IGF 2022 Day 0 Event #68 Understanding Internet Fragmentation: Concepts and their Implications for Action

The following are the outputs of the captioning taken during an IGF intervention. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid, but should not be treated as an authoritative record.



>> BILL DRAKE: So, all of the online speakers are here, and all of the onsite speakers are not. 

>> WOLFGANG KLEINWACHTER: Yeah, can you hear me? 

>> BILL DRAKE: Barely. 


>> BILL DRAKE: Could you guys maybe talk to whoever's doing the technology in that room ‑‑ because you're very hard to hear.  Anriette, given the ‑‑

>> ANRIETTE ESTERHUYSEN: We are not hard to hear in the room, and neither are you. So, I think we should just, whoever's controlling the Zoom, to see if they're able to do anything. Can the other remote participants hear us? They say they can, so ‑‑

>> BILL DRAKE: We can hear you, yes, it's just that you're much quieter than the people who are online. 


>> BILL DRAKE: All right, that being the case, then, Anriette, would you like to start? 

>> ANRIETTE ESTERHUYSEN: I'm afraid not because I have to plug in my computer because my battery is dead. So, I would really appreciate it if you can start. Sorry, I had to rush straight out of another session, so I would appreciate it if you could get us going while I set up. 

>> BILL DRAKE: All right, fine. Hello, everybody.  Welcome. Good morning, good afternoon, good evening, wherever you may be.  This is Day Zero event on Internet Fragmentation and Concepts and their implications for action. I am Bill Drake from Columbia University, and I am sort of instigator of this event.  I am the online moderator, and Anriette Esterhuysen, who is there in the room, will be the onsite moderator. And hopefully, together we will be able to bring people into the conversation from both the virtual and the physical worlds.

So, just introductions. So, I'm Director of International Studies at the Columbia Institute for Tele-Information and teach at Columbia Business School and am a longtime participant in Internet governance issues, including the IGF before it existed. 

Anriette Esterhuysen is Senior Advisor on Internet Governance for Association for Progressive Communication and was the MAG Chair from 2019 to 2021, based in South Africa. 

Neha Mishra is an Assistant Professor of International Law at Geneva Graduate Institute, where she researches legal issues in the digital economy focusing on international economic law, data flows and digital trade and the interface of trade law and emerging digital technologies. 

Milton L. Mueller is Professor in School of Public Policy and School of Cybersecurity in Privacy at Georgia Institute for Technology in the United States, not Georgia the country.  He is the Director.

(background noise)

Whoever has got their mic on, turn off, please.  Could people, whoever is having the conversation unrelated to the panel, please mute your mic. 

>> MILTON MUELLER: Can the host mute them? It's Siva.  Thank you.  Now you're muted, Bill. 

>> BILL DRAKE: This is fun!  Founder of the Internet school of governance, former ICANN Board Member and with the Global Commission on Stability in Cyberspace.  And her portfolio covers GPDs policy areas, trust and security, online content and technologies. 

Andrew Sullivan has worked on Internet issues since 2001 and has been Chair of the Internet Architecture Board. 

Xu Peixi is a Director of the Global Internet Studies Center at the Communication University of China, research interests include international communications, governance and cybersecurity, has authored three books on geopolitics of Internet governance. 

Okay, let me say a couple of words by way of background about what we're doing here. Then I will begin by throwing out questions to the panelists to initiate conversation.  And from there, we will go around for 50 minutes or so and then throw it open to the floor. To all participants weighing in, I see we have about 45 people, so we should have a good basis for an interesting conversation. 

Anriette, you're back in place, so fine. Anriette is there, and she will manage the side of things in Addis.

So, just to frame this conversation, Internet fragmentation is not a new topic, of course. Since the 1990s, there's been a lot of discussion about the tensions between territorial authority and the Internet, borders and cyberspace and all that. But in the 2010s, the term "fragmentation" began to appear and be applied to government barriers and various technical misfires and so on, but there was really a boom of interest in the topic after Snowden in 2013 and a lot of debate about fragmentation and splinternets and Balkanization, and the EU constructing a Schengen cloud and all these things. There was in this context the statement from the leaders of the ISTAR organizations warning about fragmentation.  We put out a paper with Wolfgang and myself. Milton wrote a book about it. And so, there was this growth.

But now in the past two years, the topic has really exploded like crazy. Expressions of concern and policy action on fragmentation have moved up the international political agenda. We're seeing statements and initiatives from the White House in the United States, the European Union and its members, the UN Secretary‑General has made it a central theme of his Global Digital Compact discussions. Various statements and analyses by the Internet technical community bodies, a flurry of academic papers, conferences, discussions, et cetera, a policy network in the IGF, as I mentioned, that's been formed and think about these things.

And what's notable in all of that is that there is no consensus on what fragmentation is and is not. In fact, there is sharp disagreement and a lot of people talking past each other.  It's kind of a tower of babble type situation, which is familiar to those of us who are Internet governance veterans. There are a lot of parallels with the definitional debates 20 years ago, before and during the real summit on the information society or WSIS. People were talking past each other about what is Internet governance from various angles. Some people had a very narrow understanding of Internet governance of being just about the management of the infrastructure, particularly identifiers in RootSum file, and some people said, okay, that's ICANN, and then some people said, well, it should be the ITU who does that.

Then there was a broad definition approach that included not just the infrastructure, but rules pertaining to the usage of the Internet, free information, communication and commerce. And the simple binary emerged in a lot of communications of governance of the Internet versus governance on the Internet, and there's sort of the same kind of discussion happening today with fragmentation. A lot of people take a narrow view that fragmentation only can apply to a total rupture of the technical infrastructure, while others take a broader view and see fragmentation originating from a variety of technical conditions in government business practices that block interoperability and data flows between willing partners and so on.  So, we have, again, this kind of narrow, broad debate going on, and it matters how issues are discussed and institutionalized, and the international agenda matters to outcomes, to real‑world behavior.

I'll just give you two quick examples.  In 2005, the Working Group on Internet Governance that was appointed by Kofi Annan to come up with a definition of Internet governance, when we had our first meeting, the ITU Secretary‑General came and told us that we have to adopt a narrow definition and that this, by extension, should be that the ITU's in charge because it does infrastructure. But the WCAG made clear that the IG process involved many institutions and it only mentioned the ITU once in a footnote. And the whole kind of ITU versus ICANN aspect of the WSIS battle originated from that.

The council in the United States is an influential group of policy types that came out with a big report on confronting reality in cyberspace ‑‑ Foreign Policy for a Fragmented Internet -- that says that because of Internet fragmentation, U.S. policies promoting Internet openness and freedom have all failed and the U.S. should shift towards a great power competition mind‑set on all Internet matters, but the report never defines what fragmentation is. It never says how strong it is, why it should ‑‑ what exactly is being impacted by it, and why this means that there is no global Internet and policies for openness have all failed.  So, the way people talk about these concepts can be consequential for action. So, that's why we want to try and contribute, since this is a ‑‑ fragmentation's a main theme of this whole conference. We wanted on day zero to contribute to some thinking about fragmentation, try to get people who have been working on these issues for a long time and have insights, to share some thoughts that might help to frame the discussion going forward.  So, that's what we're going to do.

