RIPE NCC services

18 November, 2015, at 4 p.m.:

KURTIS LINDQVIST: I think we will get started. Good afternoon, everyone. And welcome to the favourite Working Group of all the NCC ‑‑ all the RIPE meeting attendees, the RIPE NCC Services Working Group. I know many of you come to RIPE meetings just to attend this session and we really appreciate it.

I am Kurtis Lindqvist, one of the two co‑chairs, my fellow co‑chair Bijal is not here as she is home and waiting for the third co‑chair. But hopefully ‑‑ she is watching, she did say she is missing us all and we miss her. But anyway, she'll hopefully be with us next time.

With that, we will get going. We have the welcome, which I have just done, if you didn't notice it. We have a scribe has been volunteered by the RIPE NCC which is Emile. Thank you, Emile. I would like you all to keep in mind the work the scribes do for us, it's easy to forget that but thank you, Emile, for doing the work for us. I haven't seen any more additions to the agenda, so I think we can approve the agenda as is. The last Working Group minutes are posted, which I should have looked up beforehand, but I didn't, which are on the website. I haven't seen any comments on the minutes or heard anything so unless anyone disapproves we can go ahead and approve the minutes of the last Working Group. Done.

With that, we will move into the agenda. And we first with the agenda on the IANA update that is normally on Fridays but we moved it at least to this time to this session, this is one of the services the NCC receives from IANA, I guess. Elise Gerich.

ELISE GERICH: Thanks for allowing me to speak during this session. I did hear it was the favourite session, so I really lobbied to get here because normally it's 9 a.m. on Friday and that is not anyone's favourite session.

So, I'd like to give you the IANA update, and I guess I just push this ‑‑ so, these are the topics I am going to cover, it will be quite short and sweet. I will mention about the RDAP services, the IP address allocation that happened in September. I go to all the RIR meetings and announce what was announced on September, but you probably know by now; in fact, it may even be gone. And then I will talk about the certificates for the Reverse‑DNS, and each route server change that is upcoming soon, and finally, the results of the annual customer service survey that we do every year.

So, the slide that you see up on the slide behind me has a pointer to the RIPE URL that's in the RDAP service that we bootstrap at the IANA. What we are supposed to do with RDAP from the IANA considerations in the RFC from IETF is to bootstrap and that means that we publish your URL so if someone looks up something they can click on your URL and issue a JSON query. We don't support RDAP natively yet; that will be something we will be undertaking, but this is the first step and fulfils the RFC IANA considerations.

As, you know, on September 1st we made an allocation from the recovered pool of the IPv4 addresses and that is the link where you can find the information. Here is a little snapshot of the allocation that was made to RIPE, here the ones listed at the top. These allocations are getting smaller and smaller because the pool gets smaller and smaller. It's done automatically, this is not done on a request basis. Since there is a formula in the global policy, we just wrote a tool that just runs on the 1st of September and the 1st of March, at which point it shifts out the addresses. You can use the tool, if I go back a slide, I think it gives you the URL for the tool, you can run it against the recovered pool at any time and see if you were to run the tool on that day, what the allocation would be.

So, this graph or chart behind me shows how long we will be allocating from the recovered pool if no more addresses are added. Because right now, we apply the formula and you know we can just do that time after time after time, and you will see that we are going to run out in 2019. There was a lot of talk this morning about what to do about the last bits of IPv4 in the policy session. This is an area where you could change your global policy and we could run out faster; "we" being the IANA department. We could allocate a bigger block instead of smaller blocks and that is policy. You guys do policy, I only implement the global policy.

So, the Reverse‑DNS certificates. As you can see, the certificates that we have on the Reverse‑DNS will be expiring in March of next year. So, we had a lot of communications with the various RIRs, came up with a plan, we asked you to send us your new certificates and thank you very much, we got everybody's certificates by early November. That is this November, this month. And last Friday, we completed the roll‑over of the certificates for the Reverse‑DNS. So that is a project that is been completed, along with your collaboration and cooperation. And we appreciated your help in getting this done early.

And finally, this is the announcement of the H route server upcoming change, there will be a change in IPv4 and IPv6 addresses, so that means a change in the hint files etc., etc. This is going to be done on December 1st, so those of you in the audience who run servers and depend on maybe the H root server, please be aware that this will be happening, we have been trying to announce it more broadly and we thought this might be one of the forums where people might be operating the servers that would care.

And finally every year we do a customer service satisfaction survey. These are the results for the RIRs. We send ‑‑ we sent 4,000 surveys out to the broader IANA functions community, and we segment that and it identifies the people who received it as being RIRs, protocol parameters, domain names systems, so for the RIRs we sent 17 surveys, they go to the CEOs and the managers of the registration services, and we got ‑‑ we sent 16 out and we got 7 responses, that is about 44 percent. And we are very, very pleased that you are all satisfied with everything except one person, because that is what that 14% means, one person who responded doesn't like the quality of our instructions. So since it's an anonymous survey, I don't know who that one person is, I don't even know which RIR they came from. I will be talking to those of you in the registry services to try and find out what you don't like about the instructions, and I know I did talk with Ingrid this morning so I have her idea, but I don't know if she is that one person or not, it is an anonymous survey. But we were very pleased with the responses we got from those of you that did respond and we will continue to try and offer this high level of satisfaction and try to figure out what we can do to try and improve the instructions. That is all I have to say for the IANA update and I want to thank you very much for this opportunity to speak at 4 o'clock in the afternoon instead of 9 a.m. in the morning. If you have any questions, I'm here.

KURTIS LINDQVIST: To we have any questions for Elise Gerich? No. Thank you very much.

One thing did I forget to say at the start was that directly after NCC services, we will have the RIPE GM in here at 6:00 I think it is. That means when we are done here we would appreciate if you all can make your way out of the room as quick as possible. I know it's your favourite but you are still going to have to leave. The other announcement that I almost forget to say, historically we used to do several NCC presentations in the GM, they are no longer done in the GM but in the NCC services, Axel's is up next and the following ones reporting back to the membership but we do it in a broader base here instead.

With that, Axel.

AXEL PAWLIK: We will report on something. Hello. Good afternoon. Happy to see you all here. As I expect, this is the second favourite slot in your favourite Working Group, the favourite slot being the open mic. And I will give you some reason to use the open mic later on. So, a bit of update interest the RIPE NCC, bit of outlook into next year what we want to to and also based on what we have been doing and your feedback of course.

As you might have noticed, in our communications so far, we do see increase in membership, so rapid rather increase too, very interesting. We will talk about that and what we do with that.

Some developments at the NCC over the course of this year, looking at our sort of focus for next year and also couple of numbers for the general consumption for those who are not members and would not be in the general meeting, you deserve to get some information there as well.

Membership growth up and to the right, steepening about two or three years ago, more than 12,500 by now. It's a little bit scary but it's also good, lots of money coming in, we can do good things with that.

What we have done over this year in terms of policy, probably, well the one fact is that there is still policy to be discussed and not insane amounts but rather astonishing amounts of policy to discuss; maybe the most exciting there is the Inter‑RIR transfer policy that came through and we have implemented, so now we are looking at that and see what that does for us.

Membership contact, lots of new members means lots of members we don't know yet and we want to get to know them as well and understand what our old members do and know what they are ‑‑ what they would be looking for in our services.

