Daily Archives


Connect Working Group

Thursday 25 November 2021

2:30 (UTC +1)

REMCO VAN MOOK: All right. The computer says you can hear me now so I am going to rely on that. I see Will is here as well and here is Florence.

FLORENCE LAVROFF: Hello, everyone.

REMCO VAN MOOK: Did you have a nice break?

REMCO VAN MOOK: So we are about to get started with the Connect Working Group, give it a couple of minutes, you can all go and find your seats, or not. We will get started in a minute. Lets get this started. Good afternoon, everyone, welcome to everyone's second most favourite Working Group of the RIPE meeting, the Connect Working Group, coming to you live from somewhere on the Internet. What can I say, we mapped to find the one time slot in the day where, in winter, I get 15 minutes of sunlight into my face so if you think, why the hell is he leaning in a corner of his webcam that is why that is and by the end of the session we will be done. Here we go. So, basic rules, there is a chat window, it's being logged so proceed with caution. If you have any questions about the presentations, put them in the Q&A. There is a poll option that I don't think we are going to be using for anything. This room has one emergency exit which is on the top right of your browser window. We have a stenographer ‑‑ yes, he with do, I forgot who that is right now, but I am sure, we will have careful minutes kept. In case you have a hard time understanding anyone there is also the stenography track which I highly recommend people to read or keep track of, you can open that in a parallel window.

And with that, let's get started, shall we? Now I need to go to my second slide.

Here we go. So, the agenda: Opening, I think that's what we are doing right now. Agenda, if you have any additional points for the agenda you are a little bit late so I propose that we keep the agenda as set. Session format for this time is, it's going to be all publicly accessible, all ‑‑ it's all online so there is no real other options. I trust everyone has had a close look at the minutes from the previous Working Group session, and have dealt with them with approval so unless someone has anything really important to say about the minutes from last time within the next five seconds I am going to call those approved.

And I think that is it for opening and housekeeping. My fellow Chairs, have I forgotten anything?

WILL VAN GULIK: I think you are quite good.

REMCO VAN MOOK: Well, let me introduce my fellow Chairs. I am the co‑chair of connect, together with the lovely Will and the lovely Florence, who are all here and happy to guide you through the next hour. So, we are going to go to the first presentation, which is the Gaia‑X update, which unfortunately didn't make it to the list last time, but Pierre is here to talk to Gaia‑X, so let's find him again. I will stop sharing my slides and so you can get that started. Pierre, the floor is all yours.

PIERRE GRONLIER: Thank you very much, I am very happy to be here. So you have the ‑‑ this is intended to work through the ‑‑ I am going to take some assumption about what we do in terms of data space and I will not deep dive into the complexity of migrating and making data ‑‑ creating value, extracting value from data. One quick remainder, Gaia‑X it's about European values and digital southernty, there is so many interpretations we will talk about, control. We will quickly view how to create those data space by connecting into interoperable system and I will prefer to dive into the next slides. For having those ecosystem and to create those bigger data spaces where we can create data chain or composition depending on which we have, we have those service composition, this ability to plug and play services one with another, so the idea is that you have this pipeline and that all companies, all organisations knows how to do right now, that could be like standard extract load transform or MLX pipeline but having that in one organisation you have that in 12 different organisations, because we have 12 I CONS here.

There is a lot of papers that describe what we don't do. I would like to start with coming from a paper, NIST paper. Two organisations A and B that would like to talk to each other, what they have to do is to agree on the same trust model, what is the identity, who issues the certificate, what are the rules of engagement to on board the others. Once you have this trust plane you can figure out what types of rules and policies and what kind of services you need, most likely if you want to find the others you need to have a portal and a catalogue or a broker, etc. That most of the management thing.

