support, you already operated in this space. Whether that should be explicitly stated in the document, that's another question. That's again the balance between the level of detail. Maybe that's a recommendation. Maybe that's an implementation procedure. I think embedding this in the document may not be very productive. Like if you look at how LACNIC or AFRINIC were created, or created before LACNIC was really created the factor before ICP‑2 was in place. Like it was already operating. So, I would wonder if absence of this explicit statement that NIR can sort of progress into the RIR would impede an NIR to submit such proposal. In my opinion, it would not. Putting this in the document creates some I think conditions that might not work in the future, or maybe sort of be compared with other conditions when there are no NIR, does it say that not being an NIR prevents you from becoming an RIR? So you see, the thing is, and that's what we also heard from some of the community members. Keep it as simple as possible. We tried to do this. We didn't succeed fully. I mean if you read this document I would say that it's not simplest document you can read. But we tried.
HANS PETTER HOLEN: Be aware that we do not have national Internet registries in the RIPE NCC Services region, and I don't see an easy way of forming that. So, you know, I would not be in favour of only introducing global policy like documents that describes national Internet registries, that's a region issue, so I would try to like to say on a common denominator and since we don't have national Internet registries I would prefer not to have to deal with that potentially have the pressure to establish that.
AUDIENCE SPEAKER: I realise there are none in the RIPE NCC region today at the very least. I think what I'm getting at is if there becomes a recipe for gaining your own RIR through the use of NIRs, perhaps what should only be called out if it is called out is that having formed an official NIR does not give you any greater claim to a service region perhaps.
HANS PETTER HOLEN: While that is an interesting perspective, another interesting perspective is that if you today, as part of the RIPE NCC Services region, are contemplating changing the structure, the only other option you have is to form a new RIR. There is no option to form an NIR, right. So...
AUDIENCE SPEAKER: So, hi, I am a community member but I am coming here from dot RI Serbian country code. So, I was there at ICANN where this document began to be discussed, and now looking at that I think the progress is great. I also think that positioning how it's done now with ICANN is appropriate. In a way it's also a bottom up organisation so we should give it some trust, but obviously the way how RIRs have the decision power is good.
I think ‑‑ my thinking goes into this discussion about a level of this document. So you mentioned this quite a high level document and the implementation details will be separate. But I think we have a significant problem there. I think with he need to put more teeth into this document and I think if you try to stress test the document, even with the problems we have today, we see that there is nothing in the document that will really help us with the situation. The problem that I see is the document does not mention what is happening with the currently assigned resources, meaning IP blocks to the entities that have been assigned to. So if you have a derecognition, a recognition of the new regional internet registry, what will happen? Probably some new policies will happen. So, I think that we cannot leave that to the implementation document. I think we need some sort of high level guidance here now. So when I try to be creative because I was detached from the process so I have a kind of a thinking of a new from the side maybe it was all discussed but maybe we can grandfather the allocated blocks, maybe we can prohibit transfers, maybe we can even give some period and then the assignments have to be assigned with the new policy or return or something like that. And bottom line here is this stress testing. If we have a document now that is not really helping with the current problems, then the question is why we are doing that, it's good to have it, but I think that we need to make for theoretical RIR that has a structural problems, we need to make this document worse and more pain than fixing the structural problems. That's the something that will really help the community.
If you leave ‑‑ this discussion is taking a long time, we saw the timeline, so if you leave that to another layer of discussion for implementation document, it will take another many years. So it's better maybe, in my opinion, to bring this key problem back into this discussion and start, you know, putting it into the open. Thank you.
HERVE CLEMENT: Thank you. One more time, it's really an ongoing discussion we have an we have it since, I would say since the beginning, so thinking of the level of details, so you stressed the problem of assignment of resources, but there are other problems that can be discussed and served, so from now to today, it will be in a separated document but we know that this document and these details one day will exist, so we are sure about that. So the question so do we have to deal with that now or later? But we have that in mind.
CONSTANZE BUERGER: You are right, we have that current problem. But we have to care about a long term stability. That's the way we go forward. That's the first. And we actually try to care about the current problems. So we invite you to support us to give us some ‑‑ your ideas, take part in the sessions and so we can meet this term and so you are invited to give advices. How can we go forward.
SPEAKER: We have an online question from Andrew Campling. He said: "Excellent work and outstanding focus on consultation. Regarding acting in good faith. How and by whom is this determined? Who or what decides what process applies in the event that it is determined that one or more RIRs is not acting in good faith"?
