RIPE Community Plenary.
RIPE 86
25 May 2023.
4 p.m.
MIRJAM KUHNE: We will wait another minute or so before we start. This is the Community Plenary. I think we have all the speakers.
I am the Chair of the RIPE community. And we have Niall O'Reilly the Vice‑Chair who is sitting here in the second row, keeping an eye on you who are participating remotely. I think we have quite a number of you on the Meetecho. And we have quite a full agenda.
Let me just show that. I send around a mail to the RIPE list before a few months before the meeting asking for input and we actually got input. So we have quite a full agenda at this time. I'll give a quick update at this time, and then we have an an announcement about the Rob Blokzijl award. Leo Vegoda has submitted a short talk about the future of discussion lists, mailing lists, and then we'll have Edward McNair who is here the executive director of the NANOG community to share information about what they are doing there, and Kim Davis will give an update what's happening in IANA and in respect to our community, then we have another offer for a talk by Henrik Kramselund. And then I'm not sure if we'll is a little bit of time at the end for open mic session.
I'll start with the RIPE Chair team update. What we have been doing lately as and as a community. First of all I am happy that we finalised the Code of Conduct process.
Thanks to you all to make this happen actually. We had the RIPE Code of Conduct already published for sometime. And the RIPE Code of Conduct Task Force was working on two documents since then together with the community and they have been approved and there was consensus, I declared consensus some time ago about the reporting process and the document that describes how to appoint the team and what the responsibilities of the Code of Conduct team are.
And then also, last but not least, we actually have a team in place and you have already seen, I introduced them earlier on in the week and you might have seen them during the week but I wanted to show you the pictures here.
This is the first time we have them in place. I was so happy to get these volunteers in place before the RIPE meeting. There are three community members and three RIPE NCC staff members at the moment, including also Maria who has a legal ‑‑ who is one of the legal advisers at the RIPE NCC so that's helpful. And also we have two members, Vesna and Saloumeh who used to be trusted contacts before so that's great to bring that experience in the team. They have all been trained and active during the week and you will hear from them also tomorrow to give a short report.
Now, having said that, I'm really happy to have the team in place, we are still looking for more volunteers. It says in the process also and we consciously decided to keep the volunteer process open so there is a form on the website that you can volunteer yourself or nominate your someone else. Have a look at that, there is a link at the bottom and use the opportunity maybe maybe to talk at that time Code of Conduct members who are here this week to get a feel what it takes. There will be training, there is a lot of documentation so you won't be just sort of left to your own devices.
That will be on a continuous evolvement, I hope you are going to build a big enough pool of volunteers so that it's not too much burden for each team member.
I would really like to give a big thanks to the RIPE Code of Conduct Task Force who made all this happen together with the community.
(Applause)
We can see them all here on the slides. I think most of them are here at the meeting. All of them are here at the meeting, and they have done a fantastic job in working with the community to get this through and to address your concerns and to document it and we finally have consensus on that. With that I would like to close the task force and, they will be happy to get on with other work I'm sure.
Thank you very much everyone.
Moving on, some other things we did admit meantime. I mentioned this before, the RIPE NomCom from last time who got Niall and me in an office. They included some recommendations in their final report after the NomCom process was over and we have been going through one by one. We have updated two RIPE documents related to the Chair process, that's talking about the RIPE Chair selection process, and the NomCom process itself. So there was some updates, and we updated those documents also with consensus in the community.
And there was one recommendation also to look at principles for remunerating the Chair and Vice‑Chair. Niall O'Reilly was mostly driving that process and it was discussed on the RIPE list and that's also now a consensus document and it's published as a RIPE document. And the last recommendation in there was a, it was recommended to document persistence for RIPE NCC staff to participate in the RIPE community.
Just to clarify that and to make sure RIPE staff also feels welcome in the community. That's still under discussion, so there is ‑‑ we are still gathering feedback on the RIPE mailing list, and there will be a new version probably after the RIPE meeting and then we'll take it from there.
So there is some great progress, and I'm really grateful for all the participation and I think we as a community, we have done quite a bit over the last few months to work on our procedural debt I think is what Niall calls it, that we have maybe gathered over the years and I think it's really good to have a solid governance structure and solid documentation, not too many roles but a solid foundation in place.
So, two other ongoing activities also related to our governance is the review of the election process for the NRO, for the numbering Council for the numbering resource organisation. We shared that document, it's also under review. We'll take the feedback that we have received so far and also publish a new version and we publish that on the list.
I am also working with the RIPE NCC together on a new training course, or webinar, we don't know yet, they are the experts, they have some great ideas on how to best get this into the community for a training for new but also for potential new Working Group Chairs. I mean, you know, we always a bit struggling to get new Chairs in place and Working Group Chairs, when the term is up, always trying to get new volunteers and it's sometimes hard, and I thought it would be good to actually provide some training so that people who are interested will know what they are getting themselves into, what it takes, what's expected from a Chair, maybe learn a bit about PDP, consensus building, how to Chair a group, so that will be in place hopefully before the RIPE ‑‑ between now and the next RIPE meeting or just before the next RIPE meeting. So I'm really happy that we're getting this going, this activity.
.
Lastly, just the regularly in the background, Niall and I are both doing a lot of outreach and attending events and I presented at another regional IEGF, Niall is going to attend the TNC, then ROM and academic networking event in Albania in a couple of weeks, we are reaching out to students before the meetings and I think a number of them actually attend the meetings here to basically get new participants in the RIPE meeting.
And representing the RIPE community also in other foray.
And that's pretty much all from me. Just to set the stage here. Are there any questions now about anything I said, any clarification?
.
If not, I'd like to hand over to the next speaker.
FALK VON BORNSTAEDT: Hello everybody. I am talking about the Award Committee 2023 of the Rob Blokzijl foundation. May I just ask a sign of hands who still resource having spoken to him? I am surprised. Quite a lot. So there is very good memory on him. You see the web page here where we show a little bit about the foundation.
So, basic facts. Rob Blokzijl resigned as the RIPE Chair, I think he was a founding RIPE Chair, in 2013 after serving for 25 years for the community. He died 1 December 2015, at the age of 72.
Exactly one year later, the RIPE NCC established the Rob Blokzijl foundation to honour Rob's legacy by recognising and rewarding individuals who make substantial contributions to the development of the Internet in the RIPE NCC Services region.
We consider this a lifetime achievement. This is important at least for Germans because it's tax free if it's a lifetime achievement and it doesn't honour something, a special work, so this is my contribution as an economist. So May 2013, the first contribution was given to Wilfried Woeber, and May 2022, the next award was given to Gert Döring and this was our very well received in the community. So now we still have money and to have an extra, another award coming up on the next RIPE meeting.