One important rule that is also described in most of the older organisation architecture came across, that's the role of the fed rater and that is here to act as intermediaries in the usage plane. All this stack that I have described are mentioned in the NIST 500332 paper and to stand better what we within a to do with Gaia‑X I will invite the readers ‑‑ viewers to go and check chapter 1 and 2. And chapter 2, that's 100 percent what we want to do. Chapter 3 we kind of start diverging a bit. We talk about data spaces and being able to extract data out of those stacks, eco systems, federations, that is if we zoom in what I have mentioned before, this ability to create data chain and I am going to give a concrete example that was revelation, when you are doing exhaust pipe in a lot of exhaust pipe you have to extract element from the ground, exhaust pipe is probably ‑‑ one of the most expensive parts because that's where you capture all the particulars that are responsible for pollutions, and you need to have minerals there, element, and they are very expensive. Data chain that will be the ability for the interior lifecycle, the integration in the, the lifecycle of the car, maybe the repair, up to the recycling part of the car, that takes years, it can take years, to know for the recycling company how much there is in this exhaust pipe to be recycled and it is sold back to the manufacturer. So this will be the illustration of a data chain being able to have different factors that have different regulation market, different countries and to change the information data and to create value chain at the end to have more ‑‑ create more business. Another example that we have in this model we saw another way to illustrate that, you have an edge device with a phone, you are the data owner, you emit some data and want to do some behaviour about the users and you will do joiners SQL query, join in a place on the Cloud taking some additional information, maybe information from the website, then you might do the cleaning, the data trust ability, the data implementation, I skip that part. You will train that model and you might want later to store that to ‑‑ for enabling that, not one organisation as we do it now but across different owners and rights, imagine that you have boundary which in each of those six icons on the top right, and for each of those boundary that you cross you need to make sure you have data constant, the ability to deploy your services on the infrastructure and the interconnection between infrastructures element. That for the technical part.

There is also an entire part of the we sometimes underestimate that the legal part, the I don't see the contracting, and everything that goes with the ‑‑ provision, configuration. By the way, in a lot of exchanges we speak about portability and operability, interoperability of data, data is not inter operable, you can have date portability, and interoperability that will be between two elements, so portability is one, portability of your licence, portability of your identity and then you have interoperability between services. So the ‑‑ we have those three planes and the dilemma is to figure out how much we define and force or regulate to enable those different stacks that exists already to trust each other and to enable data exchange between themselves. So we have different type of rules and specification, Gaia‑X, we have to ‑‑ to make difference between Gaia‑X and others there, we are not that much interested by defining how an ecosystem has been built, we are more interested how to make those interoperability and trustable with each other to make them bigger so we have to enforce and to specify which is the trust model between ecosystems, how do you issue ‑‑ how do you grant rights for being certificate or Trust Anchor, Trust Anchor that will be the one providing the identity, one of the rules we set in the specification, all the identity of the BGP partners are needed for Europe, for another country the participant will have to elect their own trust anchors so we define what we call the issuers must be State issued identity but the first saying who is it exactly, that's up to the part ant of the ecosystem but at least we have the same common rules. We have the description and then we have to decide how much we can force when you describe things, for example availability of port availability, we lean a lot on the European work, the European data, and they have this notion of consent when you use data, when you are exchanging data so part of the offer when you exchange data we want to have consent field attributes where that is monitoring and to be explicitly set. However we will never go down to describe the vocably, we are transport and health and so on, that's up to the ecosystem to define that, we have to have two levels of description to be set. We have the usage plane where we have the technical interoperability of which protocol and specification are used, we have ‑‑ we are looking at three of them right now which are the GXFS, protocol and DAW‑EX. That's the last slides already, very quickly what we are doing and that's the one that I am more excited about I will say, one thing, but we have to validate and do the specification and we are making several ‑‑ when we are writing them so the first services where we are writing now in terms of implementation, source of the public and availability here on the slide, that we need to have kind of each on foot approach, where the first services will be the association itself issues verifiable credentials about membership, and we would feed that out to the service to the as an input in the civil service that will be checking the Gaia‑X compliance and we also have definition of GXFS in our specification but that will take probably one or two minutes more so I will skip that for now.