So, we try all sorts of ways to get to talk to them and we find it especially useful that we have started now on the assisted registry checks, ah, we have done 1750 over the course of this year so far, that is not a high enough number because the idea is to touch every member and talk to every member once every three years or so, that was the initial goal. And with 12,000 members you will see that again is a very large number that we need to get to.

But doing that, it's quite popular with the members, it's rather popular with ourselves as well because it leads to new insights and corrected data, it's a good thing. You have told us to go after legacy holders, contact them and see that they register their resources with us properly. I am happy to report that all of the legacy resource holders have been found and contacted and some of them liked it so much they more or less immediately ‑‑ not immediately, but registered their resources with us, 40% of them. It's not the case that the other 60% say we don't want to talk to you at all; there are some of them ‑‑ some of this process is still underway and will be more resources being registered with us the. Overall I think it's quite work‑intensive project but rather satisfying there as well.

And I hope it does what you wanted it to do.

Again, training is extremely popular. Due to our wonderful trainers they go out in all the various regions and corners of our service region. Again from there we get lots of feedback on the face‑to‑face basis from our members who attend. But on the other hand, the demand is so big we said we wanted to set up this RIPE NCC academy thing. Happy to report that that is also quite popular and many courses have been done and will continue working on that of course. Our ‑‑ the SNDs of our trainings have been doubled there.

So, local communities. Now, the RIPE NCC Service Region, as you know, is very wide, wider than this meeting room, even. And it stretches all the way from over there, Greenland, to the Pacific shores. Very wide. And we try to, again, get into contact with our members all over the service region, and we try very hard, on all sorts of various levels: The bigger regional meetings, there are the smaller meetings, there are the other events that we touched also some smaller way. We do lunches with our members when we go places, that is always quite nice. And lightweight member luch, that is lightweight before the lunch but less after. They have been happening in Minsk, Madrid, Paris, Istanbul, London and Manchester still to come this year. Smallish affairs, maybe a couple of dozen people coming, maybe even less. But it's good to get a feeling for who those people are and what they want from us.

Regional meetings we do also, we have done them in Dubai, Belgrade, Tblisi, and Odessa this year. Again, slightly bigger. We try to get them around into various bits that we maybe haven't gone to before.

Support and participation in local NOGs. We try to stimulate where they don't exist yet, the set‑up of local or regional national maybe, operators' groups but where they exist we are happy to come and present and be part of the event there as well and as those people mostly are all members as well, it's great to see them there.

Also for many years we are building the RIPE community, we want to get fresh people in and fresh ideas in and we have been doing that for, with renewed focus for quite a while now. We see that the academic involvement at the RIPE meetings with nice presentations is certainly on the rise, it's a good thing. RACI, is quite successful, six successful applicants for this meeting. We had a BoF session of sorts yesterday evening. Maybe remarkable is four of those applicants have been invited to present at the plenary by the Programme Committee, that is very nice as well. And it shows to be a somewhat sustainable thing as well, because two former RACI participants have come back here to speak at RIPE 71.

Since this is a good thing, I hear from my Board members and from the wider community had a we should continue to build this out and put some effort into this to make it even a bigger success so we will be doing that as well.

Billing procedures. Now billing is not such a nice thing; you have to pay us and we have to prepare all that stuff. We don't want to do too much work on that, and in the overall scheme of events we try to be more efficient and very effective. So, we have improved here quite significantly, I think the process time line has improved, the average payment term has improved also by, I think, about 20% from 40‑odd days to 30‑odd days, it's quite nice. We also have been asked to promote the annual invoicing cycle and have been doing that as well. And that has also been taken up so the amount of quarterly invoices goes down, less work, less tickets, all that type of thing. Electroni invoicing, who needs paper, really, these days? So that is all a good thing that we have done that.

Further update, we have heard about Atlas if you were in the room, it's growing, over 9,000 active probes, 150 anchors are in place and a mail fly another large number has come on line so maybe more than 150 by now, it's great to see that that project is taken up and contributing so much, we see interesting talks about it as well, hack on this, great thing, so happy to talk about that and we are being told to implement this and that and that and that's as much as we can do, we do that, over time. So new features there.

K‑root, we have taken up again to expand the footprint in sort of a reasonable way, that also has been a success and we have doubled the number of the K‑root clusters out there since last RIPE meeting. It's good. And generally, infrastructure, we seem to generate a whole big lot of data with all the infrastructure that we have this place, Atlas and the RIS collectors and stuff like that, we have worked on that and implemented that in terms of efficiency and cost as well, so it's looking good. Better performance and efficiency and more storage.

We were somewhat busy with the IANA transition, you have heard about that in the plenary before; I won't talk too much about it but it is quite a large process, it's not quite finished yet, we will continue working on that. We think we have done the numbers part as much as possible, but still we participate in the other communities as well, general ICANN accountability is important to us as well. So we continue to spend effort in that. I think it's a good thing.

Yeah, another thing that is nice there is that, well, we have been sitting on the sidelines in ICANN deliberately for quite some years, they don't talk about numbers that much. We bring the information but it's not entirely popular with all the folks, so occasionally we talk to fairly empty rooms about what we are doing here. This whole IANA transition thing has led to, as the names Cross‑Community Working Groups, stuff like that, our visibility, the numbers community visibility in the ICANN stakeholder groups has risen considerably and generally that is not a bad thing. Well, you see, you seen the archives, if you haven't seen the presentations there. It will continue for the next couple of months.

Focus points for next year, again obviously, the registry is the top most priority on our list. Again, RIR accountability is a good thing, we have started on demonstrating that we are accountable to you folk and to our members to ‑‑ we have written that down, we have ‑‑ we are doing this joint RIR accountability project where all the RIRs look at that stuff and publish some information about that as well, and it all happened partially ahead of the IANA transition because we thought we might need that and yes we do, and it's great to have that firmly on the way.

RIR system. The RIRs together with IANA are operating the global numbers registry is one thing. This we are doing for more than 20 years, it's very important, the stability that have is very, very important so we put lots of focus into keeping that running and looking into the future then and keeping that stability.

Outreach, I talked about that, that is important, we do more of that next year of course as well. I have been talking to by many people already that we should double the frequency of the regional meetings and things like that so we will look at that and do what we can.

So statistics, analysis of all the data that we have, is something that we want to kick into overdrive next year and get some very nice results out of that.

Local communities, again, this is important to us, we want to continue to do this. So, neutral setting, bringing what we are to the smaller pockets of our service region as well. It worked quite well, it is really, really nice to come to a place that we have never been to and there is a roomful of people and they apparently know us and are very happy that we are turning up there and we are happy they are turning up. They make us very welcome and we do our utmost to project the welcome back to this community to those smaller pockets where people might maybe never come to a RIPE meeting but still we try to make them welcome in the community, show them the mailing lists and this is how you take part and whether you travel or not, it's not that important; you are welcome part of the community.

So that is something that I think we all should be doing when we talk to those guys.

So for next year, right now, we already have plans to go to Turkey, to Albania, again to Russia, Armenia and more places as the plans sort of gel and become more concrete. Local NOGs throughout our region, we mentioned that, of course, we find many of our members in there and we really love it when they coordinate among themselves, also in those regions within their countries, so we do support them a little bit financially as well but maybe even more visibly with attendance and presentations.

I mentioned that growth is somewhat crazy. We are dealing with it quite well. Over the last couple of years we have become quite a bit more efficient and you might have seen that over the last two years, three years, we have not increased our workforce that much or basically at all. So, we do see ‑‑ we plan a little bit of increase for next year but it will not be in direct relation to the membership, still will be efficiencies and the like. So, working smarter, more efficiently, more effectively, that's on our agenda.