So how does the whole process go? We had a call for the Award Committee 2023 on the RIPE list and the website, and we got seven volunteers, that makes it much easier than last year where we had much more on the committee, and it was difficult to find dates with people in different time zones, it was quite a hazard to bring them together.
So, now I can tell you who is in the committee. So, we have Maria Hall. Julf Helsingius, Franziska Lichtblau. Desiree Miloshevic. Carsten Schiefner, Mike Silber and Jan Zorz. So thank you very much to those people for volunteering.
Now, I'm coming to the task. They have to issue a call for nominations which will not be so difficult because they can simply cut and paste the old one and make a little bit of additions so the text is ready.
The committee will then review all nominations and may seek to clarify information and talk to certain people and we should be ready early enough so the awardee can plan for receiving the prize. That's our agenda. This means most of the work will probably be done in September after the summer break.
As last word, there is still the possibility if one of you would like to give money to the foundation and we don't need to go to RIPE NCC begging for money, it would be very nice if one of you who may have some windfall profits or whatever, or wants to be generous or show his appreciation for Rob Blokzijl would donate something once in a while.
Thank you very much.
(Applause)
MIRJAM KUHNE: Any questions? Then next up is Leo who is already making his way to the stage.
LEO VEGODA: Hi. Discussion mailing lists are valuable to us, but I'm not convinced that they are valuable to the ever consolidating providers of e‑mail services. I use a number of discussion mailing lists provided by a number of different organisations. I increasingly see responses to messages that I never saw, hear about messages that never made it to the whole list, and I think this is not a me problem. I don't think this is a RIPE problem, even though I see it occasionally with RIPE, I think this is a more general problem.
I'm concerned that this kind of thing diminishes or openness and transparency. The issue is really that we're not valuable to the providers of the e‑mail services, or I don't think we are, and so they don't solve for us. The problem seems to me to have been getting worse over the last few years. I am concerned that there might come a time when discussion e‑mail lists aren't sustainable and I don't think that we should be caught by surprise.
So, this is essentially a request that people go away and have a think about this because we need to decide are we going to plan ahead and maybe do something else, introduce a different technology, and if so, how and when? And we also need to think about who the people are who are joining the community and what they expect and what they want to participate on, and are mailing lists the technology that we want to use or do they want to use something else? I don't know. But I think it's worth thinking about and having a chat about over the next few months.
That's it. Thank you very much.
(Applause)
AUDIENCE SPEAKER:
DANIEL KARRENBERG: I am a current employee of the RIPE NCC and I speak only for myself. Leo, I think you put your finger on a very sore point there. One of the things I just wanted to say is that e‑mail lists to me always had the nice property of being archived. We can still find, you know, messages from almost 20 years ago, or more than 20 years ago, and if we are replacing the mechanism, I think one of the important things is that the archives shouldn't be lost and we should use some decentralised system that doesn't have single point of failure.
The question is how do we actually organise this if we agree that this is a problem?
LEO VEGODA: I agree with you, that's why I'm asking the question. I don't think we necessarily need to reach a conclusion today. But that's why I put the words "Openness and trasparency" in there. I think they are absolutely vital. They are values that are important to this community and I agree with every word you said.
MIRJAM KUHNE: There is a question on, remote I think, Niall.
NIALL O'REILLY: Yeah. An online one.
CARSTEN SCHIEFNER: So, besides what Daniel just pointed out, the archiving function which Leo has responded to already, I also would like to see some kind of, if there is a replacement, and I would like to see sort of a synchronous replacement as when it comes to e‑mail lists I can read my e‑mail whenever I want, and if there is some kind of a discord or any kind of like a more chat functionality let me put it this way, I am a bit worried that a lot of information is simply lost because I'm not in a position to direct on that because it's out of the timeline already sort of. And then another part is I just wonder whether this community of operational people actually would really like to sort of carve in, this is a provocative question, to carve in when it comes to e‑mail services and the functionality of e‑mail services just as a food for thought. Thank you.
BRIAN NISBET: I, for one, sincerely look forward to the NCC discord or similar, I mean absolutely. The thing would I say is I don't think it's an either or in the same way we all run businesses now with SLAAC or Teams or whatever else and e‑mail. There may well be reasons for both. There may well be parts to play for both. The question I would ask is: Where are we going to be discussing in and thinking about this over the next few months?
LEO VEGODA: The RIPE list?
BRIAN NISBET: That's fine I just wanted to be clear. Thank you.
AUDIENCE SPEAKER: Martin Winter. We just had in the previous talk in our Working Group the discussion about the low activity on the mailing list and there came a lot of discussion of yes, they would rather do chat and a few other committees seem to be moving over to discord chat and have way more activity than a mailing list. Just as an input.
AUDIENCE SPEAKER: I agree with ‑‑ what the others said. That's not just what either/or. I think, like you can't replace the mailing list with realtime chat. That just doesn't work. I think it can be a good complement but I also don't think ‑‑ like we want to get new people in, I don't think having what's basically like the mailing list function but on a forum software instead of using mail is going to make a big difference. It's the kind of format that I think might be scary to so many people but I also don't think you can really get away from that when you are doing these discussions because it's like you have to do is asynchronously and that doesn't work on reclaimed chat. But at the same time I think we need both, and I think that that to some extent has been highlighted by how many people are in the unofficial RIPE community telegram. I think it's ‑‑ we have to have both. We can't replace mailing lists realistically because you'd run into a lot of issues, if we do that's not going to solve anything.
LEO VEGODA: I agree that this is a difficult problem to solve.
AUDIENCE SPEAKER: My name is Vesna from RIPE NCC. I would like to see something which is based on open standards, and possibly decentralised and federated and there is activity hub which has a lot of different applications already developed by different communities that are tang gentle and overlapping with ours so that could be one the possible solutions. Thank you.
AUDIENCE SPEAKER: I am a member of some other activity organisation that uses, so no IT, that uses mailing lists for communications, so we have basically talked about this problem before. And well I love mailing lists and a lot of those people also like it but what we have noticed for us the biggest problem was what's the speed of the communication? Because we have already had lists that were absolutely spammed with like 100, 200 e‑mails per day, and if we were going to ‑‑ if would go into a chat format people start writing in short form, and I think it's ‑‑ like, in e‑mail, I think the better thing is that people sit down, they think what they will write. They write a whole ‑‑ they try to condense all of their thoughts in a meaningful way not in a way that they will add five more e‑mails to say what they should say in one message or one e‑mail. So I think that e‑mails work really well for that. Yes, if we have maybe for the open source, as I said, I'm not on their mailing list to see how much traffic they have, maybe a chat format would increase the communication level that it's a live forum, I'm not sure if it's dead or alive anyway. But anyway I think this form is good for people to make their thoughts like that.