And basically, using the output of the first services and input of the other one we are going to be able to test and validate what we are making in the specification using those so‑called services, we also want to enable the existing project that are emerging to validate day by day as they go, where they are not they are Gaia‑X compliant and to learn from that altogether because we are also learning and writing the specs at the same time.

Other thing I mention the implementation, the ‑‑ W3C D ID, we also have German and French funded projects, which stands for Gaia‑X federation services, and we have a hackathon next week, the 7th and 3rd December where I will continue to work on those, on the services. And that will be all for me, thank you very much.

REMCO VAN MOOK: All right, thank you very much. Will is going to handle Q&A on this one you go there is one question I wanted to ask which you didn't expand on in the presentation, so what you are describing here is inter operable software layer between infrastructure providers whether that's platform as a service, infrastructure as a service ‑‑ are there any plans on the infrastructure specifically on the connectivity side for Gaia‑X?

PIERRE GRONLIER: That's the connect ‑‑ so indeed, there is ‑‑ there is a plan for that, but the first ‑‑ that's probably not very well‑defined so far. We had a lot of challenges to first define the trust model to make those eco systems inter operable and the, for example, the connection to the IX, that's seen as a service part of the catalogue to be infrastructure layer ‑‑

REMCO VAN MOOK: Thank you.

PIERRE GRONLIER: There is one big place on network, we believe to have this composition of services, the availability shall that will happen through the network so we are heavily investing here.

REMCO VAN MOOK: Thanks for that clarification. I will now hand over to Will.

WILL VAN GULIK: We have got one question that came to us from someone that couldn't be here, so I am going to read that question: How do you accept Huawei as a sponsor if you want to make a Europe Cloud? So that's the question I have got here.

PIERRE GRONLIER: That's a very easy question. We don't want top make a European Cloud. That's what I have said in the first slide. We are here to promote European values. Some people make the shortcut that says this is a European association so this is about European sovereignty. That is not what it is about. We want to promote transparency to allow the user to take decisions like the open fact of Cloud and data but part of the European value that we ‑‑ Gaia‑X now says this is good or bad, we even ourselves contradict our first statement of ‑‑ so the user needs to be educated taking decisions, and being ‑‑ and saying that we could have a European Cloud only, I will say that this is a very narrow view of how infrastructure and network works.

WILL VAN GULIK: Thank you. Thank you for that. I don't see that we have anything else for now.

PIERRE GRONLIER: In the press there is people claiming that most dominant actors are spoiling Gaia‑X, we don't give names, the last time I checked on the quarter returns on the Git, the top 30 contributors are all Europeans, so I don't know if people want to get attraction ‑‑ and attention in the press, I believe.

WILL VAN GULIK: That's a good clarification, I thank you for that. So I don't see anything else there. I guess that's all then, thank you, thank you very much, Pierre, for this presentation. And.

FLORENCE LAVROFF: Please everyone, just do not forget to rate the presentation. It would be helpful for the presenters, always.

WILL VAN GULIK: Exactly. We can go on.

REMCO VAN MOOK: Up next, we have Leo Vegoda with the an IXPDB update, that's one I haven't seen in a while. Leo, the floor is all yours.

LEO VEGODA: Thank you, can everyone hear me clearly?

REMCO VAN MOOK: Yes, we can.


LEO VEGODA: Super. So my name is Leo Vegoda and I am the IXP database project manager for Euro‑IX and I am here to give an update on what we have been doing. So, first, let me introduce the IXPDB for people who are knew to it. The idea is that this is a database about IXPs and the networks connected to them which is essentially authored by IXPs and that means that it should be accurate and authoritative, comprehensive and the data in it should be complete and timely, and that's done through an API, essentially IXPs generate a JSON file which is imported and contributes to the overall database and you can see on the right‑hand side of the screen we publish the data in a website and you can also see that the website is a little basic, so we essentially found that we have a success problem. When the website was designed, the API that IXPs use to share information through the IXPDB was new, but now we have got something like one in four IXPs sharing their data and this is really a bit too high for the current website design. It also makes it difficult to view everything that you want. We have a couple of tools that people like, one example is the tool that allows you to see unique networks, networks that only peer at one IXP, it's popular but you can only use the tools that we have created and you can make use of. So we wanted to solve that problem. And after doing some analysis, we came up with essentially doing a new publication platform, so we are keeping all of the back‑end and we are changing the front end. We have found an open data platform that supports dynamic filters for tables, dynamic visualisations, it allows integration with external data sets and an it allows people who make use of the data to essentially publish case studies, go and say I had this problem, I took this data, I did this analysis, I came out with this result, these are other results I could have come out with if I had had different priorities, that sort of thing. And so, that is something which is going to be launching in the new year.