So, expenses, again we are using expenses where possible, also spending money wisely, I think you see that in the budget as well. In the end long‑term sustainability of the RIPE NCC working for you is also priority there.

Continuing engagement, a little bit more of similar things, the ARCs, we need to do more of the registry checks, meetings come into play, mailing list and the like. Maybe the biggest thing for next year in terms of hearing from our community is the RIPE NCC survey, we do them every three years, and next year will be the end of that. So you will get the survey. A relatively small one this time, we want to minimise ‑‑ not minimise but to be efficient and not ask you for hours of work in answering the thing. But it will turn up up in your mailbox, do take note when it happens and do fill it in and give us a good feedback ‑‑ good return rate there. And good feedback is great but not so good feedback is also welcome. We need to know what we are not doing so well so we can improve that. Membership, feedback, that is what we do.

We obviously ‑ I hope it's obvious, we act on the feedback that we have been getting from previous services and informal and more formal interactions with you. We act on that and we think it's also important that we show that to you. So we will report, of course, back from that survey and any other improvements that we do.

Further focus: Good governance and accountability, I mentioned that a little bit earlier. This will never stop. But we will continue chipping away and getting it better and better, making it more transparent; what the board does, what we do, what our governing documents look like, what you have to fill in, stuff like that will continue and I think it's important not only to us but also to you that it's very clear how this organisation works to the outside world as well. It's nice if we know it, but if the others don't, then, they ask questions so they better be clear about that as well.

So, we will continue to develop and of course ramp up the registry ‑‑ assisted registry checks, we have to do that.

And maybe the biggest excitement buzz in the office is about, oh, we are going to move next year, we hope, middle of the year, it's a big project. But that is the plan to move offices. The idea is to move out of the slightly cramped and inflexible, lovely little offices and go to central station in Amsterdam, even a bit more central. We will show you pictures when it happens.

Like I said, probably second half of next year.

Financial overview, coming to a close here, you decided already ‑‑ our members decided already to cut the membership fee down to €1,400 which is quite brilliant. We do expect ‑‑ well, we sort of guesstimate there, a 12% increase in membership for next year and we try to be very reasonable and very clever about our projections this. On the other hand we don't want to be too conservative or free‑wheeling so 12% will probably ‑‑ will be what it is. We will add five additional FTEs, we try to higher them over the course of the year. Regular operational expenses went up by about 5%, I think you will see 9% overall increase in the expenses and the other bit, 4%, have to to with the change of offices and of course that is an investment that will reduce future cost, that is the idea.

And of course, for those of you who are really into non‑IP numbers but financial numbers there will be more in the general meeting and it will be presented more interesting stuff there and I will give feedback over the presentation about the activity plan and the relative changes there as usual.

If you have any questions, I don't know whether now or later but we are here. Thank you.

KURTIS LINDQVIST: Thank you, Axel.

We can do questions now if anyone has any direct questions for Axel? Lack in questions. We can also do them later.

Next up is Jari from the IETF and IETF endowment updates.

JARI ARKKO: Good afternoon. I am with Ericsson Research in my day job but here as the IETF chair, and I have Greg with me, the Chief Financial Officer at ISOC. And we are going to talk about IETF endowment and before we get into the details of what that actually means, I wanted to go through a bit of where I see the IETF and Internet technical world devolving to.

Starting with technical staff, being an engineer I am mostly interested in that perhaps ‑‑ I don't know about you, but at least from my perspective the speed of evolution seems to be speeding up or is faster today than it it was some years ago, and it seems to be speeding up, topics like the Internet of things, software development ‑‑ software defined networking, growth and mechanics of encryption, the web technology and some examples of the web technology, we recently released the HTTP 2 version of the protocol. You get fairly quickly deployed in the world, so that is ‑‑ interesting thing to see and kind of a surprise that that can happen so fast.

But we are also talking about even bigger changes. So if a few years ago someone had predicted seriously talking about replacing TCP with something else in the web protocol stack, then I would be very doubtful. But that is indeed what we are discussing now. So interesting things happening.

And also it's ‑‑ mention one thing about encryption or the growth of it, that has been a hot topic for us at the IETF, how can we increase privacy and what kind of mechanical things we can do to the details of that in order to help it. But it's also a big issue for Internet operators and ISPs and we just recently had a workshop between the IAD and the mobile network operators who are a little bit concerned that the kinds of things that they are doing in their networks in terms of network management are making ‑‑ made more difficult with most of the traffic being encrypted so we are trying to deal with that.

But technical changes, we are also seeing non‑technical change so we see more broader communities engaged in the technical work around the world, in different forums but also at the IETF. The most obvious example of that is that the growth of the OpenSource movement particularly has become very mainstream and part of the business of mainstream corporations and it's not the case today that we work on specifications and people go off and implement that in product and everybody happily uses those products. We actually work together in a far more entangled ways, during the development working on specifications and OpenSource at the same time. And lots of companies working together at the same implementation. So that is an interesting thing.

We have many non‑technical audiences that we need to engage with or talk to. Prime example of that is the IANA transition of course, that you guys also have been working with, and, you know, I find myself talking to politicians about that every second day or something. And it's kind of interesting and it's learning experience I am sure for all parties, and of course, we do hope that that thing actually gets completed like really soon now and I don't have to talk to politicians any more, well, not really. I think this is actually just an example of an area where policy and other topics interact or intersect with technical topics and we have many examples of similar things and slightly smaller scale.

We also have increased interest for participating in technical developments around the world. I guess as the whole world comes on line, lots of people realise different issues and they want to have an affect on that.

And one example of that: From our perspective at the IETF, our next meeting is actually in Buenos Aires, Argentina, in beginning of April next year, that is our first meeting ever in that part of the world, which is very exciting to see and we have a growing participant base over there, so that is good.

I also wanted to mention that we have in the last couple of IETFs being running the hack‑a‑thon, which is an opportunity to ‑‑ broach the specifications world and OpenSource world and we work together with people on various topics and that has actually been a fairly big success and lots of people participated in that and good results have come out of that today. So if you get a chance to join our meeting, definitely join it and try to come a couple of days early so you can participate at the hack‑a‑thon and build your own team or find other people or students who are interested in this topic and do something worthwhile for those two days.

And also, a little bit more generally, our ways of working are changing, both globally and then at the IETF. So globally already mentioned the OpenSource thing, lots of people seem to be using GitHub and other types of tools for collaboration, we have over‑the‑Net collaboration in general. At the IETF this has meant a number of things, some Working Groups are changing their work style. For instance the ‑‑ Working Group that developed the HTTP 2 protocol, worked on GitHub to a large extent in their workings. We have way more interim meetings than we used to have, instead of every three or four months, have physical meeting and discuss something. Now we have this model where active Working Groups are actually meeting almost on a weekly or bi‑weekly together and working together and the big meetings are just checking out how things are going or discussing bigger topics. We have remote participation for the people who are very, very busy who can't travel or can't afford to travel. Many remote presentations in our meetings these days. We have focused events, the workshop that I mentioned with the IAB and encryption, or growth of encryption or the hack‑a‑thon, these are examples of that.

