IGF 2021 parliamentary track | Online session - The Internet governance ecosystem

IGF 2021 Parliamentary Track - Pre-IGF Meeting

 Internet Governance Ecosystem

 23 September 2021

 

>>SORINA TELEANU:   All right.  Let us formally start the session again.  Good morning, afternoon, evening to everyone, and welcome to the first preparatory session of the IGF 2021 Parliamentary Track.  My name is Sorina Teleanu, I work with the IGF Secretariat, and I will be moderating this session, but before we go into substantive matters, we have a few housekeeping to make so everyone knows how to navigate through this, throughout the discussions.  

 We have the session interpreted, so we have English, French and Spanish, and captioning is also available, and I will quickly give the floor to my colleague Luis to explain briefly how this goes.  

 Luis, over to you.

 

>>LUIS BOBO:   Thank you very much, Sorina.  So basically you will find a button called "Interpretation" in your Zoom application with a globe.  You can click on it and you will see symbols of English, French, or Spanish.  You also have the option to mute or unmute the original audio should you want to hear the floor language.

 There is also live transcription, so you can click in the option next to the bottom, and you should see an option "Show Subtitle."

 [Speaking in French]  Is interpreted.  You should have a button for interpretation on the application, on the Zoom application.

 It's very easy to click on it and click on the option for French.  You can also listen -- you can also put this on silence, the original version, I mean, if you want to just only listen to the French version.

 You also have a transcription, a live transcription, in English, and for that, you have the transcript button.  You can click on it.

 [Speaking in Spanish]  Finally, I will make the same explanation in Spanish.  You can go for Spanish language so that you will listen to the simultaneous Spanish interpretation, if you wish.  You can mute the original audio if you do not want to hear the Zoom original audio in the background.

 You also have the transcription in English.  You have a button for that.  Thank you.

 >>SORINA TELEANU:   Thank you, Luis.  And hoping that was clear for everyone but if you have questions you can post them in chat and we'll make sure to assist you if you face any sort of difficulties.

 And one more thing.  This session, as you have probably already noticed it being recorded.  It is not live anywhere, and the recording will be made available upon request of participants or other Members of Parliament who might not be able to join us today.

 And as we go into the more substantive discussions throughout these two hours, we kindly encourage everyone to be an active participant in the discussion, make comments, ask questions, basically contribute to the debate.  And when you wish to intervene, just please raise your hand in the Zoom chat, and we'll make sure to give you the floor as soon as possible.

 And to briefly explain how the session is going to go, we have brief remarks from the IGF Secretariat and the Inter-Parliamentary Union.  Then we go with presentations from our key contributors, and after each presentation, there will be a bit of time for quick follow-up questions for discussions.  And then we also have more ample time for discussion around the things we will be discussing today towards the second part of the session.

 And that is basically it about the session, and we can go into the discussion, and starting with the welcoming remarks.  Andy from the Inter-Parliamentary Union, and you have the floor.

 >>ANDY RICHARDSON:   Thank you very much, and a warm welcome to everybody joining the session today, to all the Members of Parliament and parliamentary staff from different countries.

 So this is the beginning of the Parliamentary Track of the 2021 Internet Governance Forum.  The Parliamentary Track is co-organized by the IPU, the United Nations Department of Economic and Social Affairs, and the Polish Parliament that will host the in-person event in December of the IGF forum.

 So the IGF has grown into being an incredibly important place for the discussion of Internet policy, and we believe that it's vital that Members of Parliament are a part of this discussion along with governmental, civil society, and private sector stakeholders.  The IGF is a very open house, and it's very keen for parliamentarians to join these discussions.  So that's really our goal today, to make the connection between the policymakers and to try to foster discussion and exchange between different parliaments and the broader Internet policy community.

 We have five preparatory sessions lined up that will take place online.  This is the first, and subsequent sessions on privacy, on content policy, on AI governance will help to prepare the discussions during the in-person event and will also help to draft the parliamentary outcome document from that session.  I believe there are opportunities for parliamentarians to engage in a trans-national discussion of some of these issues that we understand as being very pressing in your countries.

 So the online sessions, we strongly invite everybody to take part in that and then to consider joining the IGF forum itself, which will take place in Katowice, Poland, on the -- the IGF will be the 6th to the 10th of December, and there will be a parliamentary meeting on the 7th to 8th of December, currently planned as an in-person event and also the possibility to join remotely for those that cannot be there in person.

 So with that, I'd like to thank you again and look forward very much to this discussion and to hand over to my colleague Chengetai Masango from the IGF Secretariat.

 

>>CHENGETAI MASANGO:   Thank you very much, Andy.  And good afternoon, good evening, and good morning, ladies and gentlemen.  Thank you very much for joining us here today, the first in the series for this year.  This is actually our third year that we have had the Parliamentary Track.  The first Parliamentary Track that we've had at the Internet Governance Forum was in 2019 at the Berlin IGF, and there we had 56 parliamentarians from 56 countries joining us, and they also produced a message which we called the Jimmy Schulz Call, which was named after a parliamentarian who sadly died just before the meeting.  And this message was sent to the United Nations Secretary-General and also distributed amongst the Internet Governance Forum Multistakeholder Advisory Group and the stakeholders and also the national and regional chapters of the IGF.

 And last year, though it was virtual, we did also have the parliamentarian track there was well and it was focused on building trust in a time of COVID-19 response and post-COVID-19 recovery.

 So this year we've expanded it a little bit with these intersessional activities.  And as Andy said, this is the first session that we're having, and we thought that it would be very good and also at the request of some parliamentarians that it would be also very good to have a track that explains a bit how the Internet works.  And we thought that this would also be useful to help people who are, of course -- since you do make public policy, instruments that affect the Internet, if you had an idea or better idea how the Internet works, it would better inform the policy-making process.

 So as Andy said, we have five sessions, four of them which are online and then the penultimate session is going to be at the IGF in Katowice.

 We do have funding available through the United Nations for parliamentarians who wish to come to Katowice.  So if you do wish to come to Katowice and you are from the Global South, please do contact us at the IGF, [email protected], or you can also contact Andy at the IPU and we'll get in touch with you and see whether or not we can fund you to come to Katowice.

 So as we said, this main session basically focuses on how the Internet works.  It's more a technically focused session today.  And as we know, that there is a growing trend for later sessions being put forward by national and regional parliaments that concern the Internet directly.  So we have plenty of issues about privacy, cybersecurity, content policy, the Domain Name System, and some of these regulations do have effects on it, but for both positive and sometimes negative.

 So with that, I will finish off by thanking you once again and pass it back to Sorina.

 

>>SORINA TELEANU:   Thank you, Andy and Chengetai.  And with Chengetai having already introduced the content of the session, let us go directly into it.  And our first speaker is Chris Buckridge from the RIPE Network Coordination Centre.

 Please, over to you.

 >>CHRIS BUCKRIDGE:   Thank you very much, Sorina.  And good afternoon, good evening, good morning to everyone.  It's a pleasure for me to be here.

 As Sorina said, my name is Chris Buckridge.  I'm working for the RIPE NCC.  I'll get into what that means exactly a little bit more in my presentation.

 I'm going to share my screen here, so if you just give me one second to see if I can make that happen effectively.

 Sorina, perhaps can you confirm that that's on screen now?  Great.

 So we have three presentations in this first section of today's presentation, and it's looking at the Internet -- some fundamental elements of the Internet infrastructure.  And the way we've broken it down, and I think it makes the most senses to look at numbers, routing, and the Domain Name System.  And so in a sense, that's looking at numbers that actually make up the network, the routing that gets information to different points of that network, and the Domain Name System which makes this all a little bit more usable for human beings.

 So I'm here to speak about the numbers, which it should be not a terribly lengthy presentation because, really, they're just numbers.  They should be and often are the most boring part of all this.  Unfortunately, but perhaps fortunately for this presentation, there have actually been some interesting and sometimes contentious governance issues that relate to those IP addresses and Internet number resources, as we often call them, in recent years and in recent months.

 So I want to start really with very basics.  Internet Protocol.  And this is a slide that a colleague of mine has been using for a number of years, and it basically says -- is what it says:  Internet Protocol is everywhere.

 Basically, when we talk about the Internet, we're talking about the Internet Protocol.  The fact that the Internet Protocol is what addresses the network, is what defines it as the Internet.  There are other ways of networking, other protocols, other technologies, but when we talk about the Internet, the Internet that has reached into so many aspects of our lives, we're talking about a network based on Internet Protocol.

 And so in that sense, an IP address is what defines an Internet connection.  You have an IP address, and that's how you're receiving or sending information.  That's -- That means you're on the Internet.  And that's a fundamental building block of any Internet-based service.

 Now, it probably seems a very common-sense assertion, but each address has to be unique in the context of the network.  Now, the reason -- and I'll explain a little the common sense there, but if your -- if you have an address, same as the personal system.  If you have an address where you're sending your information to and that address is both in France and in the United States, you don't know where that information will go.  That's true in the case of the Internet as well.  You need to know that only one end point, the end point that you want to reach, is actually using and identified by that IP address.

 So that need for uniqueness, that need to ensure that only one individual, one end point is using a given IP address is where a lot of the governance of the Internet number resource ecosystem stems from.  That's why this was built this way.

 Now, if we look -- this is another visualization that's often used by people, and it's called the OSI model which is basically published in 1984.  It's a way of visualizing and thinking about networks.  And what you see there in the middle, what we've called the narrowed waist of the Internet is IP, Internet Protocol.

 So what this is going back to there is the IP really defines the Internet.  The Internet has many applications, as you can see at the top level there.  It goes much wider.  It could be email it, could be phones, it could be worldwide web.  At the bottom layer, that information can travel in many different forms across many different physical networks, whether it's copper wires or fiber or radio waves, but it's the fact that the information travels via the Internet Protocol that makes it the Internet and makes the Internet Protocol so important to the Internet.

