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.
***
>> MODERATOR: Welcome. My name is Farzaneh Badii and I've been told we have to start on time and end on time. We are going to start. I am the Founder of Digital Medusa and we're doing this work on can a layered policy approach stop Internet fragmentation?
So we are going to have a conversation and we are going to have a dialogue with you after our speakers' remarks. But a little bit of what is a layered approach. Well, there are different interpretations of that, but a layered ‑‑ by a layered approach, we mean that we have the 7 layers of the Internet, which the basic part of it is the network layer where the IP address is and other critical properties of the Internet that allow connectivity to happen. They are there and then on top of that, you have the apps and other services. This was a very simplified version. You can, of course, tackle me later on about whether that definition was accurate.
But why is it important to talk about a layered approach to policy making and how can it stop Internet fragmentation? One of the research reports that I have done in the past was about Internet and sanctions, and I have used ‑‑ so the governments and policymakers want to sanction human rights violaters and this is like a legitimate policy tool. However, when they do that, they also can affect access to the Internet if the sanction applies to Internet infrastructure.
So I have used this kind of layered approach in the report on Internet and sanction to kind of show how the Internet infrastructure can be actually affected by sanctions and how the governments that actually care about the open and global Internet and want to avoid fragmentation, can prevent that from happening by having more targeted and a layered approach to policymaking sanctions. This is one example of it. I'm trying to make this as tangible as I can. I can see a lot of familiar faces, so you know the layered approach and you know the OSI model, but for the policymakers and Civil Society in general, when they make recommendations, we are trying to make this session as practical as possible to kind of look at the layered approach and see if we can actually stop Internet fragmentation and but as come up with policies, come up with content regulation policies, and other needed regulatory frameworks, of course, and accommodation of Civil Society have to adopt those.
So with us today and this workshop is co‑organized with the Internet Society, and today we have Nobuhisa Nishigata, the Director of computer communications division at ministry of internal affairs and communications, MIC Japan, and has been instrumental in convening a IGF in Japan and there is a very long bio that I think Nobuhisa can introduce himself a little more if he wants to. We have also Alisa property settlement star ‑‑ the Vice President of Cloud fa refreshing with us today and we'll make it very tangible what infrastructure is involved when we use the Internet and how we can come up with policies that consider that kind of division. We also have Jean Queralt, and I have practiced that last name like a few times. I got it right, I'm sure. And Jean is the founder and CEO of the ILO foundation a nonprofit advocating for data‑centric digital rights. And then Connie as commentator. He resisted sit on the panel and wants to be with the crowd. He's fellow of democratic tech initiative digital forensic research lab and also a veteran of the developing and analyzing Internet policy to ensure an open and global Internet. Before I go to the panels, I also I wanted to mention that Internet Society also has an Internet impact assessment tool that can be very useful in kind of like using this layered approach in order to come up with policies that do 23409 lead to Internet fragmentation. But if you also want to know why Internet fragmentation is bad, my take is that we want the Internet to be global and interoperable and everyone on the Internet and in the world to be able to access the global Internet and Internet fragmentation prevents that.
So without further ado, I'm sorry I talked a lot. We have Alisa, would you like to go ahead and start the conversation?
>> ALISA STARZACK: I am the global head of public policy at Cloud Flare, it is something very much on the Internet infrastructure side of the world. I will explain a little bit what we do and then talk about what layered policymaking might look like and why it matters. We run a big global network, we use it to do a couple of different things. Most specifically, we actually ‑‑ what a big global network means, means we have equipment in lots and lots of different countries where we can cache content to make interInternet transition more efficient and prevent against cyberattacks, that allows us to look for problems in the traffic, cyberattacks, and allows us to sort of handle them for an entity online.
So, we're very much in the infrastructure space.
I think when we get into the idea of layered policymaking. It's worth understanding a little how we see the world and what it looks like from our perspective. So one of the interesting things about Cloud Flare service is that we actually offer some of them for free, so we have something like 20% of global websites that use our services to help protect themselves against attacks.
From a practical standpoint, that means we make a lot of the Internet more efficient, which is terrific. But when you get into questions of layered policymaking it's a lot more challenging. If you think about how the Internet is set up, the very top layer is as was mentioned, all of the content. All of the things that we all Internet with online, so the websites that you interact with, the platforms that you interact with, those all sit at the top. Often when people talk about the Internet, they mean that piece of content. They don't think about the other components that actually get that content to you. Think about the DNS system where people register domain names or where so the registrars to the domain names or registries for the TLDs, the top‑level domains like.com or country‑level ones, there is a whole world of that. There is a whole world of entities that actually transmit that content which Cloud Flare is part of, and then the ISP side of the world as well, so all of the different entities and browsers that make it accessible on your computer. All of those play a role in getting that content to you.
