Thursday, 14 May 2020
4 p.m. (CEST)
HANS PETTER HOLEN: Welcome everybody to this RIPE Community Plenary. It's going to be an interesting format because what I really wanted to do when we set this up was to have engagement with the community, so we will see how we get that to work in this Zoom format.
So, I'm your Interim Chair, as you probably know by now, and with me I have Jan Zorz from the Programme Committee and two colleagues from the RIPE NCC as co‑hosts to help us on this session. So, I had a chat with Jan upfront so while I will be presenting quite a bit upfront, he will do the picking of the questions from the floor, so I will have no influence on what questions are getting asked. Jan will be the one selecting that.
So some housekeeping slides. Always write your name and affiliation only in the questions and answers window. And if you use your RFC on the live streaming page. If you want to write something in the Zoom chat, always select panelists and attendees. Otherwise only panelists will see your message.
The sessions are recorded and will be published on the website, and you can also see a live transcription that you will find on the website or the URL that you can see here.
So today's agenda:
First, we will have a RIPE NomCom update by the Chair, Daniel Karrenberg.
Then Nurani will present the 2019 IANA Numbering Services Review Committee report.
And then we have a short report from the RIPE database requirements task force by Bijal, and then, after the break, I have two different types of presentations: one on the Working Group collective and another one on the RIPE Code of Conduct, the next steps there.
At the end, hopefully we have some time for any other business as well.
So, Daniel, the floor is yours.
DANIEL KARRENBERG: Hello everyone. Can you see my slides? Okay. I am Daniel Karrenberg, I am the Chair of the 2020 RIPE NomCom, the very first one. What does a NomCom do? It selects the RIPE Chair and the RIPE vice‑chair. As the NomCom Chair, I am non‑voting. That means I don't vote on candidate selection, I am just helping the NomCom along. And my agenda for today is a status request for input and we'll have a Q&A.
The RIPE NomCom process:
RIPE community governance, so it's open to all, I have had some people ask me, hey, I'm not a RIPE NCC member, can I participate? Of course you can if you participate in RIPE, you can provide input to the NomCom and participate in any other way. This is not RIPE NCC corporate governance, it's RIPE community governance.
The process is described in two RIPE documents, RIPE‑727 and RIPE‑728. It is the most elaborate and detailed RIPE process to date. It has more words than any other of the processes we have so far.
The process is open and transparent. You can follow it on the RIPE list and on a special NomCom blog which we have set up totally outside the RIPE NCC computing infrastructure, we also have mailing lists that are outside RIPE NCC computing infrastructure, so if you mail us, it's just the NomCom plus one support person who can see what you mail us.
And the NomCom closely adheres to this process. It's not our job to invent or change the process. It's our job so execute it in the best interests of the RIPE community.
So, what happened first? Well, the RIPE NCC Board appointed me as the NomCom Chair. That happened in September 2019. So it's quite sometime ago already. It feels like years ago. First things I did was I called for volunteers to sit on the NomCom. I did that at the beginning of RIPE 79, and at the end of RIPE 79, I issued a call for nominations.
Then I ran the volunteer selection process, which basically involves making a list of all the nominees and then selecting some lottery results in the future that would determine who of the nominees would get selected. We had 36, if I remember correctly, volunteers. Of those, ten were selected to serve on the NomCom.
I also requested the RIPE NCC Board, the Programme Committee and the Working Group Chairs, to send us liaisons. The reasoning for that is basically that all these have to work closely with the RIPE Chair so they should have input into the process and they should also be able to convince themselves that the process runs correctly. And that is especially the RIPE NCC Board because they have to check that at the end.
There is also a role for the previous Chair in the procedure, and since this is the first NomCom that was the classical boot‑strapping problem, and I solved that by asking Anna Wilson to be previous Chair between quotes.
All this and much, much more you can find on blog.ripe‑nomcom.org. There is documents there, there is blog entries. If you are interested in anything about this process, this is the place to go.