We want more, though. Obviously, when we go and launch the new publication platform, it will make things look nice and there will be lots of opportunities. We want to make sure that we prioritise the development which is most important to the users of the IXP database, in particular IXPs, but not only IXPs, also the networks that use the database so that they can plan their peering and researchers and others. So, there's the IX‑F member export schema. That is there as it is. We need to get more IXPs to adopt that so that the view that is presented through the IXP database is more complete. We also need to formalise the traffic API and we are looking at other APIs, so we are looking at developing an API to share information about IXP pricing information. This is a challenging one. The offers that come from different IXPs are quite distinctly different, you know, not just peering but more ‑‑ many different kinds of service so that's a challenging one. I have also been speaking with IXPs and they are looking for information about industrial sector classification because there are many more kinds of network that are peering now than previously. In the past it was all are you a hosting network or an access network, that kind of thing? But now, there's are you a finance network? Are you a manufacturing network? Are you ‑‑ who knows what. So, we are looking at how we can enhance the data that we currently publish in an automated fashion to include more automated‑structured data and this will allow us to go and integrate better with other data sets and, therefore, support much more informed research. So, I am very keen to speak to people who are interested in how the IXP DB should develop. If you are a user or not yet a user but you would be interested in discussing what you would like, please get in contact with me, my e‑mail address is on the screen, Leo A Euro‑IX .net and I would love to have a chat with you. With that, I will hand the microphone back to the Chairs and if there's a time for questions, then I'd be happy to answer any questions.

FLORENCE LAVROFF: Thanks a lot, Leo. Is anyone having any question for Leo?

WILL VAN GULIK: Nothing in the Q&A.


REMCO VAN MOOK: So everyone, go send e‑mails to Leo at Euro‑IX .net, with your wish list and your huge level of interest and willingness to participate. Right. Let's see, where are we in the agenda?

FLORENCE LAVROFF: The colours of this slide hit me in the face every time.

REMCO VAN MOOK: I have no idea who made these.

WILL VAN GULIK: I think we were given this from some people somewhere.

REMCO VAN MOOK: I don't know. Florence, what are we doing next?

FLORENCE LAVROFF: Well the next item we have on our agenda is the Euro‑IX update from Bijal from Euro‑IX.

BIJAL SANGHANI: I am just going to share my slides. And then I will share myself as well. Hello everyone, good afternoon. Right, I am guessing everyone can hear me and see me, so ‑‑

WILL VAN GULIK: We see and hear you perfectly, thank you very much.

BIJAL SANGHANI: Thank you. So, I am here to give you an update actually on the Euro‑IX membership, I am thrilled we had Leo giving the update and afterwards we have Yolandi, I am going to talk about the Euro‑IX members. So we are a membership association, today we have 71 members, you can see them all on the presentation there. We also have a number of patrons, patrons are typically organisations that work closely with internet exchange points, support internet exchange points in some way, I am thrilled to say we had a new patron this year and that's Flex Optics, back to Euro‑IX.

So straight on to the member updates. First of all, we have an update from do. IX, who have had their fifth 100 G ‑‑ the slides are available, I am going to highlight some of the key activities that occurred.

Arista re‑seller port configuration in use since September and I want to have a shout out to Florian for that and they now have DNS node in service since October and 61 G of traffic peak on 1st November.