And this is about IETF endowment. I need to talk about the IETF finance a little bit. The overall model of that is that we are funded by the participants, about 40% ISOC, about 30% and sponsors, 25% varying a little bit, over the years. The cost will vary depending on what we actually do. We do see or we have some experience by now that reaching further out in terms of going to different places in the world or providing more abilities to connect over the Net or remote participation will be more costly. So that takes more resources. And a little bit more generally, given that we have these three funding sources, we are looking for an even broader support base to sustain the evolution of the work style that we have and the broader reach. And one way of doing that, that we are setting up is the IETF endowment and I will let Greg talk about the details of that.

Greg: As Jari has said, the endowment was an idea that the IETF and ISOC both came together with to spread out the sources of funding for the IETF for now and for the future. And that's quite a long‑term goal of both of us. So we came up with the thought of an endowment. Now the endowment has a couple of characteristics that are very important:

First, it's funded for a specific purpose and in this case it's for the long‑term support of the IETF. And no other use will be allowed for these funds. The other thing is that the funds are kept in perpetuity, so they don't go away, again it gives a longer term funding source, but it also means that you have to have a fairly sizable amount of funds in order to ensure that the resources, the interest in the earnings off of those funds are significant enough to help the IETF.

The Internet society acts as the legal home, if you will, the administrative home of the IETF. Therefore, we are also going to be the home of the IETF endowment. However, what the trustees of the Internet Society are here to do is only two things: One is to make sure that the endowment is held in perpetuity and that the principle is not invaded; that it is used only to fund the IETF and that they also find investment advice and then consult with what is called an endowment council on how to invest the funds. Once those funds are deemed to be usable by the IETF, they are turned over to the IETF and the IETF has full responsibility for use of those funds.

ISOC is leading this. However, one of the things we would like to do is work with the community so more of us work towards the funding of the IETF.

What we are looking for is for the RIRs and their member businesses to join in with us. This is going to be ‑‑ have to be a multimillion dollar effort. Again, if you figure that 5% of the funds may be available each year, what we are shooting for is at least 20 million dollar fund. So it's not insignificant but the people we are looking to is people like RIPE, to be leaders in this and to be participants as endowment councillors. So we invite a significant leadership role as well as a financial role.

And with that, I guess we will take questions.

KURTIS LINDQVIST: Thank you. Any questions?

MARTIN LEVY: So question about the endowment fund. Isn't that the job of the Internet Society? And don't they already do it? And if they don't do it for you adequately shouldn't they do it for you adequately?

Greg: I can tell you about the funding model and Jari can tell you about the IETF's position on this. ISOC puts in two million dollars a year towards the annual operating goal of the IETF, which is about 35% of that fund. We are happy to do that and we fully support the IETF. But it's their goal to get diversification in their expenditures, in their fundings source.

JARI ARKKO: Maybe I can also respond. From my perspective, I think the situation is ISOC is looking to increase ‑‑ they are supporting us, they are likely going to even increase their support, but from IETF's perspective we would actually like to have long‑term, a little bit broader support beyond those three resources that I mentioned earlier.

MARTIN LEVY: Isn't IETF essentially run by ISOC ‑‑ not run, that is not the right word, in existence because of ISOC? That is not quite right the word. There is a structure that none of us really truly understand except it seems what you are saying is wrong.

JARI ARKKO: I am not sure that follows. But the situation is that the ‑‑ ISOC is our formal organisational home, so contracts gets signed by ISOC, for instance, when IETF does something. But obviously IETF is its own independent entity in the sense we have our own leadership and opinions, community drives, discussions and ‑‑ that is the model and there is a separate aspect of where the funding comes from, which is the three sources that I mentioned.

MARTIN LEVY: Just for the record, I mean, fully support the IETF, fully support Internet Society, don't support extra bureaucracy being thrown at anything. That is where I come from.

JARI ARKKO: Yeah, I mean, I understand the position very well and the extra bureaucracy, I don't think that is present in this particular instance. It's really about actual funding.

GREG: I might say the reason we set up the endowment as it is, is it can be controlled by the IETF as far as what expenditure is used for and we don't have a lot of overhead and a lot of corporate focus on that kind of thing.

NURANI NIMPUNO: Two comments, perhaps more. First of all, I think it's fantastic to see how you are trying to evolve the IETF and find new formats for it and I think that is necessary if you want it to be more inclusive and reach new audiences but actually to keep momentum in work, that is great to see. I will only agree with Martin on the part about that no one fully understands some of these structures. And for the rest I will actually disagree. I think it's a good thing to have diversification of funds, and the way I interpret it, it's not necessarily that you are trying to get more funds; you are trying to diversify it; the IETF is there for the benefit of the greater community. And I think that community should support it, and I think the RIR community is one of those communities. So I think what you are doing makes sense; it makes sense as a governance structure and it makes sense also in sort of a greater context outside of the ‑‑ of here, so I think it's a good thing. Thanks.

JARI ARKKO: Thank you.

AUDIENCE SPEAKER: Salem NCC Board. It looks like you are looking for long‑term stability for the IETF, but are you also looking for independence? And if so, financial independence? And if so, why do you then decide to remain with ISOC if you are wanting to be a bit more independent?

JARI ARKKO: That is a good question. I think our goal is to broaden the actual financial support and as Nurani mentioned, not necessarily increase so much, but more that you actually get it from multiple sources. And, you know, that's in some sense the factual thing. The way that this is in detail organised, I think, matters a little bit less, and we do need an organisational home, as the IETF ‑‑ ISOC has worked very well for us and they ‑‑ not only fund us, of course, but also have been a very practical arrangement. IETF does actually make all of its financial decisions by itself, we have administrative committee who decides on how we fund particular things in our operations. And even there is this organisational home that someone actually signs the contract from ISOC's perspective, it's not been an independent issue for us in that sense. But of course, having broader base for funding gives you some actual ‑‑ it's not independence but it gives you a little bit maybe more room to manoeuvre or ‑‑ I mean, that is a fundamental basis for existing ‑‑ in some sense independently even if the practical organisational details are still are as they are.

GREG: I would like to add to that that we have a very good working relationship and ISOC is very, very happy and its major funding commitment is to the support of the IETF. However, this is a long‑term endeavour, and if we don't start now, then it may be too late when we need the money. Right now, we figure this is a time when you don't need money, that is when you go after to try to get that stability that we are talking about.

DANIEL KARRENBERG: Internet citizen, for full disclosure, I used to be an ISOC trustee and chairman of the ISOC Board, but I haven't been for a long time. I think I will go a little bit in the same direction as Martin Levey was. I think if you come for this, want to collect this amount of money and if I understand it correctly, you want to collect 20 million US dollars, which is 10 times the current budget ‑‑ current contribution of ISOC to the IETF, correct? The annual? Yes. I think you need to have a little bit more meat to the story. I think what needs to be made very transparent is, when you make this pitch, what is the current budget of the IETF, with numbers; where is the money coming from; and if you go ‑‑ start wanting to collect money you have to have a much better story on why the people you are asking to contribute, should do that. I agree fully that the IETF does great work, the IETF is very important and all that kind of stuff, but to me personally, this story is much too vague to actually make up my mind about this.

Also, I think, and that's my personal opinion, this is, in essence, a question of priorities for the Internet Society, for ISOC because ISOC is very well funded, and has always said ‑‑ and that one of the important ‑‑ its important roles is to support the IETF. So, you have to also answer the question, is why the Internet Society is unable to actually set up this endowment because ‑‑ and feed it, you know, initially, because once it's an endowment the independence thing is satisfied. So conceivably the Internet Society could we say we put up 20 million dollars over this time and then there is an endowment, the stability is guaranteed and you don't need money from anybody else, because what you are saying is we want money from other people so we could use some of ISOC's current, you know, resources to do other things.