I think one of the challenges when we get into the layered policymaking world is we really start to have to think about what effect regulatory structures have on how those piece it's fit together and how they don't. So I want to give you a couple of examples of what that means in practice.
So, you know, one of the things about the Internet that's so interesting is when it was set up it wasn't really contemplating the world we live in today. It didn't have privacy built in, didn't have security built in, nothing was encrypted, and so people took advantage of that to do lots of often good things, they screened for security threat, they potentially use it for content purposes. But when you start building new structures on top of that, often those things either stop working or get more challenging. And so to give you an example, people might block an IP address. The challenge in that world of blocking an IP address, so again in a country, is that if you block a Cloud Flare IP address where we have millions of domains on us, you potentially block millions of domains. So now you've created a world where a huge amount of content is not available.
We see similar things where people are trying to do something because they think it's a way of addressing content, and an effective way of addressing content, but really what they're doing is potentially cutting off components of the Internet. So when we get into that world of layered policymaking, it's really thinking about the effects both locally and collateral consequence, and then also thinking about the broader global effects.
So what happens when this is done in a different country with a different rule of law standard? What happens when it becomes the norm that that's an acceptable thing to do? Those are all sort of questions that I think we need to think about and answer.
>> FARZANEH BADII: Thank you very much, Alisa. So Nobuhisa, I believe that the Japan government is one of the major advocates for global interoperable Internet. How do you think it layered approach can be useful or what are the challenges?
>> NOBUHISA NISHIGATA: Thank you. Good afternoon. My name is Nobuhisa Nishigata. First of all, welcome to Kyoto, on behalf of the ministry, and they thank you and hope later on tomorrow you have some time to find more not only IGF things but as about the depth of the culture in Kyoto.
So then about the question, maybe starting off by the layered theme. I believe that the Japanese Government is doing pretty good job in the layer approach, particularly once it comes to the regulation. But this is only layer, 1, 2, 3. This is our domain from our ministry voice. Then maybe after 4, 5, 6, 7, maybe more people come in like she said. It's not only the telecom things. So 1, 2, 3, particularly for the domestic things and sometimes infrastructure connecting us abroad, submarine cables or satellite frequency, et cetera.
So then like we have very clear structure in our telecom business law in Japan. We are doing pretty good job, but then for the issues in infrastructure, for example, like it's kind of away from today's topic, but how we secure the Internet connection in rural areas, even in Japan or like isolated island in Japanese territory like that kind of thing. We can do that in layer 2 or 1 type of regulation or policy approach. We do that.
On the other hand, once it comes to the content issue, like maybe let me, that since my assignment in the Japanese government includes some issues of privacy or like CSAM that's kind of complicated issues. Then of course we do the layered approach in 1, 2, 3, but it doesn't solve the problems. We have to fight with the kind of the problem. I mean CSAM is more like human right that we have to solve these things, fight together. Or like piracy, some people say it's only going to come, but the harm is big. You may see some Japanese publishers in the industry, like you can see the exhibition if you go there, they are waiting for you to show how they fight against the piracy bad guys.
So, I mean like she just provided example of the shutting down the whole Cloud Flare. We don't want to expect that to happen, but still you know, we have to do something. I mean for example, maybe moving other direction. Of course, I hope that you already find that Japan is such a peaceful country and society, but still we have bad crimes going on around. And the reason is like to some extent, the Internet enabled these new type of crimes, and I mean not the whole Internet, but for example one example would be the telegram. It's a cool tool, and encryption and that's good thing, but it's quite hard for law enforcement to chase them up.
So, of course, we are ‑‑ I'm telling you that the Japanese Government is smart enough. We are not going to ban the Telegram use in general in Japan, but on the other hand, we have to find some solution or maybe going on in the technology just only for the law enforcement, but this is I would say that is the hardest part. You know, it's kind of arbitrary type of things, but as from the government perspective, we have some basic principles and first comes rule of law, right. It's a fundamental principle in democratic society. Of course, we expect freedom, open global Internet. These are the great values, I admit and I respect too. On the other hand, maybe I'm telling you you should be aware that the governments are not globals you guys are on the Internet. We have the boundaries, we do care about the boundary, you know. So, of course, like the government, we are paid by the tax so our prime, I would say subject or customer, I shouldn't say the customer, but the people within the border. Of course we do diplomacy, we do some international affair, but the primary thing is within the border for the government. On the other hand we have international organizations like UNESCO and IGF and others, et cetera. We have to recognize some gaps between the government and society, et cetera.
