FINISHED TRANSCRIPT
EIGHTH INTERNET GOVERNANCE FORUM
BALI
BUILDING BRIDGES ENHANCING MULTI‑STAKEHOLDER COOPERATION FOR GROWTH AND SUSTAINABLE DEVELOPMENT
25 OCTOBER 2013
9:00
DYNAMIC COALITION ON NETWORK NEUTRALITY
********
This text is being provided in a rough draft format. Communication Access Realtime Translation (CART) is provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings.
********
>> I think we can get started.
Thanks a lot to be here the last day at nine o'clock. That is really brave and we all appreciate.
Welcome to the first meeting of the dynamic coalition on network neutrality.
Today we will try to analyze the relations between some technical and some human rights perspective of the network neutrality debate.
In order to develop this analysis, we will start with keynote by Chris Marsden, professor at Sussex University, then we will keep with my distinguished panelists.
So Jean Jacques, I would like to call one of the fathers of the Internet, and the director of the free expression project at the center of democracy and technology, and then the senior policy director, not senior policy, policy director at access.
Okay.
Before starting our analysis, I would like to spend just a couple of words to state what is the dynamic coalition DC.
We all know that dynamic coalition are informal groups that analyze a specific topic in a long period.
This dynamic coalition has been created as an outcome of the mult steak header on human rates, organized by the council of Europe. The council of Europe has given a great input to this coalition and has been instrument toll raise awareness to educate what is network neutrality and spread this kind of analysis, and also to create a model framework that was one of the wishes of the council for Europe since 2010 declaration on network neutrality.
Why network neutrality it is important today.
We have seen here in the last days there are not just technical issues, there are not just competition law issues. There are also human rights issues. There are also consumer rights issues.
And today we are in a circa moment in which some regulation has been proposed, and we feel that this regulation should be good regulation, evidence based regulation.
And so I would like to start with the keynote by Chris to know something more about policy regulation and future of network neutrality.
Thanks, Chris.
>> I'm loud, but it's probably better if I speak into the microphone.
Just to say it is of course the book of the dynamic coalition.
My chapter in the book is page 76 onwards.
I'm not going to talk about the technical elements because you can of course read them from page 76 onwards.
But I am going to talk about some of the misconceptions about network neutrality
I was inspired by the panel on Wednesday, I don't know, maybe annoyed is the word by the panel on Wednesday.
I want to talk about some things that come out of that.
Just to say as you're probably aware, there's proposal for a European regulation on network neutrality.
It's unlikely to happen before the elections but we can have a covering about that.
I'm not going to talk about that directly
I want to talk about what ISOC said on Wednesday.
And they said this. Competition lowers a key aspect of network neutrality and provides answers to many of the emerges issues.
That is in the ISOC daily highlights. No, it's not a key aspect, and no, it can't provide specific answers
I have a book that responds.
It responds to all ISPs, big or small, mobile or fixed, need to manage traffic. Does that make sense?
Several claim to be deluged with video on HTPP, and we'll see this is not necessarily the case.
Nevertheless, it is about setting the rules of the road for ISPs.
It's not about big and small, it's about all ISPs.
So that hopefully sets to rest one myth.
Part of that myth is, oh, transparency in switching will solve the problem. That is a mirrorage. The more you try the more you fail. The U.K. has been tying itself in knots trying to increase both transparency and switching.
They have had a desperate attempt to say this is self regulation, it's as close to regulation, they are doing as much bullying without passing a law.
Compared to the United States U.K. load uses very low upload speeds. If you are trying to put something on the network, it's very hard to do so in the U.K.
There's wonderful work by elisa Cooper, you can find it by search on SSRN.com, a paper she presented on the the telecom policy conference.
As many of you know, she is on the inside architecture board and was at the ten eyes meeting in Montivideo.
This is some graphics on the U.K.
As you can see it's picked up from a very low start on broadband, very late move in the broadband market.
This is particularly shocking probably for our Korean participants to see the U.K. at the left is at about three Mega bits per second advertised in 2006.
On the right, at the edge of the screen, the actual speeds you achieve are about 45 percent of the advertised speeds, even lower I'm afraid.
We're not an example to anybody, but you're not surprised.
Let's just remember that, to recap on that, ISPs all have incentives to discriminate. Let's not fall into the trap of assuming that dominance is necessary for discrimination.
I'm a competition lawyer. Don't try to prove oligarchy in collective dominance. It's difficult to achieve unless you have people recorded saying let's fix prices in the cartel.
That is not going to happen.
Let's rephrase the definition of what the actual problem is.
Telecos should not be overtly note against their users.
In legal language we would say allow only reasonable traffic management. That is supposed to set the rules about what is note or not, or for a brand marketer, don't be ee ville or get caught being ee ville.
This is essentially what the argument is about. That is not about big or small.
So on the left you can see some telecos, on the right some of the people they would like to throttle.
We won't talk about surveillance here, but let's recall what we are engage in when we talk about discrimination between traffic types is of course about ISPs having to exercise some degree of sensor ship on their own initiative.
All week all it seems to have been talked about in the coffee breaks is states telling ISPs to censor communications for the states' wishes or tapping communications on the ISP's channel.
This is actually about the ISPs doing so themselves.
This man is the head or was the head of the second largest U.K. fixed line ISP, talk talk, and he said seven years ago, we're being note. Or specifically what he said is, we shake traffic to restrict peer to peer users
I get hate mail at home, that means we restrict the ability to play games
I have two people threatening to kill me.
That is the kind of note moment which you would have thought a regulator would pick up on.
U.K. regulator chose not to. They said he was simply making a joke. He wasn't.
It was just an illustration. He was says, we throttle. We throttle because we are managing traffic.
When we do that, we get lots of complaints from users.
Regulators need to act when they get this kind of smoking gun situation and can see exactly the kind of problem that occurs.
It's worth saying Western Europe shouldn't have a problem that requires unreasonable traffic management in any case.
First of all fixed traffic growth in Western Europe is low and manageable. The annual compound growth rate predict the by Cisco, 17 percent per year between 2012 and 2017, way below the historic growth rate of fixed Internet.
Actual fact, in Western Europe we're fine. Even mobile forecasts are falling off a cliff because predictions about video this woo swamp networks, the BBC says something like 92‑95 is actually wifi, not coming off the mobile network itself.
Of course wifi is fixed.
So the video being consumed is consumed off the fixed network on the mobile handset.
It's worth also noting for those of you from countries with only two or three mobile networks that the analysis that is being done of the actual price of data in Europe shows that it's actually the entrance that at least offer most lower prices than the incumbents.
Votafone is the giant skyscraper on the left, and the three on the right.
But we're only of course starting to implement net network neutrality.
Netherlands we referred the other day the Netherlands has a law, not actually enforced so far, so they haven't actually as far as I'm aware not have had any cases.
I'll be interested to hear if there have been. It certainly wasn't in place in terms of the regulation until this year.
Slovenia is more interesting, a law passed just before Christmas of last year.
In the U.K. just to summarize, we have had ISPs throttling traffic for over 12 years and government doesn't care.
Enforcement is actually very easy if you approach it logically.
