IGF 2018 - Day 1 - Salle VII - WS428 Spectrum for Community Networks: A "Must" That Is Hard to Get

The following are the outputs of the real-time captioning taken during the Thirteenth Annual Meeting of the Internet Governance Forum (IGF) in Paris, France, from 12 to 14 November 2018. 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 to understanding the proceedings at the event, but should not be treated as an authoritative record. 



    >> CARLOS REY MORENO: Hello, everyone.  Thank you very much for coming, and actually, it's very interesting that a question that was posed by Paul at the end of the previous session from the Dynamic Coalition on Community Connectivity is going to be the answer -- the answer is going to be somehow discussed in this whole session on a paper -- the decisions we are discussing, spectrum as a resource that community networks and small-scale operators are requiring to grow, to scale, to establish networks.  And it is based on some work that the Internet Society, together with the others, has supported.  And then we have a set of panelists who are here that will give their opinion from the paper.  I believe most of you have read it.  It was circulated within the Internet Society list for quite some time.  We received quite a lot of comments that were incorporated in the final version.  We expect it to be released at the end of the month.  We can send it to anyone who is interested in the current version just now.  And after that, first comments from the panelists, we are going to questions and answers, we will have some prepared, and Jane has done a good job creating the framework for the session.  And then open to all of you to any questions that you may have. 

    And except for Jane, which is one of the moderators, not necessarily a panelist, it looks like a man-in to me, and I feel extremely ashamed for that.  We have tried.  We have received three consultations in the last week from women that were going to participate in this panel, but I think it's important to recognize and acknowledge that there are not that many women in these spaces, and that we -- yeah, we should be reflective about that and what is the impact that it has on the discussions. 

    Anyway, over to you.

    >> JANE COFFIN: So I think before introducing the panelists -- well, I will introduce the panelists.  Steve Song from NSRC and also a Mozilla Fellow is one of the core authors of the paper that Carlos just mentioned.  The paper is up on our website, and we will try and circulate that link on Twitter a little later and happily send it to anyone that would like it here. 

    The next is Moctar Yedaly, who is with the African Union -- and each of you can quickly say more about who you are because I am going to go fast through this.  Moctar has been a great partner of Internet Society's, both on Internet exchange points core and in Internet infrastructure, and has recently helped us host a very big meeting in Addis with other regional and regulatory entities with APC and others to talk about the importance of community networks. 

    To my right is Augustin Garzon from the Enacom, which is a regulator in Argentina.  He will talk more about what he does in a sec. 

    Deepak, whose last name I always make a mistake on, so he is going to tell you his last name.  Deepak is with Symantec but is here with IEEE. 

    Erick Huerta is with Rades Communica, a key organization in Mexico that's been helping move the ball forward with regulation, policy, development, and works closely with Peter who is here. 

    Bob Pepper, who many of you know, probably doesn't need much introduction, but he is an old friend from Washington, DC; he has been with Cisco, the FCC, and currently is with Facebook. 

    So without more ado, I am going to turn over to Steve, and I just should say we are being live streamed, so please don't talk as fast as I am.  And I do want to address Carlos's point.  We did try to have more women.  We had some cancellations.  There are some fabulous people we are working with around the world, but we do need more people to understand spectrum as well, which is part of the objective of the paper. 

    Steve, over to you. 

    >> STEPHEN SONG: Thank you. 

    So I am going to try and explain why everyone should care more about spectrum, and it's a difficult one because mostly when people start talking about spectrum policy, you see eyes roll up into people's heads.  So I think the underlying premise here, why this is important, is that access has simply become too valuable to exclude anybody.  Right?  We've talked about strategies, broadband strategies, 80% coverage, but now these devices are so useful, socially, economically, culturally, and they get better every day.  That simply by standing still, those without access are being left further and further behind.  So we need strategies to connect everyone.  And the evidence is growing now that current access models, traditional sort of monolithic operators, are going to cover a large percent of the population, but they are not going to connect everyone.  So alternatives are needed.  And the two keys to access, one is fiber and backhaul infrastructure.  That's another conference.  But the other one is access to wireless spectrum.  And that is emerging as a greater and greater problem in terms of access because it has become more valuable.  And as it has become more valuable, it's become more expensive.  As the model for allocating spectrum in the world has been sort of traditional real estate model.  You know?  You put it up for sale.  And now, I mean, the price is being paid for spectrum for long-term licenses, have the unfortunate side effect of placing a giant wall around the market.

So you have virtually no market permeability, especially in emerging markets.  So you know, if there is a $50 million barrier to entering a market to doing anything, suddenly you have a serious problem.  And the paper that we wrote was an attempt to really sort of lay out some of the innovations that are addressing those challenges of the current sort of dominant model of how spectrum is made available and looks into innovations ranging from shared spectrum strategies to dynamic spectrum to spectrum set-asides and tries to paint a picture of a more holistic regulatory framework for making spectrum accessible to all. 

    So with that, I'll stop there and hand over. 

    >> JANE COFFIN: Thank you very much.  That was Steve Song, who is a Mozilla Fellow and also with NSRC.  Steve is author and also creator of village telco, so he is a person to talk to about starting from the ground up as well. 

    Moctar, the paper that Steve just described we've put out, describes the complications from a regulatory and policy perspective.  Can you highlight for us whether it's your perspective from the paper or as someone who is critical in the African Union what you are seeing and your perspective? 