There is some technologies like discord, if I'm not mistaken, but that's basically another form of mailing list. So I think that mailing lists is the right way to go.
AUDIENCE SPEAKER: I'm Janus, a student from the university of applied sciences from Utrecht and I think discord is the way to go actually. Discord is useful because you can easily start a server and make different channels to discuss their topics and easily find every subject. And you can also hop on in a voice call or a video call and share your screen so people can actually help you. So I think that will be more valuable than a mailing list.
AUDIENCE SPEAKER: Dinesh. Thank you Leo. This is a much needed overdue proposal for a discussion. I have been thinking the same thing myself for many years now. In fact I try and avoid e‑mails as much as possible myself. I know that others don't. But one of the things is that we are here talking from the point of view of what we have been used to and what we are used to. What we should be thinking about is what the younger generations are used to and how they will work and how they will work in the future. So, we need to keep an open mind here. Yes I agree that mailing lists have this facility where you can archive the mails and you can go and search through them but communication has changed, you look at mailing lists and some of them the earlier person at the microphone mentioned 200 e‑mails in a mailing list and you just get lost. And it's a case of, you know, when it gets to that point I just switch off personally, and like it's so many e‑mails, so much spam. The discussion started with something and it goes on and on and on and if I had something to reply to in the first few e‑mails and I come and look at it later in the day, there is no point any more.
So, if you look at synchronous versus asynchronous, what difference is there in that way? Anyway I just want to say we need to keep an open mind here and we should communicate with the younger generations who are going to take this industry forward and see what the best way is for them to work.
AUDIENCE SPEAKER: Niall O'Reilly channelling some remote participants. There is a comment from RoV Michelles saying "just a thought. I think some would rather more differentiate and elaborated longer contributions via mail than via chat like that seem to ‑‑ but in the end it may be both."
.
And there was also something that came up in the chat from Petra Sidler who commented "one would think if any community were able to run their own mail servers this was this one."
.
I think that's not the point that's worrying Leo it's rather that man of the recipients are served by one of the big mail providers, all of whom are implementing their own filtering around so undermining the service that we might well be able to offer better than anybody. But since we're not delivering the last ‑‑ the last phase of the message to the recipient, we lose control there.
LEO VEGODA: Exactly.
MIRJAM KUHNE: Was that part of Petra's comment or was it you?
NIALL O'REILLY: No that was my answer to Petra's comment and Leo has confirmed it.
MIRJAM KUHNE: Do you want to add to that Leo?
LEO VEGODA: No, I am really quite surprised how many comments. I expected this to go to the list and maybe rumble on for a couple of weeks or months and then maybe come back next meeting. I am amazed.
MIRJAM KUHNE: I think you might have a bit of an action point on your hands there to continue this discussion and maybe pick up the comment. Is that what we want to continue?
LEO VEGODA: We'll see what we can do.
MIRJAM KUHNE: Okay. Thank you Leo. Thanks for bringing this up I think this is an important discussion.
Next up:
EDWARD McNAIR: Good afternoon everyone. I am Edward McNair with NANOG, I want to take a few minutes to go over a few things that we're working on within our organisation. We'll talking about NANOG outreach, our education mentorship programmes, diversity, equity, inclusion programmes within our organisation and a few community updates.
So, outreach is something that I'll be honest we're still trying to sort through and figure out what's the most efficient and most effective ways to be able to support, you know, people within across North America. One of the things that we did do, and this is about a year ago, I reached out to ICANN, ISOC we also had an iteration with ARIN and we meet each month having a quick Zoom call talking about what we're each doing and is there an opportunity for collaboration or synergy where we can pool our efforts together to benefit those within the community of North America. Part of that outreach came through our NANOG you programme with them supporting us in a collaboration between NANOG, ARIN and ICANN and ourselves. And that is to reach out to ‑‑ we were invited by the city of Montgomery, the office and there was a programme talking about one part the technology itself, policy related issues and also a career focus. Had a great reception, the students got a lot out of it. It was a way to see how we can support with those areas within our region.
CaribNOG 25, so recently there was a CaribNOG meeting in Granada, again in that kind of conjunction in supporting with those other OTA reach partners, it was focused on a technical forum that worked in conjunction with the Granadian government and the focus was education and exposing young people to technology. There was over 100 students in attendance. NANOG participated from giving a BGP tutorial and also an update of what's going on within NANOG and the resources that are available to people within our North America.
AlbuquerqueNOG 01. A new NOG formed itself in the city of Albuquerque in new Mexico. We came along with ARIN to support their initiative to help them get off the ground, providing additional talks and also content. And also while we were there we got a chance to meet with some of the tribal leaders within that community. There is over 21 different tribes within new Mexico. Give a chance to listen to them, what are the things that we can do potentially to support them. Some of you may not be aware but across North America there is about 19 million people who do not have adequate Internet access. We have a large country and there is parts of it which are very rural and remote, especially when you get into some of the tribal areas and so government fund something no making it possible for them to build out networks but the question comes in: What happens after these networks are built? So, starting a conversation to see are there ways that we can help train and educate them, get them involved within our community so they can see some of the resources and opportunities that are there.
Education and mentorship programmes. NANOG is an educational non‑profit and our focus hauls has been on education but it's really been kind of in a passive fashion with he we have speakers come up and talk to our audience, not linear training. We have formed and Education Committee based upon Board members and people within our community, volunteers, to kind of see where NANOG is going to go within education, kind of building those things out.
The first fruits of those efforts was a DNS course that has been going on in the last two meetings, which will also happen in our upcoming meeting in June. In the fall we're going to have a network automation course, and we have also work with eye cock to sliver an online course that's focused on giving those people first foray into network operations.
Mentorship. Dove tailing in connecting. We have developing a mentorship programme which is the focuses to connect mentors with mentees within our region to help them kind of get to the next level of where they want to get. These are driven by the mentee coming to the mentor saying what is what I'm looking for and then working and collaborating together so they can help them achieve their goals.
Diversity, equity and inclusion. So, about two, three years ago, NANOG went to the process of formulating a Code of Conduct. That Code of Conduct is there, it exists and it's something that kind of guides our organisation. However, having a document without having any kind of support for it, sometimes can become a hollow effort so that end, we set up a DE I committee which is focused on supporting those efforts and kind of structuring where it goes. The committee is newly formed. They are going to be having some events that take place within our upcoming meeting in Seattle. Tina is here, she is part of that programme, if you want to know more about that she will be happy to share the thing they are working on.
In addition to kind of DEI, we also set‑up an Omsbud for our meetings, we want to make sure that anyone who comes to a NANOG meeting fails safe, validated, they feel like they have got a place in there. We have added Omsbuds to NANOG meetings, they are there to support, they are there to listen, and they also continue to work with our community beyond our actual events.
