These are unedited transcripts and may contain errors.

The RIPE NCC services Working Group session on 2nd of November, 2011, at 4 p.m.:

KURTIS LINDQVIST: All right. The wait is over, and the reason you all came is here again is NCC Services Working Group, your favourite Working Group. We have a very packed agenda today, as usual, and as, you know, we have moved the start of the agenda here will be part of the AGM civil duty update from the NCC here and immediately we are done here we are going to go to the room behind us next door for the AGM and Nigel has told me from the time I finish here, until, 15 minutes so no picking up coffee or going to the toilet, straight to the AGM. If you haven't registered for the AGM and haven't got one of these special tags, now is the last chance to do so, so run outside and get that done, as well.

Was that everything I have to say about the AGM? OK.

So the proposed agenda for today is what I am doing now, then we have the NCC updates, we have presentation by Andrei for the final distribution business processes contractuual relationships and the RIPE Atlas probes and going to try and keep this on schedule. Any changes to that agenda?

OK. We have a scribe from the NCC, I guess, I forgot to check who that is, actually. Thank you.

If you go to the microphone, please state your name. And I seem to have taken out approval of the last minutes, but I guess you all approved the minutes last time. Any comments, feed backs? Done.

With that, I will hand it over to the RIPE senior management which is all sitting here and I guess Axel will introduce all of you, maybe.

AXEL PAWLIK: Yes. Good afternoon, colleagues, friends, ladies and gentlemen, guests, newcomers, old timers, your excellencies. I have been in too many government meetings. I want to do a quick introduction, basically about some challenges. And if you read the newspapers you know that it is a challenging year for the RIPE NCC. I have been telling everybody who asked me that this year we are going out of IPv4 space, we have on the IANA level early in February you have seen quite a bit of attention there in the generic press, and I am in funny spot to say we will probably not run out this year yet at the RIPE NCC, but you never know, maybe. So probably early next year.

So, run?out is obviously one of the challenges to manage that in a nice and orderly and tidy fashion, to maintain stability in the process for the RIPE NCC, your organisation, and to keep all our members happy and the greater community, which is, in itself, a challenge, not that you are all particularly difficult to keep happy but there is so many of you with different interests. So what we do, we do ?? run a service as, you know, and thank you to all of you who have answered the survey and have contributed to that, also of course we do talk to you in the hallways at the RIPE meetings and other meetings and we follow the mailing list and the like. Thank you to all of you who have given us some input there, us includes staff, the board, basically all NCC folks.

So, stability in changing times is an interesting thing. We have lots of senior ?? they are not that old but they show some age occasionally. My senior management there and all of them, some of them are older that the others; some only look old. They all have taken up some challenges this year and I will start in no particular order, Jochem offer that, his particular challenge right now he is responsible for security in the company, and we asked him to look after our laptops which we left ?? so security is new, his new thing, as well as responsibility of all of our IT infrastructure. Those guys are reporting to him. So that is a bit of a challenge he took on earlier, thank you have very much. And I am very happy with the way it's been going so far. And we do have network here, that is a good thing. He is also responsible in some way for the running of things, infrastructure at RIPE meetings, but of course you have other people who do that for you. Thank you.

Serge, his main challenge this year is he has started to work with us. Of course, you know us for a long time and maybe that made your decision to come work for us a little bit easier, and you have been in the office occasionally since then; I think we had one or two days together in the office. You can occasionally see he is a bit scared, there is so many people there eager to work with you, that is a bit of a change from the previous position that you had, I believe, and so staff that is there to support you; we all support you, good thing.

So, Chief Communications Officer. Your easy task is to make RIPE meetings and all of RIPE the best it can ever be for all of the community, so in that sense, he is your guy to talk to.

Daniel is not asleep yet. Daniel has, this year, agreed to take on even more operational responsibility. You probably know already that we have had some minor reorganisations and swaps of responsibilities and one or the other person leaving, so the old information services group has been merged under Daniel's guidance into what used to be the DNS group and what used to be the science group. Those people are all making terrible progress there, and the main thing for me to convince Daniel to actually sign up to that because he is the chief scientist and something ?? wrong wrong with your arm?

Paul. You all know Paul, apart from being very entertaining on buses occasionally, he is very, very popular in various ?? all of our regions, really, and as our board members come back into my office and say, oh, my God we have been at regional meetings and governmental roundtables in the regional events that we are doing, and thank you to the Board, those guys are increasingly also going to those places and we have been pummelled by our local members there that we should come and do more. Well, Rendek is over there and I send him off to Dubai, bit of a challenge there because you are setting up still and I want to come over and visit him there and give him support, but he doesn't want me, I can't put you up anywhere. Thank you very much for agreeing to that. It's not only the Middle East but also Russian area but greater area, MENOG have been significantly benefited from your activities there so you continue to do that and you still look after all the rest of the governments, at least in our service region but also beyond.

Andrew, new team for you, database, you have taken over the challenge to look at midterm synergies between the two in?house groups that are manufacturing software, that is our business applications group, the main software group but also the database group and maybe there are some signatories there. That is the one thing. The other is the IPv4 run?out, you are responsible for those numbers things and so far, I think it has gone really, really well but you do see that we have several policies in place around this. The next guy whose telephone with music goes off has to sing along with it on stage. I am always glad if it's not me.

So the numbering stuff, I think it goes quite well currently and that is the most visible responsibility that we as Internet Region Registry has, there is more to come and we haven't run out yet. Having said all this, those guys will talk about their particular challenges and what they are currently doing and plan to do over the next couple of months.

Having said that, I do want to thank all of you, community members, guests and excellencies, for your support, that you have shown so far, and I hope you will continue to show us.

Also, of course, thanks to my Executive Board who have been very active and increasingly so over the last number of years, not only in travelling but paying attention to what we are doing and giving me guidance there. Thanks a lot for that. To all of our staff who are doing their utmost not only at RIPE meetings but also at home during RIPE meetings and also in between, obviously, for your benefit, I hope.

So, who goes first? Serge. Come on.

Serge: Thanks, Axel.

Axel: You can tell he is new.

Serge: Some of you might know me, I have been up on stage before, you probably saw me in EIX Working Group. I was the secretary general of the Europe Internet Exchange Association for the last ten years, and I decided it was time for a change so I have come over to the RIPE side.

So what am I going to be responsible for, communications, obviously communications both externally and internally, what is also going to come in my department is web services and Atlas. Let me give you a bit of a run?through of where we are.

As far as communications go, and as far as I am concerned, I wanted to be two?way effort; I mean, we are there to communicate everything that the RIPE NCC is doing, what is happening at RIPE meetings and what is going on within the RIPE community but we want that communication to come back from you guys. We have done stakeholder surveys, we have mailing lists, we have websites where you can comment via labs and so on. Everything is in place so we also want you guys to communicate with us, as well as, I hope, we are communicating with all of you.

We have taken some of that feedback. One of the things people were complaining about, the amount of paper. Some people still love paper, it's there, available, you can get paper. If you don't want the paper then you can opt out of it. So we are looking to increase that or decrease the amount of paper.