If that sounds like beginning of a song, we'll come back to that.
Network neutrality 2013, the Netherland passed a law last year, still to be implemented, Slovenia as well. In Chile and Finland we have access to the unfiltered.
The U.S. has the situation with the open Internet order but that is under challenge.
Norway has the most effective actual implementation of network neutrality even though it's so far done by co‑regulation, though I believe it's going to amend its telecom laws to make this a more specific type of co‑regulation.
Canada has rules from four years ago. They haven't really been implemented. There's been some work on usage based billing in Canada, which is hopefully got rid of some of the problem.
We have self regulation in places like Japan and the U.K., and still in France, although France is threatening to move towards law, but we'll see whether that happens or not.
So this is the Slovenia economic communications act, an unofficial translation.
The act is much more strict than the Netherland, probably more interesting in terms of a strict network neutrality approach.
ISPs are prevented from restricting or slowing Internet of traic except for the usual exceptions.
Commercial differentiation of quality of service is actually prohibited in Slovenia, really quite a strict condition imposed.
The ISP is prohibited from different connectivity prices, which is likely to affect the way in which they can offer data caps or not
I think Slovenia is the country to watch most closely in terms of implementation.
Returning to the logic, at least 50 ways to throttle the user as we know.
The question left hanging in the air on Wednesday, who should discover or regulate network neutrality.
Discovery particularly for those of you from developing countries, why not use the fact you're funding all this fantastic university research, plus of course many content writers willing to provide you with information what throttling is taking place.
Geeks can help lawyers.
Let's have a proper effort to do that.
In Europe we spend multi billions a year on network research and universities are very keen to help to discover the problem, very happy to do so.
Of course they will only find it, let's just remember, as I said, overt problems because they will only find it where it's wide spread or obvious.
Subtle discrimination will not be picked up even by the keenest research.
We can maybe have a discussion about that.
The regulation will have to have through the teleco national regulators, it must be the way.
They just need probably some retaining to remember they do have a legal function to protect freedom of expression and user privacy. They have sometimes forgotten and it's worth a little reminder to them to make sure they are not strangely forgetful of their constitutional functions.
After all, these are public communications networks, not private communications networks.
Of course enforcement with appeal to courts, why not, with all other telecom courts. And courts are perfectly familiar with balancing commercial and human rights. I don't see anything difficult.
Let's not see this as more difficult than it is.
Telecom can solve the problem, university researchers help us find the problems if they exist, and we can solve the problems by general process.
Of course all regulators can therefore find breaches.
We haven't gone mad. We're not all bonkers. We can actually deal with this problem.
You probably are familiar with Alice in wonder land and the cheshire cat.
Let's not try to make this appear a more complex problem than it is.
How long do I have left?
Okay, can developing countries afford network neutrality? You can't afford not to have network neutrality.
In almost all cases, of course the west did this as well, we sold GSN licenses much too cheap. Around the world we see this multi billion airs doing wonderfully well out of their mobile licenses.
Generally developing countries, I include the U.K. in this at least in this regard, we don't have competition, we have geopoly, at best three‑we competition amongst mobile.
We don't have fixed line alternative, in the U.K. we do, but only one whole sale, British telecom.
It's really network neutrality or benefit. When the countries tend ton selling the mobile Internet, the amount of throttling and amount of preference that takes place, an its a bit of joke to claim it's the Internet when it's actually not offering open access to the Internet.
That is all the digitally divided have in order to connect
Without network neutrality you're offering them something which is a very different experience than the real Internet
I know a lot of people, a lot of delegates here have found it very instructive to be trying to use the particularly the wifi networks in the hotels and to see what happens when you have quite strong content filtering and quite slow speeds.
It's reminder to us of user the challenges that are faced by users every day in developing countries, I should add to this, at least 10,000 villages in England where they also have very very slow speeds.
It's not simply a problem in particular countries.
So very quickly about the past, it all began in 1999 in the last Mel millennium, it's not a third millennium problem, it's a second millennium problem, based upon fires of cable wall gardens.
You remember the AOL Time Warner merger, and the submission to the FCC at that time which was called, this was meant to be a snappy title I think, geek title, the end of end‑to‑end, they wrote.
That is what they feared would happen.
ISP a big closed monopoly model.
Remember this is 14 years ago.
Do not allow your politicians and others to say well, telecom companies, politicians, this is a new problem, it's only a problem in the west, it's not a problem we should worry about.
At least 14 years old, frankly, anywhere that has broadband has this problem.
Of course the council of Europe, we mentioned did have worse about network neutrality
I wrote in '99 a record from for the council of Europe in which we talked about bottlenecks, convergence, the way these things take place, and talked about the problems that occurred
I talked about the fact that in the states, ISPs were asking the FCC to bar cable operators striking exclusive deals so the customers can enjoy the same services they now have over phone lines, which we now call network neutrality.
We knew about it back then.
To wrap up completely, you probably are not familiar with the book.
The book exists.
You can buy it, you can also find the pdf on line on your favorite pirate website.
It actually had 100,000 downloads in the first two months.
It's on a creative commons license, published by Blumberg, the Harry Potter publisher, as you will all know I would have thought.
This is some of the research being done on trying to find out where discrimination is occurring, you can see violations worldwide.
It's going to grow and grow.
I'm here actually in Bali through the network of excellent on Internet Science.
We do a lot of work on trying to put together the hard technology plus the social science to understand how networks are developing.
We had a conference recently in which we had one of our colleagues now on the Internet science project, professor Ziga Turk, he was the minister in charge of the Slovenia telecommunications act going through the parliament.
Within our research group we actually have the person who has passed the toughest network neutrality law in proper so far.
And there we are.
Just to say some of you have a flyer on the desk, it's the recent book I wrote with Ian Brown, computer scientist at Oxford. We include network neutrality on that.
It's basically a subject I think we have to ask really good questions.
We're not all bonkers and it's not all difficult. It's really straight forward.
>> Thanks, Chris, for this extremely interesting keynote
I would like also now to ask Mr. Presan, we have heard about variety of service, the problem that can trigger the service, the fact in Slovenia it's banned.
We would like to start by you of being one of the fathers of the Internet, you have the honor to have the first question.
So what can quality of service determine some interference between users or can quality of service guarantee no interference between users.
>> The question may be stated so that it has no solution always.
In this case the question is what we mean by network neutrality.
If we have first assumption more band width that is needed.
Second assumption, there's no throttling in the system. In other words, nothing like a three‑lane express way that ends up two lanes and blocking of the traffic.
If we have networks built in a very smooth way, and if users are not trying to abuse of it, then you may have no interference, but that is not generally the case.
The case is that we have heterogeneous networks where the Internet is made up of thousands of autonomous separate networks which all have some business to do, sell their bandwidth, make sure they say their major customers, lie to each other when they are peering without giving arrangements so that they may potential take advantage of the weaknesses of the neighbor.
That means the world is not perfect, and it happens that they generally try to sell more bandwidth than they do have, which means that it will be obviously some points where they will not be enough bandwidth for everybody.
Next question, do we have classes of services.