    >> MOCTAR YEDALY: Thank you very much.  Again, my name is Moctar, and within Information Society within the African Union Commission.  I must say the issue of spectrum is one of the complicated things, particularly for those who pretend to be engineers, like me. 

    Now, one of the -- I read the paper very well, and I actually -- it's a wonderful paper.  My first comment on it is I do agree with you on the fact that first of all, how you set the issue is somehow to frame it was actually very well framed for the need of spectrum and communication.  Also the pricing issue, all that.  And I agree with you the fact that it should be made accessible to everybody.  Not everyone understands what is really spectrum.  The function, what we call the location of the spectrum, which is a function of space, time, and services with implication of the different type of modulations and so on and access is really important for everybody to understand what it is all about. 

    Having said that, from technical point of view, the issue of transparency in allocation, the fact that the inappropriate rules of awarding the spectrum to different services and so on, the unpredictability because sometimes most of the regulators stay asleep on the fact that they said rumor that they would be allocating some part of the spectrum, but they sleep on it for five, six years, and the operators are waiting to see how it will happen.  The fact that the annual fees are not set really properly.  The combination between do I have to squeeze the maximum possible money from the spectrum and forgetting about the fact that could, as you said, really stop services to be developed.  The renewal -- because sometimes people get allocation, and they wait years for it to be authored.  Those are things that should be highlighted on this document.  And I will later on make some specific recommendations.  Thank you. 

    >> JANE COFFIN: Thank you, Moctar. 

    Bob Pepper, I am going to come over to you.  You are a regulator.  And having taken a look at the paper, what else would you suggest about set-asides for rural, remote, and underserved areas with respect to spectrum? 

    >> ROBERT PEPPER: So first of all, I just want to -- one of the real spectrum experts here is Dale Hatfield, who is in the back.  I didn't realize Dale was going to be here.  So this is great.  Right? 

    And also, what's terrific is that there's standing room only in a room talking about spectrum.  Spectrum is cool. 


    Right?  And so Steve, you know, it's like -- for people here, eyes don't roll.  This is like the cool -- these are the cool kids.  Right? 


    Because without spectrum, nothing is going to happen.  And you are absolutely right.  There are these two gating factors, to be able to have mobile broadband everywhere -- and that's how the world is going to be connected.  How do we connect the 3.5 billion people not connected.  There's 3.5 people connected.  It's going to be radio to the device.  Right?  So this is absolutely essential. 

    You know, I used to joke when I was at the FCC doing spectrum policy, I said there's three essential aspects of spectrum policy.  One is flexibility.  The second is flexibility.  And the third is flexibility. 


    And what I meant by that was that, number one, if you genuinely want to be technology neutral -- which you need to be -- and this goes to the point about crazy excess payments -- you know, don't have a license for 2G and then require I have to go back to mother may I to ask permission or get additional spectrum for 3G and then go back for 4G.  It's a spectrum license.  Right?  And you know, if you are not technology neutral if you say to the operators -- and by the way, this creates artificial scarcity, which increases the cost, increases the price of spectrum.  So if you say to operators, yes, you have a 2G license but you are not permitted to use that and innovate to deploy the next-generation technology, and you have to go back and pay government a toll fee before you can do that, that's the wrong incentive structure.  Right?  So that's number one.  Number two, the -- that's the flexibility.  The second is flexibility of the service and what you can do with it.  And then third is the flexibility to be able to transfer licenses to actually have the market correction so that if, in fact, you want to exit the business, you have the ability to do that and you can have a restructuring. 

    Now, there's a third piece of this way I would argue is now what we've learned is maybe a fourth flexibility, which is the whole notion of dynamic spectrum and dynamic spectrum access.  One of the things that we've learned is that there's spectrum that's been allocated.  Right?  But it's not been assigned.  And so what we really need is the ability to have opportunistic use of spectrum that is not being used.  Right?  And then again, because what we should be doing is having a goal of more spectrum, driving the unit cost down.  Do not create artificial scarcity.  Actually, we need more spectrum out there to be used. 

    That's really sort of -- most of what I have just said is related to licensed, but we have to have, in addition, much more unlicensed spectrum, unlicensed exempt spectrum, for all kinds of new services.  So some of the things that we are working on and that we have been advocating for is next-generation wi-fi at 6 gigahertz.  Right?  The FCC is opening up -- just opened up a big proceeding to make available a lot more spectrum.  Again, eliminate scarcity.  At the very high frequencies, it's 60 gigahertz, the V band.  For small cell meshing, again, most countries and parts of the world have that as unlicensed.  Unfortunately, some countries, including India, have it as licensed for part of it, and that's preventing the spectrum from being used.  And then also, some of the things that Steve, you know, we have worked on together, is open cellular and using spectrum for -- because what you want to do is take advantage of existing devices that are in the market and low cost, and the question is how do you opportunistically get access to spectrum where it's not being used in order to be able to have service to where it has not yet been built out by the traditional operators. 

    Last point is a lot of this is regulatory.  In Peru, the government looked at rural areas and said -- you know, they've tried all kinds of things, satellites and whatever.  It has not worked.  So they created a new legal structure called RMIO, rural mobile infrastructure operator, that essentially would be a wholesale radio access network that any operator could then attach to at the edge with their own, you know, VTSs or spectrum or to use that infrastructure, and that's another way of sharing spectrum in a different way, and we are partnering with Telefonica in that wholesale radio access network that will enable multiple operators to serve parts of rural Peru that today are not served.  So multiple things, but all goes back to spectrum, and spectrum is cool. 