Then like just I had an opportunity to have some speech in the same place here last month, it's 30th Anniversary for APNIC, Japanese network information center, 30 year anniversary congratulation, again, for them. Then I just talked about what was 30 years ago. So 30 years ago, just I was an exchange student in United States. I was alone having host family who was very kind and they helped me, and then I go to high school in United States.
At that time like of course, like we do some communication with my family in Japan, but other times believe me, it's just letters and phones. Right. No smart phone, no Zoom, no Internet, you know, no emails. So then compared to that, then just how beneficial the Internet has been and we couldn't think more about the recent development of technology. On the other hand, unfortunately we see some new problems. I would say thanking for having me here today would be first step to get closer to have more dialogue between the Internet and government, even in Japan.
So I would say Japan Internet environment is pretty good. It's free, compared to other countries sometimes, but still it doesn't mean that we are satisfied as a government person. We are not satisfied at the situation right now. So, but still as I said, the Japanese Government is smart enough to respect the free Internet, the global Internet. On the other happened, we need our help. This goes, of course, to the other parts of the world. So I hope that will be the first step for further dialogue. Thank you for having me, again. Thank you.
>> FARZANEH BADII: Thank you, Nobuhisa. That's actually why we're here to have this kind of conversation. CSAM for those who don't know is child sexual abuse material.
We go to Jean. Jean, you have some slights?
>> JEAN QUERALT: Yep. Good afternoon, everyone. Next one. That was nothing. (Laughing). All right, so when I was offered to be on the panel. I started going down a little bit of the rabbit hole of the concept of fragmentation and how there are so many definitions about it. I found that there were a number of elements that seem to not be very much investigated, if you want, or considered. So the term bugs me a little because it does have some implications. Think about a vase, for instance, if you say it is broken, it implies at some point that the vase was not broken. If we're going to be talking about a fragmented Internet, we need to make assumption that it was not fragmented in the past. I'm not sure that we have evidence to be able to sustain both claims with perfect certainty.
Of course, we do all have a bit of an intuitive feeling that these things are going to that direction. I would be just a little bit cautious about which direction we think it's really going.
When I hear about fragmentation, for the most part I never hear anyone talking about concept of time. Is that observed fragmentation happening in a short period of time, something that maintains across time, is it because it has been mandated by someone, is it because there is a faulty hardware or protocol that stopped working for whatever reason? Those things seem to not really much be in the conversation quite often. Next slide, please.
So, the approach that I like to take is the one that you see there in the slides, so that's the layering that I like to follow where you see on the top that there is a user ‑‑ I'm sorry, the user should be on the top but I didn't have the space. Interacting with an application through a user interface, basically, and then you would need to be able to typically generate some kind of identification, which leads you to some data manipulation, which is then transmitted. You need to be able to identify the destination machine for that, and then you have some kind of physical channel to be able to do the transmission.
Essentially, for the technical people, I would say that the UI is nothing more than an analogue digital and digital to analogue converter, it basically passes the information and the will of the user down the stack.
I do see the need, even though there is a lot of conversation as to whether we need a layered approach or not. Essentially because if we cannot pinpoint which layer a certain thing is happening, it's very difficult to figure out who you need to talk to to try to solve it. I don't really mind if we use this kind of model or any other type of model, but there has to be a model for us to unite across to be able to make sure we know which stakeholders we need to engage.
And most importantly, I would say there is a little bit of confusion sometimes and overlap between different organizations and it seems to me that we don't have a clear taxonomy of mandates, so that we again know who do we have to talk to. Nevermind the fact that some of those organizations may not be the most accessible for certain stakeholders.
Next one, please.
So, when we also talk about interoperability, the three ‑‑ the only three models that come to mind is this shared situation where everyone is agreeing on how to communicate, say protocol or ad hoc one where every single piece of a device, every software requires to have ad hoc methodology to be able to communicate with the rest of the systems, or hybrid situation which is pretty much what we have at the moment, kind of a combination of both of them. So we are talking about interoperability and sometimes I'm not 100% sure about in which direction do we want to go. Do we want to keep this hybrid model? Do we want to harmonize much more across systems? What do we want to do? Next one, please.