So, I'm going to do three rounds of questions to the panelists, let everybody kind of weigh in at the top. And then when we get to like the top of the hour, we'll open it up to the floor and get everybody involved.  And Anriette, since you didn't get to speak, why don't we treat you as being a panelist as well. Whenever you want to speak to one of the questions, please just join in the conversation. 

So, let me ask the panel then to start. Let's talk about, what is the state of the debate? There is all ‑‑ as I said, there's all these expressions of concern, calls for action, about Internet fragmentation. What's your general sense of the current moment of the discussion? Why are we having this boom of interest in fragmentation right now? And is it useful and constructive? And in particular, what's the state of play and debate in your country or region or expert community about fragmentation? Because fragmentation's being discussed in a lot of different ways.

Can I ask the panelists to please weigh in on that set of questions to get us going? Who would like to start? 


>> BILL DRAKE: Could you use the raise hand function? It's easier. Milton, go ahead, please. 

>> MILTON MUELLER: Sure. So, I think it's very clear what the state of play is right now.  I think the reason we are ‑‑ this fragmentation, so‑called debate, doesn't go away is, based on the argument that I made in this book, you know, five years ago, and that is that the fragmentation debate is really one and the same with the so‑called digital sovereignty debate. So, it not about diverging standards or technical fragmentation; it's not about walled gardens by private actors. That isn't happening. 

The Internet identifiers have not split. We still have an integrated DNS. Everybody's still using TCP IP. So, what is it really about? It's about a counterrevolution against the globalizing tendencies of the digital political economy. On the one hand, you have this incredible technical interoperability that was created on a global basis by not just the Internet protocols and identifiers, but we have an integrated market for software, for semiconductors and digital devices, and this global compatibility gave rise to a global division of labor in ICT protection to a distributed and integrated digital ecosystem. And up till now, it's fostered free and open trade in information, products and services.

Now, counterposed against this is the territorial state which wants to border and control the digital ecosystem. They do this in the name of national security or to protect a subsidized, favored constituencies, to gain an advantage in military technology or engage in espionage. So, in my view, the fragmentation debate means answering this we: Do we want to continue to foster the globalized exchanges, competition, and innovation enabled by this global compatibility, or do we want to impose the kind of customs, borders, taxes, military checkpoints, nationalistic attitudes and barriers to trade associated with traditional nation state governance? 

>> BILL DRAKE: Okay. So, for you, then, to the extent that this is a discussion about fragmentation and it's only about state policy, it's not about commercial practices, and it's not about technical incompatibilities or coding problems and things like that. 


>> BILL DRAKE: Okay. Andrew? 

>> ANDREW SULLIVAN: Yeah, thanks. So, I think I agree with Milton that there is a significant part of this that is just the attempt by nation states to sort of reimpose their framework on the networking environment. I think where we disagree is the idea that this is not merely ‑‑ or this is not ‑‑ this is perhaps related exclusively to nation states. So, I've been thinking a lot recently about what I've started to call, you know, sort of networking autocracy, this idea that you can be a self‑sufficient network of any kind, and you kind of build a wall around it. And that is true, actually, of a number of private efforts, as well as a number of nation‑state efforts, but all of them really amount to the idea that you can sort of Internet alone. And this is a silly idea, right? I mean, like, it's a terrible, bad idea, but it's one that seems to have set in to our discussion, that you can sort of carve out a piece of networking life, and then you can have it all on its own. And I think that that idea is what's underlying all of this debate, but I think it's quite a toxic notion, because, of course, networking sort of depends on interoperation, and we appear to be at the moment where we're trying to give that up. 

>> BILL DRAKE: All right. Thank you, Andrew. So, corporate behavior is in the mix for you.  Who ‑‑ Neha? 

>> NEHA MISHRA: Yeah, I agree with a lot of what was already discussed by Milton and Andrew, and I do think that digital sovereignty is often a framing for many of these issues, but at least from a legal perspective, what really interests me is how a lot of domestic laws and regulations are trying to interfere with the Internet, not always necessarily to fragment the Internet, but with the idea of establishing sovereignty. And I think a lot of people discussed about data localization measures. And they can be for a variety of policy objectives, and some of them do actually have some impact on the technical infrastructure in the sense that there could be some requirement regarding routing of data through maybe not routing of data outside the borders of the country. So, I think that's one interface that is also quite important to understand.

My claim is not that different regulations mean fragmenting the Internet, but it also requires some consideration of how governments are trying to frame different kinds of laws and regulations. So, I'll just give you one example, that of Cambodia, which recently they passed a decree where they said, all Internet traffic has to be routed through their national Internet gateway, which is controlled by the government. And the idea behind this is purely domestic regulation. They want to control the kind of content that comes in and out of the country.

But it's not here just about regulatory differences, but it also affects, to some extent, the infrastructure, the universality of the Internet. So, I just wanted to flag that as well. 

>> BILL DRAKE: Thank you, Neha. I think what we're doing so far is giving our bottom‑line views on fragmentation, rather than addressing the question of why now exactly and what's being discussed in your neck of the woods, so it'd be nice if we could try to do a little bit of both of that. 

I am not seeing, by the way ‑‑ Okay, Peixi, I'll come to you, if you want.  I am not seeing the panelists who are there in Addis on the screen. I don't know why. 

>> ANRIETTE ESTERHUYSEN: Wolfgang is ready to go, Bill. And I was going to emphasize what you just emphasized. So far, the panelists have not really answered your question. So, I hope Wolfgang and Sheetal will. Wolfgang's ready to go, Bill. 

>> BILL DRAKE: Okay. It'd be really helpful if ‑‑ oh, there, now I see you again. Wolfgang, go ahead, please. 

>> WOLFGANG KLEINWACHTER: Yes, thank you very much. 

>> BILL DRAKE: Speak loud. 

>> WOLFGANG KLEINWACHTER: The key word was used by Andrew, the interoperability. So, I think the beauty of the network which emerged from the 1970s and '80s was that everybody could communicate with everybody. And now we have 5 billion people online. And this is more or less still the unfragmented technical infrastructure of the Internet.

Bill has returned to the agenda where we had this differentiation between the technical layer ‑‑ the development of the Internet ‑‑ and the policy layer ‑‑ the use of the Internet. 20 years ago, the main debate was on the development, on the critical element resources domain names/IP addresses. But now, I think the big bubble is on the policy layer. So, that means the time has changed, and so the relationship between the technical and the political layer has also changed. On the technical layer, we have the philosophy, one world, one Internet. But on the policy layer, we have one world, 190 jurisdictions, or one world and 10,000 different companies, you know, which have wall at gardens. And that's the problem how you organize the relationship between the two layers. So, I think the dreamers of the 1990s and the early 2000s was that the technical layer will influence the policy layer so that we will move into a harmonized world. 

But what we have seen the last 20 years with all these political conflicts, that now the application layer moves downwards and threatens the technical layer. And in a worst‑case scenario, you could imagine that you have also in the technical layer 190 national segments. And then you could have a situation like we have in the travel world, when you move from one country to another country, you need a visa, entrance visa, or you know, permission to leave your country.  So, in the Internet world, it could be that you need a special password if you want to send an email to an outside server.  So, the world is full of idiots!  And some people ‑‑ Anriette said, it's a very, very better idea, but do not believe that better ideas are impossible to implement. There are people in the world who want to implement such bad ideas. And then so far, I think events like the IGF should make a wake‑up call so that the public understands the threat which is in the air to undermine this incredible achievement that 5 billion people can talk to each other and on the technical layer.