Social networking, again this is an opt?in thing; some of you aren't all that keen, some are, the ones who are can opt?in and get on with it. If you aren't, you don't have anything to do with it. We are exploring and enhancing what we have been doing with that. Probably one of the most relevant aspects that I will be working with is the RIPE meetings themselves. You have probably all heard of the RIPE Task Force, they have also delivered a few ideas to us and we have taken them on, we are thinking about a few of those other ones.

Another thing that is taking place is a dedicated plenary programme committee that Rob already talked about on Monday. I think they have done a good job this time, I have seen how much effort has gone in. Of course, this is their first go at it, so they have a few teething problems. We had a great meeting yesterday over lunch to decide how we can make it go further in the coming meetings, so we look like we are on the right track. So I am pleased with that.

What else have we got here? We are also seeing how we can help or investigating, in any case, how we can help facilitate you guys meeting each other, a lot of people come here and say I am looking for him and don't know what he looks like. We are having a look into ways ?? we have got some ideas. I am talking to a few people and asking a few others to do that as well. We are still going on with our regional meetings in the Middle East, Russia, in the Russian region. We had our first southeast regional meeting in Dubrovnik at the start of September. We had 150 people, really good vibe there and it looks like we will probably have a go at that again next next year.

Web services, new website and running about ten months ago, a lot of feedback about that. We have also tried to implement a lot of the extra features, ideas, some of the critique that we have had, taken it all on board and appreciate it. Please, keep it coming and we will keep on working with it. One thing we have added is the search functionality, now you can search on the ?? all the mailing lists as well, so type in a search, if you go to the advance search you will see little cheeky options there, play around with it. Again, this is new, we are willing to get your feedback and take on new features about that.

And we are also pushing forward with the whole single sign?on or RIPE NCC access as it's formally known, scheme. So that is also still developing.

Lastly is RIPE Labs. Again, where I am going to throw it back to you guys. This is something for the community. I mean, the NCC will put on measurement tools, ideas they have got within their development stage, so this is your opportunity to tell us, right from the beginning, what you guys think about it, what should we do, features we don't like or do like. So please visit the labs, see what is on there, but having said that, it's, again, for you guys, if you have got ideas that you want to share with the community this is a perfect way to do it. I know not everybody likes getting on stage and putting forward their ideas; well, RIPE Labs can be your stage and you can put an idea out there and if people like it, come to the next meeting and talk about it. So visit RIPE Labs.

I think that is about it as far as the plans are concerned. Again, this is my first RIPE meeting as the Chief Communications Officer by RIPE so there is a few of you that I may not have met before because I had other interests. So, please, I think most of you, I am pretty easy to approach so come and talk to me and give me any feedback, ideas you have got. I don't think I will ask Axel to get back up. I will hand over to Andrew. I believe we will hold off the questions until the end of the presentations.

Andrew: Member services update. I want to run through some action items we had on our plate over 2011 and also follow up on those, and then I will go into a couple of items which are on the list for next year; basically, main areas we have been trying to improve and the focus for our customers' services, members' services etc., was on the direction improvements with you our members.

So as Serge already mentioned, RIPE NCC access implementation, it's a single sign?on module for our services, currently we have two services running under single sign?on and we migrate them piece by piece. And in the end of November, we will start migrating the portal, for example.

Other improvements in interaction model with the registration services, in particular. Transfer process enhancements was quite a cumbersome process, we got a lot of feedback on those and we tried to change that and that also resulted in the listing service as we have it today.

One of the things we also notice was that we had a lot of mails going up and down between us and our members, many questions were raised. And what we currently do is after four mails going up and down we proactively contact the members in the sense we give them a call to try to finalise the ticket in a fast manner.

Transparency, for processes and procedures, we try to publish everything we have and how we work, so that is transparent to you guys, how we operate and how we work and what you can expect from us.

Preparation of process for IPv4 run?out. It's a very important run, of course, as Axel mentioned as well. There is a lot of process changes we are going through to manage the last bits in a proper way and to share due diligence in how we work. One of the ideas or one of the items we have been working on is one single queue, which might be a bit cumbersome for one or two you, sometimes you get another IPRA, evaluating your request, but it does give us the opportunity to really do first?come first?serve and that is one we have been working on pretty hard.

What we also did and was one of the things which was mentioned a while ago in the former survey, actually, was a web chat functionality for customers services. So we did a pilot with "customer services on?line help" it's called and it's a web?chat functionality which was well received. So that will be for 2012 one of the items we will work on.

Furthermore, with database, we have been working on user experience changes, we got a lot of requirements and requests for changes on how it all works and it looks and look and feel, so that has been one of the areas we have been working on.

And in the training area also changed quite a bit. We are working much more modular which means we can target the specific members much better than we did in the past. We have also a bit more technical input and more technical background, which helps us to create more exercises, hands?on experience and that kind of stuff, to make it more interesting for you.

Well, 2012, too many areas, streamlining the processes, it's the first area. Again looking at the roll?out of on?line help, looking at whether it would be applicable for the RS department, for example, LIR usability improvements. One of the items on my list is to improve ?? has been broken where some people would say it has been broken for a while and that is something we really want to get back on track. Improved process around inconsistent data reporting. We had a very nice meeting in the Anti?Abuse Working Group with requests on improving that process, on how to contact us and showing how we take action and all.

Introduction of webinars training. We do see a lot of requests from people and what we would like to do is get in touch with you making it much easier to get in touch with us, they will be 30 minute to 45?minute trainings via the web and you can do specific requests, if you have certain questions you would like to address, and that makes it easier to get in contact with us, as well.

And the last one on this slide is develop a new on?line learning platform. We are currently requiring ?? gathering all kinds of requirements for a new on?line learning platform to better gear to the current needs and the requests we get from the membership.

Last slide for me, strong registry, that is one of the other areas we want to focus on in 2012. Continue increasing the number of audits. This year we did around 400 audits so far and we do more ?? want to do more audits to keep the registry up?to?date and current. That also fits together with the continuation of 2007?01 which is going well and contacting the legacy space holders. And as you already heard from Axel yesterday, it's quite a challenge, but it's going really well and we had quite some good feedback during this meeting in the info centre as well.

Resource certification: That is the AGM which will follow on this, whether we need to continue on that or not, but it is one of the items which is on our list as well as migrating the ten?year?old database software to a more flexible platform so that it's easier to implement requests we get from the community and the membership on how to interact with that.

And the final one is introduction of new tools such as registration history and LIR data improvement tools, so it's easier for to you keep track of your data and improve the data as such with all being the strong registry being the main driver of that.

That brings me to the last slide, and Daniel, yes, yes, it's your turn.

DANIEL KARRENBERG: What you guys don't realise is that there is a different kind of competition going on here, it's not the least acronyms, it's being under five minutes per person. I am the Cheif Scientist so this is the Science Division update. The updates usually are what have we done since the last meeting and what will we be doing until the next meeting and in the next year. I will go very quickly through what we have been doing because I would like to spend more time on what we are going to do, but it's all fair material for questions later on.

So in DNS we have done capacity improvements to keep up with the load. We have moved the ccTLDs, which is the longer process, to our Anycast platform which has nodes in London and Amsterdam, that takes a lot of IANA involvement and involvement of ccTLDs. It's more administrative challenge than anything and we have improved our monitoring analysis and the code for that has been released in conjunction, also, with a nice video that shows you how we do this stuff.

