Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 1946

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 1946

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 1946

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 1946

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 1946

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/database/driver.php on line 2022

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/filesystem/path.php on line 143

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/filesystem/path.php on line 146

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/joomla/filesystem/path.php on line 149

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/components/com_fabrik/helpers/string.php on line 264

Deprecated: Array and string offset access syntax with curly braces is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/src/User/UserHelper.php on line 621

Deprecated: Function get_magic_quotes_gpc() is deprecated in /home/wgig/public_html/igf/website8/web/cms/libraries/cegcore2/gcloader.php on line 63
FINISHED - 2014 09 03 - WS169 - Technologies And Policies To Connect The Next Five Billion - Room 7
 Welcome to the United Nations | Department of Economic and Social Affairs







The following is the output of the real‑time captioning taken during the IGF 2014 Istanbul, Turkey, meetings.  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 session, but should not be treated as an authoritative record.   

>> DAVID REED:  We are going to get started.  Welcome to the workshop number 169, Technologies and Policies to Connect the Next Five Billion.  My name is David Reed.  If you are a weather watcher, we have got the right room for you.
It looks like we might have a thunderstorm serenading us as we go through the session here.
To begin, the inspiration or origins for this session was a paper actually that Jennifer Haroon, who is on the panel, Patrick Ryan and I completed.  If you want background reading on this topic, you can Google it, and it's on the SSRN paper, and it's called, Connecting Technologies and Policies for Connecting the Next Five Billion.
That is what we will be focusing on today.  In that paper, we had a couple simple, easy to say but difficult to research, research questions, trying to answer the question:  What technologies will be used and what are the most needed policies to remove deployment barriers in unserved areas of Internet access?
As we went through and conducted the research, we quickly zeroed in on, given the current state of deployment of the Internet on a global basis, that we zeroed in on the local access network in particular as a major remaining barrier to extending broadband to the next five billion individuals without the Internet.
This finding stemmed from several aspects.  From a technology perspective, there was a lack of a low cost network solution in the last mile.  Bandwidth is not for free, and so trying to identify the lowest cost solution was a challenge.  There was also a lack of electricity in many of the unserved areas.
That is another significant barrier.
We also identified lots of policy barriers, such as restricted access or entry into the local access market itself, often due to incumbents, difficulty to pursue infrastructure sharing in order to reduce the cost, and a lack of Government leadership overall in order to move forward.
To address the uncertainties in the paper, we rolled out a concept that borrowed from the private sector, and that is road mapping.  We developed broadband roadmaps conceptually at a high level to describe the direction that we would see over the next five to ten years for the deployment of additional infrastructure in the local loop.
What the broadband roadmaps allow you to do is to classify the best fit of some different last mile technology options by cost and the speed for access that you are targeting or trying to deliver.
Some insights from that exercise and developing conceptually the broadband roadmaps, number one is that wireless was clearly the most attractive option over wireline due to lower fixed cost associated with the deployment of those networks.
That is number one.  Number two, from the core going out to the edge, you expect the gradual migration of fiber as the specific characteristics of the local conditions would permit it.
You have core being deployed more and more, you have fiber being deployed more and more in the core and it's making its way out to the edge to a wireless type of distribution network.
On the policy front, we proposed or described a five‑point policy approach to help enable the deployment of broadband roadmaps, despite these challenging business conditions due to the cost situation.
The first was the need for shared infrastructure, particularly with power.  And an interesting aspect is that, at least looked at on a time dimension of decades, there is almost a flipping that is going on in that now Telecom is the driver for electricity and not vice versa.  It wasn't, you don't deploy electricity and then get Telecom.  But it's the need and attractiveness of the Telecom services that inspire the installation of electricity in some of these areas.  Spectrum policy was another issue to the extent it depends on wireless technology.
There is a need for much better spectrum management.  In particular there is a concern that a lot of spectrum that is there is administratively occupied; that is, somebody has the license, but they just don't use it or it's focused in the urban areas and not in unserved areas, and the need for more spectrum in those areas as well.
A third area of policy is the access and interconnection, the role and conditions of IXPs, the Internet exchange points to grow the network is a critical parameter.
Another critical parameter is innovation and stimulation for demand for broadband, very important that you tap into the entrepreneurship, spirit of entrepreneurship in order to move this forward, often accomplished through liberalized markets.  You need broadband app development or content that is deployed.  Demand aggregation might be a key point in order to foster, in order to in advance try to get subscribers that are sufficient to make and justify the deployment of the network, and subsidies like we see in the smart phones are another important element.
Finally there is a need for collaboration, multistakeholder organisations, Government leadership, like the broadband roadmap or broadband plans with specific roadmaps are what are needed.
In this paper, we laid out what we hope is a framework that can help spawn and identify issues to debate down the road.  And in true IGF fashion we hope that the IGF can help to identify and create its own policy plan five point, six point, three point, whatever it is.
But that is the contribution that helps to provide a theoretical underpinning for this session.  We are very fortunate to have just a tremendous panel here, and with a number of individuals who have literally been working on these types of issues for decades.
The layout for the panel that we are going to be doing, each one of the panelists are going to have an opening statement from the three, four minutes, maximum of five, where they are going to discuss a topic of their choice.  Then we will have a set of questions that to walk through associated with the issues of connecting the next five billion.
We will be happy obviously to interrupt at any point in time after the individual presentations with questions from the audience.  If you can be thinking about those, we certainly want audience participation.
What we are going to do is start on the far end, with our first panelist talking about a particular topic of interest from their perspective.  It's Salam Yamout, who is the national ICT strategy coordinator for the Presidency of the Council of Ministers, the Government of Lebanon.  I'll let folks know what the panelists' academic pedigree is since I'm coming from academia here.
I think it's important to point that out.  Salam has a MS in electrical and computer engineering from the University of Arizona and MBA from a university in Paris.  I'll let you kick us off.
>> SALAM YAMOUT:  Thank you, David.  Hello, everyone.  I want to start by sharing my experience of a policymaking.  You have to remember the people who are now doing policies in most of the worlds are people that have got a law degree 40 years ago.  Technology for them is not something they understand, they like, and they are very much afraid.