So, the good news is that both the Declaration on the Future of the Internet, which came out from the White House and the white paper which was published in Beijing just two weeks ago, used the terminology "interoperability" as an important element. And as long as this is safe and the big two cyber superpowers all agree on this, so we have some hope. But let's wait and see how this will be further developed. 

>> BILL DRAKE: Thank you, Wolfgang. And everyone, please, if the panelists could try to stick to around two minutes, it will be easier for us to get through the agenda and then open it up to the floor. There are people already raising their hands, itching to get into the conversation, but I'd like to get through the panel first.  Peixi, did you want to say something? 

>> PEIXI XU: Yes, Bill. I think there are these different metrics. One is the narrow kind of metric you have mentioned, and it's about the technical aspect of it. I'm still using this term to talk about fragmentation. However, I'm observing in China that a lot of my colleagues are using the term "fragmentation" to talk about the data.  I think, Mishra, you have mentioned about this. That is about the trend to legalize the digital space. I think if that is allowed by the European Union, as the digital sovereignty notion talks about, and Wolfgang has mentioned, for example, about why Internet, one world, 193 jurisdictions. I think that notion about data means that the Internet has already been fragmented, unlike the technical part of it. 

I'm also observing, for example, in China, we use the word to talk about the ideology called division of cyberspace; for example, dividing the vendors, the equipment provider into trustworthy or untrustworthy providers and also dividing the applications into trustworthy ones and non‑trustworthy ones. Of course, that is not a Chinese notion; that is represented by the Declaration for the Future of the Internet. So, there is a kind of ideological division. And Wolfgang has mentioned about the white paper that China has published. That means that a lot of my Chinese colleagues are supporting, actually, the ICANN system, the technical aspect of it, but we are worried about all the fragmentations in the different layers.  Bill? 

>> BILL DRAKE: Very interesting. Good to get a report from Peixi, because the conversation I hear is driven out of the U.S. and Europe, so I'm really curious to hear about how it's being discussed elsewhere.  I can't see the panel again who are there, but I know that Sheetal and Anriette are on the panel. Do either of you want to weigh in on this point? 

>> ANRIETTE ESTERHUYSEN: She's ready to talk. 

>> BILL DRAKE: Why are you not on camera? I don't get it. 

>> ANRIETTE ESTERHUYSEN: Don't worry about it. 

>> BILL DRAKE: Okay, but I can't see you. 

>> SHEETAL KUMAR: Thank you for having me. Bill, you mentioned that this topic has suddenly burst onto the scene, and I think, you know, if it's been in the past few years, we also have to remember, in the past few years, we've had another big thing that's happened to us all, namely, the pandemic.  And I feel like one of the reasons that this discussion has become so much at the forefront of discussions at the IGF and also elsewhere, is that we have become more dependent on being online. 

More people have become more dependent, at the same time that we have seen an increase in attempts to control user access, and that's taken away control from users themselves. We've heard examples of that. It's happening through policy in particular. And I think it's an opportunity, when you ask, is this a good thing, is it not? I think it's an opportunity for us to defend and promote and rearticulate the values that we want to see expressed by Internet governance, by policymakers, by commercial entities, indeed, by all, because there is this underlying assumption, whether you think of the internet as social media ‑‑ you know, we've all heard people say, oh, the Internet's gone wild!  Because some celebrity has posted something. It's not the Internet that's gone wild, but it's people. We think of it that way because it's such a human‑centric technology by its very nature. So, even if you don't agree with that perception, it is ultimately about that. 

And as we move ‑‑ or people cannot use it in the same way in that way it was envisioned, and as more and more people cannot, and as more and more people have that taken away from them, I think it becomes more of a question, what do we do about that? And we have heard examples of how the technical layer continues to be a space where, in theory ‑‑ or rather, in practice, really ‑‑ data flows ‑‑ data can flow, it can interoperate. All of that is well and good and functioning. But if the practice and the experience is not that, and increasingly not that for many, then we will continue to have this conversation in this way.

And I think we also heard an example of how there can be, perhaps, inadvertent changes or ways that the Internet is evolving, including to ensure that more people can get online. Where those perhaps values or different visions of the Internet that we're seeing more at the policy layer, if we want to call it that, could also ‑‑ I think you were mentioning that, Wolfgang ‑‑ we could also see that at the technical layer. So, I think we're having this conversation now because we are all so dependent on it, but we are also assuming and desiring that it maintains that vision of being an open space where anyone can connect to others, and that is changing and that's how people feel about it. 

So, I hope that's okay to start us off, Anriette? 

>> BILL DRAKE: Thank you very much, Sheetal. You know, I think what, in particular, you flagged there, that's helpful, is sort of the notion that for a lot of people, fragmentation is about user experience, about what's going on at the sort of layer above the protocol stack of whether or not people are able to exchange data and so on, which raises questions about, you know, how do we think about blockages of the Internet? Some people think blockages don't constitute fragmentation. Some people think they do. This is a point I want to come back to again.

Anriette, do you have anything in particular on how it's being discussed in of area? 

>> ANRIETTE ESTERHUYSEN: I think all I would add is what some people describe as manifestation or fragmentation, often also comes from a place of trying to find solutions for what is experienced as Internet‑related policy challenges. I think I really ‑‑ I think Wolfgang expressed that well. How do you deal with data, data requests? How does law enforcement agencies deal with demands for user data that they might need to pursue a certain investigation? They might be very happy to follow rule of law; they're comfortable with rule of law; they even respect human rights, hopefully, but it's challenging.

So many issues of taxation. How does a country ensure that it generates sufficient revenue from the profit made or operations of global Internet companies within their jurisdiction? So, I think there's not always a space ‑‑ it doesn't always come from a space of wanting to disrupt the user experience or wanting to fragment the Internet as a whole. It comes from a space of trying to find solutions for what is experienced as problems that are often not very well understood, and therefore, the solutions often don't really match the problem, and the result is this sense of a fragmented approach to Internet policy and regulation. 

>> BILL DRAKE: But Anriette, if I can press you on that just a little bit, that sounds like you're opening the door to a very expansive understanding of fragmentation, if you're saying that people are talking about fragmentation because there are problems in the Internet and they want solutions to all kinds of different problems, then isn't there the risk, then, that the current becomes the catch‑all for all kinds of things that don't actually in other understandings, fragment, segment, separate the Internet into spaces, reduce interoperability and so on? 

>> ANRIETTE ESTERHUYSEN: I think it depends who it is that's talking about fragmentation. I think different actors and different interest groups are talking about fragmentation in different ways that reflect their different priorities. So, I think if you are a company that wants to operate in a borderless world, where you don't want to be subject to national regulation because you're a global company and it's extremely expensive and cumbersome and it doesn't really work on the Internet, then any efforts to establish national regulation (audio fading in and out) comply with, you are going to present that as fragmentation, because in a sense, it does.  It fragments and complicates your business model, if you're a human rights activist who's trying to battle Internet add‑ons because they completely distort the user experience, so does censorship, you're going to cause those as Internet fragmentation. So, I think we're in a limbo space how people, in particular, state actors, are thinking about the Internet is changing, and we are using the language of fragmentation to try and describe our fears for what the impact could be of this changed behavior, particularly of state. Sheetal wanted to react. I'm not sure if I answered your question, Bill. 