On RIPE stat, we have done monthly demos and continuous demos, introduced a mobile ap for IOS which is available, if you want to play with your iPhone you can do that. For DNSMON, actually this is the most embarrassing one because we have been promising a new GUI for, I don't know how long, and we still don't have it. But what we have done is we have answered the other two questions we got from the users. We have reduced the time?lag for the data to appear from one hour to five minutes and we have enhanced the performance for everyone. The five minutes only the subscribers get because everybody else gets delayed by two hours, as you may know, but the performance is really better and we will see what we will do about the GUI.

On TTM we have surveyed all the subscribers to re?establish contact with them and find out how we evolve TTM, a meeting which was very constructive and we are planning the evolution now.

On the backend of all this we have an internet number resource database that stores all this and we have a data store, we have 850 probes, which has two more capacity and more measurement types and all sorts of good stuff.

Coming to the outlook, I already used half my time. The slightly bad thing about Atlas, we did a lot of basic development work to get stuff out there like the user?defined measurements but they are not out there yet. So we are building foundations. And that puts me in a difficult position because I can't show all that much. But what we are going to do is we are going to have more maps, we are going to have the user?defined measurements, better testing is starting right now and a better website, most definitely, and we are going to have new hosts and sponsor documentation. Watch this space until the next meeting. There is going to come out a whole lot of really interesting and useful stuff from RIPE Atlas. We are improving the community building and communication, that was one of the weak things, that is also why the Atlas website isn't what it should be. And we don't have the sponsor documentation and things like that.

We have added two very good members to our team. Many of you know /SR*EZ /TPHER and supporting her in getting our communication going and improving it, this meeting we had with the TTM hosts was a good example of that. If any of these activities that we are talking about in the Science Division you have comments about that, you can that.

We are not intending to keep you away from the developers, this is not a sales department, this is an additional resource.

On TTM, what we want to have until the next meeting is a plan agreed with the subscribers on how to move forward and start implementation of that plan. Currently it looks like we are going to evolve TTM to merge it with RIPE Atlas. And we are going to make steady progress on the other activities.

Strategic outlook, and it should be 2012 ?? there is always one error in there ?? is to consolidate the measurement services all the active measurement services, TTM DNSMON and Atlas into one active measurement area, the routing measurements next to them because they are different. We will see the full benefits of RIPE Atlas and we will have a move towards and have the first instances of unified data storage and analysis. And that keeps me just under five minutes.

JOCHEM de RUIG: I am going to take very long, actually, because I want this session to go on until the whole night long and you know that.

I will start a bit differently. I will talk about information security and IT infrastructure.

We recently ?? someone reported that one of our services was hacked and we took quite some action. We always take these kind of reports seriously, we investigate it and we also try to openly communicate about it after the event. We have information security officer, he has worked now, I think, for about two years with us, and we are continuously trying to improve our information security, we are looking at access management and intrusion detection, those kind of things, and he is quite busy in reviewing some internal processes.

IT infrastructure, well, this is, I can't say, business as usual. What we have done over the year, Brian and his team have been busy with storage and filers. As he says we have a virtual farm where most of our servers sit down nowadays, so we harvest the farm or something, and actually, at the moment we are looking at co?lo expansion, we are quite busy with setting up our capacity planning. We have two co?los in Amsterdam and we are looking for a third location at the moment, so if you have suggestion, please approach me, I am more than happy to discuss that.

And of course, a key for the IT department is supporting other services.

Finance and admin: Well, we have been busy with the budget and charges scheme especially. We will discuss it later. The invoicing and payment, what I think is key to mention here, we are still busy in looking at other payment option to make it easier for you to pay. What is also crucial is that we ?? local tax offices are quite strict with VAT numbers, so new members coming in we do a VAT check, if the VAT number doesn't match the number, we have to check it, actually get the valid number. So they are being very strict and penalising us when we don't do this and have a correct number.

We have new auditors for this year. We have our previous KPMG for six years, but in line with good governance principles, we have changed auditors, the Board requested us and this year we'll have BDO accountants looking at it.

And furthermore, we are busy with some improvements. We have office improvements. We are a bit separated from each other and we have an option to make this more unified. We have one big building with a lot of different floors so we are looking at whether we can make that a bit more efficient and logical for us.

Legal: A lot of work done mainly by Athena, our legal counsel. Of course the articles and Standard Services Agreement have been reviewed and we are proposing an update which will be voted on in the general meeting. There have been some other terms and conditions and other legal work, the listing services, for instance. We have been busy on the governance documents, remember the closure document which we have been very busy with, we are still looking at, well, the abuse handling, Laura looked at and we are looking at how to frame that. The transfer document, due diligence, audits. Engage with law enforcement regulators. Well, in the anti?abuse there was quite a bit of an update, won't tell much about it, and ongoing legal support. And that is it from me. Paul.

PAUL RENDEK: Good afternoon. Head of external relations at the RIPE NCC and they have put me last yet again, when I see the flash cards of time shining. Try to keep this as short as I can.

I am not part of the technical staff of the office, as you can see. My function has changed a bit at the RIPE NCC. When I was hired at the RIPE NCC I was hired to put together a communications department; I will not even tell you the year that that was, but Miriam was my boss and Daniel Karrenberg was director of the RIPE NCC. At that time, I actually was the first employee at the RIPE NCC that broke the bubble and had a PC with Windows on it, and Olaf Kolkmann ?? I don't know if Olaf is in this room ?? Olaf Kolkmann fought with Daniel not to let me have that machine but I ended up with it. So anyway, things have come a long way since that time.

My position is evolving yet again at the RIPE NCC; I think it's kind of done so naturally. I have moved much more into the external relations area now and I am going to kind of explain to you what is happening there.

External relations, it's a really huge area for us at the RIPE NCC. Naturally we have been doing a lot of external relations with our members but increasingly as the years have gone by in this area, we have been doing it with other stakeholders as they have entered into this arena.

I love this word, the multi?stakeholder echo system. This is something that the I?STAR organisations or organisations like us, the RIRs, IANA, ICANN, IETF, ISOC, we use this word to explain what is happening today in kind of the Internet environment. These days, a lot of external relations that we are doing are not only with our membership, that is of course very important to us but we need to have outreach with other stakeholders coming in. We are one of the I?STAR organisations that represent the technical community when we do go out there and work in this multi?stakeholder arena and we need to communicate what we do with the other stakeholders in a politically smart way because many of these are governments or regulator, inter?governmental organisations. So we are very busy. Actually, we have done a lot of learning to see how we communicate with these people and what we do share and how we cooperate with them, it's been quite a bumpy ride but I think we are making some leeway there.

In 2011 we did a membership and stakeholder survey that showed strong community support for external relations activity particularly in the area of us working with inter?governmental organisations with the likes of the OECD, ITU, Internet Governance Forum, where you see a lot of different people coming together and wanting to have as much influence as they can on Internet administration.

So what we do we do? When we are out there running around ?? incidentally, our team is huge. I mean, our whole organisation, all the areas inside of our organisation, engage in external relations activities, it's not just one person or one particular kind of area. We do quite a bit in the capacity building area, what we have seen is as a technical community there are expectations on us, making sure that we are making people ready for this big move to IPv6. There are a lot of bodies that are looking at us making sure that the technical community is doing what it can. We have heard those voices, we are doing IPv6 roadshows which we have launched in the Middle East, we plan on launching those in Russia and the central Asia area. We have a presence at all the NOGs we can. RIPE NCC wants to be there making sure we are doing what we can in the capapcity?building area all over our region really.