The overall tendency that I found is as soon as there is a problem, there is a tendency to overregulate, to jump and protecting the citizens from the harm of everything that could happen.
However, in my experience, you have to have policy, or every country is different than the other, and every country has a technology that is better for it, for example, than areas, mountain areas, desert areas, lots of poor people, lots of rich people.  Every country has its own prerogatives.  Good principles for policymakers is three in my opinion.
One is to allow competition, because competition is good for innovation.  It is also good for reducing prices, which is an important aspect to reach the last billion people.  Lower the barriers to entry.  Allow new ISPs to have a gaming network or some kind of other network.
The third thing is to try to attract private sector to invest in these networks.  In my country, they are afraid you are privatizing Telecom.  You are not privatizing Telecom.  When you privatize Telecom, you don't sell a piece of land.  You don't take a landmark or piece of land and give it to the private company.  You give them a license to operate for ten years.
Really the Government has nothing to lose by allowing somebody else to pay the money to do what they have to do.  My motto is the Government should act only when markets fail, and they have 101 ways to act in putting incentives, regulating and paying for the stuff themselves.
So the Government has many tricks in their pocket.  They have to remember that they should move only after the markets fails.
As far as technologies, there is one, I don't know if we will talk about it later, if you need, is the Internet exchange points.  Internet exchange points are a switch that is located in the country in which all local traffic is rerouted to.
If I want to send an E‑mail to somebody in my country, the traffic doesn't have to go to London and come back, like it used to do before we opened the Internet exchange point in Beirut.  Why it is important for technology for the last billion is because by reducing this going and coming back, the capacity needed, the traffic needed, and because the IXPs are appearing so their business model doesn't have to do with ISPs paying for that traffic, but they pay for half the circuit until they get to the Internet exchange point.  So it's cheaper for them to peer on Internet exchange point than to pay transit through Telecom Italia to go through Europe and come back.
It is cheaper, and so this way, by putting Internet exchange points in the country, you can actually reduce the cost and then have more people have access to the Internet.
You also spur local content, because usually around IXP you find people coming and bringing their servers.  They want to be close to the point.  The first thing, the server is empty.  Then the next thing you have is the most popular Web sites and you have Akamai server or Google server.  You have an Apple server.  The next thing you know, it is really more self‑sufficient and saving a lot of money.  That is what I had to say.
>> DAVID REED: Okay.  Sonia is next.  She is the Executive Director for the Alliance for the Affordable Internet.  If you didn't know, Sonia holds a Masters in public policy from Tufts University and BA and BS in economics and business finance from the University of Massachusetts, also fluent in Portuguese, Spanish and English.  Impressive.
>> SONIA JORGE: Thank you David, for giving that detail.
Thank you indeed.  The Alliance for Affordable Internet, for those who don't know, we are a global coalition of private, public and Civil Society organisations.
We work primarily in two ways, to give you the five‑second overview, not only through very strong advocacy at the international level for affordability to be more prominent in the policy agendas of many different countries, as well as at the development level, but then we also work primarily in two ways.  
One is through direct country engagement in countries that we are already, for example, working in, Nigeria, Ghana and Mozambique.  Some of you are present here, and I'm happy to see you and others we are working with.  In those countries we work directly by focusing on policy and regulatory reform that will lead to creating the conditions for increased affordable Internet.
We also support our advocacy through a strong research programme.  We publish annually a report called the affordability report.  The new one should be coming out very soon.
I'm telling you that, because not only the alliance has come together, about a year ago, a little over a year ago, but in the short time that we have been acting in the different regions, we have accomplished quite a bit.  One of the things that I'm proud that we have accomplished, in addition to the work that we already are doing at the country level, has been the research that we have done to shed more light as to how we are going to work at the policy level to address the next five billion that you have on the title of the panel.
So one of the findings from our affordability report was actually that of the 46 countries that we covered in our research, about two billion live under the poverty line, so under $2 a day of incomes.  And basically for those two billion people in those 46 countries, the Internet is very very far from being affordable.
There is many more.  The challenges at the policy and regulatory level are very very large.  That is precisely why we are working with countries directly to try to address what are some of the priority areas at the policy level that can address those problems, but really to reduce that affordability gap in the country.
So, for example, in Nigeria and in Mozambique, that we have representatives here in this room, we are working with the Government for making sure that they have policies in place that will allow for infrastructure sharing to reduce the cost of Internet provision and broadband access.
We are also working on spectrum policy reform, some of which we have started collaborating also with Stevie Nigeria through a workshop on spectrum policy to figure out what are innovative ways that we can tackle spectrum policy, so that not only the cost of spectrum reduces in the countries, but also so that there is more innovative users of spectrum and more options for the current spectrum that is underutilized, as you already mentioned, and David as well.  And in that way many of the populations that now, even if they may be covered somewhat by mobile networks, and the fact is that they actually are not covered necessarily by mobile broadband networks, but so that these other options on the spectrum side can also facilitate a more affordable type of access, and that relates to issues of not only utilizing the access that is not there, but also utilizing technologies like white space, etcetera, that can provide new options for more affordable access.
The other issue that I would like to add, if I have a few more seconds, is that here we are at, in Turkey, at the Internet Governance Forum, and many of us are privileged to have access to the Internet here, or when the Internet is working, and elsewhere, where we leave, when it's working, depending on what kind of speed and capacity we have and how much we can afford.
But the reality is that for the most part, in the countries that we are working in, Africa, Latin America and Asia, the percentage of the population that has access to broadband Internet is very small.  For the countries that we are working on, one that has the highest broadband penetration now is Nigeria, only at about 9 percent.
There is a lot of work to be done.  I hope that we have a chance to discuss some of the different initiatives that, in the different policy options that are on the table, to facilitate affordable access.
But that is our focus.  And finally, we work in all of these areas by bringing together all the different stakeholders at the national level.  Our work in the countries is through national multistakeholder coalitions that not only define the priorities issues that we support and focus on for reform, but also very much define how they would like that kind of reform to take place.