Update from DE‑CIX. The new self‑service portal is available for all their customers which allows on demand ordering and provision. DE‑CIX have been extremely busy with a number of number of new IXPs opening in the US and other places as well. They have reached ‑‑ Frankfurt traffic and also their Dallas traffic. And a quick update on the, the work on version 3 covering monitoring and statistics is now started and input on ‑‑ if you have any input for the IX API please get in touch with any of the IXs involved in that.

An update from Interlan. They now have 120 ASNs connected with a daily peak of 300 gigs of traffic. They have been MANRS compliant since August 2021 and they have increased security by implementing blackholing solution in Q4.

And commitment to support community projects which they have been doing for a while so that includes IXP manager and peeringDB and ‑‑ I seem to have that twice, the blackholing solution which of course is very important, which is why I have it twice.

Thank you, Interlan.

LU‑CIX, I gave a quick update on LU‑CIX ‑‑ actually I a gave a longer update last time but this time they wanted to inform the community that they are now fully MANRS compliant and have been accepted as the MANRS participant so congratulations on that.

CATNIX have been extremely busy this last year. They now have a new peering platform which is based on the Arista 7504 R and ‑‑ I am not reading that very well, but they have new switches with no oversubscription, and V O Q architecture and many for 10 gigs and 100 gigs ports available. New architecture is based on EVPN VXLAN to increase flexibility and redundancy options. The traffic now crosses over 100 gigs barrier to congratulations, CATNIX.

LINX. The man X IX appointed them as their technical partner, a new internet exchange point based in the Isle of Man. Becomes the official Microsoft Azure peering service partner and they have a new LINX nova welcome to their exchange in the US, CZ.NIC. Nokia deploys the 400 GE IP interconnection peering solution for LINX and LINX 114 which is the first hybrid event took place I think it was two weeks ago in London and I heard that was very successful so congratulations on seeing people face‑to‑face again.

And finally I have an update on INEX. Those of you who are using IXP manager will be pleased to hear version 6 was released in September, this included significant framework changes and two significant features which is the inter switch link management and database change logging.

Further on they released version 6.1 in September which focused on the route service configuration, generation on INEX's system and it took the generation time from .6 million line configuration, from ‑‑ I am going to read that again, generation time for .6 million line configuration from 99 seconds to 1.3 seconds, that is pretty impressive, well done, Barry and the time there in INEX.

178 known IXPs using the platform, I think the last time I heard they had just reached over 100 so it's fantastic to see the increase in the number of IXs using IXP manager. And they are delighted to welcome Patricia Soria to their marketing and community development team so hope flow we will all get some in person next year. And that is all the update from me.

While I was just on that IXP manager update I just wanted to add actually that IXPs that are using IXP manager actually have a default IX‑F JSON schema built in so all of those IXs, if you are not exporting your JSON to the IXP database and Leo's previous presentation, please either contact us orally owe or there's very, very good documentation on how you can actually export the link to the IXP database. And that is it. Thank you all very much, are there any questions?

WILL VAN GULIK: Thank you, Bijal. Yes, indeed, I forgot to mention that on the presentation that there is a good documentation on IXP manager for the IX‑F so, it should be easy, I should have done it but time and stuff, you know, anyway, yeah.

I don't see any questions for now. Well, I think that's mostly what we have. So nothing much more. Well, thank you, thank you very much, Bijal.

BIJAL SANGHANI: All right, thank you. See you.

FLORENCE LAVROFF: Please rate the presentations and I think that we will jump to the next item on our agenda.

WILL VAN GULIK: I see ‑‑ hello Yolandi, welcome.


WILL VAN GULIK: I guess the floor is yours for the Peering Toolbox update.

YOLANDI ROBINSON: Thank you so much for the slot. My name is Yolandi Robinson and I am the at NAPAfrica and today I will be presenting or standing in on behalf of Euro‑IX to take you through the Peering Toolbox.