Outreach and education, these are big points for us as well. RIPE NCC holds two round?table meetings a year where we have governments that come and ask us a lot of questions, they want a lot of data from us. We have been cooperating with them, I think we are finally, after many years of playing a popularity game, have established our relationships and actually what we can offer these governments or other organisations that come looking for data from the RIPE NCC.

We provide the RIPE NCC perspective in this multi?stakeholder, in this whole Internet governmental forums. We are one of these representatives, as I said, of the technical community. I think it's very important for us to have our voice there because when we do we can see that over these years we have built quite a bit of influence there and we have carved our position in what we are doing in this whole arena.

And we also talk a lot with individual governments, law enforcement agencies, regulators. This just seems to be happening more and more. We get invited by ministries to go to IPv6 days that they are throwing together with their industry and with their government and it's nice to to see that RIPE is always asked to come, or RIPE NCC to come, and if we are not asked, they ask us who in RIPE could come and participate and that is fantastic,, so it's nice to see them reaching out for us and looking for our input there.

So what is happening? Basically in 2011 if I look at these inter?governmental organisations and who we have been playing with, I have listed them. We have been active there, we have been asked many times to throw workshops, either technical helping explain technical concepts or on panels together where they are looking for the technical community's position. So that is quite interesting.

What are we going to be doing in 2012? We have got quite a bit on our plate. We spent a lot of time in Western Europe and central Europe, we have gained a lot of traction, I think we have got quite a lot of good relations there. We definitely have to continue working on those because the environment is ever?changing. But I think it's important for to us look at the rest of the service region. So part of what I will be doing is focusing on Russia, central Asia and Middle East, working with ENOG and MENOG folks, make sure we foster the right kind of relationships so we are present throughout our whole service region. A lot more roadshows and looking forward to launching the road shows together with ENOG in the Russia area. I know we have had a couple of organisations provide us with their staff for training free of charge, which is great. So they will be doing these hands?on IPv6 roadshows for that part of our region, which is super. We will be doing some more regional IGFs. I know there is talk now with the Arab League, they want to put an Arab IGF together, we have been approached ?? the RIPE NCC ?? to give our input as the technical community, organise some workshops and help them with that. Once that gets off the ground we will be moving into other areas. We show up at other local IGFs and other kind of government meetings all over the region.

One of the big things that is happening in 2012 is ITU is going to be busy again with a lot of meetings. All of us RIRs will be working together to make sure we have the correct representation there. There's the WCIT and the WTSAB that are happening back to back, they take place over very long period of time, and I am happy to see they are going to be taking place in my backyard in Dubai. I don't have to travel very far to go to there. And that is the end for me.

AXEL PAWLIK: Thank you. Now, folks, questions? Otherwise he is going to continue on with our stuff.

AUDIENCE SPEAKER: I have a question to Services Department. How much requests have you service since last AGM? What is the rate? What kind of requests you are serving? If you are not ready just now, please prepare this answer for AGM.

Andrew: The amount of resource requests you mean?

AUDIENCE SPEAKER: Requests for resources, requests for help, any requests without distribution ?? resource requests, transfer requests, help requests, and so on. Thank you.

Andrew: Yes, I will.

AXEL PAWLIK: We can get that together.

SPEAKER: I have a question, too. You are expanding with the southeast Europe regional meetings. Do you see more coming on or are you done now?

Serge: No. Well, as I said, I don't think there is another region that we need to focus on. We have had quite a bit of call to go there for a long time and it was a success and we are going to go back there again. If there is another region that comes up and says, yeah, we are going to get two or three hundred people and we really have difficulties coming to the meetings, I guess we will investigate that. But I don't see another region that would require another dedicated meeting.

KURTIS LINDQVIST: All right. That is it?

PAUL RENDEK: If there are no further questions, then I will slide right into the RIPE NCC member and stakeholder survey. I will give you a quick update on that, a bit of a background on the survey. I think that Desiree from OII actually presented on Monday, give you a round up of what was going on there, but I can give you a quick background and slide into some of the findings we had.

This was the fourth of the large?scale RIPE NCC surveys that we did. The difference between this one and the ones we did before, it was the first time we actually included other stakeholders that could participate in this, so we extended it past just the membership. That was quite interesting because we had quite a number ?? and I will walk through this ?? of stakeholders that gave us feedback and that was really great. So it's nice to see that the RIPE NCC actually has people that want to actually participate beyond just its memberships, it's great to see.

So the survey was carried out by the Oxford Internet Institute, that was to ensure independence, of course, and anonymity for the responses. We had quite a lot of responses, wow, wow wow. I went through them myself, step?by?step over the whole summer, it took forever. It was largest response to RIPE NCC survey we have had yet.

The methodology used behind this: Before the survey was even built we actually had these consultants run out to seven cities inside of our service region, in Germany, UAE, Russia, Sweden, UK, Italy and Czech Republic and basically they met with members, they were from extra?small all the way to very large members, other stakeholders, governments regulatory, bodies, LEAs and speak to these consultants and give them the input to build this survey. So they used these discussion groups as a basis to build the surveys. Not only that, it was a great way for us to get some feedback directly from those people that came to those meetings. We did this also at the other survey, it was quite popular. Therefore, we thought it was nice for us to repeat this again and we thought it was great because it resulted in quite a nice survey.

So, looking at the responses, we had 825 valid responses, 680 from our membership and 145 from other stakeholders. That is pretty large number for the first time engaging with others outside of our membership area. The responses by region, naturally you can probably guess Western Europe was the highest, followed by some unknowns, I think some people don't know what country or region they are from, that is fine. Russia and Central Asia came behind that, Eastern Europe and Southeast Europe, Middle East and other. I think other members at RIPE NCC floating outside.

AUDIENCE SPEAKER: Unknown is part of the reason when you cover the entire region, that you can't answer entire region.

PAUL RENDEK: Could well be. We gave an option for people to fill in the country for those that didn't fill it in went under "unknown."

So the day of the highest response was 30th May. We had 238 responses and that was directly related to the fact that we announced the first iPad winner on that day. Brilliant. We did give away five, it was so nice to see those were scattered around our whole service region and people so excited they never thought they would win anything, coming from far parts of our region that won an iPad, so that's great.

So the RIPE NCC services: Resource registration was the most important service to the members, that is probably not such a surprise but it was not the first and last survey, it scored 6.44 out of 7, which is a very, very high score. The average over all the services was quite high in the scoring, 5.9 out of 7. So I mean it's great to see that everybody is happy with the services that the RIPE NCC provides. But of course, for us, as a management team and as a company, I think we are more interested in where we can see the improvements for the company. So great, okay, it was nice to see people saying good job, but I was really digging around, and my colleagues as well, for where we would see what are the things we could group together of changes or improvements that the members and stakeholders wanted to see.