Community updates.
So, you guys had this conversation about, you know, mailing lists and discourse and discord, it's interesting to see the echo of the same thing happening within our community as well. It is a big challenge. People are used to doing things a certain way and when you move that cheese they kind of freak out a little bit. The one thing I find really interesting is that mailing lists, when they came out, they were the latest technology, 20 years ago. So, we need to evolve, we need to grow with it as well. How we do that is correct that's the big question. But we can't stay static. Could you manage running the organise the exact same way you did 20 years ago it wouldn't be effective. That's my soap opera. We set up a discord server, it has one component to it which is a mirror of the mailing list. If you go to community.NANOG.org you'll be able to see it you can search, find things on the mailing list but in addition to that the big thing would you remember pushing it affinity groups. We're trying to find ways of engaging our community both during our conferences and outside of our conferences. We have a series of affinity groups that are available. We can create more as we move forward and it's a way that we can kind of connect our community beyond just our conferences. NANOG is very community driven. If you go to our website there is lots of articles about people in our community. A lot of faces that you see. We're driven by community and we continue to promote community within our organisation.
So, that being said. Thank you. And are there any questions for me?
.
(Applause)
If not, thanks.
MIRJAM KUHNE: That was quick. Maybe there are questions. I think that was really interesting. Thanks. When I saw the slides I thought you only have ten minutes and you rushed through this and now we have time for questions. So...
NIALL O'REILLY: A comment of my own. I seem to remember using mailing lists 40 years ago. I guess this emphasises your point.
EDWARD McNAIR: Yeah, thank you. We do need to evolve and change, and if we look around our community it's funny looking out here, it's similar to an NANOG event, we have definitely an ageing population...
STENOGRAPHER: Speak for yourself!
.
EDWARD McNAIR: I am plenty old, I have grandchildren. It's just one of those things where you have to prepare for the next generation coming in and they want to be enabled and involved in what we're doing. I want NANOG to continue and I want my grand kids to participate in NANOG and how can we make a way that they'll feel comfortable and connected with the organisation. All right. Thank you.
(Applause)
AUDIENCE SPEAKER: Sebastian, Chair of ‑‑ but I would like very much to appreciate your, the way you are taking care of diversity and I think ‑‑ and I was also very appreciative of the way you are taking in your and the RIPE community the diversity. It's something we need to learn in other parts of the Internet ecosystem and I guess ICANN needs to learn a little bit of what you are doing in your organisation. Thank you for that.
EDWARD McNAIR: Thank you. I appreciate the comment. Now can I go?
MIRJAM KUHNE: Thank you very much. Well next up is Kim.
KIM DAVIES: Hi everyone. I have a brief IANA update for you.
So just a reminder what is IANA. The Internet Assigned Numbers Authority. Essentially our bread and butter is maintaining authoritative records of unique identifiers. We maintain about 3,000 registries of different identified fire types. The vast majority of those most people wouldn't have heard of. They are very specific to particular protocols. Software implementers that use those protocols probably have an interest in those identifiers, but then they don't have the broad recognition or utilisation by end users, administrators, network operators, etc.
.
But we do operate some registries that have that character to them. I would say inarguably the most visible registry we operate is the DNS root zone. But for this community in particular, there is three that we service. It's the global IPv4 address space, the global IPv6 address space, and the autonomous system registry.
What do all these three have in common? We manage the global address space but we certainly don't administer direct allocation of these. Instead, we work with the five regional Internet registries to make regional based allocations from these pools.
So just so give a bit of a recap from where we're at with those three address spaces. IPv4, I don't need to tell you too much about what that is. But what I can tell you is that out of the /8s, the 256 /8s, 222 are for Unicast use which is means roughly 86 percent that have address space in turn is allocated for RIR usage for further delegation.
And we don't have any more of them. Now, initially when there was a large pool of address spaces we were allocating them on a /8 basis, the last /8s that we had available in our inventory were allocated back in 2011 and there's been really no further activity on the registry till that day. There was however a global policy implemented that allowed IANA to develop a recovered pool of small allocations that had been returned to IANA or otherwise not assigned. And according to that policy, we had essentially a lottery where we pulled numbers out of that pool and evenly divided those between the five RIRs and allocated them.
That system was in place for a few years starting in 2017, but now that space is effectively exhausted. All that IANA has now is three /24s, and with five RIRs, we can't really sub‑divide it any more than that so we're holding on to them for now.
Obviously IPv6 is a very different story. The vast majority of v6 address space is simply not allocated. It's reserved for future use. One eighth of the address space is allocated for Unicast. We now, according to the global policy, allocates/12s which means that we have 312/12s in that segment. Allocations right now is on the screen. All RIRs have been given one/12, a couple of RIRs have come back for a second/12, including RIPE NCC.
AS numbers is little more straightforward. Here we allocate blocks of AS numbers again to RIRs based on their need. When they are exhausting the supply that they have, they will come back to IANA for a further set of allocations. Those that have been around for a while will recall the distinction between 16‑bit and 32‑bit addresses there. All of the 16‑bit inventory that we have has been allocated so all new allocations are made from 32 inventory. This is not a common occurrence to allocate AS numbers either. The last allocation we made here was in 2021. And we have 98% of our inventory still available.
So when we do all this and if it's not otherwise clear, we don't actually make a lot of direct assignments to RIRs because we make allocations now with v6 and AS numbers that are sufficiently large that RIR has plenty of capacity to further allocate them without regularly coming back to IANA.
But when we do do this and when we do the other activities that we do within the IANA, it's very important to us to be accountable on how we do that and the ways we do this, I won't go into the datasets, but in summary, we have SLAs that govern everything that we do. We report on those monthly we do post trust anchors surveys every time someone interacts with the IANA. We ask them to fill out a survey and we use that as a way of getting sentiment of our customers as to how we're doing. We augment that with annual surveys and there is also a series of annual community reviews including in the number resource community. Every year a Review Committee is established, that reviews that performance for the previous year.
All the information is on other website, iana.org/performance on that one.
So I'm going to switch gears now and talk about the DNS trust anchor. Another one of our key areas of responsibility is maintaining the root zone key signing key. The key signing key is at the top of the trust hierarchy for the DNS. And the way that it is disseminated means that it's very critical that it is managed in a safe and responsible way, because changing it is very complicated which I'll get to in a moment.
One thing that we do perhaps differently to most other entities that administer keys of this nature, is that we operate it in a highly transparent manner. We really thought about this system when it was devised some 12, 13 years ago, how to do it in a way that's consistent the multistakeholder model works, the way the ICANN community, the network operating community works here as well and the system that was devised is designed to give anyone in the community great insight into the how it's operated and to be able to follow along and come to their own conclusions about how it's being operated as opposed to us simply saying trust us, but doing it sort in a closed room fashion, perhaps you can have auditors there but generally speaking the model we have taken is very different.