>> BILL DRAKE: Yeah, you did. 

>> SHEETAL KUMAR: I just want to say that ‑‑ I know we're going to talk about the Policy Network a bit more, but I wanted to say that in the Policy Network discussions that we've had, there has also been ‑‑ as you said, Anriette ‑‑ a continuous reference to the point that it's not always desired, you know, the impact. It can be inadvertent, the impact.

The point is that, perhaps if certain situations arise, a certain amount of factors come together, then the impact can be a fragmentation, and that might be simply a question of unpacking that. And I think we are doing that in the Policy Network. We need to continue to do that. But the point is, it is not always something expected or desired. It can be a result, simply, of trying to do something to solve a problem. 

>> BILL DRAKE: Thank you. Sheetal, it's an important point. I know it's something in our paper we talked about, inadvertent kind of results, unintended consequences, versus intentional fragmentation. And there certainly are a lot of people who talk about technical fragmentation in that manner of unintended consequences. But it seems like everybody wants to dig down into the definition stuff, rather than what's going on in my neighborhood, so let's do another round of what exactly are we talking good here?

When we talk about fragmentation, what are people's core criteria for determining what, if anything, constitutes fragmentation? What is being segmented or fragmented from what? I mean, the notion of fragmentation implies something that was unified or connected becomes broken in some manner. So, what exactly is being broken from what? How lasting and structural does that breakage have to be in order to count as fragmentation, versus can it be just temporary or short‑term kind of thing? So, would you count ‑‑ for example, if you're talking about blocking, isn't there a difference between a great firewall that blocks 1.3 billion people from using a range of apps, platforms, URLs, et cetera, versus a short‑term Internet shutdown because of political instability that's then restored six hours later? Does scope and duration matter in talking about blockage as a form of fragmentation? So, what is the really core criteria of what we're talking about here? How do we define this construct? Can I go around the panelists again on this one? Milton, would you like to start? 

>> MILTON MUELLER: Sure. Thank you.  So, I do have an answer to that question. You know, I'm the one who really tried to formulate a very precise technical definition of what fragmentation would consist of, because you're right, we do have to pay attention to scope, duration. We can't just say because, you know, a backhoe cut my cable in my backyard and my Internet service was down for three hours that the Internet is fragmented. That's not a systemic view of what fragmentation consists of.

Likewise, there are people with perfectly legitimate concerns about other forms of fragmentation that are more products of market competition. In other words, somebody introduces a new version of a browser or some kind of modification of HTTP that leads to incompatibility on their computer. They maybe have to upgrade their software or something unpleasant like that, and that is something that, again, I would not call fragmentation, unless, again, it is very persistent and leads to a literal fork in some kind of critical part of the infrastructural software that is running the Internet and most of its applications.  And I think we'd all agree that if we had a split domain name route, which people have been predicting for years and has never happened because it doesn't make any sense, that we would all consider that to be a fragmented Internet.

But the reason I think I put so much emphasis on the jurisdictional and legal and political part of it is because that's what's happening now; that's what's really driving the debate; that's why this debate has hung around for so long. And you know, it's not like this is one of these sorts of liberal democracies versus authoritarian states things.  We see some of the world's leading liberal democracies, whether it's the U.S. imposing export controls or the Germans imposing, you know, national‑level content controls, all states, by virtue of their status as states, are pushing towards a kind of disruption of the global compatibility of the Internet. 

>> BILL DRAKE: Okay, Milton, just to be clear ‑‑ and I think your view is clear, but just for the audience who may not have heard it. You don't count any of the business‑type stuff ‑‑ wall gardens, things like that ‑‑ as fragmentation. You do think there are concerns about government policies, such as the great firewall‑type thing, but you don't want to call that fragmentation; you want to call that alignment. And my question would be, if you have large numbers of people unable to access and use the Internet, why is that not an example of fragmentation per se? 

>> MILTON MUELLER: Well, if large numbers of people are unable to use the Internet because of government action, then definitely that does count as, you could call that fragmentation in my view. 

>> BILL DRAKE: Okay. Then I think your position has evolved a little. That's fine. Thank you. Andrew. 

>> ANDREW SULLIVAN: Thanks. Once again, I agree and disagree with Milton.  The thing that I, I mean, I guess I don't care precisely where we draw the line around what is fragmentation versus some other badness that we might want to characterize as a different thing.  What I think the critical thing for me is, it's really that fragmentation or any other sort of drift away from the Internet and its fundamental approach of interoperating, independent networks is a problem. So, the Internet has never been a unified thing, right? Like, that is ‑‑ the idea that the Internet is a monolith is a terrible idea. Like, it's just factually false, and it's always been false, from the very beginning of the Internet, the very earliest networking efforts. There were definitely pieces of it that you couldn't access from every other place on the Internet. And that's never been a weird thing.

The difference is, whether we agree that we're trying to interoperate more generally or whether we're trying to hive off this or that part of networking for, you know, a given population or for a given group of people or for a given nation‑state or for a given group of users or whatever. And what I see as a more general problem is this drift away from interoperability as the primary, you know, sort of impulse and towards this kind of enclosure of users, of, you know, groups of populations, whether they're commercial users or whether they're, you know, citizens of a given country or whatever. And for that reason, I think that, yes, for instance, a shutdown for six hours is on the spectrum of fragmentation. It's along that spectrum. It's perhaps not as bad as, you know, Runet or whatever, but it is definitely within that direction. And it's this conceptual direction of are we moving towards interoperation or are we moving away from it, I think is the way to think about this, not a sort of binary matter, you know. Am I bald yet or am I fragmented yet? Not an interesting question. It's instead, am I getting balder? And the answer, by the way, is yes. And am I getting more fragmented? Also, in the current situation, I believe the answer is yes. 

>> BILL DRAKE: I very much agree with you on the getting balder part, but more generally that fragmentation is to me not a binary on/off thing. There is a spectrum; there's a range of different stuff. And here, Milton and I have a fundamental disagreement, and that's fine.  And you also have a disagreement with Milton because he's saying commercial practices, enclosures like you just said, don't count, and you're saying do.

Neha, do you have a thought for us? 

>> NEHA MISHRA: I think I like the framing of spectrum. I think that's a very good way of capturing different views on Internet fragmentation, and I do think it is beyond technical fragmentation, and I'd also like to re‑emphasize my point that if governments could, they would use legal devices to also bring about some technical fragmentation, and they do. And they do, sometimes very unsuccessfully, but they do. And there have been consolidated efforts in the recent years, and across the world.  It's not specific countries that I'm talking about.

One thing I'd like to highlight, drawing from my area of work, there is this idea of fragmentation in international law as well. And the idea is that international law is getting divided up and there are different kinds of international law that apply to different spaces, so to the economy, to the environment, to society. So, I think that that perspective's quite helpful, in the sense that I think Sheetal mentioned about the user experience of the Internet being very different, and sometimes in response to policy problems, but sometimes that user experience might give you a very specific idea of the Internet being fragmented, and I think that might be a good lens to look at it as well. 

And that would tie up to Andrew's point as well. For instance, if my entire world is the WeChat world and that's what I think of as the Internet, and that's not how the Internet looks like in the rest of the world, just as an example, then that could also be an example of fragmentation. So, the idea of spectrum I think is a very helpful framing device. 