Was that the last one? Oh, there was another one. It's fine. I would also submit that what I was mentioning about the different organizations and trying to establish a very clear taxonomy of mandates, I know that it can be looked at as we are dispersing the management of all of those layers. I somehow look at it from the perspective of the lack of concentration of one single organization where everything can be done is actually Brazilians. If you think about an actor that may have some kind of nasty attitude, let's say, it would make many more resources to attend to all of these organizations and participate in all of them, so that they may be able to affect one of the layers but possibly not the rest. So if anything wrong happens, you could try to rectify through the rest of the layers.
I think there is something to be said about resiliency based on fragmentation of organizations. Thank you.
>> FARZANEH BADII: Thanks a lot, Jean. Constantinoes?
>> CONSTANTINE: Thank you. Hi, everyone. Thank you, so, my name is Constantine or Connie, according to the transcript, and I think that one of the things that we heard and it's sort of coming across is that the layered approach to Internet fragmentation presents a lot of challenges, first of all because currently the way the Internet has evolved over the years, we are not really sure whether we can still use and talk about layers. Right. And that is because there are many actors that are engaging and they're engaging across what used to be very discernible layers in the beginning of the Internet. And also because more issues emerge that, of course, require state intervention and we see regulation.
So, I think that starting with what Alisa was saying and of course from Cloud Flare point of view the main chal seng how the regulation happening within the Internet system reflect it's infrastructure, and the ability to provide services whether it is about security, whether it is about traffic efficiency or whether it allows users to be able to use and consume the Internet in an efficient way.
Then, of course, you have the broader global effects, right. Those unintended consequences can create some sort of global effect that then spills over and we're talking about a complete breakdown and splint of the Internet.
We hear challenges that a lot of governments are having like Nobuhisa mentioned. We understand there are some things that we should not be touching and we understand that those things make the Internet what it is. At the same time, we're facing everyday policy challenges as government, and as part of our mandate and the fact that we have been elected to provide answers to those challenges, we don't know where to go. And I think the ‑‑ well, Nobuhisa brought up encryption, which I think is perhaps the biggest challenge currently that policymakers are facing, especially in connection with CSAM. I don't think that we need to debate at all that everyone in this room, and in the world hopefully, how they feel about CSAM. But at the same time because it involves ‑‑ the conversation better yet involves encryption, it makes it difficult because the engineers tell you, do not touch encryption, it is a very important protocol, very important for the security of the Internet, very important for privacy, very important for the security of people. But at the same time, the fact that encryption exists as a protocol is getting abused, being abused by various actors in order to do criminal acts.
So, and I really like actually, Nobuhisa, your call effectively to essentially this community saying that we need your help. How can you help us to find a balance or provide at least a way forward in what seems right now to be an impossible debate?
And, Jean, really your opening really hit a really nice nerve with me when you said I'm not sure there was ever a time that the Internet was not fragmented. Right. We really do not know. We really do not know because we only recently, if I can say that, we started thinking of fragmentation, and fragmentation sort of came in front of us as an issue. There might have been a time, even from the beginning of the Internet, where fragmentation existed.
I think that depending who you are and where they're coming from, they will tell you different things. The other thing and if I understand correct, and please correct me if I didn't, is that we need to be able to ‑‑ and there is a challenge on that, on identifying which exact layer we're talking about and we need to be able to do that because unless we are able to identify that layer, we won't be able to know who to speak to, how to engage with that organization or body, and what actual solutions to give.
Of course we went back to, you know, the origins of the conception of the Internet, which is the centralization and the fact that decentralization, hints of course, to more resilience, and why this is important within the context of the Internet.
I think that the one thing that I can say is that fragmentation for the past couple of years, at least, has been the buzz word that everybody is using, and rightly so in many, many ways, simply because it is a problem that can create serious challenges for the global open and interoperable Internet. And especially in the past 5 years we see government stepping in to address some of those very hard challenges, and at the same time, in their effort to do so there are unintended consequences that actually affect, you know, the infrastructure of the Internet and the way the Internet exists and continues to evolve. I will stop here.
>> FARZANEH BADII: Thank you, Constantinose. We want to have a dialogue if you want to ask a question or comment or if you need clarification of what is this Internet fragmentation, and why is it necessarily a bad and not a positive thing. Oh, I can see somebody at the mic. Great. Go ahead.
>> Hi. I'm Barry Leba. I want to comment on the Internet being fragmented by the beginning. The Internet by its nature is fragmented, it is a network of networks. I have a home network that it does not have open access to the rest of the Internet, it's a fragment. Companies have their own fragments of the Internet. This is the way the Internet was intended to work.