    >> JANE COFFIN: I am going to jump over to Augustin.  The paper that Steve, Pepper, and others have just sort of talked broadly about, does discuss the important role of government.  You are with government.  Can you explain what you've done recent any in Argentina to try and help community networks? 

    >> AUGUSTIN GARZON: Yes.  Well, first, sorry for my bad English.  I will try. 

    I agree with what others say and what they say about be flexible.  I like to think that we are open mind to get Internet to every part of the country.  We have a very big country, number in side, number 32 in population, so it's very difficult to get Internet to everybody.  So we work with services to get satellite Internet in some places.  A few months ago we made new relation for community networks so they can get a free license to start operating.  And I agree that the spectrum, with the price, you have a difficult for the small operators.  So that's a big issue to resolve.  But again, we think you have to have as a government all kinds of solutions to get to everyone.  You can get to satellite -- get everywhere with satellite Internet.  You can go everywhere with community networks.  You have to be open and get each place with the best alternative to connect everybody, but I agree that every year it's more difficult to get connected because all this governance, the location, every year it's more important to be connected, so it's very bad if you are not connected. 

    But that's a part of our work in Argentina. 

    >> JANE COFFIN: Excellent.  Thank you very much. 

    I am going to jump over to Erick Huerta now, who has had a great deal of experience dealing in Mexico.  What Pepper was just talking about, flexibility, issues related to 2G, can you use that same license to go up to 3?  Over to you. 

    >> ERICK HUERTA VELAZQUEZ: Well, I found the study great, I liked it a lot.  I think I really recommend it because it's simple.  No?  It talks about a complex thing, but it's very simple in the way it's done. 

    I just want to highlight some things on this issue.  So you have to see that the traditional way that spectrum has been regulated was not based on the nature of spectrum.  The nature of spectrum is flexible.  It was regulated about the nature of the equipment that was in those days those equipment was very hard to model it.  So you have to give them -- guarantee them some space so that they could work and not interfere with each other.  But that's not because of the spectrum; that is because of the equipment. 

    I see all the time that people, when start talking about spectrum, start talking about scarcity of spectrum.  Spectrum is a scarce resource.  Always when it comes to speech about spectrum, someone says it's scarcity.  So please, the next time you see someone saying that, say no, it's not.  Spectrum is not a scarce resource.  Spectrum can withstand a very high number of combinations between time and place without -- with those combinations could be really flexible.  It's not actually scarce. 

    So what is really important in spectrum is to use it.  It's to be used sufficiently.  So how do we -- how can we regulate spectrum so that it could be efficient, have efficient use of it?  It's to allocate this thing of time and place regarding the nature of the service and the technology available for that.  So if we take into account this nature of the service and the technology that is behind and its use, then we can come to many possibilities, and I find there are three ways that we could, regarding the spectrum, to organize its use.  One is like a schedule.  If we think in roads, if we are going to use that roads for races, we need some exclusivity because if someone crosses the road, they will be run over.  So we certainly have to say, okay, you will use it from this at the same time to this time in this certain place in an exclusive manner because someone may crash.  But if we are going to be transit and people will be running that, maybe we will need a signal, maybe we need a red light, a green light, and people can pass.  But if there is a path and there's not many cars crossing and that, maybe just we need some rules of behavior, some politeness, and say ok, just when you see a person crossing the street, stop, let the person cross, and that's it. 

    So with these three criteria, I think we can use spectrum.  No?  For instance, if there is a broadcasting there in a certain area, there is a broadcaster there, will use the spectrum and will use it in exclusivity, but what is a link or a TP link?  It doesn't go like everywhere, but it goes in just one direction, so we can put many persons using that at different times.  Or wi-fi, it's just a protocol, say okay, you go first, and they organize themselves. 

    So I think this is the challenge for the regulators.  Try to start regulating different ways spectrum is regulated.  It's no more -- if we do it in the same way we have been doing that, we are avoiding people to get connected, and the important thing of the spectrum is to be used. 

    >> CARLOS REY MORENO: Okay.  They have a question for the panel.  Yes, it's very interesting.  I agree, everyone needs to use spectrum.  That's the next thing that we need to actually connect the unconnected.  Yet most of us have been in different countries, and maybe I am going to put the two of you on the sport here from a more government and policy perspective to get more spectrum.  In the case of South Africa, we have been advocating for six years.  Yes, yes, we want to do this or that.  It doesn't get concrete.  Like this paper is out, we are discussing things.  It will die out.  What are the steps that are required?  We participate in every public consultation.  Are involved in every process.  We engage with regulators in training.  What is missing there?  If everyone agrees, what do we need to do?  Because it seems obvious.  What is the mathematic recipe for making this happen? 

    >> MOCTAR YEDALY: Is that question to me? 

    Well, there is no mathematical equation.  It's just a matter of I think traditionally, as everybody sees, specifically in the developing world, is spectrum is a source of revenue.  That simple.  And you give it to the highest bidder.  And that is exactly what is stopping things to happen.  Therefore, in order to just avoid these kind of things, we need to review our strategies for universal access. 

    If you go to the example of Africa, for instance, handful is connected Africa, 40%.  However, 60% of the population is still in the rural areas and they don't have access.  So it's very easy to say that probably connectivity is provided to the people who are living in urban areas rather than the rural areas.  It's very easy to do that. 

    The concern of our intergovernmental organization is to provide access to everyone.  This is a holistic approach.  It is not about spectrum itself.  It is about all infrastructure.  First (?) then utilization, which is actually provide access available, and then, of course, freedom, democracy to everybody. 