What happened then? Well, we received one, in words, one nomination. And that was Hans Petter. Then we had a meeting to organise ourselves. And then we called again for nominations and we canvassed, we went around, asked everybody and their mother to nominate people who we should consider and we received four more nominations: Filiz, Niall, Nigel and Mirjam, and we were actually ‑‑ personally, I was quite happy, because my expectation that Hans Petter was very well liked by everyone in the community and had a lot of support, and we would just select Hans Petter and a co‑chair.
And then, of course, as it always happens whenever you think a job is easy, Hans Petter withdrew for the obvious reasons.
What happened recently?
Well, we are recovering from the shock. We had a second meeting and I just have the main points of the agenda here. I don't want to go through them all. The minutes are published in the blog, you can find them. But one important ‑‑ two important things I want to point out is we at this point already, which is a couple of months ago, we discussed our own conflicts of interest, and we also discussed the issue of conflicts of interest of the nominees and how we should address them.
So, it is not that the NomCom is totally oblivious of things like that and it's also our job to do that.
What happened first? Well we had a meeting with each nominee, the whole committee, talked for about an hour. We also had the nominees answer a list of questions that we sent to each of them before. And we asked them to have a statement on the NomCom blog so that people who do not know them or do not remember knowing them get a chance to refresh their memories and give us useful input.
We are collecting community input and what I have to say here especially is that of course what we discuss with the nominee, the input that we're getting is all treated in confidence. We're not going to tell anybody outside the NomCom about what you might have said or what the nominees have said to us in a private meeting. We will also not minute our deliberations when we discuss the candidate selection. We will just minute the outcome and maybe some general things about the considerations that went into it, but this is a private process.
We had open office hours at each lunch break during RIPE 80, and we had one person actually turning up giving us input. So, the only conclusion that we can take from this is that you are happy with what's going on.
And we have, of course, a discussion on 'RIPE Chair Discuss', which ‑‑ where some people said they weren't happy, and we are already seeing consequences of this that don't make the life of the NomCom much easier. But I guess Hans Petter will announce that later on.
What will happen next? We'll collect more community input. We have office hours tomorrow and we are still open for hearing from you. Then, in the next couple of weeks, we will decide on a selection and after we make that selection, and that will be a reasoned selection, we will get the RIPE NCC Board to confirm this, and their role in the whole process is to have a process check. Basically they'll have to check whether we followed the process that was set out and we did that correctly. It's not RIPE NCC Board's task to confirm or reject a candidate based on whether they like the candidates or not. It's purely a check of the process.
And then we will have an early transition, and originally it was planned that the transition would happen at the next RIPE meeting, you know that Hans Petter would hand the Internet to two people in front on the stage. Now, since Hans Petter has withdrawn his candidacy and has also resigned as the RIPE Chair, there is a little bit more urgency, it wouldn't make sense to actually wait. So, we will change the timeline as foreseen in the procedure and pragmatically and have the transition as soon as practical.
Then, as foreseen in the timeline, by RIPE 81 we will have a final report, it's already partly written, and as the NomCom we will report what we did and we will probably also make some recommendations to look at the procedure and see what could be improved there.
And then the NomCom will be done and we are looking forward to that particular date and where the NomCom will be done and cease to exist.
And after that, of course, there will be an evaluation by the community on how this process went.
What I would caution the community against is discussing changing the process before that time, because it's never a good idea to change a running process. It would make the life of the NomCom very, very difficult if we now had a discussion on oh, no, we don't like what we agreed earlier, let's change it. Well the aircraft is in flight. That's not a good idea.
And finally, please talk to us. You can send mail to NomCom at the ripe‑nomcom.org. This does not touch any mail systems of the RIPE NCC or any secret entity. It's a mail server run by the NomCom for the NomCom and it's only the NomCom members plus our support staff seeing what's going on on that list.
You can also, of course, approach any individual NomCom member, no problem with that either. We have open office hours tomorrow. If you want to make an appointment, you send to support [at] ripenomcom [dot] org and you will reach Daniella. And if you need more information, for instance, the individual NomCom member contacts, and so on, you will find them at blog.ripe‑nomcom.org.