So, RIPE NCC services, the most notable request from the RIPE NCC members, they wanted to see more multi?platform services, live chat video?conferencing. Funny enough we are working on that already, so we listened to that before we got it back on the survey. Make the RIPE NCC database easier to update. More statistics and make the statistics easier to find. Both of the groups, actually stakeholders and members, were really after statistics from the RIPE NCC. Everything about the IP addresses. So I know we are busy working on that, but it's great to see the feedback and actually things that people really want to see, when we do go to meetings we do get requests for particular data that people want from us. If we can supply this, then we do do this.

More user?friendly interfaces for everything we have. The most notable request from the stakeholders for improvements, they wanted to see more statistics, of course these governments are really begging us for all of this stuff, more measurement and analysis. Oddly enough, you know, from all the deliberations we have had with these other stakeholders, they are very interested to see how they score in all this data with their countries next door who they feel they are competing with. They operate very much like the commercial firms actually, it's quite funny to see. They want easier access to our tools. A lot of them are using them now they know they are there. They didn't know previously. More help and information on IPv6. They see us as obviously an organisation that needs to be a leader there. And they wanted to see some development in the RIPE Atlas service. This came from our stakeholders, so this was quite a surprise there.

On the RIPE NCC operations side, there was quite a lot of support for our operational model, that was very strong, scored 5.66 out of 7 possible score. Members generally felt that the RIPE NCC represented their interests, and the value of being a member was also justified by the cost. These scores were relatively good and that is quite nice to see.

Some key findings, some of the areas I think we need improvement on, because again this is the part we really need to focus on, in the policy development process, there wasn't much that anyone added there; most people felt that it was a cumbersome process but fair. And I think the key findings that I am showing you here are the areas we had taken the most results that were the same, grouped them together and we have put them here. If any of you have looked at the survey and the amount of responses, there are just thousands of lines of responses there that you would see but when you group them, you come up with these key findings.

The communications, there was a lot of support for reducing printed material. Serge covered that. We are taking a look at that. And also, many respondents wanted us to use social media much more than we do. We have actually launched in and we are using social media. People want to communicate with us and want us to communicate with them more in the social media area. So we will be looking at that. That was also slightly surprising.

RIPE database. This, I mean, really everyone that listed any comment here just kept mentioning this as an extremely important service, and they wanted to us stay on top of data quality. That is a big issue there in the RIPE database and that is good for us to know. We will continue working in that area.

IPv4 administration. They actually want us to take a strong stance on IPv4 address administration. This was nice to see, everyone was saying it is the RIPE NCC's position and make sure everything is registered correctly, keep on top of us, push us. This is the areas that we saw people coming back here.

In the IPv6 deployment area, no surprises here, many people said yes, we have a plan but there is no budget or there is no customer request for this, so they weren't taking this much further. From the stakeholder side, they wanted help to understand their role and responsibilities in IPv6, so this would naturally come from a lot of kind of government agencies they want to know what they should be doing to support the deployment of IPv6 as well.
Training and education: There was a request for more on?line and multi?lingual training, and I think the multi?lingual training came from Russia and Central Asia part of our service region. They also wanted us to make sure that the trainers we do send have strong technical knowledge. That is something we are definitely looking at. And they wanted more videos and practical guides on?line. So I think that people really have taken into these short video clips that we have put up there. I know we have got a lot of feedback on them outside the survey, but inside, people were saying give us these short video clips and demos, these work for us.

As far as Internet governance and external relations, I think the stakeholders basically wanted us to reach out to other sectors in other regions, that was very clear from the responses we got from those stakeholders that filled in other ?? filled in the survey. And they also want to see us as a liaison between governments and the Internet community and I think we are actually putting our resources there, so we will continue to do so.

Next steps, I think we as a management team are going to be incorporating these findings into our activity plans and long?term strategies. We have spent a lot of time over the last months understanding what this feedback means to each of our areas to make sure the RIPE NCC functions as well as a whole. We want to improve the awareness of our services because, actually, those people that responded, even though they are members of the RIPE NCC, listed saying "I didn't know you provided half of these services I am going to have to go and take a look at these." I think we want to spend more time on awareness building inside of our membership and also other stakeholder groups. We want to solicit regular feedback. It's great to get a lot of responses back and through all of this stuff working to make sure we get all the ?? wherever we see our members and stakeholders, we can continue to find out what it is you want from us. And here is a link to our survey. That is it. Are there any questions on that?

KURTIS LINDQVIST: No. All right. Thank you.


Then next up is Andrei on the v4 ?? on final addresses. One thing I forgot to say, I forgot to introduce myself, I am Kurtis Lindqvist, one of the co?chairs.

ANDREA CIMA: Thank you. Good afternoon, part of the registration services team at the RIPE NCC. Two RIPE meetings ago I have been giving a brief presentation in this same Working Group about how we were planning to deal with the moments before reaching the last /8, so today I would like to give you a brief update of what we have been doing so far and how we plan the coming period to look like.

Now, we all know that earlier this year, IANA ran out of IPv4 address space, slightly later APNIC reached the last /8 and it's just a matter of time before this happens with RIPE NCC as well.

Now, we have been looking towards our processes in the meantime, and we decided to not change anything in the way that we evaluate resource requests because those evaluation procedures are based on policies set by the RIPE community and they were in line with those policies, so we have been looking at our business processes and seeing a bit how we could finetune them and our goal was to make them more transparent and increase due diligence even more.

With regards to transparency, what have we done? We have published the IPv4 evaluation request procedure. This is a document that tells you exactly what kind of information the IP resource analyst will request you, when you submit a request. Furthermore, we have published a graph that shows the available IPv4 address space at the RIPE NCC. This address space is available for distribution.

With regards to consistency, we have created a single ticket queue like Andrew mentioned a few moments ago. This is quite a change because, before you were used to send the request to the RIPE NCC; one IP resource analyst would pick up your request and deal with this request until completion. Now, this has changed. When you send a request, one IP resource analyst will evaluate your ticket, the message; if there is a need for more information from your side, we will send you a reply. When you reply to that e?mail, the ticket will not go back to the same IP resource analyst but back to the ticket queue, to the bottom of the queue, and when it's at the top of the queue, the next available analyst will pick up and continue with your request. And this is all done in the principle of first?come, first?served.

Furthermore, we have set up an escalation process where very large process request will be evaluated by two IPRAs instead of one. We think four eyes are better than two. And the process is then reviewed by RIPE NCC management.
Now, how do we see the coming months? I had look like, we have divided this into phases. Now, we are, at the moment, in the current phase where we are now. Now, if you look, this is typical working day at the registration services department. All IPv4 requests are handled from one single ticket queue. When a request is ready to be approved, you have provided enough information, the request is justified, we will go and approve it, at any moment in time during that day.

Now, when we will move to the next phase, Phase One, we will move to this phase when we have one month or a /10 left of address space. Now, there are some changes in our processes at this moment, because what we will see is that all IPv4 resource requests will be evaluated by two IPRAs instead of one to make sure that we have higher consistency. Furthermore, escalated IPv4 requests will be handled during the same day, at the end of the day. Currently because the request is going through multiple layers of management it may take a little bit longer but we plan to do this in the same working day.

The call?back procedure will also be slightly changed. At the moment if you go to RIPE NCC and you say, I want to talk to IP resource analyst about my request, we call you back in two hours. This was slightly changed because we will still call you back, but when your ticket is at the top of the queue so also here enhancing the first?come/first?serve principle.