    We have come to the conclusion of the way of thinking, thinking that the government and the top people are the ones who have to provide access, is obsolete now.  It doesn't work.  The Universal access fund didn't work.  We need to go to a different approach, and community access is very important.  We in the African Union want to support (?) in the premise of the African Union, ISOC chapters meeting and discussing the community issues and shutdowns.  I beg you. 

    We plan and we are working with the community to make sure that that declaration will be adopted as we adopted the Declaration of Internet Governance and digital economy, thanks to the community, and we will do that.  It's not really now the typical government approach of top-down, vertical approach, so on, but rather involving the community at the base to make sure we are providing access to everyone.  And we will be will be to work with Facebook -- and Pepper, I am making you an offer now -- for those who are coming with original ideas of access to everybody and, therefore, the spectrum will have to go with that kind of thing because if new approach for access is coming, new approach for spectrum, and I think Steve has mentioned very well we need to have a new thinking of how to effectively manage spectrum. 

    >> JANE COFFIN: I would also say that if the return on investment -- if the spectrum costs are so high and the return on investment is that still half the population of the world isn't connected, we know from GSMA that for some of the current mobile operators that for communities of 5 to 3,000 and under, there's no return on investment for traditional operators.  So we are looking at something where we are talking about like a social investment pact where we start as companies, government, civil society, technical community, companies like Deepak's at Symantec, the governments, to say okay, if this amount of spectrum could be set aside for rural, remote, and urban underserved, can that help?

    Now recently, you changed a regulation, Augustin, in Argentina.  How did you do that?  Because it's really important to get to the how.  And I am going to ask Steve and Deepak the how in a minute because Steve has some interesting ideas on new regulatory focus because the current regime, I am just going to say this -- and Raul, close your ears -- it's not working.  Let's be honest.  I have been doing this for 20 years.  Or more maybe.  But the current regulatory models are just not helping get to the indigenous communities -- I am looking at Matt when I say this.  Ucha knows, who is from Georgia, who sitting behind me, and the project we have been working on with Georgia, The Government of Georgia is putting money in from an economic perspective, from the Economic Development Fund.  So we have a regulator and others in Argentina that are making those changes.  So how did you do it? 

    >> Our success is not to collect money, it's to connect people.  That's the issue to think about it.  Nobody will congratulate us to collect a lot of money for the spectrum, but also we know that it's very expensive because the revenue that is all in the big cities. 

    Recently we are making proof with white spaces, and recently we put the 400 band to rural areas.  We have (?) asking for more than 2,000 small towns to use the spectrum.  It would be like a contest between them, but it's going to be very cheap.  And we are starting the ecosystem, but we think it's a good solution.  Just use the spectrum to collect rural areas. 

    >> JANE COFFIN: Deepak, as a company, I know you are not a mobile company, but also at IEEE, what would you say?  Rural India needs connectivity.  How would you talk to the government about spectrum policy? 

    >> DEEPAK MAHESHWARI: Number one, we should realize, as mentioned with regards to resource, the fact that spectrum is perhaps the only -- and I am repeating it -- the only natural resource that every country has been endowed with the same amount.  So somebody's got more land, somebody's got more air, more sunlight, et cetera, but this is the only source that every country has in the same amount.  Okay?  That's one thing. 

    The second thing is when you are looking at spectrum, like Erick just mentioned, yes, you may need some lanes and other things, but the fact is you also need some foot paths, you also need some common areas where nobody needs anything.  When I was in school, about 40 years back in India, you need a token for bicycles.  Okay?  There used to be a license even for radio cells, receiver cells, and that was also the case in the U.S. back, those licenses have slowly gone away. 

    17 years back, I wrote in 2001 about wi-fi and why do we need license-free spectrum in India?  People were upset about it.  Wi-fi, of course, came back in 1998.  One of the things about wi-fi was that it uses frequency hopping technology, and at that time in India, frequency hopping technology itself was not allowed.  So there are different issues, whether you need a license, whether you have to pay, et cetera, but frequency hopping technology was not allowed.  The reason was reportedly around security. 

    So it took a lot of time for us to come around that, and it didn't happen in one single shot.  So in 2003, the first regulation came which said that you could use these things in low power only in building, so for example, within this building, or only in a particular campus, let's say a particular university or something like that.  Then 2004 we got a little bit more flexibility, and it became only in 2005 that it was fully de-licensed in the 2.4 band, ISM bad.

    Later, of course, government has also de-licensed many other bands for RFID and even in 5G now, de-licensing and of course, as just mentioned, there's a need for many more bands.  That's something which can progress in many cases, but challenges are allocation, no assignment, assignment, no usage, or very, very low usage.  So that's why you do need more flexibility to leverage many of these types of technologies, and especially with the online databases currently that we have, we should be able to do many of these types of things, like white spaces.  For example, I talked about the white spaces almost a decade back.  So that's one area around these things that many times the apprehensions are so well set in the mind-set that people are not open and willing.  That's why it is important for them to show some small success.  Especially I would say start with maybe a mesh network maybe with a controlled environment, where it is one campus, so they know who the users are, et cetera.  Then slowly look at how do you chip away these regulations in phases one by one. 