So, I like to think of what we're talking about as harmful fragmentation, which is fragmentation that damages the way the Internet was intended to work. I think that's a better way to approach it. The question that I got up to ask was, I know what OSI stack layering is and all of that kind of stuff. What I don't understand is what the purpose of this session is to talk about layered policy approach and we're now halfway through the session and we haven't talked about that part. So, I'm looking to understand what kind of ‑‑ what the layered policy bit means? Are you talking about applying policy to the OSI stack layer? Are you talking about setting up a layered approach to policy? Please explain and let's have that conversation. Thanks.
>> FARZANEH BADII: Thank you. So for me, my impression is that I tried to explain this in the beginning. For example, when governments wants to regulate, not necessarily the Internet, they can come up with a regulation that could affect the operation of the networks. And this could be, for example, the IP addresses of certain countries and regions. Governments don't necessarily want that if they want an open, global Internet. They don't want to prevent access of countries to IP addresses, but the unintended consequences of that kind of policy, in this case sanctions, is that it could prevent network operators from having access to IP addresses. And IP addresses are not, at the infrastructure as you know better than me, the infrastructure. They are needed and necessary for online presence. It's not about accessing Zoom or accessing Facebook. It's about accessing infrastructure that enables you to have an online presence and have access to the global Internet.
So, I believe that the layered approach could possibly help at the moment, it could possibly help to tell policymakers and Civil Society when they make recommendations, that hey when you are coming up with these content regulation initiatives with the sanction initiatives, think about which part of the Internet you do not want to actually affect with their policy so that it actually can be targeted and you can actually achieve that regulatory purpose.
I hope that was clear, but I invite any other.
>> (Speaking off mic).
>> ALISA STARZACK: Can I add one thing. I want to make it a little more concrete. There are two different sets of problems worth thinking about. One is when you have a general regulation that applies further down when people aren't thinking about it. To give you an example, the EU has passed GDPR. Lots of very important things for privacy. One of the claims that has come out in GDPR is the idea of data localization of IP addresses. Now, if you know how infrastructure works, that's nonsense, that doesn't work practically. But that doesn't mean that a policymaker hasn't thought about it because they don't think about how the Internet works. I think there is the general regulation applying and being conscious of that. Then I think there is a secondary piece thinking about infrastructure and actually what are you trying to regulate and how. Often what happens is that people go down to infrastructure first because you have, going back to I think Jean's point, often that is a place where you can get access to something you might not otherwise have access to because we do have a network of networks. What you find is people will go to an area of access, so I think there are two components and I would sort of split them into two different ideas.
>> My name is Oliver C, co‑organizer of the session and now at the mic as an online moderator, so I'm forwarding a question from Tukio kushada, saying I feel cyberbullying is a big problem in Japan. Cyberbullying is one of the Internet fragmentation in terms. What do you think can be made to address cyberbullying from policymaking perspective. I think that's a very hard question to answer, but I'm going to add a little bit ‑‑ I'm going to add some color to the question and change it a little bit for the panel.
If you think of cyberbullying, and you think of that layered approach, how would you ‑‑ on what layer would you put that type of problem and at which layer do you think you would not be able to solve this? Is that a way you can work with that?
>> FARZANEH BADII: That's great. Anybody from the panel? So, I have one example of not good policy in order to regulate content. For example, IP blocking is not proportional ‑‑ cannot be done proportionately. If we block IP addresses, it's not just like one website. We might actually block access to a whole set of websites, and we have seen the case that happened in Europe when they came up with a regulation that for this information and IP blocking of the website, if rusha was mandate and network operators and ISPs had a lot of problem with doing that in a proportional way, and so that's where a not to do it or also the DNS blocking, I believe layered approach DNS is application but still infrastructure necessary for online presence. But anybody else from the panel would like to add?
>> NOBUHISA NISHIGATA: Thank you for raising the issue of cyberbullying. It is a big issue, but in my observation, it usually ‑‑ I mean in the beginning, maybe it's Japan's particular problem, but now I see similar issues outside of Japan as well. So, maybe we have to think about, it's kind of beyond the scope of the today's narrow scoped layered approach discussion, but however we have to think that like echochamber or future bubble type of things or side effects using this type of services. So of course, as I said, has also brought a lot of benefit to the society, but on the other hand there is some side effect. The cyberbullying issues is more closer to I would say freedom of speech and those kind of things in high‑level definitions, so it's quite hard. But maybe of course Japan is working hard it do some legislation against the humiliation over the Internet and also like some legal systems about the limited liability for the Internet service providers and those kind of things. So we're trying hard, but still I admit that's an issue.