So we will also have to explain why you are spending that money that you get in on other things rather than the IETF and why you want to us make up that difference. Thank you.

JARI ARKKO: Many answers, I will try to point out a couple of things. We do have the numbers, we can provide them off‑line, the IETF budget is about 5.5 million on a yearly basis and as for a detailed break down, we have some projections for cost ‑‑ or the development of those funds over time. Regarding the ISOC as noted, this is not necessarily a move to reduce the ISOC's contribution, it's rather going to increase as part of doing this. So, it is not an issue of moving something that ISOC already does into some other purposes, but it's rather a request for others to participate in this as well. So, you know, it's not that the ISOC is running away, he is still here and funding us quite well. You could of course, ask if they should do even more than that, that is a fair question.

GREG: I would also like to add, ISOC is heavily funded by its holding Public Interest Registry, and as we knew a long time ago public and all of the gTLDs would have a point where they weren't growing as much, so we have reached that point with them. It's not we are looking at not funding the IETF, it's that we are looking at spreading across the industry, a major event to get a capital campaign going so we have long‑term capital campaign. ISOC will always be investing in the IETF, but we think this is financially wise.

AUDIENCE SPEAKER: Hello, I am here with ISC and also affiliated with ISOC through the San Francisco Bay area ISOC chapter. My question is directly in terms of trying to expand participation for IETF for diversifying the participation, how are you doing that, is that through ISOC or are you reaching out directly to even chapters or the community at large, so that it's more diversified, thank you?

JARI ARKKO: That is another good question. The answer is we are doing multiple things. As you probably know, the ISOC has fellows and programmers ‑‑ additional people are brought into the IETF meetings. But that is not the only thing that matters. So obviously we are going further in our meetings than we have ‑‑ used to gone, that is one aspect that just having a meeting somewhere is not the thing that drives participation, it's to build on actual participants who are interested and are interested long‑term in this thing. So, part of the way we are trying to get there is to engage with potential participants and existing participants in these areas and there is many tools in there, some groups have met in those areas and have discussions with the local engineers and we have gone out of our way to reach out and talk to more communities than we would have done in the past. And also have programmes like the mentoring programme at the IETF to bring in new participants or bring the new participants up to speed as fast as possible so there is multiple tools and I guess traditional work over the network instead have come to the meetings is also an important tool for some of the participants.

KURTIS LINDQVIST: Two quick comments.

MIRJAM: I think it's a quick question. If I am not mistaken the RIRs are actually paying membership money into ISOC that is specifically dedicated to fund the IETF, I think that is a system that the ISOC came up with a few years ago to allow that specific dedication. What is going to happen with that money? Is there going to be transfers or accounted for or I didn't hear you mention that?

GREG: Mirjam that is a membership and platinum level, which is highest level we take, but it's for operating expenses, specifically for the IETF operating expenses. This is a different idea, a capital campaign to build up a separate bunch of funds.

AUDIENCE SPEAKER: Martin Swissy, RIPE citizen. Very modest proposal: Help us to help you. So if do you write some paperwork for this project, why money is needed? For what? What you want to achieve more than do you have now? So that decision‑makers will take that document, OK, you deserve it, of course fiscal things, is there any return? All of that need to be done, otherwise decision‑makers will not hear you.

JARI ARKKO: That is a fair request and we will get that done.


And next up is Wilfried and Ondrej with the Arbiters Panel.

ONDREJ SURY: Hello. How many of you know that the RIPE has Arbitration Panel? OK. That is something like half of you. So this is quick update for those people who don't know what Arbitration Panel is and what it is good for.

So, basically, there is update ‑‑ well, the Arbitration Panel has several functions, I will tell you about that later. There is some changes to the Arbitration Panel and to the Arbitration procedure that has been done at the general meeting and we also handled some cases that Wilfried will speak about later in this presentation.

So, we met last time on the RIPE 70 and if you want to know more about the Arbitration Panel, there is URL you can click on and there are all the documents available.

The arbiters panel is appointed and dismissed by RIPE Executive Board and approved by general meeting and the Arbiters Panel consists of at least 7 people, but no more than 15, and we are ten people right now. This is a list of the people and there are more biographies. Are there any arbiters apart from me and Wilfried here? If you can stand up. There is Wilfried and he is alone. But it's up here, David, James, me, Kurt, Keith, Nick and Wilfried, are the Arbitration Panel right now.

So, what was the scope of the arbitration procedure, there are two scopes, one is settlement of disputes and it could be between members and the RIPE NCC about the RIPE NCC Standard Service Agreements. Between two or more members and this was extended, the legacy in RIPE NCC regarding decisions of the Executive Board for the implementation of new RIPE policy.

So how the procedure works, there is action from arbiter but he is asked to do the arbitration. Then the parties have to submit information. After that the arbiter issues his ruling and makes a ruling and executes that.

So, the second scope is assignment of new Internet resources for RIPE NCC because RIPE NCC is the maintainer of the resources. So we are asked if, when they ask for new IP addresses or AS numbers, the Arbitration Panel has to make a consensus if it's OK.

So, this is it and I would welcome Wilfried to speak about, well, he cannot speak much because it's private but there are some arbitration cases last year so here you are, Wilfried.

WILFRIED WOEBER: Thank you very much, Ondrej. And I think one of the major reasons why I am standing in front of you is that I was the arbiter who had the most recent case on his table. The next couple of slides are going to lead me through some comments to get you some sort of feeling, like how big this endeavour is and what the structure of the cases and the frequency is.

As you see here, the last update that the arbiters offered to the community, to you, has been given in 2011, so that is quite a while ago. And as you also can see, the frequency of cases coming on our desks is pretty erratic, and we have more stable set of cases or more stable frequency coming in from the RIPE NCC asking for our approval for self assignments so when they actually take resources out of the pool and use those resources for their own projects or for their intermediate like I think one of the regional meetings, one of the training sessions was actually in need of resources, so instead of just helping yourself from your own big backyard, pile of resources, the RIPE NCC applies to the RIPE arbiters to get the second opinion and sort of to get the approval to take a piece out of the resource pool and use it for their own mandate for their own project.

So, as you can see here, one of those requests came in in 2012 and the other one came in during this year. The number of regular cases where there is actually a dispute between two members of the RIPE NCC or between a member and the RIPE NCC, they are really scarce. They are not coming in on a weekly basis. So we had one case that was concluded this year and this was the case that I had to work through. There is another case this year which started, if I remember correctly, in the second half of this year and it is still ongoing.

In case you want to have some sort of overview about the time lines, about the number of cases and the dates when they came on our desks, this is the list since 2004, so be aware this is a list for ten years more or less and it fits on one slide. So, this is the settlement of disputes. One of the things that is also interesting interest my personal point of view is that more than 50% of those cases were related to misunderstandings with regard to billing so it was not the majority of cases where there was a dispute about which party is actually entitled to use a particular set of resources or like in the case that we had this year, whether it is appropriate that the RIPE NCC terminates the regular service agreement with a member. This is the dispute resolution and the list for the applications by the RIPE NCC for resources, for their own use is even smaller, it's even shorter than that.