So, if we think about 20 years ago, obviously I was too young to know what the Internet was but at that time ISPs provided the ‑‑ that was your only choice to be able to read the Internet, so that was ‑‑ you know so you kind of forced to say that your will be managed by someone else, by another network that is in charge of the network and the routing of your traffic. And now, enterprise has changed the way they access the Internet, and have access to transit providers so this has come in terms of IXPs, so enterprises have seen the difference it makes to actually get an internet exchange instead of going with a transit provider, they are seeing the benefits of the more latency, to have direct relationship with the network you are connecting to so there is a whole bunch of benefits they have seen and they are now adapting to the peering situation that we have globally.

The Internet has become essential part of all organisations, I mean we have seen how the Internet has evolved, how technology has evolved, and it's essential to businesses to be connected, of course. And again, coming back to enterprises that have seen not only just enterprises but you know, there's a wider community that have seen the benefit of actually connecting to an IXP to get involved, to be part of that community. So the businesses need to optimise their services, of course, so that means a whole strategy change, instead of giving your business to a transit provider to manage, it means now that my business, my internal team needs to take care of our Internet access and manage that for us.

So, what are we trying to address with the Peering Toolbox? So, the growth in the importance of the Internet to business operations is driving the enterprise networks to manage their own connection to the Internet, so exactly that, because of the growth we have seen, because of, you know, networks seeing that it's important to be connected, we are now seeing more and more networks wanting to connect and manage their own services. I mean, working for an IXP I can tell you on a daily basis we have to do like a peering one‑on‑one because we see that the networks want to manage but they don't really understand how, what, where, when or where to start. And managing does come with a Rick. You need to upskill your staff because they are used to the internal, internal ways of doing things. Letting new best practices, RPKI, MANRS, you need to speak and understand the peering language. If you think about it there's a whole bunch of terms that you need to know to speak the peering language and also to understand these peering policies, what's multi‑lateral and bi‑lateral, what is the difference, how can I optimise my peering strategy? And now there's more than one connection to manage. I mean, in the past if we go back again 20 years, you had one connection to your transit provider and that connected you to the whole Internet. Now, you are kind of split so you still have one connection to your transit provider but you also have a connection to your local IXP, which has multiple connections to other networks that you want to reach. So that means you need to manage more connections, instead of just having the one connection you have more connections to worry about, more peering sessions to manage and to maintain.

And it's quite a shift, I would say, you know, if you are just used to the one, it is new to the business so you are still trying to find your way and still asking the questions but as we know in the industry, not everyone asks the questions; you know, they rather keep quiet and try to figure it out themselves but it's not that easy. What are we trying to address with Peering Toolbox? So taking control of the network growth, having the right resources, the IP addresses, ASNs, that is one of the things, it's actually funny when someone comes to peer at an internet exchange they say, right, can we please have a connection to your internet exchange? They don't have an AS number yet or their own public prefixes or IP addresses, they are not ready to peer yet. So, that is step one, we address this and help them and say depending on where you are located you can go to AFRINIC, LACNIC, ARIN or RIPE or wherever to get your resources and explain to them why it's needed and what it does.

How do you manage your peers? How do you manage that peering relationship, how do you initiate that? It's not as simple as turning up a service and there it goes. You need to maintain and see that it's growing, you need to ensure that you are maintaining that connection. What are you going to peer?

That is a good question. A lot of networks actually come to an internet exchange, they hear about the bigger names like the Microsoft or Amazon or whoever it may be and that is the main reason why they want to peer they are not seeing the bigger picture. There is a whole lot more of networks connected. How do you decide to peer with and what kind of peering session would you set up with them, multi‑lateral or bi‑lateral sessions? It all depends on the peering policies and, again, explaining that concept of peering policies, there's a lot to learn, again, coming back to where do I start? Who do I speak to? Where do I go to get the resources? These are common questions we see on a daily basis for networks wanting to connect to the internet exchange. What are the best practices? Again it's so important to introduce these new terms to these companies and make them understand why it's important, because this comes back to your security, your network security, it's important to implement it or at least understand it. How do you deal with your network challenges? So what are you doing for if your BGP is down if your traffic just drops? It's a mindset change, it's kind of giving them, you know, a basic guide on what to do, who to speak to, where to go, how do you reach out to the community? There is so much information.