 So I said this was very simple in terms of numbers.  We obviously try to make it less simple by having two varieties of those numbers, and this is where it starts to get a little more complicated.

 Most of the Internet that we use today, certainly have used for the last two decades, is based around something called IPv4.  So when the Internet was first being developed, very early on, in the 1970s, IPv4 was defined as, okay, these are IP addresses.  They are based around 32-bit numbers which means there are around 4 billion, about 4 billion unique addresses and the thought was this was enough to see us through.  We don't see this network becoming bigger that would require over 4 billion, just over 4 billion addresses.

 Now, as it happened, the Internet did grow much bigger than that, and it became clear really in the early '90s, if not before, that this was going to be the case; that there was going to be the need for more IP addresses than IPv4 would actually allow for.  So of course they developed a successor protocol, IPv5.  No, not really.  That would be too simple again.  Called IPv6.  And it was developed in the late '90s, this time around a 128-bit address.  So that means you have two to the power of 128 unique addresses, which comes down to I think 340 trillion trillion trillion addresses.  In any case, more than we need to really think about.

 It's written differently, so when you looked at an IPv4 address, as you can see here, the example is 192.0.2.130.  Probably quite familiar.  Becoming increasingly familiar is an IPv6 address which you can see written down at the bottom here using colons, and numbers but also a few letters in there.  And that's just a way for engineers to identify what IP address they're using.

 Now, the issue that was considered but not solved in the development of IPv6 is that IPv6 and IPv4 are not directly compatible with each other, which means that -- the idea and hope was that the Internet would transition from a network based around IPv4 addresses to a network based around IPv6 addresses.

 Now that hasn't happened.  IPv6 has been being deployed and -- allocated and deployed on networks since certainly the turn of the century, but it's been quite a slow uptake.  And you can see on this map, actually, how that's visualized today.  I think this was an image that I grabbed from the APNIC website back in June.  You can see that there are some patches of green there, and so this is a -- represents a percentage of the network users that can actually use IPv6.  IDN doing very well, North America, parts of Europe, Brazil doing quite well, but whole swathes of the world really quite low, very low use of IPv6 on their networks.

 Now, that's become a problem partly because the IPv4 address pool has actually exhausted.  So we're at a point now where IP addresses, which previously were able to be obtained to use on a network on an as-needed basis, have now become a scarce resource.  So rather than something that can be given out relatively freely, people are actually having to scrounge, buy, sell, IPv4 addresses.  And that's having an impact on the governance space of the Internet numbering resources.

 So the governance of this space is primarily managed through a system of what are called Regional Internet Registries.  And so I mentioned before, and I'll get into what it means that I work for the RIPE NCC.  The RIPE NCC is one of five Regional Internet Registries in the world.  And these registries developed over time.

 When IP addresses were first being used and handed out, the registry was essentially a notebook kept by one man, Jon Postel.  What we do as Regional Internet Registries today is actually not that different from what he did.  We keep a record of who is using which block of IP addresses.  We also provide that information in publicly accessible WHOIS databases.

 The RIPE NCC was actually the first of these regional registries.  In 1992, some people in the RIPE region in Europe reached out to the U.S., to Jon Postel and colleagues there and said we would like to set up our own registries for Europe and the regions surrounding.  Us that was followed by APNIC in the mid '90s.  ARIN, then LACNIC, and most recently AfriNIC which was established in 2005.

 So each of these things, and this is an important point about the governance, each of these RIRs that I have on screen here represents two things.  It represents a registry, so the RIPE NCC is the organization I work for that's acting as this registry, but it also represents community.  And so in the RIPE NCC context, we've made it a little bit easier by referring to that community as the RIPE community.  So the RIPE community and the RIPE NCC.  In the other regions, you often just hear it referred to as APNIC community and APNIC, or LACNIC and the LACNIC community.  But those two different parts actually work together in -- to ensure the governance of this system works.  And this takes us to a slightly more complicated graph here, which shows how that works.

 At the top you can see the Regional Internet Registry.  So that's the entity and the business entity which has staff, which maintains databases -- the database which actually allocates the IP addresses.  So those number resources, we hand them out to an RIR membership.  If you want to obtain IP addresses for your network, you become a member of the RIR.  You pay a membership fee.  You are able to obtain those addresses.  It's a relatively straightforward business arrangement in that sense.

 Now, the conditions, the policies under which the RIR will give you those addresses are actually determined by the community.  And this is where the multistakeholder model comes into play.

 So as you can see here, the RIR membership will be a part, perhaps a significant part, of that community that makes the policies, that decides how -- how IP addresses will be registered and distributed.  But they're not all of it.  These are open communities in each region which also include people from government; from law enforcement; from regulators; from business, not just the ISP business but also other aspects; from civil society, people who are taking note and advocating for human rights, for instance; and from the technical community, so people who are interested in what the technical processes are and how they can ensure that those work best.

 So that community actually sets the rules for how an RIR will distribute its addresses.  And the RIRs also then serve as the secretariat for these communities.  So we actually help to facilitate the meetings that take place twice a year, we facilitate the mailing lists and the forums where the discussions are had about policy, and then we also maintain the public database.

 Now, there is one other aspect here which is actually not too often called into play, and that's that we have global policy as well.  Each -- So each region sets its own policies.  Now, this was set up in the early days in what was quite a prescient mood because it was considered that each region would probably have different policy needs.  Now as it's emerged, that's been very much the case.  The policy needs and the policy situation in Africa, AfriNIC for instance, are quite different to ARIN in North America or to the RIPE NCC in Europe.  But there are situations that require a global policy, and in that case, as you can see here, the process requires that all that policy be proposed in all five regions, all five regions agree to it, and then it comes down via what's called the NRO number Council, and is then implemented by the IANA, which is the manager of the global IP address pool.

 Now, I wanted to include this partly because it shows the relationship here to ICANN, and you'll hear more about ICANN from Adam Peake later on, but it's important to start to understand the way these different entities and also the different elements of numbers, routing, Domain Name System, are actually linked together, because it can get a little complicated.

 So I want to do just one more slide here before I let us move on to the next one and, hopefully, some discussion as well, which is to sort of talk about some of the principles and the challenges that we're seeing here.

 Now, first two important to get out of the way.  The principles that this system adheres to and is built around, one, an accurate up-to-date registry of Internet number resource holdings.  Having that registry, having it public, having it up-to-date, knowing that it actually reflects who has responsibility for what IP address ranges is a really fundamental principle.

 The second fundamental principle is that commitment to open, transparent, inclusive, bottom-up development of the relevant policies.  So it's the communities that actually use and are affected by these policies that develop the policies themselves.  It's not coming from top-down from a government --

 [ Open mic ]

 It's coming from consensus of the affected community.

 So with those principles in mind, there are a few key challenges that really emerged in the last few years in relation to this relatively simple job.  One is the exhaustion of the IP address pool.  What that's meant, as I said, is that what was a system for basically handing out IP addresses as needed to the people who said, "I have a network that I want to address, I want a number," has -- it's shifted from that to a situation where people are saying, "I have a network I need to address.  I don't want to do it with IPv6."  For whatever reason.  "Where can I get these addresses?  I'm willing to pay for them."

 So that's this second point, the emergence of a market in IPv4 addresses.

 Now, that's not necessarily a bad thing, and it has been discussed and --

 [ Open mic ]

 >>CHRIS BUCKRIDGE:   Okay.  It isn't a basis of policy.  It's that our communities have discussed and have agreed on.  So there is a legitimacy to this market in IPv4 addresses, but it is quite a different approach and a different system.  So it actually presents some challenges for the system we have.  And on the next point here is one of those challenges:  Commodification of IP addresses.  The fact that these IP addresses now have a value, monetary value, are creating incentives for fraud.  So now you have a situation where people are actually trying to deceive the RIRs, deceive the system to get those IP addresses because they have value.  For the Regional Internet Registries, adapting to that new reality is something that's still ongoing and something that's been not straightforward.  It's also obviously seen a lot of additional interest from stakeholders in the law enforcement industry -- sorry, law enforcement sector and from governments, because once you start to have fraud and criminal activity, then things get very interesting to the public sector.

 So the next one is uptake of the IPv6.  This is something that the RIRs are working with their communities to address, but it is a situation that was not anticipated in the initial -- when IPv6 --

 [ Open mic ]

 -- was developed, and now everyone is looking at ways how do we encourage IPv6 adoption, how do we encourage the transition of the Internet from an IPv4 or primarily IPv4 to primarily IPv6.  And that's something that really involves a lot of different stakeholder groups.

 And then the final one I want to address, and this is really quite a challenge, is when RIRs and their operations come into conflict with local or regional regulation.  And this is obviously a reason that it's so important to be talking to parliamentary stakeholders, to others in government and regulators.

 The RIRs were established at a time when there wasn't a whole lot of government interest or attention in how the Internet was being developed, how the Internet was governed.  Governments have obviously taken much more interest in recent times.  That interest has grown as the Internet has become more fundamental to our societies.  But the way that those multistakeholder Internet governance processes interact, work with, or come into conflict with normal legal legislation, regulation, has been a challenge for the RIRs and for many countries.

 So an example of that, which the RIPE NCC, for instance, has been talking about quite recently, is the imposition of sanctions.  So you have RIPE NCC which is based in Amsterdam required to follow Dutch law, and, therefore, EU law, but that means we're also having to implement sanctions against certain individuals in countries outside the EU.  That makes for a challenging political situation and a discussion that's ongoing in various forums, including the RIR communities but also, for instance, the ITU and the U.N.