    The other thing is we are right now at UNESCO, and just very close to this is Eiffel Tower.  Now Eiffel Tower, of course, opened on 6 May 1899.  But on 5 November 1898, 120 years back and a week back, the maker of Eiffel Tower actually did a wireless transmission from there to Pantheon.  And That is one of the ways that the Eiffel Tower is still there because originally it was licensed only for 20 years.  Bus it started using wireless, got extended, and then of course, in 1910, the city administration extended for another 70 years.  Originally it was supposed to be (?) and was to be struck down, and artists in Paris were to do that.  But the important thing is that we need to look at multiple uses and different ways of doing things. 

    Now, as Jane just mentioned, on one hand, India is the second largest company in terms of Internet users.  But number one when it comes to the number of non-Internet users in the world.  That's also a fact.  You can say many things about India, whatever you say, even exactly opposite is also could be true. 


    The last thing that I want to say is this.  When we are talking about unlicensed spectrum, and as users, as participants in this whole ecosystem, we must look at the three cardinal principles that were enunciated way back when the Part 15 rules came with FCC.  Those are nonexclusivity, noninterference, and nonprotection.  What it means is actually this.  Suppose we are all walking in the corridor.  Nonexclusivity is not only I can walk; everybody else can walk and everybody has that right.  That's not the nonexclusivity. 

    The second is noninterference.  If I am walking, I should not be trying to disturb others who are also using the same space. 

    The third is nonprotection.  While we are walking, yes, it may happen that there is a crowd in between, I have to stop for some time.  It may happen that somebody may get a little bit of speed will come back or may get a little physical blockage or things like that.  We have to sustain, absorb, adopt, respect all those things.  These three values we say are the core of any unlicensed users, but I will say any civil users in terms of civility, as a matter of civility as human beings, I am saying, in terms of how we interact with each other. 

    So for example, if we have a wi-fi router, it doesn't mean we have to transmit (?) if that's what is allowed.  If I can do it at 16 DBM, that's okay.  I should do it at 16 so there is opportunity for others to also use these things.  That's where it is very important for people to see that not necessarily you have to transmit at the highest power which has been permitted by the regulators, not only that you have to use all the bands that have permitted by the regulators.  Try to do with as little power, with as little spectrum that we can so more and more people can use.  I think that's the way for us to connect more and more people going forward. 

    And I, as Jane mentioned, my day job is at as I man particular, a cyber security company.  My night job is IEEE because most of the calls and meetings are in the nighttime for me.  I am global chair of the IEEE initiative.  All the wi-fi standards, so 802.11, A, B, G, N, AX, AY, and all these alphabet soup, they are all based on IEEE standards.  So we continue to work on those areas, and we do need to do these type of things and also look at how do we use these things in a more secure manner so that people also have security and trust in these things. 

    Thank you. 

    >> JANE COFFIN: I noticed you and Carlos and Mike -- I don't know if Mike is here now -- when they were doing research on this paper, I got a ping from Carlos who said it's impossible to find anything about spectrum on the website.  So -- and I used to work at an agency -- I didn't do spectrum work.  I would never claim that, not in a cold second.  But I used to work with a lot of spectrum engineers.  There are a lot of -- 130 at least spectrum engineers in one agency doing government spectrum-related work.  So they are super focused on the work.  It's hard to know exactly what they are doing sometimes.  Right?  If you are a public person, you are a community network, trying to figure out how you get access.  In the U.S., it would be the FCC, not the only agency.  But Steve, what would you say about -- or Carlos -- how do normal community networks who are volunteer people often on occasion get access to data?  What would be your recommendations about what you would say to government, like how to help them sort this out? 

    >> STEPHEN SONG: My recommendation would be to follow a well-established tradition now of transparency and open data in government.  The telecom sector is desperately overdue for transparency in not just spectrum assignments, but where fiber networks are.  Because if you are building a community network, the first thing you want to do is find access to a fiber network.  And indeed, where towers are.  Because the mobile network operators claim coverage that is sometimes based on their desire to negotiate good roaming agreements as opposed to truth in advertising.  And so the ability to actually validate claims as we try and connect everyone.  So those three things could be part of an open data policy that could easily just bring telecom infrastructure into broader overall open data. 

    Can I carry on? 

    >> JANE COFFIN: Absolutely. 

    >> STEPHEN SONG: I wanted to address a question that was raised in the last session, a gentleman from India was asking why don't community networks scale?  And I hope to explode that myth a little bit by saying that scale is kind of like -- is the Silicon Valley end of the rainbow.  And -- (Laughter) -- and I don't believe it's necessary.  I mean, it's good in some circumstances and not in others.  And I want to illustrate it with a metaphor. 

    If you took a glass jar and you were trying to fill that jar with sort of fist-size stones, you could fit maybe three or four stones in that jar, and the stone -- the jar would roughly look full.  But if you fill that jar with water, it's still actually more than 50% water.  That is roughly the state of telecom regulation.  We regulate for these large stones, and it's never going to fill the jar.  Right?  We need regulation for these sort of pebbles and, indeed, for sand.  That's the only way we are actually going to achieve 100% connectivity for everyone. 

    So in order to achieve universal access, it's not that we need one giant community network.  We need a million community networks.  Just like we have a billion wi-fi devices.  I mean, nobody talked about building -- how do we scale, you know, to a national wi-fi -- no.  It grew much more organically.  And so I think creating regulatory frameworks that accommodate big stones, small stones, and sand is actually the strategy we need.  And indeed, this is what Carlos and Shawn and I have been pursuing in a recent submission to the South African government on the disposition now of 800 megahertz and 2.6 gigahertz spectrum there is trying to open up these opportunities for these smaller regulatory interventions that are small but can have massive, massive impact.      