And one of the key things about the model that we devised is it's not staff members of IANA or staff members of ICANN that are responsible for all facets of how we protect the key. Key areas of responsibility are actually delegated to community members. We call these community members trusted community representatives and their participation is really key to how we perform operations.
Many of these trusted community representatives or TCRs, comes to events where we retrieve the key from its secure enclosure and use t these are called key sign ceremonies, they take home credentialss with them and they keep them safe between ceremonies, and in order for us to conduct a ceremony, we require a number of these community volunteers to come together to have a quorum to enable us to utilise the device that we store the keys in.
Another thing that TCRs do is promote trust in the model. By having well recognised established people in the community come and vouch for how the ceremonies are conducted, we believe instills trust in the model. You don't have to take our word for how ceremonies are conducted and how we protect the key, because there is a number of community members that are watching it very closely and can tell you about that.
So we're now twelve years into DNSSEC in the route zone and operation of this KSK. Many the volunteers we have today have been in place since day one. At the ten year mark we concluded that it was really time to emphasise a renewal process to bring new blood into the process to get new expertise and new insight from new TCRs.
Covid put paid to that so we sort of suspended that project for a little while but know that we fully resumed operations we are very keen to do that. So this is a call for volunteers for either yourself if you feel that you might be qualified for this, or if you have contacts that might be suitable for this. We have an open call for volunteers, iana.org/TCR. When we pick them we are really striving for diversity, and that's diversity in multiple different facets. Firstly, diversity of skillsets. Having a set of TCRs that all have roughly the same expertise is not going to give us diverse oversight. So bringing people from different areas of expertise perhaps related industries, different experience is very helpful. And we also strife for gender diversity and also geographical diversity. Part of the benefit of this model as well is that the credentials are split around the world when we're not using them so that there is less risk that they all come together when it's unintended.
So I mentioned before that the KSK has to have special protections and be managed in a very deliberative way and the reason for this is that when we change the KSK it is an extremely involved event. Anyone that runs a validating resolver needs to know about the new KSK before it's put into operation. And that is requires a lot of coordination outreach and awareness building before we do that. In fact we have only done it once. The first KSK was generated in 2010. It was replaced in an event called a rollover in 2018. We're now ready to do it again, and in fact we just did the first step of a multistep process to create and roll to a third KSK.
So just last month we had a regular key signing ceremony and we generated a new key peer for the root zone. We don't consider it production ready yet. The way at administer is we have two duplicate facilities that are very far apart from one another. At the next ceremony, which is scheduled to happen in July, we will replicate that key that we have generated into the new facility. And once that's done, we're now comfortable that the key, now that it's replicated, is secure. That will then start a two year propagation period. So over the span of two years we will publish the trust anchor set with a new key added to it. We will reach out to software vendors, we will reach out to network operators and conduct campaigns of awareness building around this.
This will all culminate, if it goes to plan, on October 11th, 2025, that should be the day that we roll to the next KSK. And then we'll do it all over again. So, another thing we're trying to do is build a regular cadence in place and our target is every three years. So, the way we have designed it is that we essentially start a rollover process and as soon as it's concluded we begin the next one.
So that is our plan and we have started the first phases of that now.
So with that, thank you very much. And I'm happy to answer any questions.
(Applause)
NIALL O'REILLY: I'll jump in channelling Carsten Schiefner who says no affiliation.
First off he says "thanks Kim that was a helpful summary." Then he asks: "What about the pressure on IANA or PTI to eventually release the reserve v4 space, is it stable background noise or is it getting louder? And if the latter, who are the loudest and/or newest shouters?"
KIM DAVIES: So, perhaps surprisingly I am in a relatively comfortable position because I basically don't get to decide anything for myself. I need a global policy or some instruction from the community to let me do anything. So in this instance, theres no global policy that is pushing for that allocation, no one is certainly pressuring me, so I will wait to see how that discussion evolves into the future.
AUDIENCE SPEAKER: I have approximately the same question. Do you have any discussions, active discussions about this space, because this allocation in RFC it's almost for 40 years the space in the reserve and the reserve like for future use, so the main question to IANA, when the future comes.
KIM DAVIES: It's basically the same answer. I follow along with interest and when I see the discussions in the community, in the IETF and other forums, it's interesting to me, but I don't know of any proposal that, you know, has any stamina that would suggest that there'll be a request to IANA any time soon to assign that address space.
AUDIENCE SPEAKER: By the way, do members from any other areas try to communicate to IANA about this? Do you know any information about this?
KIM DAVIES: Not to my knowledge.
MIRJAM KUHNE: Thank you. If there are no more questions, comments... thanks, Kim.
(Applause)
MIRJAM KUHNE: And I was just remind you that have five more minutes to vote for the PC candidates that have nominated themselves earlier in the week. I am sorry I forgot to mention that earlier on. You can still go to the RIPE 86 website, go to PC election, I think it's select the first button on the page pretty much, and vote. I hope you'll still make it if you haven't done so.
Next speaker up:
HENRIK KRAMSELUND: Thank you. Hello everyone. My pronouns are he/him and I have this very short presentation and I don't think there will be a lot of questions and I will be available afterwards and I think time is getting short.
So, let's get started. It's on GitHub so you can find it afterwards and it's also uploaded in the presentation.
So, the problem statement and the thing I want to talk to you about the question that I guess a lot of you already have had from your co‑workers and friends, you work in networking so can you give me some pointers for my friend that wants to get into networking? And it's by design that I'm calling this newcomers introduction and newcomers into networking, because some people may be grownups that want to switch their careers into another job, some may be young people, teenagers, kids even and they want to get into networking and I think it seemed to be a recurring theme in this RIPE meeting that we want to attract more newcomers to our networking communities. We are missing a lot of hands, we are missing a lot people in networking and cybersecurity, network security which I'm working in so we need more people.
And the thing is that I would like for you to send me information about resources that you know about. I am just a single person and you all work in networking so send me your best bets on what you would give to that, as an answer to that question.
So, I would like to also like first of all very, very quickly say thank you to all of you for writing RFCs and text books and writing materials in the old days even, 20 years ago, ten years ago, 15 years ago, they have all been great and I have been very happy with them. But I'm pretty sure that my son would not have been very happy by reading an RFC, and searching on the Internet I unfortunately see people saying if the question comes up, how to learn networking. Go read the RFCs. How many in here agree that a newcomer, beginner in networking should read the RFCs? Put your hands up. Some people actually think that's a good idea. I don't think it's very engaging and one of the things that I think people are missing is that you need some core concepts before you can read the RFCs. Because they also have a lot of key words at the beginning of every RFC they say that they are using these specific key words. Before you can read that RFC you have to go back to the other RFC. That's what I would like us to get started thinking about what resources can we use or create and collect so we can get people to the goal they want to have finish the mission.