 So there are a number of different areas where this straightforward role of number registration has become challenging and, hopefully, interesting.  And, yeah, I'd be really happy to answer any questions either now or perhaps later in the presentation, in the session.

 Thank you.

 >>SORINA TELEANU:   Thank you, Chris.

 Do we have any quick follow-up question for Chris?

 And a reminder you can raise your hand or write our comment or question in chat.  And I'm looking.  There is nothing.  But, Chris, I think I can ask a quick, quick follow-up question, because you've spoken quite a lot about IPv6 deployment.  Is there anything Members of Parliaments can do to maybe help raise more awareness on the need to accelerate this deployment?

 

>>CHRIS BUCKRIDGE:   I think the most important thing is to be talking to your Regional Internet Registry and to your local industry, because there are really a lot of different approaches to how to encourage IPv6 adoption.  A lot of this comes down to industry:  How do you incentivize businesses to actually deploy IPv6.  And those incentives can come from the public sector.  They can come from law enforcement who has an interest in moving to IPv6, and we've seen that having some good impact in areas like Belgium.

 It's also, you know, useful to just be aware of what has worked in other places.  I pointed to India as a real leader in IPv6 adoption.  A lot of that is due to the fact that there are new mobile networks that are emerge -- that are coming online there which are just IPv6 from the outset.  It's not about transitioning to IPv6 for them.  It's about just using IPv6 on new networks.

 So in some sense, some developing economies have some advantage in that with new networks, you don't need to go through that transition period.  But as I say, talking to the industry, talking to your local RIR, who I'm sure is very keen to talk with you about it, is probably the best -- best step I can suggest.

 >>SORINA TELEANU:   Thank you, Chris.

 And the second quick question from chat:  How difficult is it to move from IPv4 to IPv6?  Is it too expensive?

 >>CHRIS BUCKRIDGE:   That's -- that's not a stationary target is basically the answer.  The price of IPv4 addresses on the market has been increasing quite rapidly.  I think in the last three years, it's gone up 150% or something.  So whereas making the transition does require training staff, it does often require some new equipment, if you need to grow your network or build a new network, you're now looking at what is the cost of buying IPv4 addresses on top of equipment, on top of training.

 So I think that is starting to shift, but every -- every network operator has to do their own sums on that and work out where the balance is.

 >>SORINA TELEANU:   And one more question -- well, actually two more questions.  I'm going to read them both at the same time, if I'm not losing my screen.

 How can we bridge the gap in the transitioning from IPv4 to IPv6 when we make the new entrants use the IPv6?  

 And then is there a global up-to-date mapping of IPv4 versus IPv6 transition?

 >>CHRIS BUCKRIDGE:   So, well, to answer the second one, I pointed to -- I had a slide in my slide deck which showed the map with the IPv6 adoption rate.  That's on APNIC's Labs website.  I'll post the link in chat once I've stopped speaking here, but it actually has some quite good information.  APNIC is doing work where they actually -- they actually use Google Ads to measure how widespread or how common IPv6 capability is in networks, and they do that down to the country level.  So that's probably going to give you the information you're looking for there.

 In terms of how can we bridge the gap in transitioning from IPv4 to IPv6, I think first it's a matter of understanding what the gap is.  I think -- as I said, now that we're looking at a situation where IPv4 is actually costing you money, IPv4 addresses are costing money when you set up a new network, it's not necessarily saying that it's cheaper to do IPv4 or cheaper to do IPv6, but there is a need for capacity building, for teaching network engineers and -- how to use IPv6 but also educating decision-makers, so managers, finance people, to say, you know, there is a need maybe to invest a little bit in this, but it's going to be beneficial in the long run.

 And actually, all the RIRs do training and capacity-building activities of various kinds.  So again, I'll come back to talk to your RIR.

 >>SORINA TELEANU:   Thank you, Chris.  And a few more questions are coming up but I'm going to copy them and leave them to the other part of the conversation so we give a chance for our other speakers to cover their parts, also.  Thank you, everyone.  Good to have questions here.

 Moving on into our technical journey on how the Internet works, Olaf Kolkman from the Internet Society.  Olaf, the floor is yours.

 >>OLAF KOLKMAN:   At which point I should unmute.  And I advise you all to click on the "view" button that is on the left-hand corner of what you're looking at now and make sure you have the speaker view because I'm going to present material behind me.

 Now, we were talking about packet switching.  And what I'm doing now is a conscious thing that's sort of describes packet switching.  I'm chopping up a piece of information in tiny little pieces.  Those are the packets.

 And what happens when a device like this (indicating) connects to the Internet, it will chop up information in tiny little packets, send those to the Internet, where at the receiving side of the Internet, which is addressed by an IP address, the IP addresses that Chris talked about, they will be recompiled, and the information is again the letter that I've just cut into pieces.  That is what packet-switch networking is about.  And everything that is between this device (indicating), this device (indicating), and the device that is receiving the information at the he other side of the Internet doesn't care about the content of these packets.  These are individual pieces of information that find their way over the Internet.  And that's exactly what I want to talk about.  Because this routing system, the way that it is designed and the fact that you have a common protocol and you can connect to the Internet from every place and you can become part of the Internet as a user and as a network builder, are two critical properties that we find important, sort of the core principles on which the Internet -- the properties on which the Internet is built.  If you lose those properties, you lose the functionality of the Internet.

 And I'm not going to discuss this framework which we have developed, which is the Internet way of networking.  If you want to read more about that, then there is a link in the chat right there.

 It's the Internet way of networking.  And it talks about -- about those critical properties, of which decentralized management and distributed routing and accessible infrastructure with a common protocol are one.

 Now, let me talk a little bit more about the Internet, because the way that I just talked about the Internet, it feels like a big, big cloud.  And what is happening, what I just said is basically a device, a phone connects to the Internet over here and a server is on the Internet over here.  A server symbol of sorts.  And they connect using IP, and they connect using IP, and some magic happens in the middle that makes the impression of a connection between those machines.  And the magic that happens is called routing.  And that magic is not performed by one network.  The Internet is, in fact, something else.  It is a network of various networks that are all independently managed, all have their own business relationships with each other, and collectively make up the Internet.

 With my iPhone, I connect to my local G4 network, which is run by T-Mobile in my case.  So T-Mobile is only one of these networks that I would connect to.  While, at the other side, if I make a call with somebody in Australia, they might be connecting to the local network over there.  When I send a mail or a packet or something, they are connecting to the network over there.

 All these networks make their own business decisions about how they create and use the technology that is available to them.  Some may use fiber networks.  Some may use Wi-Fi networks.  Some might build low-orbit satellite networks, some might use the copper cables that are in the ground.  Some might use laser point-to-point links.  It's all the individual decisions of these network operators.

 And mind you, the current Internet consists out of about 70,000 networks.  They are sort of the customers of those regional registries that I just talked about.  They will reach out to those regional registries and get IP addresses to build their networks and connect their customers.

 What is so special about this?  What is so special is that, first, that common protocol that I just talked about means that I can just go to the shop, whether I'm in Africa, whether I'm in Latin America, whether I'm in China, whether I'm in Russia, whether I'm in the UK or The Netherlands, and I can buy a device, a computer, a telephone, a camera, something else, and those devices all have that common protocol IP baked into them and can connect to their local network easily and without friction, so to speak.  That makes the Internet accessible.  Everybody can connect to it.

 But there's a second way in which the Internet is accessible, and that is if I build a new network, and new networks are built almost every day, then I can just do so.

 Let's build a new network.  Oh, I should give this network a color.  That was, at least, the intention.

 Let's build a new network.  This network, what does it need to do to connect to the Internet?  It needs to find one neighbor.  One neighbor to connect to that is willing to accept traffic and forward that traffic to the rest of the Internet.  And that's the only business relation that I need.  I pay some money.  The purple network will pay some money to its connecting network so that the traffic is being sent onwards to the network, but I do not have to make commercial agreements with all the other parties in the network.  I connect to a neighbor, and from that on, I am part, as a network, of the global Internet, and my customers can reach each and everyone else on that global Internet.

 So this critical property of the common protocol and the accessible infrastructure means that you have a global network that creates enormous social and economic value, because everybody can do this, and everybody can create opportunities and do business.

 Obviously everybody can also do harm.  We know this.  That's not a problem of the network.  That's a problem of our behavior.  And trying to address behavioral problems by breaking the property of the network itself I think would be a mistake.  We would lose the Internet and this global -- this global platform in which we can build all kinds of fancy stuff.

 Chris showed this hourglass model.  We're now talking about the waist of the hourglass.  And all the communications that people do, all the content is about writing letters that are then chopped up in little packets.  And the problems are in the letters, they're not in the packets.

 So the barriers to connect and creating the business relations are very local, and that creates a regional and a global connectivity map.

 The way that is being done, the way that this works technically is that at each connection point, there is a so-called router, et cetera, et cetera, and those routers have an idea of all the other 70,000 networks and how to reach them.  And when this network A needs to send a packet to network B, then it knows in order to send a packet to network B, it needs to send it to network C.  Network C knows it needs to send it to D.  Network D knows that it needs to send it to E and so on and so forth, and that is a continuously updated map.

 And if the layout of the network is a little bit different and there are multiple ways to reach B from A, then the network will optimize for that.  And if it is the case that one of the links breaks, then the network will optimize for that, and A doesn't need to do anything, but the packets will just flow over the network.

 And sometimes that letter that I just picked up, one of the pieces might take another route than another one of the pieces.  And that's all magic that happens underneath by an automatic protocol called PGP.

 The power of this arrangement is really that it is decentralized; that every network operator can make its own decisions about investments.