>> FARZANEH BADII: Thank you. Jean, do you want to go ahead and then Marco.
>> JEAN QUERALT: I wanted to follow up on your question. You were mentioning about maybe us talking about harmful fragmentation, and I think it's a good point that maybe we can also have a term for positive fragmentation. So, the term splintnternet goes along a long time so could apply for the harmful negative one. I'm trying to figure out and English is not my primary language so possibly not the best one to come up, uninet or cohesivenet or whatever someone can raise their hand but it might be a good idea to pinpoint a name that points to a positive fragmentation and that respects the original intent of the Internet.
>> FARZANEH BADII: Thank you. I have a comment on that but I'm not going to let Marco wait. Go ahead.
>> MARCO: Happy to wait but I know you talk a lot. It's Marco from the Dutch Government. I'm not going to share anything about fragmentation and I'm also trying to refrain from saying whether we're talking the right approach or wrong approach, but I can speak from experience as policymaker. In 2015, one of our government think tanks introduced the notion of the public core. With that every since what we try to do is make a distinction on what happens when it comes to regulating the Internet, we make the distinction of regulating off the Internet and regulating on the Internet.
In my experience, so far this has been a helpful approach. We've also been guiding two fellow governments and I do think it helps the discussion forward especially in international policy, and I hear somebody say, like some layers do not have borders, that's correct. That's sort of what we see as the public core is the layers that are practically borderless and are really hard to regulate from a national perspective. There is, of course, the content discussion. And looking at sort of areas remarks and I think that's still sort of where a lot of the challenge lies, is I heard somebody say that lawmakers don't understand the Internet. Yeah, from where I stand, I think that's also partially a good thing that not everybody is an Internet engineer. I used to be an engineer. I'm stepped over to the dark side called government, and you know keep it simple. That's why can he kind of keep it in the distinction of two layers. If you delve in too many layers it gets complex. Another thing I would like to remark on, I think it was my Japanese colleague that sort of talked about accountability. Yes, the fact that we make the distinction, doesn't mean that the public core and the people operating the public core should not be accountable. There are certain things happening on the Internet where, if you look at it from a neutral perspective, probably the only place to do something about it is that public core. That's a really hard thing to do because that's the discussion we need to have internationally, and that's usually the discussion we need to have in a multistakeholder forum to find solutions for that space. Some things are relatively easy because you know where it is and happens on top of the layer. The point I'm trying to get to is we can divide it up into multiple layers. My advice is don't do that, but as recognize that in the end from a policy perspective, what we're looking at is public interest, and the public sits on layer 8 and 9. That public interest doesn't differentiate between layer 3, 4, or 5. We're trying to make the world a better place, and your challenge is to find out which layer is the most appropriate to do the right thing to help us make the world a better place.
>> FARZANEH BADII: Which public are we talking about? Whose public interest? The national public interest, are we talking about the global public interest?
>> MARCO: Oh, that's a difficult one. Of course we always start with the national interest in mind, but I do hope that our national interest is other people's national interest, and we can find consensus in the global space of what is the public interest.
>> FARZANEH BADII: Absolutely. I had to reconcile to add that is very important for the global Internet. Do you have another? All of my friends are here. Go ahead.
>> AUDIENCE MEMBER: We will fight to see who goes first. Collin. I guess I want to follow up a little on the previous comment. You know, it's clear that regulation that affects the targets and affects layers wasn't intended and if inappropriate causes problems. The discussion has considered a whole range of different technologies. We've heard discussion of IP blocking, DNS blocking, and a whole bunch of other things. These obviously operate at different layers, operate by different people, and understanding the differences requires pretty deep technical knowledge.
Are there ways of distinguishing those layers, distinguishing the technologies, and ways of phrasing those differences and describing those differences that are more accessible and don't need the deep technical lodge to perhaps reframe the debate a little in a way which might make it easier to chose the applicability of regulation.
>> FARZANEH BADII: Thank you. Actually, I was the one still responding. I'm sorry.
>> ALISA STARZACK: We think about it and think about how it effects. It's difficult to describe as who it effects. When we have those conversations we think there needs to be a translation. Somebody who isn't on the engineering side being able to explain what the potential unintended consequences might be is really important, and frankly, I think we certainly think that industry, which has some of the technical knowledge, plays an important role. I think sometimes though, it can be really challenging to wright things the right way. Recognizing that there is a gap between the two. But writing a regulation can be really hard even when you're trying to get at something specific, even when you understand the unintended consequence, that doesn't necessarily mean there is a better way to write something.