Well, it's well‑known that if you have no class of service, everybody has the same service, that means the big consumers are restricted by the small, the small pay for the big ones, which is obviously not quite satisfactory in terms of equality or neutrality of economics.
Second thing is the network is the packet network, the IP service, as you know, has been built initially without any discrimination even though there always is some discrimination when you have to, when you have to send packets in some direction and there's not enough capacity, you obviously have to trade somehow.
You make take the short packet first or the longer first or the ones longer in the network, several criteria which obviously have to be examined in order to avoid complete congestion.
It means the network practically has always been in some way discriminating between traffic.
But if we have classes of services, it makes more sense because to have different requirements.
Things like voice, things like file transfer, things like mail, things like T.V, for example, which need bandwidth but can also be buffer, need different requirements.
When I think of a single network which doesn't discriminate according to services, think of hospital in which all patients are treated equally.
That doesn't work.
Therefore, the ideal thing is too much bandwidth, at least enough for everybody, and all the networks profiled or categorized in a way that make them completely how much use and at least ensure smooth traffic
I think that is realistic.
So I guess the networks have to be built in a way that can take advantage of the economics, in other words, if they have a product customers, they have to offer a good service to the big ones.
Of course if they don't want to deny service to the smaller one, or to deny service to those who abuse big bandwidth, they have to introduced classes with tariffs.
And just the same as high speed train and flights and many kind of product services, services may be tiered differently according to the privileges or according to the cost it makes to offer the services.
So I guess we have no other choice if we are to live in the real world, which is not necessarily very friendly to each other.
And that is what we are trying to do.
So for me for many years there was no way to homogeneous the traffic and multiplicity of the heterogeneous nets.
Some of our colleagues, you will see in my paper, have worked for years on that. But I'm not a mathematician so I have no way to tell you whether they made mistakes or not in their mathematical approach.
Apparently their peers agree they find way to predict QS, in other words, end‑to‑end performance, provided they know the criteria of the nets. That means again you need transparency.
If operators hide the characteristics, I don't see how we can do it.
But assuming somehow more or less they are pushed to tell what characteristics they use, and the user could be some law for that, trespassing, even though it's not respected, could be set into regulations.
Then there are somehow the possibility of predicting the traffic.
But predicting is always a game, saying that it may occasionally go away from the prediction. Same thing for car traffic. Car traffic uses a common resource, which is road, and predicting a certain amount of traffic, but there are some peak times.
I guess that is something that will keep happening in networks too.
So for providing a reasonably satisfactory services to users, I think it should be, instead of having a completely open system in which everything is equal, which is never true, there should be a differentiation according to what service is provided.
And there should be for the operators the very strong determination to offer in each category of traffic, in each class of traffic, they should at least offer a minimum service, a minimum QS service which is guaranteed any time, at any time, almost any time.
It can also be catastrophes on the net, things which are down, and possibly DDOS and tricks like this.
But normally there should be a minimum class of QS which should be guaranteed a new service.
Now, the service is always of course has this, it's never, it cannot, it can never be a strict observance of characteristics like response time, like meg mega bits, like twitter.
It's always statistics. Therefore you have limits, minimum, maximum.
In other words if the service is in between the minimum and maximum, it's considered as okay.
And that may vary of course with time.
Andsome if traffic, there's very few users on the net, you have better service than if you are trafficking at peak times.
I guess that is life.
So it can be improved. But it cannot be cured with a hundred percent guarantee.
Thank you.
>> LUCA BELLI: It is true some quality of service can be needed but also true that traffic management could be needed, but it's also true that we should try to translate to user control maybe traffic management, so traffic controlling, traffic management operated by a centralized operator.
So maybe the solution could be also to empower more users instead of empowering the ISP, but that opens the debate to other topics.
So I would like now to, now that we had this technical perspective, I would like to ask to Emma to provide us some inputs on what makes network neutrality also a human rights issue.
>> Thank you.
So I think a key thing in thinking about how network neutrality and human rights intersect is to think about the human rights to free expression on different human rights instruments, for example, the universal declaration of human rights article 19 says everyone has a right to freedom of expression, to impart information and ideas through any media and regardless of frontiers.
What you really see in discussions of the human right to free expression is a sense of borderlessness and also person personal choice.
I don't want to belabor the point of how the Internet is a revolutionary platform for the enjoyment of free expression. I think we are all pretty well serviced in frank Larue, the special rappatore on the united nations report on free expression, has done an analysis of how Internet influences a whole range of rights.
Access to education, participation in cultural life
I think it's fairly uncontroversial to say that the Internet is an amazing enabler of free expression.
And that is due in large part to specific character decision of the Internet, the fact that it's global and allows immediate access to information around the world regardless of frontiers, the fact that it's user controlled, it is this amazing facilitator of user choice, user content selection, and putting the power in users' hands to decide what they want toe say, how they want to target their information, what information they want to access.
It's decentralized with open technical standards and decentralized routing that really lower barriers to entry and make it easy for any person to create a new platform, a new application, put their views out there into the world.
And it's open and competitive.
It can accommodate at the endpoints an essentially unlimited number of speakers.
This really helps to create a greater parity between large and small speakers.
The way network neutrality comes in, it's really all about preserving the characteristics in the face of increasing technical capacity and economic motivation for access providers to all ter them.
>> LUCA BELLI: Folks, I think it would be interesting to know what could be the human rights implications of nonneutral traffic management.
Mr. Pasan was arguing some regulation of traffic could be needed, but it could be argued that traffic management could have some contingencies on fundamental human rights, so what could be the consequences.
>> In had addition to the censor ship consequences to flat out blocking of information, if you think of something like a content provider and access provider entering that a paid prioritization deal where a provider can pay an access provider to get fast line service to ensure their bits get to users at a high quality of service level.
And if this deal is cort of exclusively offered to certain big content providers, it's really going to edge out and squeeze out smaller operators, or smaller content providers and smaller speakers, whether they are people in the same country, the same network, or people outside of the country trying to decide do I have to make a deal to actually get my content and my views accessible to people in this, using this network.
Maybe it's not going to be economically feasible to try to send my video or send my service into that country if I know the situation there is that I have got to pay to even be able to begin to compete with some of the dominant players in that country.
So I think what we can see with those kinds of deals is the potential for creating borders and creating barriers, creating frontiers where there aren't any now.
>> LUCA BELLI: So network could be competition law friendly and also human rights friendly, so could you provide us a couple of inputs on how the best competition at the edges can promote human rights.
>> I think what we have seen with the Internet is the ability to support much greater diversity of media and content sources.
So you move away from the more limited broadcasting and cable environments where you have dominant players able to put forward huge resources to meet the very high barriers to entry to access those communication platforms.
With the Internet we see greater opportunity for small media companies, individual bloggers, to put their own voices out there and have that competition with the dominant providers in a way that doesn't require the same kind of investment and same kind of deal making is.
>> LUCA BELLI: Thanks a lot.
Wonderful
I would like to keep on this human rights path
I would like to ask to Yohi if he can provide us some inputs, his perspective on why it is important now that network neutrality regulation is adopted.
We know that at the European level, the European Commission has a proposal for a European regulation, and so why it is important that this regular legislation adopted, and maybe also some comments on what would be a good direction to follow for adopting the regulation.