 Also, the power of this is that the users of the Internet can develop protocols, can develop applications without having to coordinate that centrally.  They only have to do -- the only thing that I have to do is make sure that they can translate everything into IP and that the other side that receives the information can combine it all back into a piece of paper, of which both of the parties have a mutual understanding.  And that gives you the property of an open architecture with reusable building blocks, and it provides you permission with innovation. It makes that you can invent new things without the network needing to adapt.  And those are all sorts of strong properties that bring us a lot of social and economic value.

 So what I would like you to take away from this is that the Internet is not a monolith.  It's not one big network.  No, it's a very dynamic and constantly changing arrangement of tens of thousands of networks that all make their own business decisions and that makes sure that we have a global connectivity mesh, that we can all connect to each other as a humanity and use the network for our social and economic activities.

 And I think with that, I said all that I wanted to say, even though we have five minutes more.  But I don't think that that should stop us.

 I'm happy to take any questions.

 

>>SORINA TELEANU:   Thank you, Olaf.

 Do we have any questions for Olaf?

 Going quickly to the chat.  If we don't have anything, let me try with a question while people are warming up.

 You've mentioned some of the key characteristics of the Internet key properties, my English is going bad.  What would be the key ones that you think Members of Parliament should keep in mind when they develop Internet-related legislation to avoid in some way that this legislation has unintended consequences for the way the Internet works?

 

>>OLAF KOLKMAN:   Yeah, yeah.  So again, I would refer back to the link that I posted a little bit earlier:  The Internet way of networking.  You can also Google that.  If you Google for "Internet way of networking," it's probably the first hit.  But what we did is we developed five critical properties.  Some of them are related to the pictures that I just painted; some of them are a little bit more distant to the picture that I just painted.

 The five are having an accessible infrastructure with a common protocol.  That's what I just talked about.  That reduces the friction to become part of the network, to create new networks, like community networks and so on and so forth.  But also, it allows you to go to the shop, buy a thing that immediately connects to the Internet.  Then an open architecture of interoperable and reusable building blocks.  That's the technology you deploy on top of it.  The fact that you use IP as the common protocol means that you can do stuff underneath it.  It means that all these networks can build the networks in the way that they want.  They can use their own technology to optimize while at the top layer people can innovate and create new applications.  Improve http, make it more secure, invent blockchain, write new types of messaging systems, all those type of things.  And they're standardized in a specific way that is open and consists out of little modules that you can pick together.  That's sort of the Internet way of designing things.

 The decentralized management and distributed routing system talks a little bit about what Chris was talking about.  The distributed routing is what I talked about, but the decentralized management is that parts of the Internet infrastructure are managed in different governance institutions, but also parts of the infrastructure itself are managed in several ways without a central control.  That's the third critical property.

 The fourth critical property is we have global common identifiers.  And I'm sure that Adam will talk about those.  It's the DNS.  Making sure that if you use a name, you end up at a certain place and not another way -- place.  If you have collisions there and not a global way to name things, then the Internet stops working.

 And finally, this whole setup is technology neutral and general purpose.  It is not built for one thing.  It is not built for a specific vision or with a big picture in mind.  No, it's built for basically everything.  And it turns out that it is cheap and very versatile.  It allows you to quickly scale up.  It allows you to change the network.  If you get a pandemic and suddenly everybody works from home, that property saves the day.  You don't have a specialized network that will -- will need high level of investment and suddenly the traffic patterns change because everybody is at home and use specialized for people being at work during daytime.

 So those are the five critical properties in very short overview.  And I advise you to look at that website, which also gives you a few tools to look at policy proposals, dissect them, and see what the impact is.

 And if you have any questions, feel free to reach out to me personally.  My email address is like right there (indicating).

 

>>SORINA TELEANU:   Thank you, Olaf.  We have two more questions in chat, but let me see if people would like to speak also so it's not only us interacting by voice.

 So would you like to ask your question directly?  And I think people can unmute themselves, right?

 >> Yes, can I go ahead?

 

>>SORINA TELEANU:   Sure.

 >> Okay.  Thank you.  Good afternoon, Olaf.

 I just wanted to get some more clarification on this Internet.  Some people also say intranet and extranet.  Why don't you just call everything Internet?  Instead of some people will say Internet, and some other networks, they call it, you know, intranet, some they call -- some they have some network called extranet.  Since they're all networks, why don't we just say Internet?

 >>OLAF KOLKMAN:   I think that is a very good question, and I'm sort of stumbling to answer it.

 So most networks, I would say most network, use the Internet Protocol, but that doesn't mean that they're necessarily part of the Internet.

 Sometimes people make very conscious decisions to deploy networks using Internet technology so that they can make use of, you know, off-the-shelf type of things, but they deploy it in a network that is sort of out of this structure or that the structure is not able to reach.  And they do that for security purposes or control purposes.  And then we call it an intranet or an extranet.  Or they put up a very strict gateway between them and sort of the rest of the Internet so that not all traffic goes in two ways.  Also for security properties.

 But once you put a gateway in place that filters specific kinds of traffic, you lose that general purpose and extendability of the Internet.  You have more control and you know what is coming in and out of your network, and usually these are enterprise networks, for instance, but you lose the inter- -- the ability to interoperate with the rest of the network that might make specific assumptions about that everything is available in some network.  Sort of on a technology level, all types of connections can be made.  So if you break that, then you might find yourself building a new application, trying to connect to it, and you find that your application that you built doesn't interoperate with the rest of the world.

 So I would urge you to think about the Internet, indeed, as an interconnected network of networks in which global connectivity is available.

 I hope that answers your question.

 

>>SORINA TELEANU:  Thank you, Olaf.  We do have a few more questions in chat, but let me see if people would like to speak themselves before I go to them.

 Okay.  Let me read quickly.

 What factors could lead to the Internet becoming less decentralized, less monolithic, less democratized, and less accessible?

 Than another follow-up question to what you have said.

 IXPs are basically key to efficient routing of local Internet traffic instead of more costly international transit routes.  And what are the politics, policies, and governance issues of putting in place more IXPs?

 And final question, on which layer of the TCP/IP does blockchain reside?

 >>OLAF KOLKMAN:  I find the first question perhaps the most difficult.

 What are the factors that lead to the Internet to become less decentralized.

 There are many factors.  Some of them are economic realities of consolidation, so big companies growing bigger.

 Some of them have to do with a certain aspect being outsourced, so people -- for instance, outsourcing the DNS functions, not maintaining that themselves.

 Some have to do with regulatory measures that concentrate things in certain ways.  

 So there is not really a one-size-fits-all answer to the question what makes things more decentralized or more monolithic and less democratized.

 I think you always have to look at the topic before you and sort of dissect it and do that analysis.

 But there are a lot of pressures that I can say that do that now.

 On the blockchain, very simple.  Blockchain is built on an application protocol, basically, a peer-to-peer overlay that is based on top of IP.  So that blockchain lives on a higher layer, in essence, the application layer.

 That's a maybe somewhat technical answer, but that's what it is.

 And for that IXP question, I do want to go back a little bit and explain what the power of an IXP is.

 In this picture, I drew network A that connected to network C and said, you know, this network A needs to pay money to ship traffic to the rest of the Internet.  Now, it might be that there is a network D here as well, and there is a network E here.  And all these networks have to pay money to their upstream, as we call it, the network that is sort of closer into -- to the Internet, because that's their only connection to the rest of the Internet.

 Now, should the networks E, A, and D that I just drew be in the same locality, so, like, in the same country, then they can decide to pull cables to each other.  They draw a trench, they put in the fiber, and they connect to each other, and basically say on a handshake agreement, you send your traffic to me, I send my traffic to you, and when we do that, that traffic doesn't have to go to our upstream.  So the interaction between those networks becomes cheaper.

 Now, another way that these people could do that is say, well, we create one point where we all get together, and we dig our trenches to that one point.  And that is an Internet exchange.  And the more networks you get in your locality, your country, your city, your province, whatever, where there's economic interest and there are a number of networks that want to exchange traffic, the more of these networks connect there.  And when they do that, hard-earned money doesn't flow to the rest of the Internet.  No.  It stays inside of this circle.

 So that is the way that IXP make routing more efficient, because, for instance, the traffic from A to E doesn't have to take this long route, but it can go directly via the IX.  That's cheaper, too, because those routes cost money.  A needs to pay money to send it upstream.  E needs to pay money to receive it from the upstream.  So there's all kinds of economic incentives and interests to build IXPs.  They're very good for your Internet economy.  Besides, you have a concentration point where content providers might come and share stuff.

 So what are the politics, policies, and governance issues of putting in place more IXPs?

 Well, first, there is the economic issue, is there sufficient interest?  Are there sufficient networks to do this?  Then there is a regulation aspect.  Are those networks allowed to connect over there?  And that has to do with, you know, your local regulations, so to speak.

 And there's also a trust aspect.  It might be that in some localities, if the government organizes an IXP, that some of the networks might not want to join because they think that their traffic might be listened into.  Or that the government would shut networks down or have other ways of controlling.  So it's a trust issue.

 So there are several tools to sort of liberalize the -- or things to pay attention to when liberalizing Internet exchange points.  But overall, they make the interconnection between networks more dense.  And that is a good thing for the Internet overall, because it makes the price of connecting cheaper.

 Sorry.  That was a very long-winded answer, but I felt it was important to share.

 

>>SORINA TELEANU:  Thank you, Olaf.  And thank you also to our participants for the questions.  If you have more questions for Olaf, please save them for the last part of the discussion.

 And now we can move on to our next speaker, Adam Peake, from the Internet Corporation for Assigned Names and Numbers.

 Adam, over to you.

 

>>ADAM PEAKE:  Thank you very much, Sorina.

 Would you mind putting the presentation up, please.  And thank you for agreeing to do that for me.