Furthermore, tickets that can be approved will not be approved straightaway but they will be moved to an approvable ticket queue, to a separate queue. Those tickets will be approved first thing the morning after.

Now, on what priority will they be approved? Also, here, first?come/first?served. If you sent your request before another ?? your request will be approved first, so it's all based on the time stamp in your e?mail.

With regards to Phase 2, we will reach this when we will reach the final /8. Now, most of you are aware on the last /8 policy for you who are not, this means that every local Internet registry will be able to receive one /22 from the RIPE NCC.

Now, in conclusions, we have tried to make the process even more transparent and consistent than it currently is; especially Phase 2, though, may result in the fact that the request be answered, your request may be slightly delayed. This is something that we are aware of. However, as the slide says, at this point due diligence is the key priority. We have to make sure that we do things well and of course we will try to minimise the impact on the waiting times.

Do you have any questions?

AUDIENCE SPEAKER: When you talk about approving requests, is it just v4 requests you are talking about?

ANDREA CIMA: Yes, this is about v4.
AUDIENCE SPEAKER: You are dealing with requests as they come in in v4, as and separately and v6 separately?

ANDREA CIMA: You mean currently or ??

AUDIENCE SPEAKER: Going forward, that was the.

ANDREA CIMA: What ?? we understand that it's usually more pleasant for you as an LIR and also easier for us, to deal with the same IP resource analyst when dealing about the ticket. So for IPv4 we have this special queue, yes, one single queue for new and ongoing tickets and this is not currently applied to IPv6, for example.

AUDIENCE SPEAKER: Small clarifying question: About time stamp, is it time stamp one message ?? last message was actually sent or when it hits RIPE incoming ??

ANDREA CIMA: Yes, when it hits us, when we receive the e?mail, yes.

AUDIENCE SPEAKER: When it stops at your network?



KURTIS LINDQVIST: Any more questions? No. All right, thank you.

ANDREA CIMA: Thank you.


KURTIS LINDQVIST: You might get a real proper break between now and AGM. Next is Ingrid.

INGRID WIJTE: Good afternoon, I am also part of registration services. And I will be giving the update on this implementation project this time. Most of you have heard about the 2007?01, possibly more than you would have liked to, but I will refresh the memory for the ones that have forgotten it a little bit.

So what is this about? This policy intends to make sure that the RIPE NCC can confirm that the end users getting the resources, the independent resources, exist, and that they remain to exist, and also, that they comply with the RIPE policies in place.

The policy also describe that this has to be applied both to new assignments and to existing assignments. Now, to make it all a bit more manageable, we split it up in three phases and in Phase One, we focused on the new assignments and we started that in March 2009. And since then, we have assigned over 13,000 independent resources under this policy. Most of it is IPv4 PI assignments, the other bit chunk is AS numbers and the remaining bits are v6 PI IXP and Anycast assignments. So this phase was quite smoothly, went successful.

Then, in May 2009 we started with the existing assignments, the ones that were handed out before this policy was in place. And in this phase, we contacted our members, our LIRs, and we asked them if the resources that were listed under their registry, their LIR count, if these were still their customers or not. And we received feedback for over 25,000 independent resources and about half of those, the LIRs replied that these resources and these resource holders, were still their customers and that they intended to have a contractual relationship with them. A quarter of the resources turned out to be in use by the LIRs themselves for the infrastructure, and the remaining quarter, the LIRs responded that those were no longer their customers, so they would not enter into a contractual relationship with them.

In March 2011 we closed Phase 2 and we started Phase 3 and we started targeting how we call them, the orphaned assignments, and those are independent assignments for which we did not receive contractual documents and registration papers from sponsoring LIR. And an important thing to note here is that in this phase we also included all the independent resources that were listed in closed registries and in the past they used to continue to live under these closed LIRs. But for now ?? for this project, we included them in here, so 7,000 resources were added to the list.

So, how did we develop this process? Well, we are going to contact the end users directly and we are going to contact them in batches and we have prioritised them on the resource feasibility in the global routing table. So we started with the most visible resources. Those might be the most easiest to deal with. And for this, we are using different levels of contact information in the RIPE database that we can find there and we are sending them an on?line form in which they can provide feedback on the resource usage, the holdership, which LIR they intend to have contractual relationship with. And those feedback forms we're following them up manually.

So, what has the progress been so far, the results? To name, we have started with almost 19,000 resources that we had to contact. And since March, we contacted one?third of those, one?third of the resource holders, and we have been sending out 12,000 e?mails out to them, and 25% of those bounced, e?mail addresses no longer valid, so we had to send them further e?mails to the other levels of information that we could find there. Out of those, 40% responded with feedback, and we have then sent about 4,000 e?mails trying to get the information, information from them, verify the details that they were sending us.

For 1,600 resources, we have completely completed this process, and these are divided in three groups; a big part, almost, well, 40%, turned out to be in use by the LIRs for the infrastructure. They requested these before they became LIR but never moved them into their own registry file. 20% of end users signed contracts with either sponsoring LIR or the RIPE NCC. All the registration data, everything cleaned up and case is closed.

Interesting enough, the remaining 40% were actually being returned to the RIPE NCC because they were not being used. So far, we have received a little less than a /16 in total for IPv4 PI, about 300 AS numbers and three /48 v6 PI.

What can we take from this? Well, Phase 1 and 2, those were successful, went smoothly, finished. And Phase 3 is a little bit more challenging but we expected that before we started. And the reason for this is that a lot of these resources were requested up to 15 years ago. We are contacting people that have never heard of us, have no idea who the RIPE NCC is and why we are contacting them all of a sudden. And as we said, a lot of the contact information is no longer up?to?date. So, this was not surprising.

As I mentioned, we got some resources back. This is nice side effect of this implementation. However, this is not the reason why we are undertaking this. The main reason why this policy was put into place, is to check the registry to see if the information is still up?to?date and make sure that it remains that way, and we see now that by doing this, by contacting all these people, we are significantly increasing the data quality of the registry that we have.

And that is it for me. Well, I will take questions if you have any.

AUDIENCE SPEAKER: I have a question regarding this nice letters we received in our welcome package regarding legacy space. Is it also covered under this or is it completely different process?

INGRID WIJTE: No. Legacy resources are out of this project. These are PI or independent resources as we call them, that were issued by the RIPE NCC.


AUDIENCE SPEAKER: Jan. Is there any way for an end user who is PI holder to find out whether the assignment is orphaned or it's ?? let's say, OK, so it has a contractual relationship approved by the NCC?

INGRID WIJTE: Yes, they can. They can send an e?mail ?? well, this one is ?? they can ?? that is a good question, actually, because the e?mail address we use is by reg ID. But if the end user wants to know, they can either ask an LIR to contact us, because if they have a contractual relationship, they are in contact with an LIR. In most cases they will know, they sign a contract with somebody. And if they have no contact with an LIR, as last resort, I will presume they can contact us as at NCC [at] ripe [dot] net and we will move it to the correct group to answer the question. But the preferred option would be through an LIR.