While A4AI does have as its foundation a statement of best practice, policies and regulations, we very much believe that it's important to understand the realities in the context of each of the countries, so you can, starting from what we see as a best practice, but also be flexible to customize that option, that idea, that solution, to the situation that a particular country may have.
I'll stop there, so others can ‑‑
>> DAVID REED: Okay.  Thank you, Sonia.  Next is Jane Coffin, director of development strategy at the Internet Society.  Jane, you have to cough it up.  (chuckles)  What is your degree in?
>> JANE COFFIN: I am from the same alma mater as Sonia.  But you would not believe that I studied Greek and Latin, archaeology and art history.  (chuckles).
Then I went on to political science work.  But the focus was things like the classics, the books that you read about the Iliad, Odyssey and the other, and archaeology, which is, this is a beautiful country for that.  I studied in Greece, did things like that.
But I joined the Telecom wave 18 years ago, and then did more work in the Internet space, both overseas, living in countries, working with regulators and policymakers up front with them, in their offices.  I have background in that as well, which was a good segue to the Internet Society.
I'll give you five seconds about the Internet Society.  Like Sonia, so you know, many of you have heard of us probably, but we are a global nonprofit with teams around the world, and we focus at the microcenter of technology policy and development.  We have a great economist on staff here in the room with us, Michael Kende, who has written reports that many of you may have seen.  
We have been bringing on more of that rigor to look at the economic development aspects, as Sonia was suggesting, because we want to put data to hypotheses, particularly with respect to Internet exchange points.  You can see from data we had in our report that Michael helped put together on Nigeria, Kenya and Internet exchange points, there is economic value, there is economic development aspects to the infrastructure, which many of you know, intuitively working in your countries.
I'm going to focus now on IXPs in more detail than Salam did.  But we work closely with colleagues around the world.  I want to stress that we work with partners.  We have partners from the RIPE NCC, Mike Jensen in the room here, directly, I'm not going to say this properly, Kashabata from Georgia.  We go where people want us to go.  We work from a development aspect, developing Internet exchange points, are working with others to take them to the next level.
I'm going to focus on three pieces of this, technology, economics and governance, quickly.  From a technical perspective, when you do build out an Internet exchange point, you are creating a hub, not only a technical hub where people can come and peer as Salam suggested, they are not paying as much because they are usually peering free.  There is small cost like cross connect ports for a building, but you are building an ecosystem.  You have fiber, cables, but you also have people coming together to form a community of interest and to learn from each other.
Often you can see when you are running traffic as the IXP manager or operator where other ISPs may not be running their traffic efficiently or effectively.  You can communicate with them and say, look, maybe we should be working on better technical throughput for your network and how to do that.
We also offer training with other colleagues related to that technical throughput for traffic.  But you start to build a community of engineers that become tight with each other.
That IXL is sort of a social home as well.  We often find technical engineering for IXP is easy, 5 to 10 percent of the effort.  Some say 20.  I have a debate with Philip Smith about this.  He says 95, 5, 95 percent social engineering, 5 percent technical.  We could set up an IXP here in the room if we had the gear.  It is that human side of bringing people together who compete with each other who may not want to be in the same room, but then they begin to realise this is something important for the country.  It is good for our development, good for the network development.  Things progress.
It is, gives you energy when you see this happen.  We have teams in Africa who are doing great work with the African Union, over 25 best practices workshops in the last two years, 15 technical assistance workshops.  We have moved in Latin America, has been a good benchmark for us.  We have done work there.
It's a work in progress.  We are not the only group that does this type of thing.  As I mentioned, the regional Internet registries do great work.  Packet clearinghouse does excellent work around the world.  It is a team effort.  Things don't always succeed.  Sometimes they fail.  But from a technical perspective, you pick it back up again and see where you can build the community of interest that is so important.
I won't rest too much on the economics because that could come out of the conversation.  But you do see cost savings.  Ecuador is an amazing example.  They are paying a hundred dollars per megabit per second for international transit.  After the IX was installed, traffic was running a dollar for local traffic, one megabit per second.
We have this outlined in a report that we released with the support of one of our grantees in the fall of last year.  It is November 2013 on Columbia, Ecuador, Brazil and Argentina.  It shows you some of the savings on cost, but also on how the traffic management was increased.  They went from 400 milliseconds for traffic throughput down to four or ten.
There is amazing engineering, traffic engineering that is facilitated there.
From a business plan perspective, and this is growing as I speak because we have colleagues in Africa who told me, ten years ago there was more a volunteer spirit.  It was more of a start‑up mentality.  Now we may not see as much of the start‑up mentality as you find in the past, because people see the Internet as how many apps can I design and become a millionaire, versus building a core infrastructure of the IX, which is critical.
The IXP doesn't make money like a gateway might.  Some people think of it as a place where millions of dollars are going to be made.  That is not the spirit of the crux of why you would put one in.  But we are often seeing the cost savings from a perspective of everyone coming together.  There is peering versus paying for transit.  You are saving money as an operator.  Your quality of service goes up.  
From a Government perspective, this is a unique bottom‑up usually governance model of volunteers coming together, network operators figuring out what to do.  Government normally isn't involved in the top‑down management of an IXP.  We are seeing governments becoming interested in Internet exchange points.  We are suggesting a collaborative model.  You are seeing a shift from just that volunteer spirit to be on a board.  CGI.BR has an excellent model that should be replicated in some places.  Some of the regulators in Africa are asking how they get involved.
Rather than doing a top‑down management perspective, we are promoting a leveling up their participation where you advise them.  They become part of a team to take a look at it.  Salam, you probably know this from Lebanon.
In any event, we look at the governments and talk to them about how they can play a role in the enabling environment outside of the IX, because it is an ecosystem.  Where can Government help facilitate, as Sonia's team has done, to articulate the realisation of cable back haul from the landing station.  There are miles of fiber.  There is a high cost in many countries.  You don't send traffic over land.  You send it by submarine cable, because it's too expensive.
Where can we as entities in this room and people involved in the Internet community help bring down costs, or success to policymakers and regulators, what the value chain of effort is.  
Another report we put out called Lifting Barriers to Connectivity, which Michael was a primary author of, points out the different places where you can push, pull and talk to regulators and policymakers about bringing down cost in that first mile, middle mile, last mile.