Now, trying to explain this into one meeting, I mean you can imagine this can take an hour, two hours, it could take a couple of meetings. And what we are trying to do with Peering Toolbox is give the solution. So, Peering Toolbox is a community‑focused learning tool. So, we got together with a few of our community leaders and have combined some basic information that we feel that networks need to know, and to inform them as to what is your next step, where do you start, what to do?

So, the peering ‑‑ what is the Peering Toolbox? It's by the community, this is is actually a community project where everyone in the community is welcome to reach out, give some input, give us suggestions and share with us your experience in the peering community, because that can actually help another new network wanting to connect to find their way in the industry. The organisations we have involved includes LINX, NAPAfrica, Microsoft, Amazon and INEX. So, this is not to say that is it, you can get involved if you would like to get involved with Peering Toolbox and help us put together this base, this information base, we would love that. The aims is to provide a learning structure is best practices information for the new entrants. So again coming back to when we to a network there is a lot of places you can actually, you send to information so for instance, for ASNs, IP addresses and so forth, you can send them to the RIRs, for best practices, you have RPKI, MANRS there's a whole bunch of places, we are trying to bring that all to one so you have a one‑stop‑shop to go to get your information to get started. The toolbox will act as a reference exactly that, that us as IXPs, as well as other networks, can help build the peering community to educate others wanting to join the community to say, if you are getting stuck, you can go to the Peering Toolbox, they do have the information you need and if it's not there, you are more than welcome to reach out to the team or to someone in the community that can assist, depending on what you need to know.

There is already a lot of good work that has been done with the community so we do have the website up and running. So we want to pull it together to make it available in one place, exactly what I said, so there's so many websites and resources we can use to refer networks to get started, to know what's going on, to learn the language and all of that, we want to put it in one place at one reference point for our networks to get this information and get, you know, peering in the community.

So we are not creating a peering model, so again, this is not to say whatever we put into the Peering Toolbox, that is the way you do it. This is merely a reference, it's a guide, it's just assisting to you get started or if you are struggling with a specific problem, just to overcome it you know, just to see what ‑‑ if the community has this issue how has it been resolved or I don't understand RPKI, I hear it a lot, what is it about? So this is not a peering Bible, it's a reference point for the best practice information.

The Peering Toolbox can reference your documents so please if you have information, if you have some interesting information for us to put into the Peering Toolbox that you have seen has helped other networks connect or improve their networks, please reach out to us, we have a peering e‑mail address which is called Peering Toolbox and lists at Euro‑IX ‑‑ please, if you would like to get involved, even if you want to give your personal opinion or suggestion or whatever it may be, we would love to hear from you so please, if you have any suggestions as a company or as an individual, we would love to hear from you. This is a community‑based project.

So coming soon, we are going to create a learning structure. So, the learning structure will be broken into smaller pieces so what we have seen or the feedback that we have gotten from the community is that we have one reference point to go to, but where do I start? How do you distinguish between advanced information to beginner information or intermediate information? Which we understand, it's a bit daunting to start right off the bat so we want to create levels of information so if you are a beginner just coming into the community, it gives you the basics, do you have an ASN, what is it, where do you go to get one? All the basics to get started. We have the intermediate, this is if you are connected and you just want to take it a step further, how can I optimise my peering relationships at an IXP?

Then we will have the advanced things, and that is maybe where I am going to start perhaps with my learning, can my learning on Peering Toolbox is to understand what is the advanced things that I can learn, you know, in terms of the peering ‑‑ or the peering community that we have.

And we are thrilled that Phillip Smith is joining us as the product manager. We will also be holding focus group on the 8th December, should you be interested to join this focus group we would love to have you there, so again, if you can reach out to us on the Peering Toolbox e‑mail which is the Peering Toolbox at Euro‑IX .net we would gladly send you the information for this meeting so we can get some input, we can share some ideas, we can get you involved, again this is a community project and we would love to team up with our community and just, you know, get out there, get this toolbox, the Peering Toolbox just growing as we are going.