WILIFRIED WOEBER: RIPE Database Working Group trying to follow up on this question, actually. We have seen an idea or a proposal to actually extend the registration record for IP addresses and AS numbers to include an indication about these double sponsoring LIR so I would rather go tomorrow to the much smaller group of the Database Working Group, take this opportunity to sort of pull the community in this larger room, whatever the community thinks this would be a good idea to progress that proposal. Could you just tell us some indication of thumbs up or down or throwing bubbles at me or whatever you think is ??

KURTIS LINDQVIST: If you are in favour, hands up. Against ??

I see ?? I see ten hands.


That is ten in favour, two against and 200 doesn't care. Did everybody understand the question, maybe you want to repeat it?

AUDIENCE SPEAKER: Is there any expected date of the end of the Phase 3?

INGRID WIJTE: No. It will go on for some time. We have now targeted one?third. A lot of these ?? well, the policy also states we should make a reasonable attempt at contacting these people so these resources at some point will, if we can't get in touch with anybody, we will move into the registration process as this policy also describes, but for ?? so that is one?third out of the way, but it will go on for some time. There is still some 12,000 resources to deal with, and the visibility of these resources with each patch is going down, so the likelihood of getting a quick response of these people, so it will take sometime. I can't give any indication of exactly how long. It depends how difficult it is to get contact with these people.


KURTIS LINDQVIST: Do you know how many of these resources you haven't been able to contact is actually advertised ?? how much of the resources that you haven't been able to contact is advertised?

INGRID WIJTE: Well, the ?? what we have contacted so far through the on?line forum, those are high visibility, that is the resource that we started with, and so we have contacted 6300 resource holders. This is not only visible resources. What we have done also as a side activity is, when the workload was lower, is that we contacted the unannounced resources that were not visible at all in the routing tables and those actually amount for the resources that came back to the RIPE NCC. So, these 6,000 resources are mostly the visible resources and in parts the ones that are not visible ?? that were not visible anywhere.

KURTIS LINDQVIST: So it's not space you can reclaim easily?


KURTIS LINDQVIST: OK. Other questions? No. Thank you.


KURTIS LINDQVIST: And then last, we have Daniel to talk about Atlas, and he actually has quite a bit of time on his hands.

DANIEL KARRENBERG: I plan to do this in ten minutes, and definitely I want, at the end, not a 200 undecided whatever. I am going to ask for feedback and for support and I hope we can maybe take the remaining five minutes to actually get some of it.

So, I am going to swim against the stream a little bit here because I am going to argue for spending more money in these times, and as I said, at the end I hope for an interesting discussion and maybe shows of hands.

My proposal is to over and above what we already plan, to do in active measurements, to do a major leap in it, and give each RIPE NCC member at least one Atlas probe and here is why.

Atlas probes are these things, I think everybody knows by now. This is the state of deployment right now. The green curve that you can see there is at 870 and we are actually picking up the green curve in relation to the red curve which is the ones that we put out there, so we are going after the people that are lazy and have not installed them yet.

We also developing those maps, those things that are useful for the community. As I said in my report earlier, it's a little bit difficult because we need to do a lot of groundwork to actually get these things out there but we are getting there. So this is an example of one that is operational right now. It gives you the letter of the closest DNS route name server in terms of PTTs from those services and you can do many more of these.

Next one actually uses some DNS queries to find out which instance of an Anycast resource or Anycast service is actually being used in this case also DNS to K?root and you can see from this maps which one is being used.

That is ?? may give the impression that we are DNS specific with Atlas. No, we are not, this is just because this is our built?in measurements are a bit DNS heavy, you can measure almost anything with it.

What is in the current budget? Well, the probes up, we have very specific targets. By the end of this year we want to have one K probes up, that is 1024 and we are at 870 right now. So I am quite sure we will make it by the end of December. Half year, halfway through next year we want to double that to 2 K, that is 2048, and by the end of 2012, we want to double it again, to 4 K. That is all in the current budget that is before the executive Board.

With the user?defined measurements we have, what we have now is that there is going to be a better test among probe hosts and sponsors, around ten testers to do ping trace route and DNS measurements, test some raw data access and ping visualisation, we want to release that this year. And if all goes well and we are quite confident, we will be able to release it late in the year, otherwise very, very early next year. And by the halfway next year, to have more measurement types there and really nice visualisations. We want to come up with new map products, one ?? every one or two months and don't take maps so serious ?? so literally; it can also be other things that give us a situation awareness of what happens in the network.

We are building communities, so for the probe hosts we have better documentation and things like that and really intensify the two?way communication there. That is all within the current budget. But, so I think what we have proven is that we can deploy this infrastructure. It actually works. The architecture works, we get results out of it. But there are also others in this space and the real benefits come with size, so what I am arguing for, a little bit against the trend, is that we should be much more ambitious than these plans.

A short reminder how it went ?? what is the original intuition. Some of you have seen this. This is a simulation. We just take dots representing IPv4 resources, we geolocate them and we colour them randomly, but in one country, that was Estonia at the time, so if you have 1,000 probes out there, which is roughly where we are now, and we assume that all the probes that we have in Estonia suddenly lose contact or have some other unusual signal in them, with thousand probes you see very little and you see one red dot up there. With 5,000 probes, let's remind you that is about the level that is currently budgeted to be deployed by the end of next year, you have this kind of a picture. Yes, there is something but you can't really quite see it. If we go to 10K probes the picture becomes a little bit more clear. If we go to 20K probes, well, here the naked eye can actually see that there is a signal up there somewhere in Estonia. And if we go to 50K probes it becomes even clearer. So that was the intuition. Again, what we currently plan to have by the end of next year is about this density.

The other reason to be ?? what I propose to aim for is around 20,000 probes. Why 20,000 probes? Well, I use the term buying power here, but that is not really, as a colleague pointed out, that is not really quite clear. The thing is, if we want to buy specialised hardware, if we want to ask the hardware vendor to do a special thing for us because it works better for the proceedings, then this is the threshold where they can redesign a chip and a product around it. That is a minor reason, but still it's around 20K. There is about 50 ?? 15K ?? 15,000 ASNs in the RIPE region that are actually active that we see in RIS and they are about 8,000 it should actually read right now, RIPE NCC members. So to give you the amount, the background why I come to around 20K. 15K ASNs, that means in the stop ASNs, the small ones, we could have one probe in each one and in the bigger one, we could have multiple probes. It's a significant step in the resolution of the machine as I have shown you in the simulations. It is actually, I think, the resolution that is required to have the real gain here, the reliable situation and awareness to the small AS level and the real things like perspectives on DNS, so basically seeing whether, somewhere, somebody mucks with DNS or somewhere mucks with SSL. If we had this kind of machine we could spot the digging out thing; if we saw it, we could verify it very, very quickly.

This is also something that changes, would change Atlas to a really noticeable activity above the many sea level of what was going on. I think it would create tremendous goodwill with the stakeholders and the membership, at least that is what I read out of the surveys and of talking to some of you, and it would avoid us being surpassed by others; we would be sort of ready to the market.