>> Thanks, Luca.
I'm the policy director at access center.
We're an international NGO that defend and extends the rights of service around the world
I think Chris did a good job of highlighting how network neutrality is something that is happening around the world right now.
Some ways running ahead and some under threat.
And I think one of the things this highlights, network neutrality means different things to different people and different governments.
And so I think that perhaps some of us familiar with the European commissioner for digital agenda has proposed a market regulation, she says it will guarantee network neutrality consumer rights. It still allows for discriminatory practices.
It prohibits ISPs to block, slow down, down grade, otherwise prohibit for specific applications.
That makes these meaningless for allowing for commercial agreements.
Simply put, if there's a fast and slow line, there's discrimination.
The notion if ISPs are restrictive about practices, then discrimination is acceptable, is ludicrous.
Transparent law is still wrong.
As Chris has pointed, this will allow users to make informed choices, fails to take into account the nature of the European telecom markets and most markets arounds the world.
The way we approach this in the report is the language around discrimination.
The language of the ISPs to intentionally apply restrictions to user access to the Internet.
Generally speaking this can take place through blocking of application services, slowing or throttling speeds, blocking websites, and preferential treatment on platforms.
We have heard about reasonable traffic management. This is how the discriminatory practices are often justified
I think there's a fine line between preventing saturation by slowing down and degrading the quality of competing users or violating user rights to receive information
In our paper we try to talk about what do reasonable traffic management reports look like
I think traffic management is reasonable when it's deployed for the purposes of technical maintenance of the network.
Namely to block spam, to deal with viruses, service attacks, to minimize the effects of congestion whereby equal types of traffic should be treated equally.
This is established in the dutch law.
Traffic should be only used during exception the moments.
When traffic management are in place to pursue other purposes or used on permanent basis they should be considered unreasonable.
(Reading document).
They undermine the proper function of the online marketplace
I think the traffic management also raises another human rights concern.
Access to information, in addition to that, the freedom to receive information, the increasing use of perpetual and unjustified traffic management raise questions about the privacy of applications.
In order to implement a variety such as blocking, shaping, filtering, ISPs are employing tools such as depack inspection technology. This allows them to examine the information traveling and recognize the type of packet.
That is important for spam and mitigating technology of the network, but it can fall outside the scope of securing the network.
This highly intrusive tool can be used not only for practices but also to monitor and indeed copy all information flowing over the network, which I think we have heard quite a bit about now in the last few months.
Again, by inspecting communications, I mean, ISPs breach, and this is a violation, we would argue, of article 8 of the European convention (reading).
The been of the data protection supervising, we would say these can only be provided with safeguards which state what can be done under what circumstances.
In the preface to our paper, the founder of the worldwide web argues, some are concerned mandating network neutrality through regulation would impose regulation and inhibit broadband deployment.
One vision is that the industry and government alike can domain taken the neutrality on a voluntary basis.
However, recent events insist to trust future and present governments with some so crucial, whose abuse is so important, is naive.
Generally the Internet thrives on a lack of regulation but the core freedom of speech and rules against dim discrimination are indeed enshrined to create benefit.
Why not network neutrality.
Just as democracy depends on free speech so does freedom to connect to any party is the fundamental social basis of the Internet and now the society based on it.
History has shown in the absence of discrimination many companies will particularly if there's profit to be made.
Especially on issues concerning human rights network neutrality should be no exception.
Strong legislation will only provide citizens access to the Internet and also (reading).
To enable and promote human rights, we believe Europe needs strong comprehensive legislation now.
Thank you.
>> Thank you for highlighting that traffic management, discrimination, could be allowed for some integrity and security reasons, to preserve integrity and security for the network, on a temporary basis.
The problem is to deploy such measures on a permanent basis and also to deploy intuitive measures as we have seen in the inspection.
So I think it could be also interesting to analyze some social aspects of network neutrality
I would like to ask Francesca to highlight why it's important to study network neutrality from a public sphere, sorry, public sphere perspective.
>> Thank you very much, Luca, for the invitation and delighted to be here and to be part of this group project.
So what I'm about to say in the next few minutes is the result of joint work with German colleague who could not be here today.
What we have done, we are trying to approach the network neutrality topic from the perspective of a public sphere literature and studies.
So why is that? Because regardless of the ways in which network neutrality is defined, and a few speakers have already pointed out, it can mean different things to different people, it has to do, it is at the very core of the question of whether distribution channels can be used as a means to discriminate, to control and to prevent communication in some way.
So another way, content and user behavior can be controlled through the architecture of the physical layer and the code layer of the Internet.
So the discussion on network neutrality actually touches a number of fundamental values that range from public interest to freedom of expression, freedom of the media, and the free flow of information.
And communication policy authorities actually frequently appeal toe all of these values in order to legitimatize their interventions in media systems.
The implementation of these values from a normative point of view is seen as the precondition for media to create the public sphere, whether on line or off line and fulfill its function in society.
Of course the concept of public sphere has been elaborated in many different ways by different authors.
However, the concept developed by yergen Habermas is widely recognized as being influential. Card according to him the public sphere links people in which public opinion can be formed.
His concept on public sphere gives a lot of importance to deliberation and POS itself that functioning deliberation provides that access is provided to all citizens.
You can see the emphasis on access makes the concept of the public sphere particularly useful for the network neutrality debates.
And we particularly used beda dagman concept of public sphere because he developed into an analytical tool to study the role of the media and Internet more specifically, the public sphere.
So he distinguished three analytical dimensions of the public sphere, the structural, the representational, and the interactional.
The first refers to the organization of communicative spaces in terms of legal, social, economic and even web architectural features.
And all of these patterns impact Internet access in different ways.
The representational dimension directed attention more to the media output and raises questions that concern fairness, pluralism of views, agenda setting, ideological biases, and other evaluation criteria for media content.
Finally, the third dimension, interactional one, focuses on the way users interact with the media and with each other in particular online sites and spaces.
We have been using the analytical framework to identify studies relevant for communication studies and use them as entry point into a particular set of network neutrality issues.
The strat eral dimension has been a way to address the issues related to access for Internet structures, both individual and collective entities.
The representational dimension has led to the question of how net neutrality relates to online content and its distribution.
The related issues are content, diversity control and censorship.
And the interactionel dimension has been used for the spaces of communication online and particularly focused on closed system or world governments to illustrate the potential for actions liberation to be impeded or lost.
There's some overlapping between the dimensions, but we hope that it provides communication studies informed perspective on these debates.
I'll be happy to get any questions on like how we address this in more detail in the chapter which starts around page 36, I think.
Thank you very much.
>> LUCA BELLI: Wonderful. Thanks a lot for this explanation
I would like to check if there are any questions in the room before attacking the model framework
I think Eduardo, then the gentleman there, and another person there.
Eduardo.
Can we bring the microphone there, please.
.
>> Okay, thank you for wonderful presentations, very useful for me.
First, I'm Eduardo ain Antonio, director of the freedom of access, I'm teaching human rights on the Internet, and one of my classes is network neutrality. Of so the network neutrality is already something that I have been working.