There is one thing that gets into my mind because things are hectic this week. This page still lists ‑‑ unless the other has corrected it already ‑‑ Tina Mars is a voting member of the NomCom and Tina has informed me that, for personal reasons, she has to resign, so please do not approach Tina, she is busy with other things.
Let me finish with a few remarks about how to help us.
This is not an election. So, a plus one kind of, oh, I'd like you know, Jenke or Peter is not going to help us. What helps us is a reasoned input saying, like, you know, I think this person could do this job for this reason. That would ‑‑ that's the helpful thing.
Also, some input about the points that are raised in the RIPE Chair Discuss list or in the general community discussions is always welcome. We will treat all input in confidence, and there is one more thing is, we have had some people approach us saying I'd like to make an anonymous comment. We are not I think a hundred percent agreed on what we should do with that, but personally, I really, remuneration, really appeal to you not to do this because it's really difficult to take anonymous input into consideration because it could be anything, and it could be anyone, it could be a bot, it could be some person talking to us five times. This is not ‑‑ I personally don't find that really helpful. So if you ‑‑ if you are concerned that your input might leak out, please, please believe us, we take everything really, really in confidence, but we really need to know who you are in order to place your input and be sure that we're not going manipulated.
And that brings me to the Q&A.
JAN ZORZ: Hi everyone, there is one Q&A. There is one question. Let me see. It's from Erik Bais:
"Is it possible for RIPE NCC employees, non‑management, to provide public input on the process as well? And if not, why not? As one of the nominees is currently an NCC employee, would that be a conflict of interest in the balance of power with the current or future MD of the NCC or will the employee be asked to be the NCC role? It would be viewed very strange if this going to look like an open door in my opinion."
DANIEL KARRENBERG: Those are really two questions. The first one is basically the process doesn't prevent anyone from making input to the process. I don't think we want much public input because then it turns into an election. But that's fine. But certainly, the NomCom will listen to anybody who makes a useful input to the community.
Then, of course, Jan just hid the second question from me. I got it. This is ‑‑ basically, RIPE 727 says any natural person is eligible for selection unless they have already been selected twice for the role that they are nominated for.
And this is the procedure we are following. So, there is no possibility for the NomCom to disregard any candidate. The language is absolutely clear. And since the NomCom has not discussed the candidate selection yet, it's not appropriate for me to say any more.
JAN ZORZ: I hope this answers the question. And there are no more questions so far.
DANIEL KARRENBERG: That is surprising.
JAN ZORZ: That is surprising.
DANIEL KARRENBERG: I'll give it back to Hans Petter and if people have questions in the meantime, maybe Hans Petter can take them at the end of the Community Plenary, or I am happy to take them at the end of the Community Plenary.
HANS PETTER HOLEN: Thank you, Daniel. I was actually expecting more questions as well. But since we have none, we can have them at the 'any other business' at the end of the session, then I will hand over the floor to Nurani to present her report.
NURANI NIMPUNO: Thank you. And I guess this is where we see the limitations of this online virtual webinar format of the Zoom meetings. These discussions are of course a lot easier to have when you can speak freely at a microphone.
Can you see this presentation I am wondering? Because I can't see myself.
HANS PETTER HOLEN: We can see you but not your presentation yet.
NURANI NIMPUNO: Okay. Hopefully that's better.
HANS PETTER HOLEN: Excellent.
NURANI NIMPUNO: Thank you. My name is Nurani Nimpuno and I am presenting a presentation here on the IANA Numbering Services Review Committee report for 2019. I am the Chair of this Review Committee. And just to give you a bit of a background, the reason this Review Committee was put together was that it came out of the IANA stewardship transition in 2016. So that's before the numbering services were managed through an SLA between ICANN and the five RIRs, and as a result of this process, that SLA was then shifted to ‑‑ sorry, it was before with the NTIA, was now shifted to an SLA between ICANN and the five RIRs.
And part of that proposal called for the establishment of this review committee.