So, as an example, I am going to use the Nmap port scanner which I think can be used for both good and bad and I think it's a fantastic tool and it's something that can automate a lot of people for network people, security people. But if you want to learn port scanning and Nmap and if you try to rewind that you need a lot of prerequisite knowledge. You need to know but TCP IP, you need to know what is an IP address, what is a the port, what is a service, what is HTTP? HTTP today is not only HTTP it's HTTPS with TLS and you also have the demand and you also have the DHCP before it can give an IP and you need to know what a router is and you need to be able to know what you are on the Internet and everything.
So, a newcomer in my sense should be given some tools that will allow them to progress until they reach their goal. And it's the first steps that may be very, very hard, and if we hand them an 800 page book or an RFC document which is too hard to read for a beginner, a newcomer, then they will not progress and they will find another way to work in another area of business.
So, it's very, very difficult to get started because running a port scanner will also require you to have an operating system. What is an operating system? And my myself I have many times said people should try out Kali Linux, that can be run today as a VM which is the easiest way so you don't have to find all the hardware support in Linux, which is still a problem, so I tell people to create a virtual machine with Linux, but wait what is a virtual machine? And so it all keeps ongoing back to a lot of core knowledge that is very hard because we have worked in the industry for so many years. When I do trainings I tell people to go into user. But actually I mean go to the directory call USR, it's not USER, it's USR. It's stuff like that is a thing we need to learn and I have a core, small set of what I call recommended network technologies to learn, and I am not sure ‑‑ I am pretty sure we cannot agree on what should be the smallest set of technologies to learn first. But this is my opinion about what to learn first.
And by the sign I am leaving out datasets, I am leaving out wi‑fi, I'm leaving out IPv6. Why am I leaving out IPv6? Because I did my thesis in 2002 about IPv6. I love IPv6. But I think it's okay sometimes to give people resources and options which are not the full story. The full picture but a lot them to actually understand parts of it and complete something because when you complete something, you gain that sufficiency I think it's called in English, perhaps that you can do this now, you can do a ping with IPv4 and that's fantastic. Then later on when you get more interested and you have learned the core concepts you can probably learn very quickly to ping with IPv6 and then start learning IPv6 and hopefully implement it some day.
I would like the beginners to go all the way from say ARP to TLS because that's from a low level to a very high level and going from the low level will allow them to get started and running networks and running their systems basically on their own. And the TLS part will allow them to go all the way to say a web service and running a web service today is so complicated and you might not think it because you have had 10 or 15 or 20 years to adapt to TLS and all the funny stuff. Don't get me started on e‑mail which is another layer cake which is huge. I really hate e‑mail and I run my own mail server.
What I'm looking for and what I would very much like for you to help me get is some resources like this one which is one of my favourite tools on the Internet, Julia Evans has done this very, very nice drawings which are not the full picture but they are enough to get you started on and they are more engaging, they are more easy to understand, they are more friendly, and newcomer friendly. So stuff like this is what I'm looking for.
I'm looking for tools that help you present real world data, because Nmap is presenting real world data. This is a take from SE which is something I saw in the DNS hackathon is a fantastic example of an old school tool which is very, very difficult to understand, it's something that, sorry to say it sounds like I'm hating all technology, but I hail dynamic and I'm using host myself because I find is too difficult to work with dig, it's too much text on the page. This is highlight the tools you need to know and that's a fantastic example of more easy to learn way of network tools.
Also this book how the Internet really works, an illustrated guide is one that ‑‑ it's becoming one of my favourites. It's something that I can hand out to people of all ages actually, in Denmark we are pretty good at English so it might not be working with a kid ten years old but certainly a teenager would be able to read this and get interested and more engaged hopefully.
So, real books are also fine. I am very, very happy to read books, recommend books, but they should perhaps be shorter. They should perhaps be more practically orientated to keep people engaged. I learn from coma and Stephens and all the other books that you may recognise, and going from all the protocols from A to C may not be something that keeps people engaging into learning this. Where this is like a 3 of 0 page book, it has at least a screenshot on every page, so it's more easy to get started, working with this. It also introduces these core protocols that I had on my earlier slide. So this is, for me, an example of something that I could give to people.
And I know there is a couple of resources, some of them were actually proposed in the chat today, I forgot who proposed it but I know Brian Carpenter is having a project on GitHub and there was something from the UK task force I think, so there are some books being written now which are more updated and which also include IPv6 by the way, so they have those chapters too which is also essential for real world applications.
So, my e‑mail address is here. And I also created a repositories called learning paths on my GitHubs and you are very welcome to send me some pool requests. I have started to add a few resources but I don't want to put in too much already because I am hoping that you will have some engaging, fun, learning and nice materials, and you are welcome to propose your own materials, and let's collect these and make sure that we can get some newcomers into networking.
(Applause)
MIRJAM KUHNE: Thank you. You don't have to rush. We actually have plenty of time. So be prepared for a lot of comment and input and questions. I think Franziska you were first.
FRANZISKA LICHTBLAU: Recovering research assistant. So, I have worked with university students, especially networking an a lot. I would like to point out two things. First great thing you are doing, we need entry level material, we need to point people in directions. That's all fair and scare. But we also need to talk about how we talk about the material to newcomers. And especially this is a call to the people in this room and everyone listening to this. Remember that we are all working with these technologies for ages. We all know about the intricate design choices in the UNIX system hierarchy. In IPv6, whatever, and if we explain this for 20, 30, 50 times, we get tired and we may say oh God, another person who doesn't know that. Remember, those are new people. It's not their job to know everything yet and try to explain it.
The other thing. Dig does have a short option, so you may want to look into that.
And RFCs. Yes, I absolutely think new people should read RFCs. And I sat down with a lot of university students and actually explained them how to read an RFC. Because that is a skill as well. You can ‑‑ they follow a structure, and they are the next best thing to a source of truth. You can debate implementation quality, but that's a different story, it's the best next best thing to a sort of truth that we have where we can show to people this is how it works. It is not a method discussion. This is how it works. You can have a look how shit looks on the wire and that is important for them to know as well.
HENRIK KRAMSELUND: Thank you very much. Maybe I issued clarify I do not disregard RFCs as information and I am a teacher in Copenhagen at the Copenhagen school for design and technology, which is just under university level, and in my courses I make sure to present both ASCII drawings and RFCs and hack assigns and PDFs and papers and academic two column papers because I think all of these different resources and the ones that we have available are fantastic, but they are not the first things to present people for.