And just an announcement.
Our center just launched yesterday a paper in Spainish, a policy paper on network neutrality with some recommendations. Because in Latin America, this is an issue and many countries start after Chile's law, start thinking, having regulation.
We also translated into Spanish the framework.
So it is in the website, on our website and the website of the coalition.
We think it's a very very useful starting point for discussions.
My questions
I agree with you that network neutrality has different means for different governments. Sometimes misleading or, you know, people put network neutrality on misleading spot light. They talk about network neutrality but they are not really talking about network neutrality.
But my questions are more, I think are specific on three issues.
One is related to traffic control, traffic management, which we agree is necessary at some point, should not be discriminatory but it's necessary.
My question is who should control the nondiscrimination of traffic management
I think we found very complicated answer.
Yes, the easy answer is to say well, the telecommunications offices in the government or some of those offices should be controlling to be the ones that control.
But sometimes they don't have the capacity to do so.
So my question, is this better to have some sort of control policy control, random control, or some kind of reactive control, somebody make, denounce or something, then the agency can start work.
My second question is about the establishment in the rules of the network neutrality principle.
What do you think, it's better to have a law, a formal law that established network neutrality, or it is enough with executive orders or implementing decrees that implement telecommunication laws?
This is something we are experiencing. In Argentina, for example, in one track the Senate is discussing network neutrality law. On the other track, the president issued an executive decree that implement one telecommunication law that included network neutrality.
We are in favor of a law. But I want to hear your reaction.
And my final question is about, again, traffic management. And I ask who, and now I ask how.
When we ask how, again, going back to Yohi, DBI is the issue, but DBI could be very dangerous.
So what is the limit there for controlling the traffic.
Thank you.
>> LUCA BELLI: Thank you for the interesting questions.
Who would like to go first?
Thank you.
.
>> Not sure I understood the second one.
For the first question, I think it's a real simple question. Who controls indeed.
First, I think that networks must be transparent. In other words, the characteristics of the net have to be transparent.
That can be a regulation. It can be put into law, some kind of coercing institution.
Second thing, it had an is to be measured.
If you cannot measure the QS, there can be no meaning. Therefore there have to be instruments that verify that the end‑to‑end characteristics defined in the QS are indeed met by the service.
Now, if it's not met, what happens.
Well, indeed, that is becoming a little bit more difficult.
But usually when it always occurs in some end‑to‑end terminations, it usually points after some statistics you can point to a particular area on the Internet where the trouble occurs.
Now, it is a specialized job
I think the agencies, usually the country where you have telecom, several telecom operators, you have sort of a state control agency which is supposed to be independent, and which typically make measurements to the characteristics of the service.
That should solved.
The second thing, of course, you cannot lie, because they have also competitors and competitors observe what is happening. And if they would lie consistently, they suddenly would be caught quickly.
The third thing is the user should be able to determine whether the QS he pays for is actually met.
And this is possible. Today we have various tools which you know, for example, can measure the bandwidth and the bandwidth you are currently receiving, or can measure the number of packets or the signal, for example, using the voice communication.
This usually requires some special software which people can acquire, which can be given to them, or it may require some extended box which makes measurements out of the net, because it may require end‑to‑end acknowledgment of signals.
But it may not be simple for anybody to do.
But it certainly is very possible to do for advanced users or for technicians who are assisting users.
So I guess unless you have, you know, a certain agreement of tools for measuring the traffic, measuring the characteristics, indeed, QS doesn't mean much.
And in each case there's no point in complaining about distortion. If you get the QS you pay for, who cares about whether the other services, what other users are doing.
It's their business.
It's up to each user to complain if the service he gets is not correct.
Like anybody, people have today with their providers in any case, accountants, traffic, airline traffic and so on. They can complain if they don't have the good service.
>> LUCA BELLI: Thanks for this answer.
It's also true that there are tools that exist, but they are frequently not understandable for the average user.
So I think Nisha can easily complain but the user will not be able to understand.
So I think the resources too are important.
>> I must say first of all, it's great honor to be on the panel with professor Nusaf.
It's a wonderful opportunity to hear from one, from the father and one of the fathers of the Internet of course.
I want to say a couple of things about this.
The first is that one has to set down certain rules to tell the telecos what they can't be doing.
And I was really delighted to hear profession son Usan to say one of the problems are not, we should not assume their honesty, but we should put this place tools that allow us to demonstrate they are not being honest, particularly by using the competitors' analysis of these things, and of course advanced users. Use the geeks to catch them out when they are doing the wrong thing
I think you're quite right, a formal law is necessary in order to establish very clearly the baseline from which telecom operators have to actually carry out these tasks.
As I say, inn in many European countries we have tried all manner of nonlegislative means of being suggestive, but in natural fact they seem to be sending the wrong signals rather than the right signals to the regulators and the operators.
The third thing is, think we mustn't be scared of deep packet inspection. It needs regulating. We need to make sure that it's controlled the way that it's used.
But most fixed operators and almost all mobile operators use deep packet inspection
I think it's foolish to start having a discussion about whether or not it should be introduced. It's too late. Five years past the debate about whether or not it should be introduced.
It has been introduced, whether we like it or not, it has been introduced
I just want to make one comment also. The European council conclusions on the tele‑Compaqage which includes network neutrality, says they welcome the package and encourage the legislature to carry out an investigation with regards to timely adoption, which I think means it's being killed in the current session. Timely adoption doesn't suggest any urgency.
I don't know if that is the case.
I'm less skilled at reading council conclusions than some people in the room.
This issue about quality of service, we must allow it to be introduced, it's being introduced with specialized service
I didn't get time to get to that in my presentation, but clearly the definition is going to be extremely important in terms of what is the open Internet, what is specialized services with guaranteed quality of service.
And this is something which we will, I think, already be a growing topic as we go forward.
>> LUCA BELLI: Thanks a lot.
A reply about Johi and I'll keep the question for after the model framework.
Otherwise we will not have time.
>> Because Eduardo asked such great questions.
I'll try to keep it short and pithy
Because we are entering a goldenage and in terms of making the network understandable by average users, I think it's game changer in way to the question of sort of compliance and enforcement.
So I think who should control the traffic, management is probably the national regulatory authorities on the telecom side, but I think looking at DPI use as set forth in the privacy concerns, think the data protection authorities and others, human rights institutions, do have a role to play as well
I think that we, as Chris touched on, I think we have seen around the world that anything short of hard law fails to actually compel changes in behavior, even when we do have hard law, a lack of enforcement has failed to change core behavior.
And so we need regulators, we need hard law to empower the regulators, and we need the regulators have enforcement powers.
If you look at some of the regulatory approaches in, for instance, Norway, they don't have the capacity to pass sanctions for violations
I think that is why we pass for hard lay.
Chris, to clarify, I don't think DPI shouldn't be used. I think it's critical to maintain network security.
Here too I think we need to have oversight and transparency and audits by the data protection authorities to ensure that there aren't abuses on the network.
And I think that just to finalize, finish up, I think going back to my first point about measurement, we can see three network graphs, right, when traffic management is, here is the occasional disruption, congestion, let's throttle things a bit to make sure everything keeps going, versus every night at six p.m. traffic kind of drops, or this protocol is banned entirely ate all points.