>> BILL DRAKE: Thank you very much, Neha. That's very interesting. Since I can't see the panel, I'm just going to call Peixi, then we'll turn to folks in Addis, and the three of you there can just go across.  Peixi? 

>> PEIXI XU: Yes, Bill. I think it may be useful to define, for example, the narrowest version of the Internet fragmentation, or we could call it the Milton Mueller Version of Fragmentation, which refers to the separate rules and associate kind of naming of elaborate systems.

I think I'm very comfortable with this narrow version. However, I think we can also imagine, for example, the broadest version of the Internet fragmentation, which may refer to, for example, the separation of the global digital system or ecology that consists of, for example, the supply chain of chips consisting of applications, consisting of cloud services, consisting of undersea cables or consisting of the which country the tech companies are listed or where the data is stored. So, I think we can also imagine the broadest kind of fragmentation. That is, anyhow, that is a kind of exaggeration of the concept. Bill? 

>> BILL DRAKE: Thank you. Let's go ahead ‑‑ again, I can't see the panel, but I know you're there. Two minutes, please, Wolfgang? I'm just going to call your names out in the abstract. Wolfgang, do you have a thought about ‑‑ my question was, what's the criteria for defining what fragmentation means? What is being split from what? What is being fragmented from what? You know, can we drill down to a core construct about this? 

>> WOLFGANG KLEINWACHTER: Okay. Sometimes it's good to remember history. So, I think in the 1990s, the question of fragmentation was the alternate route, but the problem was that the legacy route had millions of users. The alternate route had just 10,000 users. And so, the alternate route collapsed.

Then, in the 2000s, the French government introduced the idea of an object naming system that said, okay, the DNS is for the Internet of people.  But with the growing Internet of Things, we need an additional and alternate system, and so they introduced the owners. Remember GS1 and all this. And I think it was during the IGF in Nairobi in the year 2011 ‑‑ 11 years ago ‑‑ when we had a very intense debate about the Internet of Things and the ONS, and we finally agreed, and even the French government accepted this, that the Internet of Things is just an application on top of the DNS. Through the DNS, TCPI protocol survived this effort.

Then, Russia introduced in the conference in Dubai of WCAG in 2012 the segment that was not adopted by the ITU, but the confusion was, what is the national Internet segment in a globalized infrastructure? So, Russia was unable to explain it but adopted a law which allows Russia to decouple from the global Internet. It's a stupid idea, but they can do it, if they think it's useful for them, but it would not have affected the global Internet if Russia decoupled itself.

Now, in the Ukraine war, the Ukraine Minister wrote to ICANN and asked it to take .ru and .rf and Moscow out of the route. And fortunately, you know, I can, you know, was very clear that say, support Ukraine and understand their position, but they didn't do that. And says, okay, we are a neutral steward of the critical Internet resources, and we will not split the route or take something out of the route and contribute to a fragmentation debate. So, that means these is some historical experience which show that, you know, things are happening. But at the end of the day, as Milton has said in the beginning, you know, we have still the TCPI Protocol, we have the DNS, and it's working. So, even the new IP proposal China introduced a couple of years ago in the ITU is now more or less, I would not say off the table, but it has taken an interesting development.

Now, Chinese papers speak about the certain special networks on top of the DNS, or you know, to manage the weaknesses of the DNS latency security. So, for autonomous cars, for instance, probably for surgeries, you need special networks, but these are special networks, and it's not mean to the fragmentation of the Internet.

And so far, sometimes it's very natural that you see the emergence of special networks on top of this. So, this is not a danger for the Internet.  The real danger would come if you would undermine the interoperability. Thank you. 

>> BILL DRAKE: Thanks, Wolfgang. Let's see. Sheetal, yes. Go ahead. 

>> SHEETAL KUMAR: Thank you.  So, I wanted to just bring in the Policy Network at this point, because we have been having these discussions. And I would say it was full of chaos and fog at the very beginning and perhaps still is, in some ways. But what we have been trying to do is listen to everyone, and distill from that an understanding or conceptualization of what could constitute fragmentation. And I think that as we develop that ‑‑ and I encourage those of you who can come to the Wednesday policy session, which is at 9:30 ‑‑ we are seeing the user experience aspect there as well as the question around what does fragmentation at a technical layer look like, and is there some sort of intersection around that? And how does the governance of the Internet impact both of those aspects? And so, what we're trying to do with that, I think, is also what we're going to get to, and I've been hearing this in the discussions, the webinars that have led to this distilled‑ish framework, is the need to be more specific about what combination of factors ‑‑ scope, duration ‑‑ what point on a spectrum would we then qualify something as an instance of fragmentation?

It could be, on our last call, someone mentioned, perhaps when it comes to user experience, it could be a combination of measures that collectively result in separate ecosystems of applications and use that mean that you don't really have access to the global Internet in practice. So, collectively, it's separate things then result in that. I'm not saying that that's what we're going to come up with and agree on, but there's something around scope and duration and spectrum you've all spoken to, which I think will come up there. So, I think we have come to this more middle ground position, perhaps, that we intend to take to you all on Wednesday. And of course, there are other ways to engage with this and hear from you as to whether this is a helpful framework for understanding the issue and, of course, working on it and developing solutions where we collectively feel that we need to. 

>> BILL DRAKE: I think group efforts to try to sort things can be very helpful.  That's what happened with the Internet governance definition, too, in a way. So, I hope this goes well.  Okay, finally, Anriette? 

>> ANRIETTE ESTERHUYSEN: I think we should open the floors. 

>> BILL DRAKE: Okay, we can do that. We've got a half an hour left.  I do see that there are some people ‑‑ I mean, there's been a hand over here from Siva for an hour. And I see people talking in the chat as well. The only thing is, I would like to try to make sure that we at least get some input from the panelists on the sort of lending question of what is to be done, you know? What should Civil Society members, governments be doing, what is the role of international governments' agreements? So, I hope that maybe when we are responding to points raised in the open discussion that maybe we can get some people to add some thoughts on that, too. But sure, why not? Let's open it now because there are people ‑‑ and I also have noticed in the chat quite a few people ‑‑ Demetry and Pedro and others. So, raise your hand, say who you are briefly, and try to limit yourself to a minute, minute and a half actual question or something and tell people who your question is to. 

Okay. Siva, you've had your hand up for an hour, so why don't we go with you? 

>> SIVA: Thank you. And I want to respond to your basic question: Why are we discussing fragmentation now? And so, I thought ten years ago there was signs of fragmentation. Today there are cracks. Tomorrow it will be broken pieces if we do not act. So, that's probably why it's time‑sensitive now. But ten years ago, it was one or two geographies that were causing fragmentation. We were talking about China, Russia.

But today, something that many countries do, willingly or unwillingly, leads to fragmentation. Even in the case of countries that have a good understanding of Internet, it is some legal regulatory measures or taxation measures that that unwittingly cause fragmentation. And particularly, there is a sovereignty concerns and some sort of hunger for surveillance, which is also causing fragmentation.

One particular point that I want to mention is that the Internet is global. We want it to remain global. But there is no global legal environment in any geography for a global corporation, global Internet corporation to function, and there is no global business zone or no offshore zone, and that sense of order ‑‑ and even there is an absence of a global judicial process or an Internet judicial process. These are all what is needed and what needs to be created by the community and the Internet organizations to make sure that our global governance process is trusted and there is a suitable framework for global Internet. That sense of a framework, absence of that, is also a cause for fragmentation. Thank you. 