AUDIENCE SPEAKER: Hello, Edward McNair, NANOG. What I'm going to say is my thoughts and others just kind of our organisation as a whole.
I think the trouble is we have done a really, really good job of creating networks that function very well. And I think a lot of young people just take it for granted that it just existed, it works. And I know in North America there are not a lot of colleges with programmes in network operations or network engineering. The big thing is to be a programmer. And, you know, in talking to kind of addressing your point about getting, encourage people, NANOG hopefully should become at some point if you are in a college or computer science, I would love to be a household word. If you want to know about networking you have got to go talk to the people at NANOG. It's difficult to do. We're a small organisation. How do you get yourselves out there? How do you make sure that information is available? So, my question to you: How do you see organises like RIPE or NANOG or other small NOGs getting that message out there, marketing yes, the start is to create these simplistic way to get hooked but how do we reach out and communicate that to others?
HENRIK KRAMSELUND: I'm not sure I'm the right one to answer the question but I can give my point of view of course. I am also a part of a DKNOG and very happy to go there every year and it's not a very active NOG and not has a lot of members like NANOG and I think the RIPE website and the RIPE technology blogs and all of these things, academy, I took the certification here, that's very fantastic and I think it is needed to have these areas of business or these different ‑‑ we are the network community, so we should have that task. And then the basic schools and learning and education should prepare people for whatever way and path they select in their life. And I think we have a responsibility as a community to have into ways and describe it and have the IPv6 act now when this came that website if some of you remember that. I'm not sure if I responded or answered ‑‑
EDWARD McNAIR: I have one more kind of tie into that. Because programming is the big attraction for students going into computer sciences, do you think that there is a potential to create a kind of course work that if you are a programmer, this is how you would get into network engineering, what you would need to know to shift your skills. Everything is moving towards automation so we need nor software developers in the networking side of things.
HENRIK KRAMSELUND: Yeah. I am thinking also of a term that we are using at least in Denmark around me, and some of my friends we're calling it hidden curriculum I think it would translate into. But we have some kind of hidden curriculum all around, and we assume that people will know how to identify, so we will ask people to you should edit this BGP decould have, but what is an editor so you have to select an editor and an operating system and a lot of other things to be able to actually function in these areas. So what was the question again?
EDWARD McNAIR: The question was, is creating a course that would allow someone who had studied, you know, programming software to shift those skills over into network operation.
HENRIK KRAMSELUND: I think that's definitely something we should do more. It seems that since people are getting older, they are switching careers in their life, so people in the old days only had one career in one job for all of their life, basically, and today we are switching more and I'm hoping to switch academics in Denmark over to a cybersecurity because they know how to write long documents and a lot of security people don't know how to write long documents. So, definitely. But I think we should maybe I'm not sure what is the best way, but having some kind of checklist, some core concepts. If I can go back to this one. This is my list of things that I think you should look at if you want to come into networking, and if you are a programmer and you already know all of these, great. Then you are probably ready to read some of the more advanced stuff and go directly into the developing networking code and looking at the open source projects within networking. So, my idea would probably be to have some kind of checklist that people can refer to.
EDWARD McNAIR: Thank you very much.
MIRJAM KUHNE: Let's go to Florence first.
AUDIENCE SPEAKER: Florence. No particular affiliation. Thanks a lot for your presentation, it's really cool to see that some people matter about this kind of stuff so thank you. A question for you: A lot of people in the community will appreciate I'm sure this initiative. But there are also receptive to documentation which is not in English but in their mother tongue. Would you be open to also collect information which is not in English?
HENRIK KRAMSELUND: Definitely, definitely. I think that would be so cool. And it was a problem I was struggling with because people are giving me resources when I haven't been pushing this a bit when I have been talking to people and people have already started to push in Swedish and I will probably be able to find something in Danish. So definitely yes, I think we should have that in different mother tongues because especially young people will not, in Denmark, be able to read technical English. So, you know, that, yes.
AUDIENCE SPEAKER: Thank you, I will be happy to give you a hand there.
MIRJAM KUHNE: Great. Thanks. That was a good question to clarify.
AUDIENCE SPEAKER: So, I don't have a question per se I just want to share my experience, as a part of my jobs I do on boarding of new employees. So also like students who just finished university or developers who are trying to get into networking. So I tried a different approaches over the years. One of them being giving them documentation and giving them materials from the web to discuss. And actually I found it is not the best way so I'm actually talking from a perspective of person who wants to get new employees into the field as soon as possible.
And basically what I found for the best approach to be hands‑on giving people more complex tasks to do. Then the primary tools I give them is GNS 3 and Wireshark, I tell them connect to PCs try pinging them. And just if they want to read the documentation, I actually never give them RFCs to read, never, I don't want them to leave the company.
And basically just point, maybe some resources and then leave them to do the problem solving. I tell them okay you need to connect to three routers, you need to first put stick routers there, then use OSPF and then in the end use MPLS. Every time I tell them look at Wireshark or look at the packets are going. I find this kind of a network something really good for keeping people interested because one the first employees I on boarded them I gave them documentation to read for two months and in the end they left the company because it wasn't really interesting. So, just my two cents.
HENRIK KRAMSELUND: And great two cents. People are learning in different ways. People have found out that some are more visually and some are more audio. I cannot read ‑‑ listen to body casts, I will zone out and not hear anything, but slides work very well for me and hands‑on is required for some people. They actually have to do to be able to remember. So, good comment.
NIALL O'REILLY: I have two remotes from different people. The first one is from Carsten Schiefner who suggests "how about copying the summer school on Internet governance concept to kick start the interest of future network people".
.
The second one is Marco Silinger of the university of applied sciences upper Austria. He says "Henrik, thank you. Great stuff. We shall experimenting with teaching networking in an OZ top‑down approach. The idea is starting with the applications with the stuff people know so the stuff is more accessible and then show them top‑down how this is all implemented in the network."
HENRIK KRAMSELUND: Sounds really good, and probably something people already have, already use, most people have a browser so if it works in a browser, people will feel more confident doing it. I really agree that Wireshark is a fantastic tool and it's one the tools we are using so definitely people should get exposed to Wireshark quickly but with some sort of guide.
AUDIENCE SPEAKER: Robert Kisteleki speaking as an individual with the programming background got into networking relatively early on. I'm looking at your slide and I'm really wondering about the shelf life of some of these things. Just like nowadays it's not reasonable fashionable to talk about class pool addressing. Maybe HTTP will disappear, or should disappear from this slide because QUIC is replacing it. And the much related increasing complexity, I started programming CGIs 27 years ago and I could do it in C because I understood the protocol fully. I don't think there are too many people in this world who understand QUIC alone. I don't know where this is going but it's going to get harder and harder to get people into protocol space purely because the complexity is such that humans will not be able to comprehend in very few short amount of years. And I don't want to throw in the word AI, maybe that's going to save us or doom us, I don't know yet. I have no conclusion, but just stuff on my mind.