    >> CARLOS REY MORENO: Okay.  I think, yes, before the other panelists actually get to speak, we would like to hear from the floor and to take some questions.  So we have four, let's say, one, two.  Five.  Okay.  So please, then go that way.

    >> AUDIENCE:  Thank you.  Picking up on Steve's first point about the wall based on spectrum licenses.  There's been a lot of proposals around things regulators should do, but the business interests of providers seem to prevail in most regulatory environments, that not many regulations are implemented that are opposed to the predominant carriers and who are able to take liberties with alternative business models.  Anything that might possibly threaten their own returns on investment they simply oppose.  Not necessarily because it will; because it just might. 

    So my question is rather than try to convince regulators, how do we convince the holders of these valuable licenses who have more success convincing regulators about policy than people who want more for everybody. 

    >> In South Africa, there is (audio distorted) I have monitored the spectrum.  I have proved that it is unused.  I have approached those operators, and they say no.  And there is nothing that the government can do.  They say no, period.  It is a policy decision to enforce what they say.  (Audio distorted).  2057.  Whatever.  Sorry. 

    >> JANE COFFIN: We just need you to use a microphone because it's being live streamed. 

    >> AUDIENCE:  Thank you.  I wanted to touch on a new problem -- actually, it's not a new problem but still ever existing.  That is cross-border interference.  One country can decide, indeed, to allocate additional spectrum for mobile broadband, but if there is a neighboring country that uses it, I don't know, for TV, broadcast, or sometimes military purposes or whatnot, it will be (audio feedback)

    >> (Off microphone)

    >> AUDIENCE: See, exactly, already, interference.  So basically, while depending on geography or let's say from countries small or big, then it can be actually a very huge problem.  We know eventually these things are sorted out and some harmonization decisions are made, but still it sometimes takes a lot of time, during which either you cannot use the spectrum or can use it only partially.  So do you have any suggestions from these kind of situations, how can we improve the cross-border coordination?  Thank you. 

    >> JANE COFFIN: I am also going to take a small prerogative.  There are small commissions that are set up over boarders.  There are commissions between countries that are often set up to handle these disputes and/or not a dispute but a happy thing.  So Steve is from Canada, and I am from the United States.  The FCC and the Canadian regulator have had 40-year-old cross-border commission -- same with Mexico -- where Erick is from.  And when the U.S. was changing from analog to digital TV, you can imagine what that did for the border areas.  Yeah?  So Mexico and Canada moved faster to change over.  But the key thing was they worked with each other for years in discussions.  So this also touched upon one of the biggest problems in the world right now, particularly in Sub-Saharan Africa, and in central Asia, is cross-border connectivity.  If you don't talk to each other, those problems are going to continue.  I have lived in two landlocked countries where this is a huge issue.  Power levels will be turned up on the border on purpose to mess around with the spectrum in the other country. 

    So the question is sometimes we need governments to come together with industry and others, our organizations, to say if you care about connectivity, which equals socioeconomic development, then you've got to find a way forward. 

    So I am going to -- who is next on the question?  But this is a really good question, and thank you for answering it. 

    >> A quick one.  From the African point of view, we have set up parameters through which these kind of interferences should be avoided.  That's one thing.  And we have the policy of harmonization of the user spectrum across the continent.  Now, the challenge is all the resources are there, this mechanism is there, but enforcement is still yet to be implemented. 

    >> (Distorted audio) one of the reasons why community network should scale is to get past situation where smaller community networks have to get the bandwidth from telecom companies.  I can understand you can also get bandwidth from IXPs, but in some cases, the situation is that you have to go to a telecom company to set up a community network.  And the other reason is that as he was talking about, the ability of telecom companies, large telecom companies, to have a voice in regulation.  When community networks scale, they do have the ability to have a voice as much as a telecom company has, so these are the two reasons. 

    I actually wanted to ask a question to Jane.  What are the countries that have exemplary policies and spectrum regulations?  Is which are the countries that have opened the spectrum space or broadened their policies?  And what countries share with other companies so that such a good practice becomes universal?  After all, there was so much talk about the revenue from spectrum being so high, and is it really high?  Is $50 billion or $60 billion a sum of difference --

    >> I am going to go with yes on that. 

    >> JANE COFFIN: I would actually suggest Dale Hatfield.  Can you raise your hand, Dale?  Dale is one of the most preeminent experts in lots of different aspects of teaching everything at the FCC.  I am going to say Dale might be a person to talk to.  I am not going to put him on the spot now.  I would say quickly some countries we know about, Argentina, Brazil, Mexico, Colombia, forget the -- South Africa, Georgia.  So there's quite a range.  Steve may be able to help you later with some more that are in the study. 

    But Dale, over to you.  You had a question. 

    >> AUDIENCE: I don't think I can directly comment on what was just being said, but I wanted to raise three points.  One, too much our work is based upon worst-case interference situations where you assume that there's going to be a disaster and it's very high probability and therefore, you tend to claim it.  That causes the spacing to go out a lot further than it should be.  And my colleague, Pierre De Vries, has been advocated risk-informed analysis, where you really look if this interference occurs, is it really going to do any harm that we worry about? 

    The second, I think, still to this day we don't remember -- everybody in this room knows we can have billions of transmitters on the air.  They don't interfere with each other.  It's only when you put the first receiver out there.  So a lot of the problem here is on the receiver side, and receiver technology has not -- well, I won't say it hasn't kept up, but we don't focus enough on receivers because the root of a lot of these problems is the receiver doesn't have adequate performance. 