I'll stop.
>> DAVID REED: Okay.  Thank you, Jane.  Next is Steve Song, researcher at the Network Startup Resource Center, has a degree from University of Toronto in?  You didn't tell me.
>> STEVE SONG: I studied at the University of Toronto.  I didn't get a degree.
>> DAVID REED: You and Bill Gates.  (chuckles).
>> STEVE SONG: I want to start by talking about what I think are, there are three trends that inform my thinking about the future of access.
My focus is particularly on Africa.  1994 was a big year as a South African.  Not only was it the first Democratic elections, but also the first mobile network operator licenses were handed out in 1994 in South Africa.  That was the beginning of a revolution and in wireless access across the continent.  I don't think anyone's wildest dreams back in the mid '90s were a fraction of what has actually happened with mobile networks across the continent.
However, mobile networks have become a victim of their own success to some degree, because spectrum licenses cost nothing in 1994.  They were given it because there was a relative surplus of spectrum.
Now, in 2014, we are in a time of apparent scarcity of spectrum.  And because spectrum has become so valuable, being auctioned for millions or billions of dollars, that has become a huge barrier to the spread of access, both because regulators find it very hard now to find an equitable way to actually hand out spectrum.  It's very difficult to run a spectrum auction that isn't being gamed by the wealthiest players in the auction.  And it means that the sums of money being spent on those auctions are ultimately being passed on to the consumer.  If we are talking about connecting them, the next five billion, that is a problem.
Second trend, fast forward to 2009, in early 2009 there was a single cable, 40 gigabits, single cable connecting the entirety of sub‑Saharan Africa to the Internet.
Five years later, there are ten undersea cables connecting sub‑Saharan Africa.  Really that is just the tip of the iceberg, in that the arrival of those undersea cables has sparked a tsunami of terrestrial fiber build.  There is not a country in sub‑Saharan Africa that doesn't have a fiber backbone.  
If anyone wants to see this brochure, this is one fiber operator that runs fiber from Capetown to the Ethiopian border.  Fiber is spreading faster than anyone would have believed possible.  This is good for two reasons; not just the obvious reason, which is fiber being the digital equivalent of a deep water port in that you can really move serious amounts of bandwidth once you are on glass, but the second thing is, that it actually changes the way you think about rollouts.
The reason you had to give out a national spectrum license previously was that, to be a Telecom operator, you had to do the full stack, had to build a national backbone network, you had to build the handsets and everything in between in order to roll out a network.
The profusion of terrestrial fiber networks now are offering new opportunities in terms of access, in that you don't actually have to offer the full stack now.
That brings us to the third trend, which is the phenomenal success of low‑cost wireless technologies.  I don't think there is a person in this room who isn't carrying a wi‑fi device.  They are not licensed.  They were not designed as a rollout strategy.  Wi‑fi evolved because it could.
These devices, we call them mobile phones.  But in the industrialized world, 75 percent of all smart phone data travels over wi‑fi.  We might more appropriately call them wi‑fi phones, if we are just going to go with the biggest data carrier on the device.  And there are more wi‑fi chips shipped globally than mobile phones.
It is a huge underrecognized trend, I think.  There is opportunities for that to extend into new areas with dynamic spectrum management, which is a kind of midway ground between completely unlicensed spectrum and licensed spectrum.
I'll talk more about that later.  Thank you.
>> DAVID REED: Thanks, Steve.  Our next speaker is Jennifer Haroon, the Principal in Access Strategy and Operations at Google.  She received a BS from Duke and MBA from Berkeley school of business.
>> JENNIFER HAROON: Thanks, David.  One trend when you talk about connecting the next five billion that David mentioned, and all of my panelists mentioned in some way, and the way Google thinks about it, is there is no silver bullet in terms of connecting the next five billion.
That is both on the technology side and on the policy side.  In addition to the fact that there isn't one silver bullet, there is also not one organisation or one constituency that can achieve connecting the next five billion.  It takes the work of governments, Civil Society, and private sector working together.
When Google thinks about this challenge, it really tries to address this fact that it doesn't believe that there is a silver bullet.  So we work on a variety of efforts to try to test out ways that we can help advance Internet access.
For instance, we look at business models that maybe haven't been tried before or that we can tweak.  Examples of that include Project Link, which is a wholesale metro fiber network that we built and operate in Uganda, where we are enabling local providers, i.e the local mobile operators and ISPs, to take advantage of fiber built once, so that each of them doesn't have to invest separately, as Steve mentioned, or even the Google fiber effort we have in the U.S., where we are looking, we roll out fiber in a nontraditional way for the U.S. where we look at fiber hoods and where there is a lot of demand as we roll out.
We also think about innovative technologies.  There are obviously the technologies that exist today, and I couldn't agree with Steve more that wi‑fi is I think an underappreciated technology.  But there is a possibility that there are other technologies yet to be defined that can be more efficient, more affordable in connecting more people.  Examples where Google invest there is in Project Loon, where we are looking at high altitude balloons as a way to connect people, and also Titan Aerospace.
In addition, we think about the policy issues that I think have been well mentioned on this panel already.  But there in particular it is again about working with partners.  Google is a part of A4AI Dynamic Spectrum Alliance, and trying to work with all these industry and Civil Society players to think about ways to solve this issue.
The last thing I wanted to touch upon that I think was not addressed yet is the capacity‑building aspect.  That is both capacity‑building in helping do the work that Jane mentioned in terms of training network engineers, but also on the, what I would say the demand side of Internet access.
That includes investing in tech hubs and innovation hubs around the world, because it's no use to just bring the infrastructure.  There then has to be the demand, the local content and the use of the Internet.
I will pause there.
>> DAVID REED: Okay.  Thanks, Jennifer.  And last but not least, Vinay Kesari is standing in on late notice.  Virat isn't feeling up to participating today.  Vinay is a lawyer specializing in ICT law and policy, works with a law firm in New Delhi, and also advises the National Internet Exchange of India.  He is a graduate of Nalsar University of Law in Hyderabad.
>> VINAY KESARI:  Good evening.  I think that considering that I'm the last speaker here and I'm replacing a good speaker at the last minute, I'm going to keep my comments short.