The granularity of measurement and accessibility and intuitive ty of measurement is getting much better, and I think that will really empower both users and regulators.
>> LUCA BELLI: Thanks.
So I would like now to start to analyze a little bit the model framework.
Could we project the model framework on the screen.
And also if you want, you can also have a copy of the report of the dynamic coalition.
At the bottom of the room.
So you can directly read it on paper if you like reading on paper.
If you don't reading on paper, you can also check the model framework on the networkneutrality.info, on CC by license, so you can find it.
And so it is page 103
I would like to introduce the model framework.
As I was stating before, the process of the model framework has been stimulated and initiated by the council of Europe with the 2010 declaration.
The government announced the strategy of the council of Europe explicitly for deliberation of human rights principles to protect network neutrality and the declaration suggest elaboration of a framework to provide guidance to council of Europe members in order to protect network neutrality.
As it has been highlighted clearly, the internet is a collection of thousands of autonomous systems.
And these autonomous systems are interoperable and compatible because they use the same standards produced through an open and participatory process and transparent process.
The purpose of the dynamic coalition was to be instrumental in elaborating this model framework in the same way in the same way as the standard process of the IGF.
An open inclusive process, we have tried to transpose the IGF standard process to policy making.
We have tried to have a specific model that can provide guidance in how to protect network neutrality in accordance with human rights because specific standards are what allow a lot of autonomous systems to be interoperable and form the Internet.
So the purpose of this model is to allow a lot of autonomous, sorry, let's say independent national legal system to be interoperable and to protect network neutrality in an efficient fashion.
The first step was to define what is network neutrality and what are the limit of network neutrality.
So as we were hearing before, several people have different conception of what is network neutrality, so we have defined the principle of network neutrality, which is the principle according to which Internet traffic shall be treated equally without discrimination, restriction or interference regardless of its sendser, content, and Internet user freedom of choice is not restricted by failing the transmission of Internet traffic with particular content, services, applications or devices.
So what is important is nondiscrimination features of this principle that do not allow to target specific type of content, of services, of applications, or to block according to seeking or sender.
However, there are limits to this principle. As freedom of expression, it is not an absolute right. There are limits, some limits and some responsibilities.
The network neutrality principle has some limits and some responsibility.
In article 2, in accordance with the network neutrality principles, Internet service providers shall refrain from discriminating, restricting or interfering with the transmission of Internet traffic unless such interference is strictly necessary to give the provision.
And the provision can be an exception to nondiscrimination. For instance, blocking some kind of content that is against national legislation.
Preserve the integrity and security of the network. We can see that is something essential, essential exception, for instance, in cases of service attack or to prevent the circulation.
Prevent circulation of unsolicited communication for marketing purposes.
It is the case to allow the ISP to prevent the circulation of spam when the end users didn't give their prior consent.
Comply with explicit request from the subscribers. That is the case in which subscriber himself wants the ISP to block some content, for instance, to receive Christian Internet or to receive child friendly Internet, but it is the case in which it is the subscriber that is being the subscriber, paying for the Internet, he wants a specific kind of Internet.
And to mitigate the effect of temporary and exception the network congestion.
Primarily by means of application agnostic measures.
So measures that do not target specific content applications or services.
And when these measures do not prove efficient by means of application specific measures.
So if application agnostic measures are not efficient, application specific measures could be used.
Also article three of network neutrality principle shall apply to all service offered by ISP.
We have seen discrimination can take place not only at the last mile but at the intersection level.
If we want to truly protect network neutrality, all ISPs should convey traffic in nondiscriminatory fashion.
The network neutrality principle need not to apply to specialized services.
This is an important exception that allows the provision of services according to specific quality of service, but these specialized services are defined
I would like to draw your attention to the definition of specialized services.
If you go to article 9 indent f, the expression specialized services refers to electronic communication services provided and operated within closed electronic communications network using the Internet but not being part of the Internet
I really hope that the European colleagues later when they will review the European commission proposal will consider this definition because it is essential to preserve the open and best effort Internet, that specialized services be kept as something separate that is provided in closed network and do not interfere with the open and best effort Internet if we want to preserve the open best effort Internet.
And coming back to article 4, Internet service providers should be allowed to offer specialized services in addition to Internet access service, provided such offerings are not to the detriment of access, service, performance, affordability or quality.
Specialized services should be encouraged. That is an exception to network neutrality, it has to be encouraged because it can foster create activity, innovation, but the provision of specialized services should not diminish the performance, affordability or quality of the open Internet.
And offerings to deliver specialized services should be provided on a nondiscriminatory basis and their adoption be voluntary.
So users should not be obliged to adopt specialized services.
Article 5. Subscribers of Internet access service have the right to receive and use a public and globally unique Internet address. This provision can sound odd for someone that is not a technician or has a technical background.
This provision specifically targets the carrier grade network address translation that basically exploits a unique public address and unique IP address, and then provides thousands of subaddresses.
But this carrier net has presented several problems for users, not allowing users to enjoy services like voice over IP, or I would rephrase, not they do not allow, but they present a lot of technical problems.
The issue is if you are a subscriber, if you are paying for your subscription, you should be free to use all the services you want, all the legal services you want, and it does not have to be up to the telecom operator to tell you you cannot use that.
If you are a regular Internet user that is enjoying free access to the internet in a library or cafe, in that case you maybe could be access to a carrier grade net. But if you are paying for your Internet connection, you have the right to have a publicly unique Internet address.
Any techniques to inspect or analyze Internet traffic shall be in accordance with privacy and data protection legislation.
This recalls the European data protection, sorry, the Europe data protection supervisor opinion that was stating that, in 2011, was stating that technique, intrusive techniques should be granted in accordance with both privacy and data protection legislation.
So that is extremely important.
And we stated, by default such technique should only examine header information, not the content of the packets.
Because analyzing the content of the packet is like a postman that opens a letter and sees what is inside and decides how to deliver it.
That is a violation of privacy of electronic communications.
So if that is put in place, that has to be scrutinized by the national data protection authority.
The user of any technique which inspects or analyzes the content of communication should be reviewed by the relevant national data protection authority to assess compliance with the privacy and data protection communications.
So intrusive techniques may be used as long as they are used and implementation is reviewed by the international data protection authority.
Article 7, Internet service provider shall provide intelligible and transparent information with regard to traffic management practices and user policy, notably with regard to the existence of access and specialized services.
We have seen transparency is of utmost importance, but we know that simply providing information that is not intelligible, cannot be understood by the regular users, is equal to not providing information.
The information must explain how traffic management is adopted and which kind of traffic management is adopted, should be intelligible, meaningful, understandable by the users.
And especially the information regarding the existence of specialized services and public Internet should be clearly stated. Because we have seen that specialized services are encouraged, are allowed, but they do not have to diminish the quality, affordability and performance of the open Internet.
When natural capacity is assured between Internet service and network services, it should be stated.
Article 8 is the enforcement mechanism of this model.