And we call it a committee, but at the time we talked about calling it a team or something less formal because it really has a very small role. So the idea here was really to enhance the accountability of the IANA and the relationship between the RIRs and IANA, so to have a community channel for input into this process.
So, TTL review committee consists of three people from each RIR region. And the function is to advise and assist the RIRs in their evaluation of how the IANA numbering services have been carried out. So there is an SLA. It is really up to the RIRs to see whether or not this SLA has been fulfilled. But this is a channel for the community to express any opinions or concerns should they have any.
And the process, the way it works. It's very simple process.
The IANA actually publishes reports every month on their services. We ask the RIRs to jointly put together a performance matrix to summarise this. And we then take this matrix and we give it to the community to comment on.
From the RIPE region, we have Filiz Yilmaz, myself and then there is always the staff, RIR staff representatives. In the RIPE region, it's Nikolas Pediaditis and it's the same set‑up for all five regions.
So what we do is that we engage with our communities, we put out this RIR matrix and we also ask for input but we also see does our role to make the community aware that there is this process and this review committee.
So, what happened in 2019?
Well, we collected all the reports for the year. We asked the RIRs to put together a performance matrix. They did so. And we sent this out in all the RIR community mailing lists and we asked for any comments on that.
We then looked at that and reviewed all of the input and we also tried to make an evaluation ourselves. And this results in the report that we put out.
So, in 2019, there was one IPv4 allocation, that was actually an automatic allocation to LIRs due to the post exhaustion policy. There was one AS number request and there were three IPv6 allocations and they looked like this. This is all on the IANA website. And we have tried to be faithful to the format that they provided so that it's clear to the community what happens.
So, when we have put out the dates in the matrix that actually reflects when the requests have been processed.
So, in March, this is the transactions, these were the ones in May and June. Any ones that are not included here, there was simply no requests to IANA. And in November, there was IPv6 requests from ARIN.
So, the report concludes that all requests have been fulfilled accurately and on time. And there is no indication of failure or even near failure by IANA to meet the SLAs and we haven't seen any concerning or interesting patterns or anything else that we feel that we should comment on.
There have been no concerns raised by the number community. There was one comment from Leeman to the effect that this was not a surprise and noting that the SLAs had been fulfilled. So the review committee concludes that the performance of the IANA numbering services operations are within the SLA and meets the need of the Internet number community.
So, this concludes my presentation. It's not a very exciting one; it's, in fact, quite a boring one. But we think that boring is good and we hope that you agree. This is a very simple task that IANA performance and that in itself is not a bad thing. But, of course, we still try to work on improvements. We have done this now for three years, I believe, so, for example, we have talked to the RIR teams about finding a nicer, prettier format for the matrix, we want to improve the way the report looks, and look at if we are communicating properly, are people actually well understood with the process for this?
If you have any feedbacks or thoughts or observations, we welcome them here, but also feel free to contact me or anyone on the IANA review committee.
We think it's good that it's a boring and simple process but that doesn't mean that it cannot improve ‑‑ be improved, we think that's important to state as well.
Here are some references for you and that was it from me. Thank you very much.
HANS PETTER HOLEN: Thank you very much, Nurani.
JAN ZORZ: Talking about boring, there are no questions.
HANS PETTER HOLEN: So hearing none. Then I suggest simply moving on, since I potentially have more interesting things to discuss in the next slides, who knows?
The next one up is the Database Requirements Task Force. Bijal, take it away.
BIJAL SANGHANI: Sorry, I think I tried to take over your screen. Okay, hello everyone. It's nice to be here. Of course seeing everyone in person would have been nicer, but here we are.
Okay. So I'm going to give you a very quick update on the RIPE Database Requirements Task Force.
So, the RIPE Database Requirements Task Force was formed in October 2019, and this was an outcome of the big picture BoF that took place at RIPE 78 in Reykjavik, Iceland, that seems like a lifetime ago as well.