I'd like to address two issues:  One, the network infrastructure that is required to deploy new networks, and secondly, key policy issues that's as of roadblocks.
Now, while addressing this, I think I will largely dwell on the Indian experiences, but try to extrapolate from this to try to locate issues that have wider relevance.
In India, when it comes to facilitating broadband access, the major project that the Indian Government has chosen to embark upon to make this happen is something called the National Optic Fiber Network, which is usually shortened to NOFN.
In general, the Indian Government, when it comes to Telecom policy, operates on the basis of a national Telecom policy which is published every few years.  The first two focused on, the first focused on bringing in private investment into a sector that had been a Government monopoly up to that point.
The second focused on building out mobile networks.  The third and current policy focuses on ensuring that broadband reaches every Indian.
One of the major ways in which the Government is trying to achieve this is building out a fiber backbone, using public funds, which can then be used by anyone on an equal access basis.
The funds that are being used for this, I won't get into the details, but essentially we have something called the Universal Service Obligation Fund, which is basically a large pile of money that's accumulated over the years, largely from mobile carriers.
This is now being invested in building out this National Optic Fiber Network.
Now, the target of this network is essentially to make sure that about 250,000 villages in India are connected by fiber.  The idea is to lay this fiber backbone up to a certain point in each village, and then ensure that conditions are right for private service providers to provide last mile connectivity.
And therefore, this is in some ways a reasonably unique project, in the sense that here you have public funds being used to build out only your basic backbone, and leaving last mile connectivity up to private providers almost exclusively.
Now, there are a number of challenges related to this.  And we need clarity on some of these before the network is built out.  We are still in a fairly early stage of building this network out.
Firstly, this is a classic case of a public/private partnership.  There are, the case can be made that there should have been more stakeholder consultation before the project was actually implemented, the reason being that, the business case for building out the last mile is still a little unclear.  
And this is something that is essential to the success of the project, because I think that everyone is clear in their minds that the Government is only going to fund the buildout of the backbone itself.  It is not going to get into providing last mile connectivity through this network.
And the second major issue is related to payment for the use of this fiber backbone.  Will it be a pay‑as‑you‑go model?  Will it be some other type of model?  Again, there haven't been extensive consultations on this.
As a result, there is a fair bit of ambiguity.
I think one lesson that can be learned from this type of experience is to make sure that you have stakeholder consultations that happen well in advance of embarking on a major infrastructure project.  I think that no one would argue that this isn't a good project.  This is a great project.
I think that it can revolutionize things in India.  But there are quite a few details that need to be ironed out.  What is interesting about this project I think is that what they are trying to build here is essentially a neutral backbone.  The hope is that this will create, that this will spark the creation of an ecosystem, which will in many ways revolutionize the way communications happen in India.
Now, the second issue that I'd like to address is the kind of policy roadblocks that can come in when you are trying to facilitate broadband access.  I think that one of the, and this carries on from one of the points, from the earlier point, about the National Optic Fiber Network, I think it's important when charting out policies to enable access to make sure that you are not preventing yourself from using the most appropriate technology in the most appropriate location.
I think that, for example, fiber works well in many types of situations in many topographies, but not all.  So if you want broadband access assured to 99 percent of your population, you need to make sure that there is a wide variety of technologies that are easily available to service providers in order to provide these services to the consumer.
Satellite technology, for example, is a great way of providing international access in remote and sparsely populated areas.  India has quite a few of those, especially in border areas.
These are areas where even laying down a publicly funded fiber backbone might not result in Internet access being provided to the end user.  But in India, currently providing Internet access through satellite is practically impossible, because of various policy challenges.
We don't have an open sky policy for satellite, as a result of which it is basically commercially infeasible to provide Internet access to end users through satellite.  Satellite as an Internet access technology is used only by niche areas like banking and the military.
I think the other thing to keep in mind is to make sure that when you are looking at a broadband access policy, you make sure that you are not shackling yourself to a particular type of technology or particular set of technologies, but you make a rational determination as to which is the best technology for a particular region.
I think that is something that every developing country particularly needs to keep in mind when public funds are being used to push a certain type of technology.  Thanks.
>> DAVID REED: Thank you, Vinay.  Question time.  Question for the audience first.  How many of you sat through the policy enabling access growth and development on the Internet, the earlier main focus section?  Can you show hands?  Not very many.  A few.
I guess not as many as I thought.  I thought what was going to be the easiest question might be the most difficult for those of us who participated in that panel, since it was three hours long.  Besides one take‑away being not to do three‑hour panels, (chuckles) I'm wondering if, very briefly, 30 seconds to at most a minute, if any of the panelists have a take‑away from that session that they might want to share with the audience.  If not, we will move on.  But yeah.
>> SONIA JORGE:  If I may, to me, the take‑away, in addition to many things that were interesting that were mentioned, but kind of disappointment, let's call it as a take‑away, I think many of us here in this room and many of us who were in that room were looking for more concrete examples and more concrete options and forward‑looking opportunities on how we as practitioners in this field, as people working with policymakers in all the different sectors, how can we make things happen.
For me, one of the things that I would like us to also discuss here and start a discussion with all of you, is to not only share some of those experiences that some of you are having in your countries, in your organisations, but also look at the specifics.  
What does spectrum policy reform really mean in terms of what we want to see?  We want to see obviously reduction in the price of spectrum licenses, but we also want to see very concrete regulations around allowing a license, license spectrum to be used, in a way that is freely without complications and without the fear or the threat that you may be violating some kind of regulation, or innovating with technologies that are available already, but many innovators in Asia and Africa, Latin America are not able to put in practice, because policy and regulations are impediments to those opportunities.
To me, what I would like to hear and us to discuss here is who is doing what, and for those examples of where we see those things happening, can we share with each other.  Spectrum is one area.  Taxation is another.  It was mentioned in the other panel.  But what specific taxation, what does it mean to change taxation?  Is it lower taxation on the devices?  To me, that is not the only question.  It is one important question.  But there is taxation on infrastructure inputs.  There is taxation, kind of taxation issues that colleagues in Nigeria face, multi‑taxation problem that has become a convoluted process, and the Minister Johnson mentioned earlier today, universal service funds that were mentioned already by our colleague next to you, and there is many here in the room that are working in that area.  