>> BILL DRAKE: Thank you, Siva. And again, I'm asking if, going forward, when people take the floor, if you could identify yourself and try to keep to like a minute and a half.  I see a lot of people in the room there in Addis, and I want to get people included. Anriette, are there any people raising their hands there that would like to speak? 

>> ANRIETTE ESTERHUYSEN: Lots of people. If you give me a moment ‑‑

>> BILL DRAKE: Lots of people, excellent!  Let's get some questions from the group there. 

>> ANRIETTE ESTERHUYSEN: I'm not going to give you a minute and a half. I'm giving you no more than a minute. So, please, can we have this gentleman and then Raul and then you and then the hand at the back. Go ahead. Please be brief. Introduce yourself.  We're getting assistance. A volunteer is here. 

>> BILL DRAKE: Who's controlling the camera? Why can't we see whoever's speaking? 

>> ANRIETTE ESTERHUYSEN: No one's controlling the camera and there's no AI in this room, so ‑‑

>> BILL DRAKE: Oh, all right. 

>> AUDIENCE: So, quickly ‑‑

>> ANRIETTE ESTERHUYSEN: Please introduce yourself. 

>> AUDIENCE: My name is Jose. When we talk about fragmentation, we usually make examples of government's action or political authorities, like what we have seen today about Cambodia. I think as Sheetal mentioned, or about the national (?) or six hours complete shutdown Bill pointed out. What if the giants like Google, searchable like Google, which has some sort of monopolistic power decides for any reason to limit the results we see in its searching for some specific region, for example, or they activate the safe search for some other countries like what we have seen in some countries in eastern Asia? Or a satellite Internet procedure. For some regions, they decide to give even somehow free access to Internet, but for other regions, they limit it because they are negotiating with the governments. What is your idea of those region‑wide discrimination of those global platforms? Do you consider them as an example of Internet fragmentation as well? Thank you. 


>> BILL DRAKE: That was an excellent question. Anriette, let's just take a bunch of questions together, and then I'll go back to the panel and they can ‑‑

>> ANRIETTE ESTERHUYSEN: I was in the process of doing that, Bill. So, just give me a little bit of time. I'll hand back to you when I'm done with moderating the questions in the room. 

>> BILL DRAKE: Great. I can't see you. 

>> ANRIETTE ESTERHUYSEN: That's fine, trust me. 

>> RAUL ECHEBERRIA: Okay. Just about what is fragmentation. For me, fragmentation is when you get two different or different responses to the same action in different parts of the network. So, this is as simple as that. I'm really concerned about ‑‑ I think I agree with the points from Milton about what he said, and what I'm seeing is that many law proposals in different parts of the world are laws that have already passed, that deciding about content removal or blocking, filtering for different reasons that I will not argue here, but lead to action as that are not explained and are based in the way that are described are based on a huge ignorance about the Internet architecture. So, it's just saying to the Internet providers or the platform, saying "You have to remove this content." 

And in countries where there are hundreds of ISPs, the big risk is that hundreds of different actions can be taken, leading to the future fragmentation. We've got to the last part of the question, what we can do. I think that we need a benchmark. We need in this, to develop in this environment some agreement, kind of agreement with governments, too, that we can show to the other parts of the governments, saying, these are the things that cannot be done because these are the consequences.  And this, I think that could be a great progress if we can do. Thank you. 

>> ANRIETTE ESTERHUYSEN: Thanks, Raul. And over there, if you can just pass the mic. Just introduce yourself, please. 

>> Laura: Thank you. I'm Laura from Brazil, here with the Youth Brazil IGF. Thank you for the plan. It was great. I would like to ask this. Some of the panelists compare the issues of today with another moment of the history of the Internet, especially from its beginning of Internet governance and how it was possible to arrange a system that works towards compatibility/interoperability, and apart from fragmentation. I think Wolfgang mentioned that how we manage to make ‑‑ to forge a system that allowed 5 billion people to connect. And I would like to ask, I'm curious about this question. Why did it work back there, do you guys think? Why did it work back there and it doesn't work anymore? You mentioned a lot of factors, but I would like to precisely assume one hypothesis, because of what was discussed here.  Did it work back there because the state was not so involved or interested in this aspect of policy‑making about Internet? Did it work back there because the Internet was not so strongly spread and was not so strongly implicated with political agenda or not? Do you think there are other aspects that should be highlighted when you compare nowadays and back in the history of the Internet, and Internet governance in general? Thank you. 

>> ANRIETTE ESTERHUYSEN: Thanks for that. Bill ‑‑ oh, we have the last question from the room. 

>> AUDIENCE: Thank you. My name is Suozzi from Nigeria. I want to follow up to the question that the last two or so had asked, which is to look at another mention, where non‑state actors could also provide an impose for fragmentation. If you take what the new owner of Twitter is doing, it is provoking two different responses. On the one hand, many people are leaving Twitter to other smaller platforms, and that for me is a sort of fragmentation. But on the other hand, because Elon is using his practices for private ‑‑ it is up to governments to say, "Well, since you want much profit out of subscribers, we will also have to tax you" and begin to tax other platform providers and begin to roll out policies that can impact on the digital space, and there are, contributing to more discussion, more date and more reasons of the question of fragmentation. So, I'm wondering whether these actions of non‑state actors whose motive is primarily profit can also be put under the satellite as part of the reasons for the fragmentation debate currently. Thank you. 

>> ANRIETTE ESTERHUYSEN: Thanks, Suozzi. And I think we have two hands. I see Andrew's ready to respond to some questions. Should we just take the last two hands? We have Izaan who is with us virtually. So, if you can go quickly and then I'll hand the floor back to the moderators to end the session. Izaan, you have the floor. 

>> IZAAN Khan: Sure, thank you. Can everyone hear me all right? Izaan Khan, Youth Ambassador for the Internet society this year. I just wanted to say quickly that implicit throughout this entire discussion is the fact that we consider Internet fragmentation to be having negative connotations, in a sense, but the focus of the discussion today has sort of been sort of measuring, like how red the sky is without asking the question of why does it matter that the sky is red? And in the sense, why does Internet fragmentation actually matter for us as the user?

And I think a very important point is to know we have something called Internet universality based on the Rome principles. We have rights, openness, access, and multi‑stakeholderism. And I think every single measure that people consider to be fragmentary or not should be evaluated with respect to those principles, because then it will let us know whether it actually affects the fundamental spirit and ethos of the Internet, and then we can evaluate the risks associated with that.

For example, if something hampers interoperability, why does it matter? Does it lead to the risk that those will be violated? And I think that would add important context to this discussion now, because otherwise, we can run left, right and center thinking about fragmentation without considering what the real implications on the user might actually be and whether they are justified or not. Thank you. 


>> BILL DRAKE: Okay, thank you. So, we have had excellent interventions, six interventions on a range of key points, and I'd like to just go around the panel and ask them to pick up on whichever ones they want to touch on. And if your comments should bleed into the question, what is to be done, that also would be very good. I think Andrew had his hand up first. Andrew, go ahead. 