And that is it. Can you believe it? That is it. So thank you so much for for listening to me, I hope it was interesting and, yes, if you have any questions, I am here.

WILL VAN GULIK: Thank you, thank you very much. That was really, really nice. It sounds like something I wanted to do for some time so I need to send you an e‑mail, I guess.

YOLANDI ROBINSON: Is this on your to do list again?

WILL VAN GULIK: Something else, that's fine I will survive. We have got Nurani who wants to talk to us, I guess.

NURANI NIMPUNO: Hello, everyone. Great to see that you are pushing forward this initiative because it's such a wonderful initiative and I just wanted to actually, well, say thank you to you for taking this over but also to actually call ‑‑ call out to the community, because, really, I mean, this is ‑‑ we are driving at the Euro‑IX because it's where the community comes together but the Euro‑IX is not the expert here so we need the inputs from the IXPs but we also need the inputs from networks and that is, you know, that's all the rest of you who are in there the room somewhere and I also know that some of these ‑‑ some of you actually write sort of guides for how to peer and I know some of the big content providers have done that because you are struggling with the same thing, so you have things like that to share or if you want, you know, if you need anonymise that so get this touch because that's something that everyone would benefit from, resources that you have but also problems that you see, when you see these are the problems that we encounter all the time that's incredibleably useful information for us to get in so we can create resources around that. So that's all I wanted to say. And thanks.

WILL VAN GULIK: Thank you very much. Okay. Okay, so yes, that was ‑‑ we have one Q&A. So, yes, I have got one Q&A question from Rudiger Volk with no affiliation: Is there a publicly information offering say URL or is this only a request for input?" And I think there is an URL that was in the slides ‑‑

YOLANDI ROBINSON: Bijal has shared it in the chat box and again, if you have any suggestions, ideas, comments, we would love to hear from you.

WILL VAN GULIK: Thank you. Okay, that's actually great. Okay, I don't see anything else, and we will soon be reaching our time. So thank you very much, Yolandi, it was good having you are and I guess we will all see you soon.

FLORENCE LAVROFF: Thank you for your enthusiasm, take care, bye.

WILL VAN GULIK: That's amazing, we are on time, we are on time.


REMCO VAN MOOK: We are indeed. So, that brings us ‑‑

WILL VAN GULIK: I have one ‑‑ I just notice that we are having SwiNOG next week in Switzerland, in Bern, and there are a few slots available so you should consider that if you are available to move and to go to a live meeting. Just that. That was my advertisement so far, for everything that I don't organise.

REMCO VAN MOOK: So while we are at it, RS NOG is also next week, the Serbian NOG, where I will be presenting. It is a virtual event though, so you are all welcome to come and watch that as well. Anyone who has any other events or online thing that they want to plug because this is your 30 seconds to put that in the chat so I can read it out. I'm going to go on to say to all of you, please rate the talks, there is ‑‑ the RIPE 83 website is there for you to do that. I would like to thank our scribe, I would like to thank our chat monitoring, Marco, and I think that is it. I don't see anyone typing in the chat. So, that was your ‑‑

FLORENCE LAVROFF: My lovely Remco, of course, we should just like offer our dear audience send us any suggestions for the next topic for our slides of the next RIPE Connect Working Group, of course.

REMCO VAN MOOK: Not just a slide, also if you have topics or other things you would like the Working Group to discuss, please send an e‑mail to the Connect Working Group mailing list, last time I checked it is functional, it doesn't see a lot of use, but by all means, put something on there and let's have a talk. And I hope to see some or all of you again at the next RIPE meeting, which hopefully is going to be in person or ‑‑ and/or hybrid and may or may not be in Berlin. So, with that, I am going to close the session. Thank you all so much for your time, you have just received another two minutes back of your life. Thank you so much and see you next time.