>> AUDIENCE MEMBER: I'm thinking about the bullying thing. It's a perfect example of where we need to look at this. Part of me says it's entirely a layer 9 thing, you know, you crack down on bullies at the personal level. But the Internet has changed the way that works. When I was growing up, if somebody bullied me, it was that person against me. And that person might get a few of his friends to join him in bullying me. And so now maybe it was 5 or 6 people against me, and the administrators at school could deal with that.
Now with the Internet, you have hundreds of thousands of people bullying me over the Internet potentially. So that does change things, and we have to look at how that works. But at the same time, it's still a layer 9 issue. It's, do you stop the bully from accessing the Internet at all? I think it's a very difficult thing to think about. So I like the idea of trying to think of that in layers, and I guess that's the comment that I wanted to make. I don't have an answer to it.
>> JEAN QUERALT: I want to make a comment. I feel sometimes there is this kind of cognitive disnans when it comes to the technology and expectation that we have in other industries. Unless there is a architect in this room, nobody really knows why this building is not collapsing on our heads, and we don't really care because that job is done and we just have put the trust into whoever is responsible for this to do the right thing.
And so trying to explain certain technical things to the general public, yes, for those who are curious, of course, you shouldn't be closing that door. It's just it's kind of assuming as well too much pressure for the final public to go up to that level of technical knowledge.
If I'm using a phone, it shouldn't be my responsibility to know the amount of ionizing radiation I'm getting from the antenna. There is mechanisms for that for me to not have to worry about it. Just as when I go to a restaurant, I'm not checking as to whether the water is drinkable or not. To a certain degree, the people who are involved in those industries are those who you would consider frontliner, and I would say we need to make sure that the technology that is frontliners has the tools for that. Often enough, I don't see that happening.
>> FARZANEH BADII: Thank you, Jean. I just wanted to add that for the content regulation and conduct regulation issue that should happen, we have been advocating for that should not happen at infrastructure layer, and like the Cloud services and CDNs and DNS resolvers and IP blocking, so and by we, I mean like the technical Internet community. Because that cannot be done proportionately. But on social media platforms, you might be able to take certain action, and for digital trends and safety and kind of like come up with take‑down mechanisms and stuff like that.
The problem with down that at the infrastructure level is that you cannot ‑‑ you might actually hamper some innocent person's online presence by doing that. So, I just wanted to add that. I see Andrew. Right, Andrew?
>> AUDIENCE MEMBER: Thank you. I'll take it as my cue to ask a question. Andrew Kamplin a consultant and also Internet What Foundation. Just to give a couple of reflections. I would respectfully say that the industry has got a lousy reputation for giving reasons why something can't be done, starting with privacy. So, each time it comes up, there is a raft of excuses to say why not. And then when GDPR is enacted, it sort of seems to find a way of accommodating it.
So, I think you could reply what we sort of discussed to say, okay, the industry is expecting legislators to do things in a way which allows for the efficient running of the industry. Which I can see why the industry wants that, but most industries have to live with regulations imposed on them. They don't get to dictate the terms that they regulated, and I would respectfully say that maybe with the maturity of the Internet we need to face up to the fact that, you know, governments are accountable to their sort of voters not to the technical companies. So, for example, to say you can't do IP blocking because that would be more difficult for CDNs, would be one approach. The other approach would be to say, if the governments decide you do need to block X, it's up to the CDN to work around that so they can still function.
So I think you just ‑‑ the industry is trying to have it to dictate the terms of how the environment operates in, and why should it be special compared to say car manufacturers?
>> FARZANEH BADII: Thank you, Andrew. We have a last like two minutes. Please be short in your remarks. Go ahead.
>> AUDIENCE MEMBER: Okay. I will be quite short. Rafiki. I want to ask a clarifying question, from MIC. So you said that MIC, also was not taking or try to regulate lot. But maybe I want to if you can clarify more how it's left more to the industry, the operator, to teleco, Internet service provider in different areas like the anti‑piracy and riding MIC or CSAM to work on country measure and so and so. Maybe, it's kind of a light approach, but it might also raise some ‑‑ it raises some questions about accountability and particularly it's not quite multistakeholder model and it's more of the kind of private sector taking the lead here. And at the end, also, kind of presented how Japan is leading and maybe respecting the right and so on. But what are the safeguards for the future? So I wanted more clarification. Yeah, thanks.