The competing national regulatory authority shall be mandated to regularly monitor and report on Internet traffic management practices and users policies in order to ensure network neutrality, evaluate the potential impact of aforementioned practices and policies on fundamental rights, ensure the provision of a sufficient quality of service and location of a satisfactory level of natural capacity to the Internet.
Reporting should be done in an open and transparent fashion, and reports shall be made freely available to the public.
So the question who should monitor and how, well, this is the answer.
It is the national regulator that should monitor how traffic management is implemented, which kind of is implemented, and also to monitor that these traffic management policies, which kind of impact these policies have on network neutrality and human rights.
So as we know, it is not particularly difficult to train some officials that work for national authority in human rights, in fundamental rights, to extend their competing. Just that is to be clearly stated and defined how to do it.
And also recalling an anecdote of south American regulator that was sitting thousands of complaints on network neutrality without taking any step, we have clearly stated that reporting should be done in an open transparent fashion and the reports shall be made freely available to the public so the public can know which kind of reports have been done, and put in place, the network neutrality put in place appropriate, clear, open, efficient procedures aimed at addressing network neutrality complaints, and all Internet users should be entitled to make use of such complaints procedures and fundamental authority, and respond, and authority should respond in a reasonable time and be able to sanction the breach of network neutrality principles.
All end users should have a right to file a complaint.
As Mr. Yusan was saying, the users have the tools, and they have a right to signal discrimination and file a complaint.
As I was saying before, recalling the anecdote of the regulator sitting on thousands of complaints without acting, the regulate should respond in a reasonable time and be able to use necessary metering to sanction the network neutrality principle.
End‑to‑end, finally, the authority must have the necessary resources to undertake the aforementioned if we want them to be effective to receive the complaint, to act, to monitor, there have to be resources.
We cannot pretend an authority does these overwhelming jobs without having resources.
Now here we state how it should be done.
The national government should effect the necessary resources to do this work.
Also definitions in article 9, but as we are running out of time, I hope you will understand that we have started with five or seven minutes of delay.
So if we exceed a little bit the time, maybe you will understand
I would like to open the floor for comments and questions.
So I think there's a gentleman there.
Could we bring him.
>> Thank you.
Mike Hanson from the Association For Progressive Communications APC
I was wondering if the framework might be made even more effective by broadening the scope beyond the technicality and mechanics of the Internet traffic management to consider one of the key issues that sort of has been alluded to but I don't think really focused is, which is the relationship between the network operator and the content provider, the commercial relationship.
By this I mean everything from, we hear about network operators who are pressuring content providers to pay them for the delivery of traffic, to the sort of fast lane concept, probably the most well‑known aspects of this relationship.
Then also even to an area which kind of falls outside of this, I think particularly important for developing countries, which is the trend we see where content providers are forging special relationships with particularly the mobile providers and other providers which charge on the basis of bytes.
So now for example Facebook is freely available on many mobile phone networks in this Africa, so they are able to access the Internet but all they are really looking at is Facebook because they can't afford to look at anything else.
So that is my one question.
The other question is on the efficacy of actually relying on the regulator to be able to enforce.
My particular concern relates to recent case where Verizon took the American regulator to court and basically said, you have no jurisdiction over what I do with my network.
Those are my two questions.
Thank you.
>> LUCA BELLI: Thank you for the questions.
Is there someone that has an immediate reply?
Chris.
>> I think that in terms of the details, they should respond to some of the concerns about the content providers relationship with the ISP
I think we should bear in mind the content provider is quite often actually in terms of the large commercial content providers, is pressing for an exclusive relationship with the ISP, so of course these problems can go both ways.
It's not nasty ISPs ganging up on content providers.
It can go the other way as well.
We need to be aware of that especially with specialized services
I have a comment myself.
Luca knows this
I was worried about the way the word exists in part 4 in specialized services because I particularly think that to talk about specialized services not being to the detriment of the open Internet is not strong enough
I think there's a zero sum game sometimes being played in terms of investment, and if one is investing in specialized services, that can be to the expense of open Internet services, and that might be to the expense of open Internet services in terms of increasing speed.
In other words, I know there's a big debate in Germany about the fact that Deutsch telecom was talking about capping users initially at 384 kilobytes per second then two meg bytes per second, over a period of years.
You saw from the U.K. graph that the speed of increase of Internet services has been actually quite impressive over the years.
So I think if people are investing in specialized services, they should also have a requirement to invest in the continued improvement of open Internet services.
Otherwise we get this dirt road problem you were talking about.
So I haven't even answered your question properly.
I wanted to make that point.
Forgive me for abusing process.
>> LUCA BELLI: Thank you, Chris, for your answer.
Anyone else on the panel want to reply?
>> Just on your first point to say that I completely agree. These questions of pricing and what kinds of commercial relationships contract providers are entering into, especially with mobile networks, are a key concept we should all be thinking about. The concept of zero rating services, not thinking about caps on mobile devices, we are seeing from major companies like Facebook and Google, and Wikipedia has a project like this as withel
I think there's going to be some difficult discussions over when do, when a potential, what might seem like a network neutrality violation is being done in the service of greater access to information, you know, what is the right response there
I think that would be a really interesting conversation to keep moving with.
Then on your point about Verizon's challenge to the open Internet order in the United States, think this kind of picks up on what Eduardo was asking and Johi's comments about the need for a strong framework supporting network neutrality. The fight in the U.S. is all about does the communications commission have the authority to treat Verizon essentially under a common carriage rule, and that is something they are pushing back against
I would height light Verizon is further challenging the order and the concept of a nondiscrimination requirement as violation of its own first amendment free expression rights.
Regardless of what legal framework we have around network neutrality, we're going to see the ISPs, at least some of the networks, pushing back on essentially every front they can and arguing that, as Verizon has been arguing, they are like a newspaper editor, they have the right to decide for users what kind of content they go
I think we will see a whole range of arguments brought up by the network operators against nondiscrimination principles.
>> I agree with everything that was said.
But I would add that just to rif off something mike said, it might be beneficial to put something into the model framework around against sending party pay and to ward against that. Think that is something we are seeing pop up in a variety of places.
To the point of whether regulators have actual power to enforce and so forth, I think adding a point on remedy into section 8 might be useful as well under the U.N. guidance on human rights, corporations do have responsibility as well as governments to provide access to remedy, and providing that right of action might be quite useful.
>> First the question is whether this little booklet is wonderful, the organizationers of the meeting has been fantastic in putting up in a book all the articles.
Are they going to be on line?
>> Yes.
>> Okay. Just for the French speaking people, we have put the French version of my article on the open dashboard.EU.
>> Wonderful. Obviously, you can, as I stated already at the beginning, you can find the paper copied there.
If you don't like to read paper or if you want to share with the risk of the world, the report is at network neutrality dot info. You will find it in the events pages where the description of this event is shown.
And you can share with the rest of the world. It is a creative commons license.
And the model has been translated into Spanish thanks to Eduardo and it will be translated in French because another record with the model will be communicated, has already been but is still to be translated to the council of Europe so that the model could be considered and hopefully maybe recommended by the council of Europe
I would like to remind that the coalition has been instrumenting to develop this model and this record in an open, transparent collaborative fashion, but the whole process has started thanks to the con sill of Europe and will keep on with the council of Europe.