 So my name is Adam Peake.  I work for ICANN, the Internet Corporation for Assigned Names and Numbers.  And I'm also a member of the IGF MAG, the Multistakeholder Advisory Group.  So I work with Sorina and Chengetai and very much look forward to seeing some of you, I hope all of you, in Katowice later this year.

 Is the presentation available?

 Or I could -- Thank you very much.  Sorry about that, Sorina.

 So, yes.  That's who I am.  And as I said, I'm a member of the MAG and look forward to working with you in Katowice and developing the parliamentarian track.  It's something that we're very excited has been added.  Well, not just added, but working towards improving this year.

 May I have the next slide, please.

 So, ICANN.  This is the organization I work for.  What we do is we're responsible for the unique assignment of names, numbers, and that's the domain name system.

 As Olaf mentioned, numbers, protocols, domain names need to be uniquely assigned or we will see collisions.  If they were not uniquely assigned, you would not end up at the place you would really want to be.

 And the other thing we do is translate between the technical infrastructure that Chris and Olaf described, the Internet Protocol addresses, and domain names, which we as humans, as people used to alphabets and so on, find more convenient.

 If you were to type that number you see, 192.0.43.22, into the browser of your phone or laptop, you'll see that it actually resolves to a name.  And so the first sort of function of the domain name system from a very simple point of view is that we're thinking about a way that is easier for us to remember and navigate and so on and so forth.

 The other part of this is that the domain name provides an interface to the technical infrastructure, which is evolving all of the time.  

 Think about the first time that you accessed something like Google, Google.com.  It might have been ten years ago.  It might have been longer.

 And over that time, the services that Google provides, their infrastructure, all the routing networks that Olaf went into, the development of IPv6, you don't need to see that.  All you need to remember is Google.com.  That remains your stable interface to the domain name system, to all of the routing that happens behind it.  That's two important aspects of what ICANN is responsible for.

 We're not solely responsible for that.  I have to say this very clearly.  We have technical partners.  And two of them, of course, here, the Regional Internet Registries.  Olaf is one of the technical lead developers for the Internet.  The Internet Engineering Task Force is responsible for the development of standards, of protocols, and so on.  So we have formalized relationships with our technical partners to make all of this coherently come together.

 May I have the next slide, please.

 So Internet identifiers.  For ICANN, this is basically what it is.  As Olaf has described, it's a mesh of networks as operators agree to communicate using predefined protocols, TCP/IP in this case.  And a protocol is a set of rules, a system of rules, that allows entities in a communication system to transmit information.

 The one you're probably most familiar with would be HTTP, so the hypertext transfer protocol, which establishes connections between the server, with information, and the HTML pages, which define what should be displayed on a Web page when you get that information.  So the protocol enables communication for the World Wide Web, HTTP.

 And it's the network of identifiers for the individual computers or hosts on the Internet so that these can communicate.  And so at ICANN, we see these as names, numbers, and protocol parameters.  And as we've said, they must all be uniquely assigned.

 And if I could have the next slide, please.

 So here we see a simple representation of that.

 Chris mentioned IPv4 and IPv6 addresses.  You see the example I mentioned earlier, which will resolve to ICANN.  And then the IPv6 address, which, as you can really see, you might be able to remember a few IPv4 addresses, but you're going to have no chance of remembering that and typing it into your browser window.  So that's where the names really do come in.  People need to use names.

 And the mapping of these names, IP addresses to a name, is name resolution.

 So if we go to the next slide, we can start to see how some of that works.

 So the example of www.intgovforum.org, of course, which is the IGF website which is run by Chengetai and his team.  You can see the different components of this in -- well, we call it the URL, but it's the Web address.  So you can see https is the protocol, identifier that will be use.  www is the Web server.  EuroDIG is incorrect.  You should see "intgovforum" there.  I'm sorry.  That should have been updated.  My mistake.  And that's the second-level domain name.  And then that's something that Chengetai, or in the case of EuroDIG, that should be Wolfgang, one of our later speakers, would have acquired from a company called a registrar.  And I'll mention those in a moment.

 And then we have the top-level domain.  These are TLDs.  And these are primarily in what we call the generic name space administered by ICANN.  Then there are others which are country codes.  We'll come to that in a moment.

 Then we have the example of dot.  It's a dot you that won't use and don't type when entering these Web addresses in your browser, but it designates the root server.  And root servers, as we'll see on the next slide, contain information.

 Next slide, please.

 This is the sort of hierarchical lookup system for the domain name system.

 At the top of the hierarchy, we have the root.  Below that, top-level nodes.  Second-level nodes.  And third-level nodes, or top-level domain names, second level domain names, and third level domain names.

 The root, as a database, contains information, IP addresses and so on, that will tell your -- the system on your laptop or on your phone the location of -- on the Internet of where the information is about -- well, it will contain information about .NL for the Netherlands, which is where I live, .COM, .COFFEE, which is one of the new top-level domains. 

 And you can see some strange numbers on the left-hand side.  This is a code that will actually represent -- in this case, I think it's Hong Kong, in Chinese script.  It's a translation system so that we can use Internationalized Domain Names so that people can use the domain name system in scripts other than the typical ASCII, Latin-type alphabets.

 So the root server will have Internet Protocol address information so that the lookup system will know how to find more information that's held in the servers that control, for example, .COM.

 And under -- and .COM will then have all the information needed for those second-level domain names, the second-level nodes.  The example shown here is "example."  So it will be example.com.  And the .COM database will have IP addresses that would tell your lookup system what information is required so that you can access information held by "example."

 And "example" in this case would be a domain name that one of us would go and acquire so that we can run a website and run our services.  And underneath that, we may want to run different types of -- types of services itself.  And this happens at the third level.  So you can have a Web server.  And that is something that you will decide for yourself, the types of services you want.  You'll have a mail server.  And some of you may be familiar with websites where they actually provide different functionality at the third level, for example, europa.eu, which is the main website for the European Commission.  At the third level, they use all of their departmental names.  So the department for social services or whatever it would be are all designated at the third level.

 So you can see at the bottom of the page, this is sort of broken up, www is the -- that's the third level.  The second level, intgovforum, the website for the IGF; the top level, .ORG; and then the root.

 At the moment, there are 1514 TLDs in the root.  This number changes a little bit as top-level domain names are added and removed.

 The next slide, please.

 This shows that there are zones here.  And the zone is the different designation for -- in the hierarchy.  And these are administrative boundaries.  So the policies are set by each of those who control or have been designated these top-level domain names.

 So the policies for .ORG are decided by the registry operator for .ORG.  And underneath this, you can see the intgovforum.  And the policies for this and what services are provided are decided by the person who's registered that.  In this case, I think it's Chengetai or Chengetai's team.  And he decides what services he wants to make available under this hierarchy.

 Next slide, please.

 So just to go to the root servers, in the lookup process, the root servers play a critical role, because they identify the locations for information of all the top-level domain names.  There are 13 root zone files.  The information and databases is the root zone file.  And they're hosted on 13 identical root servers.  They're all managed by -- these are managed by different organizations.  Actually, there's one that manages two, so they're managed by 12 different organizations.  And, very importantly, these 13 servers have identical copies around the globe.  And there are -- as of last night, when I had a look, there are 1,404 instances, perfect copies, of these root servers.  And so that there's no problem with trying to access one of the 13, as we use the Internet, the distributed network that Olaf describes allows us to access root servers that are present nearer us on networks nearer us, and so on.  And the functionality is exactly the same. 

 ICANN is the operator of the IANA functions, which Chris mentioned, through a subsidiary of ours called the Public Technical Indicators.  And these manage -- the services that the IANA and PTI manage include the management of the root zones database.  We have, through PTI, which, as I said, is a subsidiary of ICANN, we have service-level agreements and MOUs with the Regional Internet Registries, with the domain name operators, with the Internet Engineering Task Force, to ensure that as this function is performed, ICANN is performing it correctly.  And it also binds us to each other in a sense of governance and shared goals.  So this is incredibly important as we think about the -- the connections that are made between the different technical operational aspects of the Internet.

 An important aspect of this in terms of governance is that until 2016, the United States government had a -- had a role in overseeing some aspects of the IANA function.  These were administrative oversight roles and did not get into operations.  But it was seen as something that was a somewhat archaic result of the Internet's development.  And in -- the 1st of October 2016, the U.S. government role was removed and the responsibility for the operation of the IANA functions was transferred to the ICANN community and to the partners I mentioned that work with the IANA through the various relationships that have been formed.

 This transition was something that was done as a global process.  The rules, the new accountability and trans- -- and transparency mechanisms, the relationships that were established between the RIRs and PTI/IANA were created at that time.  And it was really a very good example of multistakeholder approaches to governance in action.  We see one of the most essential parts of the Internet being moved from -- with a minimal oversight of the U.S. government, but into a pure multistakeholder approach, where all the policies are developed through bottom-up and inclusive processes.

 And that we will have a look at in a moment.

 If you can go to the next slide, please.

 So this is what the domain name industry itself looks like.  There are three main components here.  There's the registry, be which is the database of domain names and registrants in the top-level domain name, so that a registry would be the operator of, for example, .ORG or .COM, or .NL, and so on.

 There are registrars.  These are the agents that act between the registrant, which is us, we who want to acquire and use a domain name, and the registry, so that the registration information about me, if I'm getting a domain name, is available in the registry and they know how the traffic should be routed.

 There can be resellers under the registrars.  They don't all -- and these are contracted with the registrar organization itself.  Or the registrar -- sorry, the registrar will have a direct relationship with the registrant, or it may be a relationship through the reseller.

 Next slide, please.

 And this is one of the key parts of the work that ICANN does, is the work with the contracted parties about developing policies for the domain name system.  So registries and registrars have a contract with ICANN to perform their duties and functions.