>> ANDREW SULLIVAN: Oh, thanks. I don't know that I did, but anyway, I'll talk now.  So, the concern that I have about ‑‑ well, I think it reflects the last point that was just made. The reason the Internet expanded the way it did, in my view, and the reason that we have the Internet that we do is because of a shared illusion. We agreed implicitly that what we were going to do was interoperate, and that was how we got the Internet. People keep forgetting that the Internet is not one thing, right? It's right there in the name, the "inter‑net," so it's a network of networks. But we've forgotten that because our experience of it is this smooth, single thing. And as long as we had in our heads the idea that interoperation was somehow fundamentally the good thing, then we automatically tended towards that kind of interoperation. But having abandoned this idea that the sort of shared goal of interoperation is automatically the one that we're pursuing, we're drifting away from the potential that the Internet offers.

And what's been interesting to me over the last couple of years, thinking about this ‑‑ excuse me ‑‑ is that the Internet is technically quite robust because of the way that it works, the way that things, just you pick them up and they kind of work together. But actually, politically, it's quite fragile. It really does depend on this idea that the pieces, that everybody's going to assume this same positivity of interoperation. And I think that that assumption is the one that we've been abandoning gradually. And so, to answer this implicit question that you were asking, you know, what is to be done? Either we will re‑embrace this idea that interoperation is fundamentally a good thing ‑‑ and I, of course, believe that it is ‑‑ or else we're going to lose the Internet. There isn't a possibility of, like, national Internets. That's a nonsense idea.  What you get when you have a fragmented Internet along country lines or along technical, so‑called ecosystem lines ‑‑ they're not ecosystems, they're parts, but set that aside ‑‑ what you get is not the Internet, or not an Internet. What you get, actually, are just separate networks. We had that experience. We had CompuServe, we had GEnie. We don't need that problem. We need to recommit ourselves to the idea that interoperation and interoperability is a fundamental good that leads to better human outcomes, and that, I think, is the thing that we really need to focus on. 

>> BILL DRAKE: Okay, thanks, Andrew.  Milton, let's try to be also responsive to the specific questions as much as we can as well. Okay, Milton, go ahead. 

>> MILTON MUELLER: Sure. That's exactly what I wanted to do. I thought the gentleman who, in the room, who asked the question about the cooperation between the large global platforms and the governments was really hitting the nail on one of the main drivers of fragmentation, or what some people want to call fragmentation, and that is, again, it's a classic case study of what I call alignment.  It is the platforms are saying, hey, we've got a globally interoperable platform here, we're connecting people from all over the world. And then the, let's say the Thai government or the Cambodian government or the German government say, hey, we want ‑‑ we just passed a law, and we want you to enforce certain rules that only apply to German citizens, right, or to Cambodian citizens, or whatever.

And then, we have discovered to our dismay that the platforms are perfectly capable of doing that, that they can, in fact, configure their systems to discriminate against and differentiate between users in different parts of the country. Now, there are also subversive and heroic ways of avoiding that, you know. You see people in China using VPNs to get access to things they're supposed to not get access to. But fundamentally, that's why I think we have this problem.  It is at the first order.  The biggest issue in fragmentation is this alignment problem in which governments are trying to align their jurisdictional authority with the functioning of the Internet.

One other point related to the question from the Brazilian woman about why things are different now. I don't think it's about imaginaries at all. I think what happened is very clear from a political economy standpoint, and that is, the governments did not know what was happening. They were taken by surprise. It was a de facto creation of this global space of compatibility before they understood it, before they knew what to do about it. You look at what was happening in the early '90s and mid‑'90s, up until about '98, is that we simply created this, writing on the back of telecom liberalization and the liberalization of value‑added services, which the Internet was classified as. So, suddenly, you had this global interactive space and this new capability before the governments had approved it. If you had asked them to approve it in advance, they probably wouldn't have, and it happened as a default, as a de facto accomplishment. And for the last 20 years, they've been sort of slowly realizing that they have lost control and trying to reassert control. So, that's my theory of why things are different now. 

>> BILL DRAKE: Okay, thank you. Before we drift into the imaginaries discussion, let's go around, though, and pick up the questions that have already been asked, because we had six very good interventions.  Neha, would you like to respond to any of the questions that were raised? 

>> NEHA MISHRA: Maybe I can just briefly respond to the gentleman, Siva, who first made the remark about global governance or some kind of a global governance solution. I mean, one thing I wanted to just highlight is that Internet fragmentation itself is multilayered in many ways, so the idea of having a single global governance process or institution just responding to all concerns is probably not going to happen. But from my perspective, I think there are certain solutions that international law can offer. I think of national human rights law, but also from the perspective of international trade law, particularly with now a lot of trade agreements including rules that prohibit data localization, require cross‑border data flows with some exceptions, and it can be a dialogue for another day what those exceptions might actually mean in practice, and also whether the Internet governance community feels comfortable with some of these dialogues happening outside of the Internet world in close trade organizations. But I think, certainly, there is some potential for international law to offer solutions, at least inform certain global governance processes, but I don't think it would be a unified solution. 

>> BILL DRAKE: Neha, if I could follow up with you quickly. In addition to trade agreements, there are new instruments, digital economy agreements, which are a bit more expansive and allow a multidimensional approach. Do you think those could play a role here? We see governments, particularly in Asia, pursuing those. 

>> NEHA MISHRA: Yeah, I think the digital economy agreements are a better attempt in the sense they're solely focused on digital issues. It's not trading apples for digital services or trading, you know, automobiles for opening up the market, telecoms market. So, in that sense, they are a better attempt.

The other strong point about digital economy agreements is that they have a lot of building blocks. They are looking at emerging digital technologies. They are looking at best practices. And I think the capacity of these digital economy agreements to co‑opt multi‑stakeholder processes is far more than traditional trade agreements, so I think, yes, digital economy agreements are ‑‑ definitely can be a game changer, but I'm not sure how many countries outside the Asia‑Pacific are going to jump onto this idea

>> BILL DRAKE: Hopefully, engage, rather than co‑op multi‑stakeholder. Peixi, any thoughts? 

>> PEIXI XU: Yes. I'm sensing that our audience members, or the questions, some of them are worried about the state actors, some are worried about the private sector. And personally, I'm observing that at the moment, that many actors, state actors ‑‑ or not state actors ‑‑ are accusing each other of having the intentions to fragment the Internet, for example, and that leads to a possibility that, for example, one state actor may take a real action to fragment the Internet in the name that others are going to do it.

Then I am thinking that if we take, for example, the narrow version, or the technical version of the fragmentation, we can find actually a solution to that. For example, we can propose the principle of no first use or no first fragment or no first to take action, into the language of Global Digital Compact of the United Nations and make it further up, make it internationally binding, for example. If that case, we can have, somehow a closed circuit about thinking about the technical version of fragmentation. And meanwhile, we can settle some disputes or worries from the state actors and the private sector. Bill? 

>> BILL DRAKE: Interesting. So, like arms control ‑‑ no first use announcements. That's an interesting parallel.  In the room, Anriette, the three of you there want to jump in on the questions that have been raised by the six people? 

>> ANRIETTE ESTERHUYSEN: Yes. I think Wolfgang and Sheetal would like to respond. 

>> BILL DRAKE: Volume is bad. 