    Then third is just a warning that's been concerning me recently.  I love this notion.  I have spent my whole life -- I started as a ham radio operator.  I love radio.  But we've got to remember, I think, that radio is an inherently open system.  You can't put a Faraday cage around a radio to isolate it and expect it to work.  It doesn't.  It's inherently open.  Therefore, I think we need to think more about jamming, spoofing, and those kinds of activities and the enforcement that goes with it.  Otherwise we are going to build radio networks built upon something that can be hurt pretty badly by people who are malicious actors wanting to jam or spoof what we are doing. 

    >> MODERATOR: So we have four more hands up.  We are running out of time.  I know we would like to hear something from the panelists as well.  So Raul first. 

    >> AUDIENCE: I will take over the mic.  Related to something that you asked, Carlos, and Steve touched upon it.  You say why we don't do that if everybody agrees?  I was thinking my personal experience in talking with government and when we met in Argentina for WTDC, that talking with many governments, private sector companies, many telcos, everybody was sensitive to this but in a positive way.  We agreed with the principles that we were promoting that the regulatory and framework language would be a catalyzer for connectivity and not an obstacle.  But the problem is when they get all together inside of the room and they are afraid of being true novatives.  So they are not willing to take the risk.  I think the way to solve that is what you are doing here, and it's very good to have Stephen Song from Argentina here and Moctar because they are the people that will show, together with the Government of Georgia, Mexico, and many others, they will show to the rest of the world how things can be done without hurting anybody, without hurting private sector, without hurting investment, but allowing to build an environment to connect people.  This is what we have to do, to continue.  Having good experiences and showcases experiences in order to motivate governments that private sectors that probably are afraid, that's unnecessarily afraid of innovating. 

    >> Okay.  I took some notes about things that I think are dangerous to install as truths.  One of them is this idea that incumbents or big companies always have more power regarding regulation, and why should the dark side, as some people call them, have more power?  This is one point.  One important point.  Because regulation is there to guarantee the rights of the people and to defend those that have less power in the society.  So if we are considering as a truth that those who have the power, the economic power, have the power to regulate, then we are doomed. 

    Another thing I consider not good is to say that the next 3.5 billion people will be connected through mobile networks directly to their devices.  This is not only said here, it's said in many different places, that from our perspective, that is not the Internet that we want.  We want people co-creating the Internet.  We don't want people accessing a concentrated Internet.  So that's another thing that's dangerous. 

    And third thing -- I have more, but I will finish here -- when we say that the state uses lots of income, if you don't sell spectrum, the truth is that spectrum is paid -- always paid by the users.  Yes, companies are always having the users pay that cost.  And users are not only paying the cost of spectrum, but also the financial cost of that.  Yes, because companies have big financial cost to pay for that spectrum.  So you are actually transferring all of that to the users when we shouldn't. 

    >> (Off microphone).  Can you be brief, please? 

    >> Yes, I will try. 

    When you said that the allocation of spectrum should not be combined to technology, and allocation is an allocation for 2, 3G, LTE, et cetera, this is breaking the net neutrality, that's okay.  The question is if it doesn't break neutrality when we talk or discriminate between rural and urban?  I am saying so because this has been used, at least in our case, to harm us.  We were said as long as you stay rural, do whatever you want.  But when you start getting into the urban areas and especially when we started dealing with fiber, then it got higher, higher.  And here -- and I am concluding -- we are marginalized even from a policy perspective; will we never reach the new minimum size to be sustainable, then we cannot evolve. 

    >> You took most of my points.  I was going to say I think we get stuck a lot of times in this sort of either/or, it can only be this way or it can only be that way.  I think it can be both at the same time.  Soar with doing a lot of work in Mexico.  So we have spectrum set-asides for rural areas.  We have a guaranteed right to spectrum as community networks.  And that hasn't reduced or in any way affected the investment in large networks.  Those guys are still building their networks, still doing whatever.  So it's sort of like the regulatory space is the space of capture.  That's kind of how it's always been treated.  And I don't think we need to continue strengthening that viewpoint.  We need to break that down and break it apart and figure out how other people can enter into the space.  A lot of times folks from communities say I would like to build a network, and they just can't do it.  They are not allowed.  There is no space within the regulation to even begin to do that.  So we need to start from there and say other people should have the ability to do these activities.  It's as basic as that. 

    >> JANE COFFIN: So we'd like to give each of our round table panelists a final word.  Steve, we will start with you, then go around the room to Moctar, Augustin, and end with Pepper. 

    >> STEPHEN SONG: Okay.  This is for everyone in the room and in particular anyone watching on the Internet.  The teachable moments in your countries are the moments when regulators offer consultations.  Right?  When they issue a call for input on their licensing strategy, on the regulation strategy, it's complex.  It's difficult.  It takes a lot of getting to know.  There are people in this room, we want to help you.  Right?  We want to assist you in that process.  So reach out to us.  Our names are here.  And we'll help you. 

    >> MOCTAR YEDALY: From African Union perspective, few recommendations for everybody.  Set modest prices for spectrum.  Prioritize the allocation and assignments, both of them.  Help operators, including community operators, to mitigate the risk of implementing their network.  And specifically another thing, a long-term vision for the spectrum. 

    Now we, the African Union, are starting already this month a 5-million-euro project with ITU to address this issue of a more organized method of network.  We hope all that has been said here will be factored and of that we will end up with something like that could serve all the community.  Now, with regard specifically to the community network, there is a resolution of the WTDC 17 that probably very positively we can actually help the network to move forward.  Thank you very much. 