The task force is tasked to produce a RIPE document listing all the requirements for the RIPE database and of course their rationales. So, this doesn't just mean that we're looking at the technical elements of the database. You know, we are looking at the bigger picture, and obviously the database has changed immensely since 30 years ago when it was originally started.
So, who are we? These are the task force members. I think you probably have seen most of us around at previous RIPE meetings at least.
The task force, as I said, was formed at the end of last year. Since then, we have had nine meetings, two of those were full day face‑to‑face ones and we have had seven meetings online.
One of the working drafts that we have been working on was posted on the RIPE lists and other mailing lists on the 6th May, which coincided with the date of the euro RIPE database task force BoF.
So, what do we do? We carried out a survey. We had 206 responses. So thank you to everyone that responded. We were very pleasantly surprised to see the number of participants at that.
We had a number of ‑‑ we had some very good feedback from the survey but we had a number of unexpected results. IPAM for one, RPSL and geolocation. I think everyone knows these topics and that they have been discussed in detail previously as well.
So, you know, other interesting discussion points that we have talked about in the task force is, you know, one of the feedbacks we had from the survey was, you know, we think the data accuracy is important, but the database isn't accurate. And then we had others saying that it is accurate, so ‑‑ and also, you know, what does accuracy actually mean? You know, without any context, it can mean anything to anyone. So, you know, we have had some discussions on trying to figure out when we say we want accurate data, what does that actually mean?
Other things like personal information. Do we really need to have the person objects or any sort of personal information in the RIPE database?
Another topic that came up that we were surprised to see was the IRT objects, are they actually being used by the community and ‑‑ or is there essentially somewhere else that, you know, this work item can be used?
So we have identified initially five purposes for the database, and they are to provide accurate registration information of Internet number resources, facilitate the communication about the resources, the usage of the resources, publish the routing policies by network operators, so the RIPE Internet Routing Registry, the provisioning of the reverse domain name system, and RPKI, which is the new one, because it wasn't originally part of the RIPE database, but it complements and references the RIPE database, so we have put it here for now and obviously there is more discussions that we need to have on that.
Other uses that came up in this survey was research into network operations and topology, Enum, again geolocation, iPAM and the transfer of the IP resources.
So we had some really good constructive feedback both at the BoF and on the mailing list. So thank you everyone that attended the BoF and gave feedback and also to those that have sent us messages on the mailing list. You know, I think we definitely have to take these back to the task force for further discussion. I think some of the interesting points that came up, for me anyway, was, first of all, the comment from Nathalie during the BoF on IPAM, this was a real surprise on the number of ‑‑ number of the participants that took the survey that were using the RIPE database as their IPAM solution and during the BoF we asked if, you know, was this their primary ‑‑ for those that do use it, was this their primary or secondary or just a reference? And it actually came out that a lot were using it as primary, which again surprised us until Nathalie actually explained that it seemed that there is an IPAM solution that actually updates the RIPE database at the same time. So people just think, well, we'll just update the RIPE database so it's one place to update. So that answers that. But, you know, does it make it right? You know, we don't know yet.
You know, we need to look into, again, there was a lot of good feedback on RPKI and how that might change things. And Nick presented in the Routing Working Group on RPSL and what potential next steps there are for that.
So, while we haven't been able to meet face‑to‑face, we have had some really good feedback and discussions, and I would like to continue those. I hope that we continue to get the constructive feedback that we have had already and, you know, I think it's time for the task force to kind of go back and see where our next steps are and then I'm sure you'll hear from us soon after that.
So, that's it from me. Thank you. Are there any questions?
JAN ZORZ: I see no questions in the queue, Bijal. I think you were quite clear.
BIJAL SANGHANI: Okay. That's good.
HANS PETTER HOLEN: This is extremely efficient. I had expected many questions regarding this very important issue, but maybe it's just an indication that you are doing extremely good work, so thank you so much for the work and I'm looking forward to seeing the result.
DANIEL KARRENBERG: I'm not really sure whether lack of questions is efficiency.
BIJAL SANGHANI: I am actually surprised as well that there are very few questions.