>> WOLFGANG KLEINWACHTER: To react to the question from the lady from Brazil. You know, 25 years ago, the Internet was a technical issue with some political implications. Today it's a political issue with the technical component.  25 years ago, they were analyzing a good book when they argue, code is law. So, the code‑makers are seen as more important than the lawmakers. And you could see it now that the lawmakers are firing back. 

And 25 years ago, this was a million‑dollar business. Today it's a billion‑dollar business. And it's not just the price; the governments have a stronger lock of it. So, the way forward is, you need a better cooperation between lawmakers and (audio fading in and out) code‑makers so that lawmakers and government understand, you know, this technical infrastructure, how it works ‑‑ DNS, TCP/IP. But code-makers also need to understand national concerns, whether it comes to digitization or security issues and cybercrime and things like that. So, that means the way forward is a better multi‑stakeholder collaboration where everybody in his separate function, you know, works hand in hand, and this needs trust. And unfortunately, this was lost in the last couple of years. That means we have to rebuild trust in the global Internet governance ecosystem. Thank you. 

>> BILL DRAKE: Thank you, Wolfgang. 

>> SHEETAL KUMAR: Yes, thank you. I just wanted to respond to three ‑‑ I think it's three of the questions specifically. You asked the question about why now? I think in one word, it's power. And state ‑‑ And by the way, I'm not speaking as a co‑facilitator of the Policy Network at the moment, but from my personal capacity. I think state and commercial power ‑‑ whatever it is ‑‑ there is a power that is being exercised in different ways. And we're seeing that through the policies and measures that we're talking about here, implicitly or explicitly. But power is fluid. It is exercised in many different ways. We often hear about people power, for example. 

And I think that as we are talking about this topic, a really important point that was made earlier is when and how do we act and based on what? And for that reason, reasserting certain values, whether they are, for example, interoperability and openness, where we are seeing a shift away from that and using spaces like the Global Digital Compact, like the UN, also at the national level, to reassert those values. And values have a material impact, and there is no better, I think, example of that than the Internet itself. 

So, if we continue to reassert the importance of those values and perhaps create and define a framework where we can measure where certain measures, whether they're by state actors, whatever actors they are, commercial or otherwise with power have, then I think we can continue to shape, and I hope we can, the open Internet that we all want to see. And we had an example there of the ROAM Principles as a framework, we have the International Human Rights Framework. There are frameworks that we can assess against. So, looking forward to continuing this discussion the rest of the week. Again, really encourage you to come to the sessions on Wednesday about this topic as well. Thank you. 

>> ANRIETTE ESTERHUYSEN: Thanks. I'll just make my sort of last remarks and then I'll hand it on to you. I think I ‑‑ you know, a lot of what's said really resonates with me. I think Andrew's remarks are really to be taken seriously. The value and the goal of an interoperable Internet conceptually informed how we approached Internet governance initially. And I think we are losing that. 

I would add another concept to it as well. It's not just about interoperability, technical interoperability, but it's also about interconnection and interconnectedness, the power of the Internet to connect people, to connect continents, to connect countries. And I think that this might not be at risk of fragmentation at a technical level ‑‑ we still have interoperability, but I don't think we should underestimate the impact of the kind of ideological fragmentation, the kind of ideological polarization that Peixi talked about, the power of the lawmakers over the code. I don't think we should underestimate that. I really think that by taking this interoperability, this conceptual and technical and philosophical interoperability and connectedness for granted, we are putting it at risk. 

And I agree with Sheetal and what others have said now as well ‑‑ we need instruments. We need ‑‑ the project of multi‑stakeholder inclusive governance of the open Internet is actually, I think, only just starting. And I think that for me is what we ‑‑ we've been playing around. Now we really actually have to ensure that Internet governance and the way in which policy and regulation is being made does not put this interconnectedness and interoperability at risk. Back to you, Bill. 

>> BILL DRAKE: That is the most optimistic end point I've heard in a long time, Anriette. I'd like to believe that we're just starting with multi‑stakeholder, rather than hitting a wall.

But we've got five minutes left and we had previously Niels had raised his hand and several people responding to him in the chat. Niels, are you still here to pose your question, or no? 

>> NIELS TEN OEVER: Of course!  No, more than happy, but Anriette's point was so happy because it meant that we will be discussing this for many more years, as we've been doing for quite some time. I just wanted to add like a relatively short historical note and that is that fragmentation and interoperability are a spectrum going on in transnational operations since the telegram. That is why the International Telegraph Union was established in 1865, right? So, I think we should really make sure that what we're seeing now, and also the emergence oligopolies, we've seen this in the telegraph, in the radio and television, but we as the Internet governance Forum and technical people tend to think we are the first persons to go through this. But no, we have done this.

But also, this Internet thing we're talking about is not completely the same thing all the time, because the original end‑to‑end principle kind of went overboard with the introduction of network address translation, when not all end points could be accessed at the same time, so that introduced the difference between clients and servers, between users and producers, but also IPV4 and IPV6 are not interoperable. It's not all TCP/IP anymore, because now we also have Quick. So, like, this whole understanding that it's one homogeneous network with one homogeneous stack is also not true.

But previously, there were also interconnecting networks that ran on different stacks, that was sent mail servers did. They translated between different networks. So, even fragmented networks can be made to interconnect. So, there are a lot of ways to introduce political and technical friction that can also be overcome in different ways. So, we just need I think a better understanding of the way in which states, corporations, and technologies, continue to reconfigure the relations among each other. And for that, of course, we do not just need international relations or technology or economy, but integrated understanding of both on all levels. And for that, we need several perspectives. And for that, I'm super grateful to everyone, to you. 

>> BILL DRAKE: Well, thanks, Niels, but you didn't set up the sociotechnical engineer bomb that was going to make heads explode, but that's okay. 

>> NIELS TEN OEVER: So, one more thing. Okay, so, we got together around this idea that we're going ‑‑ that more Internet is good and that more connectivity is good for everyone. And this was a very like end of '89 thing and the Internet was invented with the launch of sputnik. With the fall of the Soviet Union and the Washington consensus, more connectivity is good for everyone. But now we've also seen, as Sheetal said, with the power analysis, the power between different actors. So, this whole idea, this imagery that got everyone, that more connectivity is good for everyone, has actually limited application and sociotechnical imaginaries don't yet take this reality into account. So, we should really try to understand, what are these different technological configurations to reconfigure relations between users, the market, and the state. 

>> BILL DRAKE: Okay. So, ending on a social constructivist note. That's fine.  We can convene just for the debate amongst the academics. We have one minute left, so let me just give a closing and say, I think this was a very useful conversation involving a lot of people from around the world, which is always an excellent thing. Of course, we did not resolve anything, but perhaps we bounded the range of disagreement in a little bit more self‑conscious way. And so, that's useful. I hope everybody found it helpful. There have been other sessions before, recently. I mean, I just did a webinar on fragmentation a week ago at Columbia. That's on the web as well. And Sheetal mentioned that she's got a session coming up with the Policy Network at IGF. There's a main session and there's a number of workshops. So, conversations on fragmentation will continue.  And hopefully, we will come to some greater level of convergence, or at least principled disagreement.

Thank you all very much for your participation. Thanks to the panelists. Thanks to the people in the room. I hope you had a good time.  And Au Revoir, until next time. 

>> ANRIETTE ESTERHUYSEN: Thanks very much, Bill.