How do we make them more effective?  How do we make universal service funds, instead of being just another opportunity for those who like to be critical, to criticize, funds are being accumulated and not being disbursed, what are the steps to make them effective and useful to subsidize, not just infrastructure, but to subsidize the kinds of things that many of us raised in terms of demand issues, content, etcetera, and even awareness, consumer awareness, campaigns, projects to make sure that those who don't have access can get ‑‑
>> DAVID REED: Longer than a minute now; very good answer, Sonia.
>> SONIA JORGE: Good concrete examples, but there is many here we can learn from here too.
>> DAVID REED: Anybody else want to add anything?  No.  Okay.  We have a question here.
>> AUDIENCE: My name is Osama from Digital Empowerment Foundation based in India.  I want to share a story.  Sonia knows, Steve knows our story.  We have been working using unlicensed spectrum in India, about ten location.  Our experience is that private sector is not interested in connecting last mile.
They are not helpful at all.  They don't want to give us a dedicated lease line for further distribution, because they want to distribute themselves.  They didn't want to distribute themselves until the time the paying capacity is there at the last mile.  They think that we are basically taking lease line at a lower cost and distributing and making money ourselves.
They are neither giving to further distributed, nor they are going last mile, until the time there is a paying capacity.  That is one chunk that needs to be discussed, that we cannot connect five billion unless you find paying capacity the way private sector wants them to pay as a ISP or teleco.  This is a hard fact on the ground that we have been experiencing.
Even if you are using unlicensed spectrum, you are going, because there is a lot of last mile activity that requires to go work with the communities, which is not like going, buying the connectivity out of the box.  You buy on the shop and the rack, you know.  
Here, you go knock on the door.  They say you don't know English, you don't know India, you don't know what is access, but you are still interested, come here.  There are many things you do with them before they get connected and take advantage of it.
This is one thing that public‑private partnership or anything, it is all on the paper.  They are not interested in connecting last mile.  I'm doing a very strong remark, knowing that what we are doing on the ground.
The second thing is that National Optic Fiber Network that the Government of India has given from USO fund, universal service obligation for connecting 250,000 village councils, in the last three years we have not been able to reach even 600.  Private sector again is not interested.  They are not part of it.  They have been invited by the Government of India.  I sit on the Advisory Committee.  I know it.  I attended five, ten meetings.  They are not interested.  They are saying it's not a Government job, it's our job to connect them, why our Government using tax paid money which goes into USO fund to be utilized by giving connectivity to the last mile, which is our job to do.
>> DAVID REED: If I could have Steve, if you could react to the unlicensed ‑‑ you are saying wi‑fi is the underappreciated here.  In some sense, he is wrapping technology with the business model.  How do you capitalize on wi‑fi to address issues we are seeing in India with the unlicensed and the lack of participation in the private sector?
>> STEVE SONG: Partly, you have to get out of the way, in that a lot of people assume that wi‑fi is free to use anywhere.  But the regulatory approach to unlicensed spectrum makes a huge difference.  In South Africa, there are over a hundred wireless ISPs.  They have a very strong industry body association that lobbies on their behalf.
There is very liberal policy around unlicensed use.
One country, North Zimbabwe, it's possible to use wi‑fi for a hot spot, but point‑to‑point licenses are unavailable.  You cannot ‑‑ you need a license to do it.  As regulator said to me directly, we don't give those out.
Further north in Malawi, you can get a point‑to‑point wi‑fi license, but it costs you $1000 a year for a license.  For a technology that costs under a hundred dollars, a device, $1000 a year license is punitive.
Partly it's getting some of the basics right around the existing wi‑fi, and then looking forward to new areas.  The FCC in the U.S. have unleashed another big chunk of five gigahertz spectrum for wi‑fi use.  This is tremendous broadband potential, looking at the five gigahertz range for expanding wi‑fi, and then I'll talk about it later, but looking at dynamic approaches to spectrum management.
>> DAVID REED: I saw a couple hands.  We got a lot of hands.  Quickly.
>> VINAY KESARI:  To respond to the point about the USOF and NOFN in India, as I mentioned, definitely it could be argued there should have been more stakeholder consultation before the project was rolled out.  But getting the private sector on board is a function of making sure that you present the right kind of business cases to them, and make sure that they figure out the right business cases; for example, NOFN, which can be used for mobile backhaul, which is a huge requirement in India.  As usage grows in India, even service providers who already have existing backbone networks will need more bandwidth.  
Rather than invest in building more fiber or building new networks themselves, they could use NOFN to provide mobile backbone on existing networks.
>> DAVID REED: Where is the microphone?  Whoever is closest to the mic to start with here.  We got a lot of questions.
>> To add perspective, I'm in Nigeria, operating.  I heard what you said in India.  It's not very different.  But in most of these developing markets, where we have people who are unconnected, the reality of giving access and the economics are really based on the local macroeconomic conditions.
You can give free wi‑fi, but you have to back all the traffic to connect it to anything.  Someone has built that network with some that they haven't fully recovered, so they are blocking it.
Part of what you have in most developing parts of the world is the regulatory systems are not that sophisticated as in the advanced economies where these things work.  Rule of law is not quite the same.  Indicators of risk in doing business are different.
Really, where solutions are technically feasible, without resolving those local ecosystem issues, you don't get the kind of rollout or development that you really require.  
We built a submarine cable, and our biggest challenge is we can't build the impact, not because, the expectations in terms of this was a critical missing component and this did not exist, so once we bring this in, everyone who has latent demand to the Internet is going to be able to get it, and then you find that retail operator that's built their network for mobile voice is not prepared to put the electronics on to facilitate data, because the possibility of data is different from the base case they built.
It cascades down the line.  I think Sonia asked a great question, in terms, what does it really take.  Each market really needs to sit and look at it.  But some of the global policies from more advanced societies just don't recognize the challenges.  No matter how many billions, I hate to say are put at it, it doesn't get through to the people who are trying to connect.
>> DAVID REED: Behind you, we will work back a little bit.  Is there any reaction from anybody on the panel as to that comment?  Jane?