>> NOBUHISA NISHIGATA: Let me try to answer you. We are not trying to introduce some regulation over piracy or CSAM directly. It's quite hard in the technical perspective in how you build up the law as a text, like a code. We have like, you know, it is not just so easy. Even at our premise here today, but just if he says that the piracy is not good, we have to keep the industry, do something, guys, right. He may want to say that. But still in the technical ways, thinking how you construct the legislation, it doesn't work all that way. We know that. It's more you know we have to check up on many things in democratic society, for example rule of law, freedom of expression, et cetera, et cetera. It is not easy. Then we are smart enough to know that, so then we are asking ‑‑ but somehow, I mean can you do the piracy without the Internet, but the skill is much, much smaller than what we see in Internet piracy. Then another point we need, like we have ‑‑ we want to do something just only against CSAM, piracy, some bad guys. Of course, we want to foster and help them develop further from the good side, and I would say more than 99% in Internet community are the good guys. Right.
>> FARZANEH BADII: Exactly. And on that positive note, we can go to the next comment. I'm sorry, we have 5 more minutes and I promised I will be on time.
>> AUDIENCE MEMBER: No worries. My name is Azon, member of the Internet Society standing group. I wanted to talk about my Internet fragmentation comments but you can hear me tomorrow in the policy discussion. A question directed towards, I guess, all of you. Ultimately the balgt for a global Internet is battle for global value, which we don't have. We keep talking about, you know, democratic societies and multistakeholderism but not every single society is built like that. One thing that I definitely note that everybody over here also knows but never dares to say is well the very same mechanisms that we control content and applications and services in one jurisdiction will be used in other jurisdictions as well. That exactly is the problem that I feel needs to be discussed a little more, which is why we need to talk about what layer of the stack is this going to affect, not only the jurisdiction where this may be legitimate, but other jurisdictions as well.
And we've seen, and I want to end with a link that was presented in last year's IGF which is Protect the Stack which talked about the role of infrastructure providers, and I can see you smiling over there, because there have been instances where infrastructure has been deployed or used as a layer to sensor content or take down content, and Cloud Flare has been on the other side, we have TV farms and mastadon and so on and so forth and so just want to know your thoughts on all that have and what we should do going forward in the multistakeholder setting? Thank you.
>> FARZANEH BADII: Thanks a lot. I want to use the last two minutes to think about what sort of processes we want to use to discuss these solutions like the layered approach and other solutions to prevent Internet fragmentation. But Marco, very short.
>> AUDIENCE MEMBER: Very short and very final thought come from RFC 1925. Yes, it is always more complex than you think, but as watching this, be careful not to add multiple layers of indirection. I hear a lot of people pointing to each other, so then to technically on the last question, I think dialogues like this help with that.
>> FARZANEH BADII: Final thoughts? Where can we go from here? How can we take this conversation and advance it forward? No? You don't have anything? (Laughing).
>> ALISA STARZACK: The last set of comments and questions I think is right. I think we need to figure out how to make this conversation feel accessible, and also to understand that, yes, every nation wants to legislate within its borders, and sometimes when you start getting into the world of infrastructure, you are legislating far beyond your borders, and that is where some of the challenges come. I guess that's what we're getting at at the layered policy concept. If you are legislating to things that affect outside of your borders, you should be thinking about that reality from a practical standpoint because that means other people should legislate outside of yours ‑‑ or inside of your borders as well that are not accountable to you in any sort of way, and I think that's where some of our challenges come.
I will say on the bigger question, we have lots of challenges online. I guess this is a shout out on that, the optism side. The best way to solve all of those is to think about how we work together, both from government to government, government to industry, Civil Society, all of those have a role to play, and I think that's really where we want to sort of move forward.
>> JEAN QUERALT: Quickly to wrap up. Oops. As next steps, I would want to encourage that we start working on outcomes‑based taxonomy, both for human rights as well as data management. That would be an interesting step to take.
>> NOBUHISA NISHIGATA: Last and final words. I'm impressed by the Dutch Government and good for them that they have such staff like him, he knows both the policymaking plus the technology. Like to me, just to be honest with you, like I just talked about the telecom business in Japan, but it's originally aim at phones. Then the Internet came, then of course we tried to catch up and update our regulation, but the Internet was much faster than our effort to do some of the regulation legislation work. That's the truth, I would say. So then we had to catch up. And then, of course, we respect the freedom part of the Internet, so then now it's time too that we have to catch up by other types, new types of solutions, which is going to be brought not only by the government, but as with you guys helping us, the technician, industry, Civil Society, everybody.
>> FARZANEH BADII: Great. Another positive note to end the session on. Thank you, everybody.
(Applause).