That's just the statistics to give you a feeling this is not something we do on a daily or regular basis. There is one thing that I would like to maybe put also from the procedure point of view in front of you and in context. This is the list of arbiters which are on duty at this moment, and the procedure actually allows two modes of operation and one of the modes of operation is that the parties having a dispute, agree on one arbiter, trying to manage the case and to come up with a ruling. If this is, for whatever reason, not felt to be appropriate, there can also be a set of three arbiters being assigned to the case. But that's ‑‑ I can't remember that we ever had a case where this was really the ‑‑ requested, so usually it's one arbiter. And in order to spread the load amongst the group of us we have a sort of rotation mechanism. So if you haven't had a case for quite a while, you move up in this list and when the next request arrives on the table of the RIPE NCC by way of a form on RIPE NCC website, then the person on the top of the list is, by default, asked to take care of this case. On the other hand, the complaining party also has the option to ask for a particular individual to be the arbiter in that case.

So, that is the reason as you can see here, why I am at the very bottom of the list because not ‑‑ not because my name starts with a W but because I was the last one to work through one of those cases this year.

Going back to the sequence of the slides of the presentation, we also had informal meetings for the last two or three years, always now and then, next to the RIPE NCC ‑‑ RIPE meetings here where we exchanged information about our experience with some cases, and this is one source of part of the proposal to change the procedures, because we found that the time lines which are prescribed in the procedures are sometimes quite a bit tight and it sort of pushes whether you want that or not, it pushes the parties which have a disagreement and it also pushes the arbiter or the arbiters. So, one of those proposed changes as outlined in this URL here on top of the slide, is actually trying to take care of that in giving us and the complaining parties a little bit of more freedom to wiggle and to have a little bit more time to come up with documentation and to work through that.

So, just a summary of the proposed changes: One of them is the time frames. The other one is that, in the current procedure, the RIPE NCC is expected to act as some sort of secretariat, as some sort of support infrastructure for the arbiter, and as we have seen in the summary here, settlement of disputes, the majority of cases, more than 50% actually involve the RIPE NCC as one of the parties in the dispute. So there was a feeling that it's maybe not really appropriate in each and every case to have the RIPE NCC to provide the secretariat function, when, at the same time, the RIPE NCC is one of the parties in the dispute. So, the clerical support in the new version, if it gets adopted, would also allow to have a third party to provide that service to the process and the rest is editorial changes and clarifications and that sort of things. The proposal has been on the website for quite a while and the final discussion will happen right after this session in the general meeting and it's part of the things the membership is expected to vote on. And that's almost part of the questions and I think I am reasonably within time. So, I can actually go back back for a minute to this slide, because the number of cases which are listed here are the cases where it really started to become a formally documented, formally requested case, where one of the arbiters actually took the trouble to together with the two complaining parties to work through the procedure and come up with a ruling but the only mandate we have as the arbitration also and ‑‑ and also within the procedures, the only mandate we do have is to look at complaints or disagreements in the framework of the service agreement of the RIPE NCC. So, some parties, and I can easily understand that, some party get the feeling, always now and then, if something goes wrong on the operational plane, if someone on the routing plane, then the Arbiters Panel ‑‑ arbitration procedure would also be a vehicle sort of to take care of route hijacks of mistakes in router advertisements and that sort of things, so those things are never going to become a formal case with a formal ruling. But still the group of arbiters, of course, tries to take care and to help in these situations, by talking to, for example, ISP and the complaining party, but you as the community, you should be aware of the fact that we may be able to help but we don't have a formal mandate and the RIPE NCC does not have the means actually to sort of enforce a particular decision or a particular recommendation. The only thing the RIPE has this mandate to abide by the ruling when the RIPE NCC is one of the parties in the dispute. And that gets me to the end of this presentation. Thank you.

KURTIS LINDQVIST: Thank you, Wilfried.

Any questions for the entire Arbiters Panel, I guess? One.

AUDIENCE SPEAKER: And I was wondering, so you said that it is not enforceable in court, the decisions that you make?

WILFRIED WOEBER: No, what I said is that if it does not involve a dispute which is regarded to the service agreement or to the registration of resources, the RIPE NCC does not have the mandate and the tools to enforce it. But, there is also part of the procedure, if a formal case is worked through and there is a decision by the arbiter or by the set of arbiters, then the decision is communicated to both parties, having the dispute, and either both parties comply with the decision or they have, if I remember correctly, a time window of two weeks or something like that, to not accept it and not to implement it but to apply for ‑‑ to court. So this is not final in the sense you don't have any mechanism to get a second opinion or to sort of push it up higher into the legal enforcement or court system.

AUDIENCE SPEAKER: So this is really non‑binding arbitration?

WILFRIED WOEBER: It's binding in the sense that the RIPE NCC agrees to do the thing which is compatible with the arbitration and we also ask the or require the other party in the dispute to agree to abide by the decision. But we don't have any legal mechanism to enforce it.


AUDIENCE SPEAKER: Ondrej again, I wanted to thank the secretariat people at RIPE NCC for all the support we have from them. So thank you.

NURANI NIMPUNO: Probably a very uninformed question: Is there a roll‑over procedure for the appointees in the Arbiters Panel? They are appointed by the Board and approved by the GM, but is there a renewal sort of process?

WILFRIED WOEBER: For the set or the next one to be assigned to the next case?

NURANI NIMPUNO: For the set.

WILFRIED WOEBER: For the whole thing? No, it's the responsibility of the management of the RIPE NCC and the General Meeting to either remove arbiters, maybe because if they are not doing their job, or whether they are still on the list, but if a case comes up, not being sort of available to do it or stepping down and we had quite a few of those.

KURTIS LINDQVIST: I was going to say there have been roll‑overs.

WILFRIED WOEBER: Yes, but no fixed mandate for about three years and then you have to step down.



KURTIS LINDQVIST: Next up is the Operations update.

ANDREW DE LA HAYE: Good afternoon. I'm Andrew De La Haye, COO, RIPE NCC. And I will be giving an operational update today. Usually I go into quite a lot of detail on numbers, if people have the ‑‑ if people want some additional information on detailed numbers please come and see me outside after this session. But I try to keep it a bit more high level and give you bit more trend analysis on these numbers and what we see happening in the RIPE NCC.

Operational teams ticket volume, actually we have seen as Axel already mentioned and tremendous membership growth, double digits this and next year. But fortunately with quite some efficiency gain and process improvements we have been able to stabilise the overall ticket load. I do have to say we see a shift in ticket load in customers services and registration services from the more general allocation related tickets to more transfer related tickets and fraudulent cases and in that case I also have to say that those tickets are way more complex than the regular allocation and assignment requests we got in the past.

From an improvement perspective, I will touch upon a few of those and Alex will go into more detail afterwards. We worked on the membership, the new membership process to make sure that we get all the relevant information upfront, so that you are not con fronted with multiple questions throughout the process and that was very helpful and we got really good feedback on that. Also we are creating RIPE database objects throughout the process for new members automatically.

Number resource request forms we have changed as well and there it's mainly the streamlining of the processes and an API for ultimation, although that one has not been used a lot. Furthermore, we work on integration of services and creating seamless process, for example through our single sign‑on which we have.

There were also operational improvements very much related to the customer survey, the members' survey we did in 2013 and as already mentioned by Axel, next year there will be another survey and we would like to tell you ‑‑ you to tell us which areas you would would like to see improvements in so that we can make sure that we take those into account. These are three examples of changes we made, RIPE database object creation through the IP analyser, a very basic IP management tool we have in the RIPE NCC; new maintainer password re‑set functionality where there were quite some questions we got from the membership, we also had quite a high ticket load on password re‑set, hence we made changes in those processes. And there was also question for example, better documentation on transfers that have taken place, and all the transfers are accessible via one page.