Other questions.
Okay, I hope you will be merficul and stay five minutes more.
Thanks.
.
>> Yes, thank you.
My name is ennesto. I come from the Norwegian telecom re regulator
I think it's very good work you have done to put together this model framework for network neutrality which can be used on a global scope. So that is very good. I think it's very good guidance to actually implement and ensure the network neutrality.
Just an observation and comment connected to the proposed regulation in the European setting, of course, from the Norwegian side, we would very much like to continue the co‑regulation on network neutrality in Norway because that works there.
But we are findful that we are part of a larger community, and of course the European internal market, and of course we also, because of the EEA agreement, we also have to sort of implement the regulation in Norway as well when it's passed through the European parliament to council.
So, but of course, we also agree what Chris mentioned also that the specialized services needs to be specified so that you can allow that to sort of, not to influence and to ensure the proper network neutrality.
Just a short observation as well.
It's very interesting also, I think, not in this room, but in the past at least, many from civil society, technical community, academia, have sort of been very much against and have problems seeing the government role in the Internet governance and regulation.
Of course now we have strong sort of wanting to have hard law regulation on the Internet.
So that is interesting as an observation.
Of course this is very clear that the government do have a role in the different aspects of Internet governance.
So I think this is a very good example of sort of identifying the different roles of the different stakeholders in the Internet governance framework.
So thank you.
>> LUCA BELLI: Thanks a lot. Thanks also for highlighting that governments have a fundamental role, especially in protecting rights, and should be first starting process and involving other stakeholders to have efficient process
I think, sorry, Anna was waiting.
Sorry.
Thanks, Chris.
>> I'll try to be short.
Some comments, Chris mentioned how and actually was in also it did, establishing how you need to watch out for how it balances. When you introduced specialized services, how the other lanes sort of balance so there's not a dirt road.
And I wanted to note that both the Dutch andSlovenia law and the European commission, they consider the possibility of allowing telecom authority, sorry, regulating authority, to establish a minimum service, regulating a minimum services to be guaranteed.
Maybe not time for discussion, but just a reflection how that can affect the evolution of the network and the two‑lane effect.
Also there was a comment by Emma saying that some of the effects of network neutrality can be that small content providers are squeezed
I agree. But I have to also put into perspective that there are many other things going on right now that squeeze small content providers.
So we would have to put that into perspective an see how much more is it squeezing.
Also even consider if it could actually be a means of empowering some small content providers
I know this is maybe more complex than current time allows.
And just a third point.
Empowering users. That sounds great. Most users do not want to be so informed. They do not want to make so many choices. And they just don't want to control so much stuff. They want their device to work and their information to get to their device.
So we have to watch out how much weight we put on the user in terms of workload.
>> LUCA BELLI: I will just take one last question, then some final remarks, because we are really up, I think, time‑wise.
>> Just my comment, observation, rather than, on Christian, we have just to speak of neutrality in terms of rule making, regardless of whether it is hard law or soft law.
From my experience first, I introduce myself, I'm from peace net Korea, so for last couple of years we have grappled with Korean government to make some public policy for network neutrality.
Quite frankly, my feeling is the major debate point of network neutrality was how we can define the scope of discrimination.
We are talking about discrimination, but in many different languages we are using, we are seeing discrimination sometimes is thought in a measurement or control, or estimate management.
So in many different languages we are talking about discrimination.
The real issue, the issue is how can we define the whole scope of those discrimination or measurement or management.
The central concern of the network neutrality debate, in our society the main topic was how to define the reasonable scope, reasonable traffic management.
>> LUCA BELLI: Thanks, sorry, I have to.
>> I will finish.
>> LUCA BELLI: Thanks a lot.
>> So a couple of seconds.
>>.
>> LUCA BELLI: You have two seconds to wrap up.
>> Okay.
So my point is network neutrality debate is concerned with some kind of a power game among interest groups.
So it should reflect those power games.
So I think the fundamentally important power tools for end users is some relations with rights aspects of end users, whether human rights or communication rights.
Any how, some kind of rights should be clearly defined in terms of end users' rights.
That is my point.
>> LUCA BELLI: Thanks a lot for your comments
I would like to ask to the members of the panel to provide some, maybe an answer to Anna's questions with some final remarks so that we can all go and enjoy some coffee.
>> I have talked too much.
We have ten minutes to get out of the room.
Just to say my presentation is on the website.
Anna and I can always talk, have a coffee over drink
I quite agree the language of network neutrality is not always conducive to user rights.
One of the good things about the Norwegian model is user rights are very central because it's less formalized in some ways.
On the other handle, we know Norway has maybe almost a uniquely benign atmosphere to have made this agreement in.
So I think that we probably have to have this tedious legal language which looks less like user rights and more like this wonderful framework.
So yeah, we need to work on it.
>> Very quickly, a call for interdisciplinary around this issue.
We are I think I good example on how like perspectives can compliment and this can help to highlight issues we are all familiar with as legal, technical, social scholars, reappearing in forums on this issuing and at the same time learn from the past to address them.
>> One question about the characteristics of minimum QS or maximum QS.
Typically the ITU has defined a recommendation for QS from telecommunication networks which suddenly is now a bit obsolete, not obsolete, but does not include the IP part of the Internet.
And that suddenly is a gap to close.
But it's normally this institution that the QS are defined.
About discrimination, the problem in discrimination is mostly in the eye of the beholder, and it is difficult to determine the very specific figures about that.
But does exist, but that is a societal problem.
>> Thanks, just to agree that I think the questions you raise are really interesting and much too complex for the time we have.
But I thank everybody for the wonderful conversations we have had and just emphasize how important it is to, when we are talking about these issues, it's easy to get bogged down in telecom policy and thinking about giant companies and networks pitted against each other.
The importance of really elevating users' free expression rights in this debate going forward.
>> I think you raise a lot of important questions. I would say I think we should have users have choice and power to make some decisions. Whether some might not be able to exercise or realize the full potential, we should put that decision in their hands and not make the for them
I think we have seen countless sort of studies, tools, demonstrating that this is a discriminatory practices are happening.
We have seen that models that fall short of hard regulation are failing to keep out. Even when we do have hard regulation, often it's not being enforced
I think we really need strong, binding regulation with strong enforcement capabilities to defend sort of network neutrality and prevent discrimination.
>> LUCA BELLI: Wonderful.
Thanks a lot.
Thanks to everyone for having had the strength of being here on Friday morning.
Without maybe coffee.
Thanks, I would like also to thank the council of Europe that has initiated this process
I would like to thank all the panel because you have been just wonderful.
Thanks to me.
(Chuckles).
>> (Applause).
>> Okay, last but not least, I also would like to thank Microsoft for the printing of this book because if you have a good wonderful printed version, thanks to Microsoft.
Thanks a lot.
And have a wonderful last day of conference.
(Session ended at 11:53).
********
This text is being provided in a rough draft format. Communication Access Realtime Translation (CART) is provided in order to facilitate communication accessibility and may not be a totally verbatim record of the proceedings.
********