    >> AUGUSTIN GARZON: I want to say two things.  The first one is about the spectrum.  I think that we, as government, we need to think one framework for rural areas and small cities and another to big cities with big companies.  And that's there are actually different ways to approach those issues. 

    The second one is what I said, what is said about flexibility and open to get to everywhere with the best solution, could be spectrum, to be fiber optic, satellite Internet, but in countries such as Argentina, we have to work that way to connect fast, the faster way, everybody. 

    >> DEEPAK MAHESHWARI: So three quick things.  Number one, Steve, you mentioned about database.  So India is the first country in the world where the data for 600,000-plus, every mobile tour and every antenna on tower in terms of who the operator is and which band, whether it's 2G, 3G, 4G, et cetera, that data is available on a map in a public domain.  Okay?  So that's one.  I will share the details with you later.  That's one. 

    The second thing is we need to look at the size of the wi-fi economy itself.  This year, if the wi-fi economy had been an independent country, it would be the ninth largest economy in the world, 1.96 trillion, close to Italy's GDP.  The 2023 projection is 3.47 trillion.  So it's pretty huge in that sense.  That's the second thing. 

    And the third thing is, within IEEE, as we continue to advancing technology for humanity, we are also looking at how do we bring in security and privacy by design as well as by default by making these type of decisions.  Thank you. 

    >> ERICK HUERTA VELAZQUEZ: I would comment the first question on how we don't -- how we do not fear to threaten the actual operators.  The thing is that they will always feel threatened.  No?  I work for the industry.  I started, and I remember negotiating law, certain acts from the industry.  And even in things that they were not in danger for them, they would be beneficial for them, they would oppose.  That's what I think, it's very important the decision of the government.  They have its own position, different probably from the users, different from the regulator, a point in the middle that says I know that this is going to be good for the market, it's going to be good for the competition, and I think we are living that example in Mexico.  When we get this, at the beginning we asked one of the companies to use the spectrum, they didn't want it.  Now that these regulations have come over, two of the biggest operators approached us and said how can we work with you?  Do you want us to help you in a certain way?  Because more users are always better than nonusers.  At the end, these people brings to each other, so the market gets benefit, but it's important that the regulators stand in the best benefit for the people and for the market itself.  So not to protect the industry, not to try to protect them, because at the end, if you try to protect the operators, you are acting against the industry, and that's very important, the regulator stands for the industry, stands for the market, stands for the users, and not for protecting the operators. 

    >> ROBERT PEPPER: Yeah, I completely agree with that.  In fact, it's consistent with what I was going to say, which is one of the things about spectrum policy is focusing on competition, not competitors.  So the market needs -- because empirically, what we have seen is that when markets get opened up and there are new entrants, we've seen investment, innovation.  You can just go back to all of the naysayers that said that wi-fi would never work.  Part 15 would never work.  Right?  The world would end.  Of course, that never happened.  And so it's actually about aligning.  So I think focusing on end use. 

    It goes back to the other thing I said, which is, you know, where spectrum is -- even if it's been assigned but it's not being used, that's where the new techniques for opportunistic use of dynamic sharing, right, and going back, Deepak, to your point, which is it's no protection, no interference.  Right?  There are ways to do that.  Yeah, I mean, I think that that actually is great because that also -- that begins to attack some of the scarcity things. 

    And you know, Nico, actually agree with two of your three points, but I think maybe you misunderstood what I was saying.  I used the phrase "wireless," not "mobile," which is important.  Nobody here in this room is connected through Ethernet.  Our devices are connected through wireless, but the first thing that happens when it hits the access point, it goes into a fixed network.  So I used to joke about mobile operators were talking about their mobile networks.  I would say I have no idea what you are talking about.  No idea.  In fact, there's no such thing as a mobile network.  The network does not move.  Except maybe a mobile satellite.  Right?  The reality is we are talking about wireless links into fixed, and you need really robust, good, open backhaul, right, from the point of the access point. 

    Again, I agree with you.  We don't want the world to be exclusively connected through traditional mobile operators, although they are going to do maybe 70% of that.  But what's going to discipline the market is if there's the ability for people to compete with them on the 70% and then fill in -- this is, Steve, the sand that fills in around, right, the big rocks to fill in the remaining 30%.  And so it's not an either/or.  Right?  Sometimes we get trapped in false choices.  What we need to do is use every technique possible at our disposal and all different ways -- licensed, unlicensed, opportunistic, dynamic -- to get the spectrum out there, reduce the scarcity, and get service to people.  Right?  And then afterwards we can talk to Matt because Matt's done some amazing things -- seriously -- in areas that people said would never get connectivity.  Ever.  Right?  And Native American lands in the U.S.  And for indigenous people.  So I think there's -- I am really optimistic that if we do this right we can actually close that gap. 

    >> Thank you.  Thank you very much, everyone.  I think when we first started discussion about this paper, it was about opening up a conversation that was totally closed.  I think this conversation is opening up.  Having all of you here, from panelists to audience to the people on the Internet, in discussing and thinking and considering these new options about how what is not being used efficiently can be used to better create connectivity, I think we agree.  Thank you for the support and, yeah, I think the goal has been -- I will have my magic bullet in terms of what is the next step, but I think the next step is it needs every one of us to take this conversation forward in our constituencies, in our countries, in our frameworks.  It is an opportunity.  We need to believe this is a way forward to actually act in every single window that opens for us. 

    So thank you very much for listening and for participating.