 And if we go to the next slide, it's a slightly complicated image of these relationships.  But you can see at the top in rather small letters that there is -- for registrars, they have a registration accredit- -- sorry, a registrar accreditation agreement.  And that includes the terms and conditions that they need to operate under.

 And there is a registry operators' agreement and service provider agreement, which is also with ICANN.

 The important thing here is that these policies, the agreements and the policies that go into those agreements, are developed through the ICANN multistakeholder community.  It's not ICANN staff that do this or the board.  These policies are developed through a bottom-up process.  And it could be anything from new consumer protections to more technical aspects.

 There are also agreements between the registrant and the registrar, so that the company I get the domain name from will have certain requirements and services it must provide me.  These might be things about how quickly they respond to questions and so on.

 But, again, all of this is developed so that the community is deciding what policies the contracted parties should abide by.

 And if we can go to the next slide, please.

 So this multistakeholder model is at the heart of what we do.  It's global.  It's a decentralized governance model where individuals and industry, noncommercial industries, and, of course, government, come into ICANN on an equal level to develop these policies.

 Contracts and policies, as I said, are developed by a community process.  It's bottom-up in the sense that policies are initiated from the community members themselves and take it up into various processes, ending up with the board to make those decisions.

 And the other thing, it's not just about the domain name system in the sense that it's about the credited parties.  ICANN's budget, ICANN's strategic plan are also developed in the same way.  So at the beginning of each year, the strategic plan from the previous year will be discussed and developed with a plan for -- that will go forward into -- into the next years.

 I should say this is actually a five-year plan, but it will be evaluated each year.  And then the budget for that will also be developed.

 And one of the interesting things about the accountability process here is that of course the community is involved in developing these strategies.  It's involved in deciding what the budget should be.  But if the board decided for some reason that it wanted to have a different budget, then there are mechanisms to prevent that happening.  The ultimate mechanism for that is a power for members of the community to come together and actually remove board members for reason.  And this is one of the reasons that such a -- such a mechanism could be brought before.  It's not expected to happen.  There's no particular reason why the board would ever want to ignore the budget process that itself -- it itself is part of.  But these are important accountability mechanisms within this multistakeholder model.  So the ability to call the board to account, it's a -- it's the ultimate mechanism.  And we hope, of course, it will never be used.  But these are important to have in place for accountability to help prevent capture and other things.

 So how does this work?

 There are supporting organizations.  And these are responsible for developing and making policies in the areas they represent.

 Chris touched often one when you spoke about the Regional Internet Registries.

 And then there are advisory committees who advise the ICANN board on policy development.  And in certain cases, they can raise issues of policy development.  But usually policy is initiated by the supporting organization, because they have the competency, as it were, in that particular area.

 So if we look at that in a little more detail, we can see -- sorry, the next slide, please.

 This is a sort of diagram of how it works.  We have a -- the global multistakeholder are volunteers who are the active policy developers.  We have directors, the ICANN Board of Directors.  These are representatives who are drawn from the community, and they will look at the policies that the community develops.  And then we have the ICANN look at, evaluate, and move to implementation the policy the community develops.

 And then ICANN org, which is the ICANN staff group.  And our responsibility is in implementation of community-developed policy and in supporting the global community in the work that they do.  So it may be looking after calls.  It may be writing papers.  It would be helping them generally to make sure that they're able to work as volunteers and perform these functions.

 The next slide, please.

 So these -- this is how it really breaks down.  The supporting organizations and the advisory committees, board, and the organization, the organization being the staff, which is what I am part of.

 And let's have a look at the next slide, please.

 So this is the support -- a slide showing who the supporting organizations are.  And as I said, they have competency for the -- for their areas of expertise.

 The Address Supporting Organization represents the groups that Chris is a member of, the Regional Internet Registries.  And as he mentioned, they do come together -- while policy is developed within the regions, they do come together to work on the global level for Internet Protocol address policy.  And they do that generally within ICANN.  They also have a role in advising the ICANN community through membership of the ICANN board, so that there are -- there are members representing this experience, knowledge, and community into the ICANN policy development process.

 There's the ccNSO.  And this represents country codes.  And -- the country code name supporting organization.  Country codes are -- and I'm sure you're very familiar with them -- these are the top-level domain names that are operated within a country.  The examples that I have sheen here are .BE and .NL.  Of course, there are many, many others.  I think the number today is around about 309 country codes, which is more country codes than there are countries.  They also represent some territories.  And they also operate IDN top-level domain names, internationalized top-level domain names.

 So, for example, while there is a top-level domain names .JP for Japan, they also have a -- an IDN country code representing the country name in Japanese.  

 So the ccNSO develops policy at the global level, similar to the ASO.  All in-country policy is developed as a sovereign entity, the country code manager and team will develop their policy with their local community.  So ICANN does not have any say in that.  We do -- they do come together when there might be an issue that is global, for example, the development of Internationalized Domain Names and the standards around those.

 The core of the work, really, the ongoing daily work, of ICANN takes place through the Generic Names Supporting Organization.  And they develop the policies for generic top-level domain names, .COM, .ORG, and all the new generic top-level domain names that you might have seen, for example, .AMSTERDAM or .PARIS, or .COFFEE, as I said, .SHOP.  And these are the policy development processes I mentioned for the contracted parties in particular.  There are various constituencies in this -- in the GNSO representing not-for-profit organizations, business organizations, intellectual property organizations, and the contracted parties themselves, and ISPs and so on.  So we have a structure beneath the GNSO, the Generic Names Supporting Organization, where different stakeholder groups are represented.  And they develop policy for the generic top-level domain names.

 And the next slide, please.

 Here we see the advisory committees.  And there are four of these.  

 The At Large represents the interests of individual Internet users.  And the job here is for the At Large group to think about what are the interests of individual Internet users in ICANN policy.  And their job is to make recommendations on that policy to the supporting organizations and the processes that those supporting organizations run, and to ICANN board on policies that come from the supporting organizations.

 There is a Governmental Advisory Committee.  And this is where the governments of the world come together and provide advice, particularly on public policy issues related to international laws and international agreements.

 At the moment, there are 179 members of the GAC, so 179 national governments and territories are represented, and 38 observer organizations, from intergovernmental organizations such as the OECD or Francophonie, and so on and so forth.

 The GAC plays a particular role.  In particular, its advice has to be considered in a special way by the ICANN board when it's receiving formal advice from the governments.  But it's -- and its role is particularly noted in that way.  

 There's a Root Server Advisory Committee.  And they are the root server operators from the 13 root servers -- root servers.  And they ensure that policy does not negatively affect the security and integrity of the root server system.  And they also may advise on other policy that affects the technical operations.

 And then the Security Stability Advisory Committee advises on matters related to security and the integrity of the Internet's naming system.

 So that is really the last slide.

 What I hope you can capture is that there is an opportunity for anyone to participate in these community structures.  Particularly within the Generic Names Supporting Organization, there is a home for anybody to help develop policy.  And this policy can start within one of the stakeholder groups, for example, for Internet service providers and connectivity providers.  They may recognize that there's a new policy needed, and it will go through a process of development, which will be formalized with a working group that has a charter and membership.  These charters and the policy development process that emerges from this is something that the board -- when this working group finally recommends a policy as an outcome, one of the things that the board will do is ensure that the processes have been followed accurately, that those voices that were included must have been heard, if there are minority opinions, then how have those been addressed, so that there's integrity to the policy development process.  And there are well-established and known rules for the policy development process.  These will be things that you would be very familiar with in good practice for the development of policy and regulation legislation.  And we have built that into the ICANN system to ensure both integrity of the policy itself, but also that it's transparent, that it's inclusive, and that it's fair.

 I think the last slide, please.

 Which is a little bit please engage with ICANN.  We are very interested in to work with parliamentarians.  One of our current goals in our staff strategic plan is to be a provider of mutual technical advice about the Internet, about operations affecting the domain name system.  And as you can see, we work with organizations like the Regional Internet Registries, Chris, and the Internet Society, Olaf, to make sure that this happens.

 There's technical training.  And there's also -- my colleagues are more than happy to talk to any parliamentary group about issues relating to the domain name system.

 So thank you very much.

 

>>SORINA TELEANU:  Thank you also, Adam.

 Do we have any questions?

 Yes, I see someone has a question.

 Can you unmute yourself?

 Or, Luis, can you help?

 >>LUIS BOBO:  Yes.  Anyone can unmute themselves.

 If you want to talk, please. 

 >>SORINA TELEANU:  Would you like to ask your question?

 

>>NEEMA LUGANGIRA:  Oh, hi.  Hi, Sorina.  Sorry, I wasn't hearing.  The network was not very good. So I'm Neema Lugangira, member of parliament from Tanzania.  And we have recently set up a loose group called Parliamentary Caucus on Internet Governance.  And when Adam was speaking here towards the end, he mentioned that they have opportunities to provide capacity-building to members of parliament on these different areas.  So I just wanted to find out from him if our caucus wanted to engage with him to get some training to better understand these areas of Internet governance as well as the areas that he just spoke on, is there a possibility?  And if so, how do we go about it?

 Thank you.

 >>ADAM PEAKE:  Thank you very much.

 Absolutely.  I have a colleague who works in the same division as I do, global stakeholder engagement.  And I would be more than happy to introduce you to him, Neema.  He is actually based in Nairobi, in Kenya.  And I'm sure he would love to get in touch.  Yes, absolutely, we would be very able and willing to have a talk, develop a program, or whatever it would be that your colleagues would be interested in.

 And I will put my email address in the chat so that you can contact me directly.  But it is on the slides.

 So, yeah, absolutely, this is something we are very interested in doing.