Allocations from the last /8: The intent of the last /8 was to have /22s for new entrants and facilitate v6 adoption. We see quite a strong correlation at the moment between the membership growth and the amount of /22s we are allocating. We do see also an increase in non‑traditional LIRs as we call them, so you have to think about enterprises, banking, industry and those kind of members, so not the traditional ISPs so to speak. And 60% of our existing members have a /22 at this stage, but an interesting number is that only 40% of the old LIRs which are the ones that were member before run‑out in 2012, only have /22s. So, if you translate that into a different sentence it would mean that most /22s are going to new entrants and new members.

We worked pretty hard on the legacy policy implementation mention bid Axel. These are some high level figures, I don't want to go into too much detail here, if you have any questions please let me know. In general, the amount of legacy resource in our region is about 11 .7 /8s, we contacted quite a lot of those resources or the holders of those resources, and in about 2000 cases we got in contact with them and they have been registering those resources, which is a grand success, I want to say, for the registry and the quality of the registry.

From a number perspective, that means that 43% of all legacy address space has been registered in the region and under some kind of contractual relationship, either directly with us or sponsoring LIR. This percentage might differ a bit in the future because legacy holders have the ability to cancel any contract, but we are also discussing with quite a few legacy holders to register their address space at this stage. So it can go up or down but this is generally what happened over the last year.

ASN assignments, this is something I would like you to be aware of it, was discussed in Address Policy earlier today. We are going to start issuing referenced ASN numbers which was discussed prior to this in database Working Group and routing Working Group. Currently, we have about 2,900 of those 16‑bit AS numbers and if we look at the current allocation rate, that gives us about two‑and‑a‑half to three years of address pool in AS numbers. Interesting I think is also to mention that two‑thirds of the AS numbers we hand out are 32‑bit.

Not for getting the IPv6 allocation trend, this is the allocation trend but similar you will see with assignments on v6. It's a stable amount of requests we get in per year, around 250 per month. And actually, recently this month, we allocated the 10,000 block and that means that about 75% of our members have an IPv6 allocation out of those 75%, 40 to 50% is seen on the Net.

Then, an interesting one, also quite a lot of discussion in Address Policy this morning, for those that did not go, it might be interesting to look back to those discussions, is the amount of allocations and the amount of ‑‑ or the transfers of IPv4 allocations. On average, the teams are dealing with 200 transfers a month at this stage and actually we have two inter RIRs transfers in progress, which means we received the request from ARIN to take these two into account and we are currently working with the holders to make sure they are taking place.

Next to the amount of transfers we see, there is mergers, acquisitions and consolidations and we have seen a tremendous growth there, more than 260% growth over last three years, which is quite extensive. And stuff the team need to deal with. And most of these mergers and acquisitions and consolidations are quite labour intensive.

That brings me to the resource transfer process and the changes we would like to make. We are working on making it easier for to you transfer, marriage or do the acquisition process with the RIPE NCC, we would like to reduce the number of transfer types, we will simplify the process through a wizard, we reduce paperwork for LIRs and we will make it easier to maintain an accurate and current registry, which is one of the main topics. I do have to admit that we have to find right balance between securing your resources ‑‑ and I will touch on that later ‑‑ and making it easier for to you transfer resources.

Because resources registration hijacking is taking place as well as we see some disputes between transfers and LIRs accounts, first the hijackings, we are increasing the effectiveness of our processes and by doing so we come across quite some of these cases, we do more due diligence on those, we improve the maintainer re‑set process to ensure there is less issues and we are educating resource holders how to protect their resources and to make sure your contact information in the portal is correct and up to date.

Number of investigations, some general numbers, we have 51 cases currently ongoing and 45 cases we finalised this year so it's about ten cases a month we have.

Then there is the disputed transfers and LIR accounts and this ties back to the hijacking. We do see disputes in transfers, it's not a lot; we did about 4,000 transfers so far, and it's a handful of disputes we see. But those disputes have quite some major impact on those it concerns. Mainly, the disputes come out of outdated contact information and LIR contacts that were no longer working in a company and transfer resources out. We are working on improving our due diligence process for that as well so we are looking constantly in the process where we can make changes to make sure this is not happening but we do see some of these cases and it's also one of the items we would like to bring across time after time, that it's up to you to keep the data up to date and how important it is to keep this data up to date.

Assisted registry check, that is one of the processes we are running. One of the RIPE NCC key responsibilities is to maintain an accurate and high quality registry, and hence the checks we do. We get in contact with the members, it's quite a painless process, but one of the things we do is we highlight what the contacts are that are currently registered, for example and we give some other pointers to the LIRs and to the members.

Striking is that as of today we did about 1750 of these ARCs and in one out of two cases contact data has not been updated correctly and there was quite a shock, because it can have quite some consequences for you. So please have a look at that and we will try to help you through the processes if you have any questions on updating or changing your registry data. Other information we share with LIRs and members as such, there is lots of questions on getting more IPv4 addresses of course and how does v6 work, and we give them answers and guide them through the processes.

So all in all, a brief update on where we are and if anybody has more detailed ‑‑ need for more detailed information, please feel free to contact me. Any other questions?

AUDIENCE SPEAKER: You were discussing this morning in the Address Policy Working Group a number of issues, so there is a feeling in the community that there are two processes which are being abused, and we are talking about multiple layer per company and mergers and acquisition which basically make the, in some cases make the policies completely clear ‑‑ and are these issues and problems they pose, are they supposed to be discussed here in the Services Working Group or in the General Meeting? Because the understanding is that they are more related to a business process ‑‑ more related on how RIPE NCC does the thing. Not policy itself.

ANDREW DE LA HAYE: I can report on certain numbers, of course. But I think that this is really a membership topic on how to deal with this and I think the General Meeting in hopefully half an hour, will be the right spot to discuss that. Thank you.

NIGEL TITLEY: RIPE NCC Board, yes, it will be discussed at the General Meeting, likely.


And then the almost last presentation for the day is Alex with the ‑‑ not with questions, was it change in ‑‑ improving services, sorry.

ALEX BAND: I am the Product Manager at the RIPE NCC. Basically, I want to hook into what Alex and Ondrej have talked about earlier and do a little bit of a show and tell, show you what we have done in order to make our services easier to use, make your life easier and get stuff done quicker.

Let's see, because this is sort of how life used to be. We had forms and you had to copy and paste them and this is, of course, pretty error prone because everybody can make a mistake, you have have syntax in there, anybody can make a typo and on the receiving end, the people who have to process this data, they have to do lots of copying and pasting into the back‑end systems, getting the information into the registry, getting certain information into the RIPE database and all of this can make ‑‑ ‑ be error prone and that kind of thing. So, in improving all of the services that we offer to you, what I am going to show you is mostly the services that we offer to the outside world, at the same time we are working very hard in hooking in all of these services and the back end as well, making sure that we streamline all of the processes that registration services does, that customer services does to make sure they can work as efficiently as possible. Because ultimately, if you look at the membership growth that we have and the amount of people that we're handling all of the incoming tickets with, we are actually doing really, really good as Axel has mentioned, and I would like to show you how we have achieved that.

First thing I want to show is you the signup process and it started very simple because on the left‑hand side we just give you an overview of everything that you are going to need in order to complete the process so you know what it is that you are going to need, it will just take you ten minutes to fill in all of the details. And and you just have to create an account and if you don't have a RIPE access account yet, you fill in your name and e‑mail address and we will create an account for you.