HERVE CLEMENT: That's a good question, that's why we introduced in the process third parties. So if we come back to the details on operational process with something so we can answer, that's why also... definitely we agree that that's something that must be impartial or independent of course but we have to define the term of independence.
ANDREI ROBACHEVSKY: I would think that that might go into the implementation procedures for auditing where you really sort of detail of what kind of parameters you look and how you determine for instance good faith as well. But again, and I think that's back to the previous session whether we want to sort of go and embed this in this document, or that we should rely on sort of community wisdom that when a situation arises we can sort of discuss and create those procedures.
The thing is goes back to LACNIC for instance, creation, I think if you look at this document as a sort of instruction how to create, recognise nice or derecognise RIRs, it raises all these types of questions. But in fact it sets the boundary conditions, in my view, and all this kind of nitty gritty details that were agreed by LACNIC before the ICP‑2 was in existence even, right, so LACNIC was dethat facto operating before ICP‑2 was adopted. So, that's why ICP‑2 doesn't go into this nitty gritty detail. It does but in the form of discussions. It says well it could go this way, we can go this way. In this document we have chosen a different approach. We do not show options, it's not an implementation guide. It's more a, you know, sort of policy, a framework that sets the boundary conditions leaving to the communities or to the ‑‑ I don't know... ability to develop those implementation details.
ATHINA FRUGOULKI: Thank you. I realise I did not introduce myself the previous time I was on the microphone. I am the Chief Legal Officer of the RIPE NCC, and I appreciate a lot the question that was made about the good faith, yes, it's a very ‑‑ it's a term that is used in the legal language and it's up for interpretation every time it takes into account various circumstances on a case‑by‑case basis and so on. However in this particular case in the provision 2.8 good faith, we try to focus a little bit on the following. We say all actions required under this document shall be undertaken promptly and within reasonable time lines ensuring that processes are not unduly delayed." So when we tubing about good faith, we focus a little bit on responding on a timely manner and giving opinions in a timely manner, acting in a timely manner. And if this is not the case, then there might be reasons to claim that an RIR does not act in good faith. Thank you.
RANDY BUSH: I mention this not because I represent them, but because I want you to know my biases. It's just good manners to do so. If you want complexity, the IETF meets next month. And I really dislike complexity. The first time I gave a talk, and I believe it was at RIPE, on how just at the time, this is the late nineties, just three RIRs were driving me crazy because of the given procedures, the different da da da da, I don't have to go ‑‑ many of us experienced this, the lack of real progress on coordination with the current number of five RIRs, does not make me want more RIRs. It makes me want to work on the problems of making the whole community work together, and for those others in this room who do global scale operations, we all feel this pain.
So, I do see the down sides of the unanimity point, I also see the up sides. I am an engineer, I am less worried about the process of whatever we're calling it to turn off an RIR than in fact the actual engineering cases we have had is a failure of an RIR, not the community deciding to eliminate that RIR.
So I want to deal with the real problems we have, not the corner cases that we can imagine if we want to imagine corner cases and do complexity, as I said, IETF meets next month.
HERVE CLEMENT: Thank you, Randy. I don't have the exact words in the document which is under consultation now, but we have something so we meet ‑‑ so regarding the number of potential regional internet registries we just say that so it is not to impinge, or an easy way to work, I will say, it's not exactly like that, but it's ‑‑ it meets what you say.
AUDIENCE SPEAKER: Evening. Alexander speaking for myself. Thank you all for coming, and thank you for the splendid work that you have done. Is it perfect? Most likely it's not. Does it cover everything? I think it doesn't. But the most crucial question is: Is it better than the current document? And my answer to this is yes, it's definitely better. And that's all we need. We cannot always get it perfect a hundred percent on the first try but it doesn't stop us to improve. So thank you. Two minutes left. Time for one question.
AUDIENCE SPEAKER: Wolfgang. More a comment than a question. My impression is that most of the technical community is not really working a lot in the governance area of stuff. So what I think would help a lot to get more input from the technical community is if you presented a walk through through the process, starting from what goals you want to reach, why you want to reach them, and then walk through the draft, explaining we choose to use this text with the goal of implementing these top level goals.
HERVE CLEMENT: Well noted. Thank you. So if not, so, we really thank you to have spent time to be with us.
.
(Applause)
.
And of course as Andrei records we are here all the week, you can talk with us during the break. There will be a presentation during ‑‑ Andrei there will be outputs on the, a resume of the outputs of the discussion on Thursday during the Plenary session also. Thank you.
(Applause)
LIVE CAPTIONING BY
MARY McKEON, RMR, CRR, CBC
DUBLIN, IRELAND.