 The other thing that I would highlight is that I'm quite sure Tanzania does have a representative in the Governmental Advisory Committee.  And we can make sure that there's a connection made there.  And this is something that I would say to all of you.  I'll put a link into the membership much the Governmental Advisory Committee for each country.  And it would be interesting perhaps to try and talk to them.  They will tend to be from a -- the digital ministry or ministry that's responsible for digital in your country.  But they are the formal interface in the policy sense with ICANN.  But for you as parliamentarians, please reach out to me, and I will be very happy to introduce you to our national and regional representatives on staff and arrange this type of activity that Neema just mentioned.

 

>>NEEMA LUGANGIRA:  Thank you very much.  And I also saw that Olaf from ISOC mentioned the same thing.  If I can have both your emails, then I will take it from there.  Thank you.

 

>>SORINA TELEANU:  Thank you, Adam, and Neema as well.

 I don't see any hand, no other question in chat.  So that was the wrap-up of our technical or more technical discussion about how the Internet works and how this connects with Internet policy work that most of you are doing as part of your parliamentary work.

 Is that a hand?  From central Africa?

 Please unmute yourself and speak.

 Go ahead.

 >> (Indiscernible.)

 Thank you for giving me the floor.  I want to thank you -- thank all the persons who are present to this call.  And I want to thank you all.  We would like to thank you for the interest that you have for central Africa.

 >>INTERPRETER:  The interpreter apologizes, but the audio is quite choppy.

 >> We are very interested by the governance of the Internet.  The Internet has a great influence on the world.

 The Internet is a means of communication, and it's part of our daily life.  It's present everywhere.  It is -- we can hope -- we -- we -- we can also discuss all the dangers that it represents.

 >>INTERPRETER:  The interpreter apologizes, but the audio has been cut off.

 >>SORINA TELEANU:  (Inaudible) -- speak English.  

 I would kindly ask you, if that was a written statement, maybe also paste it in chat for everyone to be able to see it.

 And from what I managed to pick up from your intervention, I would also kindly encourage you to follow the other sessions of the parliamentary track as we will go more in track into substantive issues and some of them that you have just mentioned in your intervention.

 Many thanks, again, and we hope to see you for the other sessions as well.

 And we have not so much left of this session, and we still would want to cover one more element of Internet governance, if I may say.  And we have Wolfgang Kleinwachter from the University of Aarhus with us for a brief overview of what is Internet governance, how we got here, and what to prepare for in the future.

 Wolfgang, over to you.

 >>WOLFGANG KLEINWACHTER:  Yeah, thank you very much.  Thank you, Sorina.

 And I think it's a very good idea for the IGF and the ITU and UN DESA to prepare parliamentarians for the IGF.  As Chengetai has said at the beginning, it started in Berlin.  It moved to the virtual meeting.  And I think parliamentarians are more or less a stakeholder group in its own.  I would call it a stakeholder group sui generis in the whole Internet governance ecosystem.  Governments are a fixed part in the stakeholder mechanism, but parliamentarians very often have a different opinion.  They have to deal with day-to-day things in their parliaments, and more and more issues from the Internet are relevant now for national policy-making.  And so it's a very good thing that this parliamentarian track is now more strengthened within the IGF.  And this will be a process which hopefully will go beyond 2025.  And I will come back to 2025 a little bit later.

 And to start with the more technical aspects of Internet governance is also very good, because you have to understand the underlying technology if you want to understand what's going on in the Internet governance ecosystem.

 You know, the Internet governance ecosystem has no definition.  So I have compared this several times with the rainforest.  So you cannot manage or control the rainforest, so there are so many different animals and plants in it.  So that means it is really a very, very complex system with different stakeholders, different layers, different properties, different regulatory system which more or less coexist or sometimes fight against each other.  And it's much more than names and numbers.  But names and numbers are more or less the heart of the whole system.  And so far, again, it was very, very useful to hear Chris, to hear Olaf, and to hear Adam that explained the background.

 But all this was discussed already 20 years ago in the framework of the World Summit on the Information Society of the United Nations.  So I'm involved in this since more than 20 years.  And what I see is that now a new generation of people, it's not only a new generation of programs, it's also a new generation of people who have no exact knowledge about all this history.  And so far, it makes sense to get a little bit more familiar with the history when it was discussed.  Because what I see is that we have a number of new problems, but the structure and the conflicts behind the problems are more or less the same we had 20 years ago.

 Twenty years ago, Internet governance was seen as a technical problem with some political implications.  But today, it's a political problem with technical components.  And you have seen this just the last couple of days in the 76th General Assembly of the United Nations, that President Biden, that the U.N. Secretary-General Guterres, that the President Xi of China are referring to Internet governance-related issues.  It's called now today cyber or digitalization.  But all this is part of Internet governance.  And so this has -- this issue, which was a minor issue 20 years ago on the big policy agenda is now at the center of global diplomacy, and, unfortunately, also of global conflict.

 And so far, it's worth to remember what were the conflicts 20 years ago and how this was discussed, because the first thing was what governments in the World Summit on the Information Society wanted to know is, what is Internet governance?  There is no definition.  And so because some people argued this is just a technical aspect, and others argued, oh, it's all the broader issues.  And the interesting thing which came out on the Working Group on Internet Governance, which was established by Kofi Annan, the Secretary-General of the United Nations at this time, was to say, okay, it's a complex issue where we have on the first thing the -- we need the involvement of all stakeholders.  So it's too big to leave it in the hands of governments, and it's also too big to leave it in the hand of the private sector.  And what you need is the involvement of all stakeholders.  That was the first component.

 And the second thing was to make a differentiation.  In fact, you know, we have a technical layer, and we have a policy layer.  So this is interlinked.  You cannot separate this totally.  But these are two different shoes.  And in paragraph 34 of the Tunis Agenda, which was adopted by 193 member states of the United Nations, you have this definition which refers to the multistakeholder model.  That means you have to have all stakeholders involved and which makes this differentiation when it speaks about the development and the use of the Internet.

 So the development of the Internet is more the technical dimension.  And the use of the Internet is where the public policy-related Internet issues.  And, by the way, after -- now 20 years later, 90%, probably 95%, of the problems we have with the Internet are coming with the use of the Internet and not with the development of the Internet.  And I think this is also very important, because today, policymakers very often you know, beat the horseman, when they mean the horse.  The problem does not come in the political problems we have, with fake news, with cybercrime, with a lot of other things which are related to the Internet now, so these problems are not coming with the technical components of the Internet, the management of IP addresses, domain names, and protocols.  They are coming with the use, or let's say with the misuse of the Internet.  And so far, you have to have this differentiation from the Tunis Agenda translated into today's policy meaning.

 What I see as an observer is that many governments do not understand this differentiation, and then they want to interfere into the management of this, what we call through public core of the Internet, the routing, the domain name system, the IP address.  But you cannot settle content-related issues by changing the policy for IP address allocation or for domain name allocation.  So these are big problems, what we have today, from cybercrime, to fake news, and many other issues of the economic dimension, you know, freedom of expression, development, infrastructure development, and all of this.  So these are big problems.  And this is for policymakers in the parliament the most pressing problems of today.  And they have to deal with it, no doubt about it.  But, you know, if they want to solve problems, they have to look into the stakeholders which are using the Internet and not in the groups which are providing the service.

 So the CEO of ICANN has introduced a couple of -- probably already years ago, this terminology technical Internet governance.  The acronym is TIC.  And this makes sense, because what the IP addresses and domain names and protocols are meaning, they are mutual resources, which all -- can be used by everyone.  And the most important thing for the functioning of the Internet is that you have access to these resources and you can use them easy and cheap.  And this is provided by the various stakeholders which are in this technical community which were described by Chris, Olaf, and Adam.

 So -- And it's also an interesting observation from the recent two years that, you know, in this incredible crisis which came, this pandemic, this COVID-19, and with so many day-to-day operations has been moved into the Internet, that day-to-day function, so it was not a big problem, you know, to have new websites, to have new services.  So that means the technical infrastructure, surprisingly, worked very well.  So that means it could be used by -- by anybody, and it was able to accommodate these millions and billions of requests coming from all over parts of the world.

 There was certainly misuse.  And we see an explosion of cybercrime also related to the COVID-19 pandemic.  But, again, you know, this is not related to IP address management or domain names.  This is related to criminal gangs, to organized crime, sometimes even state-sponsored groups, you know, terrorist groups and other individuals, you know, which have criminal behaviors.  And this is a different thing.  And, again, to have this differentiation between the day-to-day operation of the technical issues and the Internet-related public policy issues is important for policymakers also with regard to the future.

 And so far, the explanation which came out on the three presentations, that the Internet is a layered system, you know, the hourglass model which was presented by Chris, and -- So that means you can regulate.  And that's the beauty of the layered system.  You can do policies on each of the different layers.  So that means if you deal with issues like cybercrime or content-related issues, so there is no need, you know, to extend such policies for the management of IP addresses in domain name system.  That means you can do your policy related to content, crime, or whatever, without, you know, interfering into the management.

 I think the biggest risk for the future of the Internet is if governments have an interest or would make the mistake to pull the technical Internet governance into their countries.  So that's -- the domain names and IP addresses becoming the football and the controversies among governments.  We have seen this also in the U.N. General Assembly, the speeches from President Biden and President Xi and others.  There are political conflicts.  So everybody says we do not want to have a new Cold War in cyberspace.  But, you know, there are -- you have different policies.  And you should be aware and should be realistic that, you know, cyber is a field of conflict between the big powers now, big cyber powers, economic conflicts and political conflict, military conflict.  We see weaponization in cyberspace, new weapons emerge and all this.  So these are big policy issues which has to be settled.  But this has nothing to do with the management of IP addresses and domain name system.  And this is very clear.