HENRIK KRAMSELUND: But I agree that the complexity has risen, so wonderfuley, like I mentioned mail before and when I started doing mail we didn't have SMTP LS we didn't, but now we have, now we have demark and it is PF and deacon and ARC and all the things that have been added over the years.
ROBERT KISTELEKI: And we didn't even mention the DNS camel.
AUDIENCE SPEAKER: I think the problem is more basic. So my first introduction to computer networks did you think what it was, it was called OC something, and to understand the problems there and why the OC model existed it took actually multiple years,
And what I realised is actually it doesn't make sense, that's how I understand it because it was designed from 20, 30 years ago, and it had multiple problems that needed to be solved and that's how it was designed. It actually doesn't make sense if you think about it. So when you introduce new people to it and new people are what is this and why does it exist? And that's when you start losing the new people because there is a lot of theory and people who are like are why. That's my comment.
HENRIK KRAMSELUND: Yeah, I can comment on that. Because one of my favourite pictures which I'm using a lot in my slides is a comparison where I have the OC model on one side and the Internet model on the other side and they are not completely parallel because they are not the same and OC was before Internet. Internet protocols came in '83 or so. But I think the idea which I tried to transfer with the OC model is that we cut up large complex problems into smaller bits, and it's the same with programming. We cannot do a huge applications so we will do small functions and we will split a function if it's more than say one page, two pages, so I think the OC model should not be taught and we should definitely not spend our brain capacity remembering all the layers. But as a concept I really like it and I like the Internet model better of course.
AUDIENCE SPEAKER: Still when you showed both now you confuse people more because they are like... they are more confused about why this, why that?
HENRIK KRAMSELUND: If you try to teach IPv6 at the same time you get even more confused because ARP is in one ways and isn't v6 is in the another place in the model so that it confuses people even more.
AUDIENCE SPEAKER: The problem is more basis to attract new people from new generation.
HENRIK KRAMSELUND: I'm looking for when I get new materials to try it out on people who have no networking experience because I am biased a lot.
MIRJAM KUHNE: I think we have the last two comments. We have a few more minutes for any other business. We're starting to run out of time.
AUDIENCE SPEAKER: Hi. I'm still a student, I am 26 years old. Just so you know where I am and when I was dealing with these things I did my bachelors' university of Ljubljana, I did an exchange in Berlin and right now I am a masters student. Well, thank you for bringing this up because I think it's important to talk how to approach it. Although there are parts that I don't agree. I am not really on the side of simplification of topics, and now when I compare a lot of courses that I have taken and I am still taking, I actually like most of those that had books. That had long books that you could look into that you could before the exam, study, go through it, find a topics, find the chapters and we were learning networking by the book of, that were written by Kudeson and Ross I think they are one the most known books and I would still want to teach in that way.
On the other hand, one thing that wasn't mentioned right now and one the basic questions to figure out if a student understand or anyone if they are understand networking I think is the layered approach. Like, I know so many people that makes a Layer2 and Layer3, and I think it's good to know like if you really want to get into Internet to understand it by heart, to really tell people look, this is here, this is there, this is how to compare which parts work where and there are parts that even we here don't understand like a lot of the technologies and on layer 1, about you maybe if you know where it should be and what it should be its function, we can even like those gaps in knowledge can be fine.
About RFCs. RFCs are technical documentation. I think they are great the way they are for the need that they are there. Like if you really want to look into tech will documentation, to see how exactly it works, all the cases, I think that RFC, even yes, they are ugly, but they work fine and you don't need somebody, if you are reading it, you already have interest to read it, which is enough to get through maybe the ugly formatting which is not that big of a problem then.
But maybe to go to another part. Why I got, why has networking attracted me the most out of all the topics that I could learn at university? And that was it was really hands‑on. A lot of other things are getting more and more abstracted. Also networking, but the fact that you had a networking I could open things in Wireshark, maybe see how SMPT works, it looks like a chat, but it's really easy to get the students interested when they see how things work, for which in many other parts of computer science engineering you can't, even compilers, how things are compiled are hard to understand. And while in networking you can see exactly what happens. So I think more practical examples in that way and I prefer books, I might not be the best example, I know it was one of the only students going to the library to take the books out, but here I am.
So, that would be all for my part.
MIRJAM KUHNE: Thanks for the input.
HENRIK KRAMSELUND: Thanks very much. You touched upon a very interesting subject of what interested you in networking, because that's a whole other how do we get more people have had in this world of networking. But that's another research project and Ph.D. I guess.
AUDIENCE SPEAKER: Maybe we should get young people that are still studying or like I don't know, I am 26, there are a few people that are younger than me and others and ask them what is the thing that got them hooked?
MIRJAM KUHNE: I was going to say because we are running out of time. We're having another session today after this in the break where we are actually going to focus a bit more on the topic. We are focusing on diversity in tech, next generation, youth, what can we do and this fits into right it. So we'll have some opportunity to pick this up. Also from Edward will be there to talk a little bit more about what they are doing in terms of outreach to universities and to students and there will be some other communities presenting what they are doing to get next generation in. So, if you are interested in this topic, please also come to the next session.
The very last quick comment. Dmitry.
AUDIENCE SPEAKER: I didn't plan to give a long one. I'll give one pedantic comment but please don't use ARP twice and network discovery zero time. I think we should not add IPv6 layer it is not that hard, it is not that hard on parallels, but seriously my now ten year old son and used to be search year old son he managed to install Linux on the VM himself and he managed to mount the device under the VM so he can actually change the partition type, you know, to appropriate so he can do some interesting stuff and mount it on the host so he can get format and bootable Mac OS device. He writes his own pathing code. I'm not converting him to a network engineer but I am going to guide him how to. Hands‑on. RFCs are okay, maybe we need a quick summary of RFCs, like people summarise the books, and yes the summer camp is a great idea. I think we should not be afraid of the people not being able to read RFCs but maybe make them more accessible and more hands‑on and use IPv6 first please. Seriously.
MIRJAM KUHNE: Well that was a great discussion. Thanks for bringing up that topic Henrik. Thanks again.
(Applause)
.
We have a minute and a half left if there is any other topics you want to bring up during this Community Plenary session, or otherwise, there will of course an opportunity next time. This was a mix of topics. I was grateful to get some input on the list when I sent out a call so I'll do that again and hopefully we will have community discussion on the list and also at the next session and please stick around for the diversity session after this, after the break. Thanks to all the other speakers too.
(Applause)
LIVE CAPTIONING BY
MARY McKEON, RMR, CRR, CBC
DUBLIN, IRELAND.