>> JANE COFFIN: Often we see that you need a champion, one or two people to push an IXP from a technical perspective and get a social engineering right.  You need that with regulators often as well.  You need a champion in Government to push.  
If it's not the regulatory body, often, if it's one of us as a partner coming in to talk to the Prime Minister's office, or the President's office.  We saw that happen in Romania, when the Prime Minister's office said this is important, it is going to happen.  Rollouts took place.  Licenses were more ‑‑
>> DAVID REED: I'm going to try to move through all these questions as fast as I can.
>> AUDIENCE: I'm Antonio from the Philippines.  I'm a blogger.  What do you think would be the role of the state or the Government?  We are talking using the language of markets, and companies.  Do you see any role for states or governments specifically in establishing broadband networks?  Are this important?  That is number one.
Number two, what would be the role of fighting monopolies and trusts, because those are concerns in developing countries.
We feel that the absence of regulations for, against monopolies and trusts do not serve common interests like rolling out broadband access.
>> DAVID REED: Somebody want to respond to that question?  You can help me back here.
>> You can take several.
>> There are so many questions.
>> DAVID REED: We won't get to them.
>> Let's take some more and then group a bunch together.
>> DAVID REED: We will take some more.
>> AUDIENCE: Hi, Michael Kende, colleague of Jane's at the Internet Society.  Shifting gears, we put out recently a report, global Internet report that is available on our Web site, and one thing that I found, one group, the World Internet Project had done surveys of non‑Internet users, asking them why they weren't on the Internet.
There were about 20 countries.  I would say they are basically upper and middle income, no lower income countries.  The questions divided into, why aren't you on the Internet, and one set of responses was, I'm not interested or I don't understand it.  The other was, I can't afford it or I don't have access.
In every one of the countries, the majority was not interested, not that they couldn't afford it.
Many countries have 3G networks throughout the country, and much higher penetration or coverage than people taking the access.  So I wonder if we also should be looking at getting people interested, getting them the content that will get them online, similar to the type of content that we look at, some of us have available in our home countries about our home Government or home cities and things that interest us, may just not be available and may not bring people online who could afford it and who do have access.
>> DAVID REED: Other people wanting ‑‑
>> AUDIENCE: Hi, Mark Summer from broadband in San Francisco.  We help service providers in emerging markets get into the market, mostly focused on enabling wi‑fi on licensed type technologies.
I'm pretty excited.  It is an interesting time right now, because we have so much cables coming into sub‑Saharan Africa especially but as well in Asia.  We are starting to see more of the middle mile fiber come to villages and secondary towns.  The key is to be able to access that in a sustainable way for a service provider.  For example, in Tanzania, it doesn't make sense to access the fiber unless you are buying 150 megabits at a time or more, because the economics are bad when you buy less.  But that is a huge chunk for a small population potentially for a service provider to start.
Those are the kinds of things where I think there is a lot of enablement can be done with the right regulatory environment or the right economic enabling framework, networks like the one in India the gentleman mentioned.
We see when that happens, we can build as service providers who provide an alternative access, with really interesting economics to their populations.
>> DAVID REED: If we can bring, there is a couple folks up front here who want to...
>> AUDIENCE: I don't believe wireless alone is enough.  You need backhaul and for backhaul you need fiber infrastructure that is accessible, which means that if there is only one party supplying it, it won't work.  You need a working market.
The question to the panel, this was brought up in a session this morning, that was supposed to go on net neutrality, is if you are to build that infrastructure yourself, which means you have deep pockets, which means if you build it, you are going to keep it to yourself.
The other alternative is have the Government subsidize it, have the Government incentivize, build it.  And to hook up to what Salam says, yeah, Government should step in when market failure.  Market failure, Government steps in.  What was brought up in the panel this morning was, yeah, but that is Government stepping up is actually disrupting the market, in that sense creating market failure.
I was wondering what the panel thinks about that particular dilemma.  Should you wait for the market or be disruptive?
>> DAVID REED: Maybe, we have had three, can we have the panel respond to a couple of this before we ‑‑ Jennifer or, we haven't heard from you.
>> JENNIFER HAROON: Maybe I'll comment on the last question also related to Mark's comment.  Like I said earlier, I think there are a lot of constituencies that need to come together.  And in particular, talking about back hauling, what we did with Project Link in Kampala is decide to build a fiber network.  When we looked at the issues, they have access to underseas cables.  There is a wonderfully competitive last mile with multiple mobile operators and ISPs operating.
But what was missing was that high quality, basically unlimited capacity back haul.  Most providers were using microwave, which most people here know has capacity constraints.
One of the ways we thought about this issue of competition is that we run a network that is open to all of the local providers, as long as they are licensed.  We also have a flat rate card, so we call it in our team the bar test.  If all of our customers entered a bar, showed each other their rate cards from us, it would be the same.  There wouldn't be any surprises.
The other thing that we really wanted to encourage was usage.  It is a little funny to me that the industry charges everything by the bit.  That is how all of us pay our cell phone bills.  Instead, we offer our local provider customers unlimited capacity, and it's instead based on a monthly fee, because as I think we all know, fiber capacity doesn't cost more when you use the next bit.
At some level, maybe you have to upgrade the equipment a little bit.  But it really isn't a per bit cost.
Those are some of the business model changes that we are trying with Project Link.  I don't think that it's just a choice between Government or one player.  In our case, we have a license from the Government of Uganda, but it is not an exclusive license.  Someone else could come in and compete with us with a very similar model.
>> DAVID REED: Thanks, Jennifer.  We have Salam and then we will go right here.
>> SALAM YAMOUT:  I want to answer the question of the role of the state, monopoly and trust.
Telecom is a commodities, public commodity.  That is why it is regulated.  Unlike bread or water or when you buy clothes, the Government does not regulate the people who sell you jeans.  However, the Governments all around the world have decided to regulate the Telecom business because it's a public commodity.
That is the role of the Government, could happen at many stages.  The first one, if they have a vision especially, they can give incentives.  The Government of Australia built a network, core fiber network to go into the desert, because businesses did not go there.