 And I think this is also important.  If you start now looking forward into the year 2025 -- and this is my last comment.  So in 2025, there will be a review of the Tunis Agenda of the World Summit on the Information Society.  The conference is called WSIS + 20.  And this needs a good preparation.  So we have now 2021, so it's only four years to go.

 I think the whole World Summit on the Information Society was stretched over four, five years of discussion.  So it would make sense, although within the IGF and also among parliamentarians, to look forward and to say, you know, if we want to reach something, if we want to make the next step, a stumbling step forward, the next opportunity is 2025 to reach proper consensus among the governments.  So it means the discussion has to be started now.

 And in some bodies, like UNESCO, the ITU, we have seen already, you know, some draft resolutions where the organizations are invited to start the preparations.  So that means I want just to ring the -- not the alarm bell, but to ring the clock and to say, okay, if you want to do something in 2025, you should start early and based on the understanding we have achieved in the Tunis Agenda.

 So Tunis Agenda was 20 years ago.  And as I said, things have changed, and a lot of things are changed.  But the basic conflicts, the basic structure didn't change.  And the existing mechanisms have proved that they are working.  And COVID-19 was an argument for the noninterference of governments into the technical management of the Internet as it was accepted in Article 55 of the Tunis Agenda, which says, you know, that the existing mechanisms are working well.

 So with this invitation to look forward to 2025, I finish here.

 And certainly like the others, I am happy to take any questions.

 Thank you.

 >>SORINA TELEANU:  Thank you also, Wolfgang.  We're exactly at the top of the hour, when we should be wrapping up the session.

 I'm going to quickly apologize to our scribes and interpreters.  We will take a few more minutes to actually close the session.  Thank you for still being with us.

 Do we have any questions for Wolfgang or for anyone else, for that matter, since we're wrapping up?

 >> Yes, I have a question.

 

>>SORINA TELEANU:  Liva, I think Adam has just responded in chat to that question, unless you have another one.

>> I had a different question with regards to -- related to the parliamentary track.

 >>SORINA TELEANU:  Let us -- I will give you the floor in two minutes on that.

 >> Oh, okay.  Thank you.  Sorry.

 >>SORINA TELEANU:  No worries.  And since we do not seem to have any other question, I am going to ask quickly all of our speakers to respond in a tweet or as short as possible and in some sort of (indiscernible) back to everything that you've just said to this question.

 What will you say policymakers would need to keep in mind when devising legislation which is relevant in some way or another to the digital space, the Internet, but also beyond it, to avoid these unintended consequences that we've spoken about throughout this session?

 Again, principles, key characteristics, anything you would like them to keep in mind while you wrap up the discussion.

 And please keep it short.

 Chris, I see your hand.  Go ahead.

 >>CHRIS BUCKRIDGE:  I'll keep it very short and just say open communication.  I mean, as technical community, we go about the open Internet and open networking.  But I think the other thing is to keep that open communication between the stakeholder groups.  And it's not always easy for governments, regulators, parliamentarians who have lots of different policy areas to cover.  I think events like this are a good first step to make clear that there are people on that technical community side and the industry side who are ready to engage, ready to offer explanations, ready to brainstorm about policy solutions.

 So, yeah, the more communication like this we have, the better.

 

>>SORINA TELEANU:  Thank you.

 Please, Adam.

 >>ADAM PEAKE:  Absolutely, communication between the different groups is essential.  And I know that all three of us representing the different groups that we're from are very, very open to that.

 I think it's very important to remember something that -- well, what Olaf explained about the 70,000 autonomous network, the interconnected network of networks that makes up the Internet.  It's very easy to think of when you see it in the media that people talk about the Internet as if it's Facebook.  And this goes to Wolfgang's point about technical Internet governance.

 In 2005, 2006, when we had the definition of what Internet governance is, we were speaking about, of and on the Internet.

 So of the Internet are the technical functions.  And on the Internet, the services that layer on top of it, all the applications and so on, the content of the packets that Olaf described.

 And I think it's worth noting that over the last 18 months, as we've all gone online all around the world, shifted our work on the Internet, the actual technical layer, the IP packets that are being assigned by Chris, the different routing that Olaf described, the policies developing new top-level domain names, all of that functioning has happened perfectly, it's occurred online.  And this layer of the Internet has not -- has scaled perfectly, responding to all the demands that the pandemic has put upon it.

 And we can only continue to do that by making sure that we communicate amongst ourselves and understand all the issues that are happening here.

 So thank you very much.  Great to see you.

 

>>SORINA TELEANU:  Thank you, Adam.

 Olaf?

 >>OLAF KOLKMAN:  Yeah.  All of the above.

 What I often say is, different players at different layers.  That concept of having different people in groups and entities responsible of different parts of the Internet I think is critical.

 And I have been unapologetically spamming, I would say, the chat box with the Internet way of networking, which describes those critical properties of the Internet.  And to me, that's the short version of the answer.  Those are the properties that we think need to be preserved to keep an Internet, a global connected Internet.

 There are a couple of additional things that you need to do to make sure that it is open, secure, trustworthy.  But those are the basics.  If you don't protect those properties, you make -- you have the chance that you re- -- you lose this beautiful infrastructure that has connected humanity, this amazing thing.

 That's my contribution.

 

>>SORINA TELEANU:  Thank you, Olaf.

 And Wolfgang?

 >>WOLFGANG KLEINWACHTER:  Yeah.  Thanks the Internet when it emerged was based on trust.  So that means people just wanted to reach out to communicate and they trusted each other.  So we had in the 1980s and '90s something which was called a netiquette, so rules more or less everybody accepted and followed.  

 So, unfortunately, this has gone.  Today, we live in a world of mistrust.  And the Internet will work properly only if we can rebuild trust among the various stakeholders.  And so this starts with the governments.  So that means, at the moment, I see a huge mistrust among all governments.  The Russians do not trust the Americans and Europeans, and the Chinese do not trust the Russians, and the Americans do not trust the Chinese.  So I think this is really a big issue for the functioning of the Internet.  If you have a system which its functioning is based on trust and we have mistrust in the world, then you have to do something.  And so far, you know, methods for confidence-building, you know, demonstrating you can work across differences in ideology and economic (indiscernible), in policies.  You know, the world is big and can accommodate a lot of different ideas.  So -- but -- And the Internet so far has demonstrated that this is possible, that you can work across political lines and provide services, you know, to the benefit of all.  

 So the Internet is seen as a public good.  So it's good for everybody.  And ICANN's slogan, you know, "Internet for everybody," or ISOC's slogan, "The Internet is for all," this is the challenge.  And so far, you know, it would be good if the community, including parliamentarians, would engage in more trust and confidence-building in the Internet governance ecosystem.

 Back to you, Sorina.

 

>>SORINA TELEANU:  Thank you, Wolfgang.

 And thank you to all our speakers for today's interesting discussions.

 On that note, following up on what Wolfgang has just said, we hope the fooling sessions as part of the parliamentary track will help build for communication, more opportunities for collaboration and maybe rebuild this trust that we keep talking about.

 Neema, over to you.  I know you had a question.

 

>>NEEMA LUGANGIRA:  Thank you very much.  

 First of all, this session has been very important, very, very enlightening.  So I'd like to commend you all and thank you for preparing it.

 I noted on the parliamentary track it was talking about issues of data, of privacy and legitimate use of personal data, issues of content, policy, you know, balancing freedom of speech, et cetera.  And one of the areas that I wanted to pose a question is, oftentimes, you know, society expects parliamentarians to be accessible through the Internet.  But the -- you know, the downside of it is also there's been an increase in online abuse or, you know, mob attack online on parliamentarians, particularly women parliamentarians.  

 So I just wanted to get an understanding, is this an area that also the Internet Governance Forum would also look at to see how can we be capacitated as members of parliament, one, to make sure that our different countries have the right policies in place for, you know, data protection, but also looking at how to make the Internet a safe space for all without forgetting the leaders, especially political leaders who are prone to it.  And also how to capacitate as parliamentarians on how to properly use the Internet, you know, and all these different jargons, et cetera.

 That's what I wanted to pose.  Thank you very much, Sorina.  Thank you.

 

>>SORINA TELEANU:  Thank you, Neema.

 And to follow up, as we go into more substantive debates on policy issues, we will go into this as well and deal with content quality and Al other challenges related to Internet content.  I'm very glad you raised that up because it's also giving me some ideas into what to put in that session.

 On that note, thank you.  We're really, really running out of time.

 I would kindly remind everyone that we do have four more online sessions as part of the track and everything is available on the IGF website.  I'm going to share the link once again in chat so you have it.

 Please make sure you are registered for all of these online sessions but also registered for the IGF taking place in December, whether you are going to Katowice or not, registration is key.  So you have to do that, no matter what.

 The next session will be on 7 October.  We will discuss privacy, data protection, and some of the challenges related to this in the digital space.  And once again, I do kindly encourage you to participate.

 And one final kind request.  To help us improve for the other sessions, we have a very, very quick feedback form.  I've just shared the link in chat.  It would be great if you could spend just literally just one minute to tell us a bit about how this session was for you and what we can improve for the other four online sessions and also the sessions happening in December.

 And on that note, I will wrap up this session unless someone would like to add anything else.

 I see nothing.  And many thanks to everyone.

 Someone was unmuting?

 Or maybe not.

 Again, thank you, everyone.  Have a good rest of the day wherever you are.  And see you again on 7 October.  And many, many thanks also to our scribers and interpreters also for staying overtime.

 Good-bye.

 

>>ADAM PEAKE:  Thank you very much.  Good-bye.

 >>CHRIS BUCKRIDGE:  Thanks, all.  Good-bye.

>>OLAF KOLKMAN:  Good-bye.