Now, during the signup process, you know, we have things like auto‑complete and form validation, all of the basic stuff would you see in a modern application to help you and guide you through the process, and if at any point you get stuck you can hit the live chat button and talk to one of the people at the RIPE NCC to help you out, so you can complete the process without having to interrupt your session.

Also the up loading of documents has been completely integrated and this goes into our back‑end system. We have entire documentation system that stores all the documentation you send to us and previously this was an email attachment and we had to upload it into a certain system and this is completely seamless and automated within our internal systems as well.

So, in summary, what we did with regard to the signup process, account creation is completely streamlined, we do form validation, auto complete, uploading your documents, live chat is integrated but maybe most important of all, what used to happen when you sign up as a member, would you get an e‑mail and this e‑mail will be like welcome you are now a prospective member of the RIPE NCC, we are going to be sending you a large stack of member and you have to put down lots of signatures and here is a crash course in understanding the RIPE database, you have to create an organisational object, a person object, a maintainer, you are going to to have to understand the concept of taxis, etc., people were like, what? What is this? You can imagine that for somebody who is completely new to this organisation, that you are like what did I get myself into now? So it's like a six paragraph plain text e‑mail trying to explain to you what it was that you had to do. But that is no longer the case any more either, because once you fill in all of the details in the application form, you just hit submit, we still send you the paperwork because we still need real signatures,but once we've accepted you as a member, all of these objects are created in a database automatically, based on the information that you have provided in the application.

So, then you are all happy because now you are a member, and then you go and request some resources so that is the next step that you are going to take. It's a bit hard to see on the screen shot, but again what we do because you are logged in with your RIPE NCC access account, we know a lot of things about you, so we pre‑fill the maintainer, we pre‑fill the adminC, you just have to select the size of the allocation that you want, and say in which country you are going to use it, and from that moment on, everything is rather simple. We try to restrict all of the options that are available to you, for example, if you are an existing member and already asked for your last /22 that option simply isn't available. And again, because ‑‑ because we restrict those options, and we make these things easier on the user side, on the back end we don't get requests that ultimately can't be fulfilled. So that saves one more ticket and makes our work more efficient and makes it easier and less frustrating for you.

So, once you have actually requested your allocation, then it's time to sort of make assignments. And the thing is that a couple of years ago we integrated single sign‑on into the RIPE database but it was purely on a functional basis; could you go into the RIPE database, edit your object and automatically or add a SSO account to your maintainer so you could also use that in web updates but it was also a very complicated and convoluted process and we are working on that right now. In fact that database functionality is in currently in release, so we are going to deploy that next week, all I am going to show you now, you can see it live in the release ‑‑ of RIPE database and next week you can use this.

Let's see. When you now edit an object in the RIPE database we will simply ask you for your current password that you use in order to update your objects. Once you have successfully completed that we will just say, oh by the way, now that you are logged in with your SSO would you like to associate this SSO account with this maintainer so you can use it for future updates? So you no longer have that dependency on MD 5. Again, you have no idea, if you look at the ticket load that our Customer Services Department deals with, just regarding MD 5 password resets, it's not even funny. It's not even funny. Because if you look at it, a maintainer is like this anonymous box and it contains a password that is not linked to anything or anyone, it is just a password that somebody chose and shared with their colleagues. So it's this, if somebody requests the password re‑set, yeah, doing the fact‑checking whether it's actually a legitimate party asking for the re‑set, making sure that we don't re‑set the credentials and give new credentials to the incorrect person, that that is causing a lot of work for us, still today. But it is slowly ultimately over the years decreasing because every time we do an MD 5 posit re‑set, we replace that with SSO account and that is something that you can do yourself, and of course, you get all of the niceties of modern authentication like two‑step verification and that kind of thing.

So, making SSO an integral part of what have we do in the RIPE database is also part of this strategy of seamlessly aligning all of our services. So essentially now you can just go to, log in with your access account, do something in the LIR portal, check your Atlas measurements, update objects in the RIPE database without having to fill in a single set of credentials again. I realise what I am talking about a lot is what is happening on the website. So our mission moving forward is to make sure we have this authentication system available in all of the other update methods as well so you can use this over e‑mail or API, etc., etc.

The other thing that we want to do is sort of lift out the maintainer because a maintainer in the RIPE database is a really, really abstract concept for people. It is like this thing, it's this look that is used to protect other objects or something, I am not quite sure what it's supposed to be. A lot of people ‑‑ a maintainer is a ‑‑ it's hard but you also have two flavours, you have this sort of personal maintainer that you only use to protect your own objects that you only have access to, you also have the shared one that you use to together with your colleagues to, for example, manage the assignments in the RIPE database. And making this distinction clear is also something that we have actively implemented in the RIPE database.

Plus we are now showing you all of the available authentication methods. We give you things like auto complete on names, auto complete on organisations, so it's really easy to find you as a person and to find you as an organisation, etc., etc.

The other thing that is causing a tremendous amount of Customer Services tickets and a lot of frustration with users is that once you have this link of two objects that reference each other, it's next to impossible to delete one or the other if you want to get rid of it. So, now, if you select to, for example, delete your personal object, if the only reference that exists is to a maintainer and a maintainer only reference back to that person so it's this only reference to each other, just say OK, if you are going to delete one, we will just automatically delete the other as well. The same with unreferenced data, you think it's unreferenced but really there is some lingering objects in the RIPE database that block from you deleting your object. Now, we will just actively list which objects are preventing you from deleting or sort of completing your deletion. So, that is now linked, you can easily go to it, change your references and go back and delete the object that you want.

And then this is the last thing I want to talk about, is we are now also putting all of this streamlining into the registry changes, because it's currently for ‑‑ for Registration Services the largest ticket generator, it's people dealing with complex name changes, transfers of documents, consolidations, mergers and in a lot of cases when a user comes into the process, they go like, I just want to change my name, can you just change my name and can you make sure that that is reflected everywhere? We go like, well, why are you changing your name? Can you ‑‑ is it just changing the name of your own company and then you may be going into a process where you think it's a company name change and then halfway they go like, no, actually, this is because we are merging with another company, oh, OK, so it's a merger, so different rules apply and people would have to go back to the initial process. So we are trying to make sure that we ask the right questions, making sure that people are guided into the right process from the get‑go and this is tricky to get right. So, what you are looking at here are mock‑ups, this is still in development, we are trying to find best way of getting this out there. So we ask you what it is that you are trying to to, name change, merger, acquisition, consolidation, etc., etc. And when it comes to transfers, we will make sure that we offer you an interface that only allows a ‑‑ lists the transfers or the resources that are eligible for transfer, making sure that you don't submit a request that ‑‑ for transferring resources that you can't even do. So all of that is coming up later in the year. We will first create a wizard for you to deal with the simple cases such as plain name change of the organisation itself and making sure you can transfer your resources from one party to the another is something we will roll out as we go from there. You can expect the first functionality to be available by the end of the year. This is something we are trying to do to make your life easier and in the meantime it makes our life easier. I hope you like it. If you have any questions, please let me know and you can also find us in the hall. Thank you very much.


Any short quick questions? No. OK. Thank you. That brings us to the very last agenda point which is the open mic phone and this is your chance to provide feedback, comments, proposals, suggestions, to the NCC and the services that the NCC provide to you as members. Anyone? No, thank you all, we will see you all in Copenhagen and please leave the room quickly. Thank you.