Here comes the additional cost, and this is very rough budget, so if we say we want, by somewhere in 2013, have 20,000 probes out there, it's about a million for the hardware and 700K for the fixed costs, and it would actually mean 12 people to do it and do a lot of other things. If you want the full story it's on RIPE Labs and there is a recurring additional cost here that is around 800K a year. So that might sound like quite a lot of money. On the other hand, if you agree that we should be ambitious, then I think you need to speak up. It's not going to happen, the trend ?? unless a significant group of the membership says yes, this is something we want, yes, we want to be ambitious in this area, and rest assured, you know, I think we should be. But I don't think neither myself or any of my group would take it personally if the membership says no, it's not really interesting. But I feel I have to ask this question because we are at a crucial point where we have to make the decision on whether we want to have something that is useful or something that is really useful. And there go my ten minutes.

So, way forward. The way we organise this is quite clearly that the RIPE NCC Executive Board decides on the budget. So, if you agree that we should be more ambitious with it then you should make yourself heard, comment on RIPE Labs, in this meeting, comment in the general meeting; and of course, if you don't agree, do the same. Also, I am not asking for a budget change immediately. I don't think ?? we are not at the point where we actually can show enough products to be convincing. What I am looking for is reflection on your part as far as your members think about this, is this something that we should be doing, and then in the time between now and the Ljubljana meeting, some opinion building inside the membership. Then the way we organised, once that becomes clear and enunciated, both pros and cons, then in the next year at some point, the Board could evaluate the discussion in the membership and take the necessary decisions in order to do this. So it's not like, oh, we have to decide this now. No, I just want to expose it and have some discussion about it and if you want to read some written words about this rather than slide pack or what I just said, it's on RIPE Labs.

That is all from me.

KURTIS LINDQVIST: Questions? I have one. Where does the money come from?

DANIEL KARRENBERG: Well, if it's the RIPE NCC does this, obviously it comes from the membership. If we ?? we can obviously also think about generating additional money, but I think if we do these kind of things, I think they are investments and one can later on maybe recoup some of it, but to start it with a quote of business plan that says we need to get that many extra amounts of money, I think it wouldn't be right. I think if the membership, if the community wants this, then we should pay it from the membership fees and from the current budget. Mind you, what I am asking for here is not a huge amount in relative terms as far as the budget is concerned, also not as far as the reserves of the association are concerned. But the money comes from you, that is what the answer is.

BRIAN NESBITT: Brian Nesbitt, HEAnet. Well obviously, the money is one part of this. I don't think I have ever heard anyone turn around and say, yes, we are monitoring the Internet enough, we can stop now. I think that the Atlas project is fantastic in what it is at the moment, the scope of it there, the untapped things that could be done with it are hugely interesting, and I think this is a fantastic idea. So from the point of view of money aside for a moment, voicing support on that, please, yes, go ahead with this if at all possible.

DANIEL KARRENBERG: Unfortunately it's not money aside.

BRIAN NESBITT: I realise that. The money will have to be talked about this, but just from a conceptual point of view, yes, please.

KURTIS LINDQVIST: Any other opinions right now? I am going to to be the counterpoint to Brian. As a comparison, say we happen to one of the route servers being monitored by RIPE Atlas so far and we have as many staff running what you want to monitor, I mean we have 12 staff running the route server with less people ?? just put things in perspective. My personal opinion is I agree with Brian, I would like to have everything measured. Do we want it measured for 1.7 million euros? Maybe not. I think that is bit premature, we first want to have the liberals of the probes we have, raising that many we would like to outsource measurements to Atlas, which we can't do yet, come back in a year or two years when we have seen what Atlas can do today.

DANIEL KARRENBERG: What you are saying is I haven't seen the goods, I have no confidence?


DANIEL KARRENBERG: That is fair, and I said that is fair. I think ?? and that is why I am bringing this up now. I am not asking for this to be decided now; I hope that it will not be a year before we can make those decisions, that is why I bring it up. My expectation is, and I will say this openly and be ready to be proven right or wrong is that by somewhere in the middle of next year I want to have convinced you, and if our cycle is that then by the end of next year we decide something, then before we can implement it, it will be the middle of 2013 and I think that is too late, that is why I am standing up here, just to explain this again.

AUDIENCE SPEAKER: Michele Neylon, Blacknight. You will get my name right one of these days, Kurtis. I'd agree with Brian, I think this is a fantastic idea and I would be supportive of it. Obviously, the costs and how that money is being spent is something that would require further information and analysis because you are talking about 700,000 recurring per annum which does seem a tad on the high side. You don't want to end up with an ICON?esque situation where the budget keeps growing exponentially per annum. But the concept is really cool and I am not a geek and I get it, I like it.

DANIEL KARRENBERG: Obviously it's not a small amount of money and definitely an ever?growing budget is not what I want. I don't want ever?growing work for myself. On the other hand I felt it important to make very clear that we are talking about substantial amounts of money. I didn't want to be apologetic about it or basically playing it down. I would rather play it up. And of course that needs analysis certainly by the Executive Board because they have to make the decisions, but I also think by the community, to a certain extent, and yes, I welcome that discussion.

JAN ZORZ: Internet citizen. If you would ask me, if we would go to 20K probes in Atlas network in state, that it is now, I would probably say no, but the user?defined measurements, this feature multiplies the value of the Atlas probes network so this is the best thing that happened to measurements so far. And I would say yes, go for it, but on the budget, you should probably go to the GM as here. So, yes, the user defined measurements is really, really big add on to this network and I am first in line to be a better tester in this.

DANIEL KARRENBERG: Believe me we are working very hard to make that reality and to convince Kurtis and you.

AUDIENCE SPEAKER: Sam Wiler. When you first sold us Atlas ?? and I do host two Atlas probes and I do think it's a useful network and I agree with the last speaker, it would be more useful when you have user defined measurements, but when you first sold us this it was to be the world's biggest measurement network, and today, you are saying that there are others in this space and we need to stay out in front of them. If there are others in this space why are we doing this? Are we really seeing some real numbers that might help?

DANIEL KARRENBERG: That is a discussion, yes, that is definitely a discussion we have to have. The others in this space are basically at the moment more interested in broadband, last mile kind of stuff, which we are not, but it might be that we are taking second seat there but that is not my main argument. My main argument is more like we want this and we need this.

AUDIENCE SPEAKER: Sure. The second is about the budget I think this might be more appealing if the numbers were smaller and as I look at the per unit cost, I think, that alternate probe you are showing up there, the red one that I know you need some custom development on, I just bought ten of those for about 15 euros a piece, seeing 50 euros, maybe there is some room to come down on your budget.

DANIEL KARRENBERG: Yes, there might be some room there but this cost also includes packaging and chipping and the whole logistics behind it. And as I said, I put those numbers high in order to make clear that we are talking about a major investment. Of course, once we go into the detailed design, we will try to save as much money as we can but I don't think we will save half that money, and I can explain it to you privately, and I will obviously also explain it in ?? once it becomes clear that we ?? that there is a chance of this going forward, then all the detailed work needs to be done. But I don't want to take the fifth step before the first one and so I put them relatively high.

AUDIENCE SPEAKER: Very good. Thank you.


DANIEL KARRENBERG: Well, thank you for bearing with me.


KURTIS LINDQVIST: So before I let you go, I want to say one thing I did like about Daniel's proposal, I am ?? there was a proposal with a cost associated and not often we see that and I really like that.

With that, we are done, unless anyone has any other business? Nigel has told me that he will start at 6 o'clock sharp. So, you have 22 minutes, by my watch, to get out of here into the other room and you might be allowed to go to the toilet in between. There you go.