Other countries control the number of licenses, to decide how many players can compete against each other.  Thirdly, also governments have given incentives to help companies go to the areas where there is no business case, like the gentleman from India is talking about.
So the governments has a lot to play, and as Mark was saying, how aggressive they are, how laid back they are, if we see around the world different models, but at the end of the day, the Government plays a role in not allowing that Telecom market to become a monopoly, and to keep some trust and some quality of service for the people.  And they have a lot of tools to regulate that market and do stuff work.
>> DAVID REED: Okay.  Thanks, Salam.
>> AUDIENCE: Hi, from Independent Technology Initiative.  I have a question for Salam and Jennifer; it's the same question actually, just different aspects of it.
Salam, you mentioned that we need to lower the barrier of entry and attract private sector and to build things, and that Government should only get in when markets fail.
My question is regarding what we consider a market failure, because as I see it, if we take it to its logical conclusion, we are going to end up with only private digital spaces, private spaces in which we can, we communicate, we share things, we engage in things that we think are necessary to democracy, but they are going to be only private.
So would the lack of public digital spaces be considered a market failure in your model?  And part of the reason I ask this is, Jennifer mentioned that Google are now providing connectivity in the form of Google fiber, for example, and Google Loon for the next five billion.
What wasn't mentioned is that to use Google fiber, you need to use your Google account, your one Google account, the same Google account you use for gMail, same account you use for Picasso, for all the other Google properties.  At this point it is not just connectivity.  It is a certain type of connectivity that comes attached to your Google account.
I would love to know if, Jennifer, you see any elements of, I guess what I could possibly only call a new type of colonialism where we are going, we are going to connect you to the Internet with Google Loon, but here are rules that are attached to it.  It's a Google connection.  We have seen Facebook do this with  They provide a subset of the Internet as a free Internet connection.  But it's for accessing Facebook actually.
Both of those initiatives are going to a group of other people who are the other and saying, here we go, we are bringing you the fire, here is the white man bringing you the fire.  What are your thoughts on those aspects?
>> DAVID REED: We are going to get a couple more questions from the audience.  We will circle back.  Jennifer, you have time to think about it.  Right behind you?  Here.
>> AUDIENCE: My name is Vanansa, I'm building from what you asked some time ago.  You asked the take‑away.  I really wanted to take away something, because a part of me, recognizing the quality of our speakers, also I was very much interested in knowing what technologies and policies to connect the next five billion.
So I'm not sure if I'm taking away something still on that.  My suggestion is probably in the future, maybe today it is not possible, it would be interesting, it would be an interesting subject like this one to discuss what are the strategies to cover the five billion.  We also have indicators to cover the five billion, assuming that we do not have more babies, so we will not have more (chuckles).
And then, I think we also will have sort of success stories from which we could build, taking into account what was said.  So I think from one of this meeting to the next one, maybe this institution could work out based on their programme some results that can be shared, both in terms of a strategy and indicators that can help to resolve this issue.  
Mr. Chairperson, it's almost 20 years now I am discussing the issue of the last mile.  This is what we are discussing here.
I think it will be good if we have some outcome.
>> DAVID REED: We need to get folks over here as well.  We will come back.
>> AUDIENCE: Hi, the Chairman of the Web payments community group at the World Wide Web Consortium.
I know the discussion today is around technology and policies for connecting the next five billion people.  The reason that we are talking about that is we believe that if we connect those next five billion people, there are going to be some positive outcomes for them.
But the discussion today has really focused on the exact building out the hardware kind of, the networks and whatnot.  We haven't focused on applications, what are technology and policy decisions we should make around applications that those next five billion should get.
For example, should the Internet have a core way of doing payments, a way for the world's unbanked, the people that financial institutions aren't interested in, in building their networks out to; should we create a mechanism for people to store and send money to each other, built into the core architecture of the Internet.  
Should we have an identity‑based mechanism built into the core architecture of the Internet, so governments that have a hard time building identity mechanisms out to their population have this capability of just reusing a good privacy protecting identity mechanism that we have built into the Web.
These are problems that we are actually working on at the Worldwide Web Consortium.  But I haven't really heard them discussed as technology and policy discussions such that we should be having.
I think there are two layers to the discussion here.  One of them is how do we get last mile stuff done, how do we build out the networks, and then what applications technology and policy discussions should we have around applications we are going to deploy?  Because this isn't the first time we are pushing the Internet out to people.  We have experience in building the Internet out to people.  Now lets learn from the mistakes we made over the last 15 years, and figure out how to do it better this time around.
>> DAVID REED: I'll let Jennifer respond, then Steve, then Sonia.  In response to the one question, I'll ask each of the panelists for a one‑sentence take‑away they would like the audience to take away from the session today at the very end, as the last question.
It won't be as comprehensive as you asked, but it's a good suggestion for the very end.  Jennifer and then Steve and then Sonia.
>> JENNIFER HAROON: So, there are sort of three responses that I want to give that are related.  When it comes to the innovative technologies we are developing, there we are focusing on developing the technology, and seeing if it will work.
So that we have yet to think about the potential business models for using those technologies.
In the terms of Project Link and Kampala, we are serving the end users ‑‑ we are not serving end users.  We are serving the last mile players who are bringing Internet to the end users.
There is no restrictions on using that, using our connections.  It is up to the last mile players and their end user customers.
Then related to that is we believe the ISP's role is to connect the end user.  It is the end user's decision what they want to do on the Internet.  That is true in the case of Google fiber.  Our customers connect Google fiber and they can connect to anywhere on the Internet.  There are no restrictions.
>> STEVE SONG: One of my frustrations at the IGF is that all of the conversations start with a very narrow place and then span out to embrace almost everything.
I want to suggest, while I'm so tempted to respond on the zero rating issue and, after the five sessions we have had on neutrality, we are going to leave that for the time being.  And I mean what you are talking about, with Web payments, is layer 5 and 6.
The Internet is made of 7 layers.  So far, in all of the discussions we have had to date, we have never dipped below layer 3.  TCPIP layer 3 and 4 were concerned about net neutrality.  We don't talk about the importance of the actual access layer of the Internet, and there is no incentive, right?
There is tons of incentives for digital entrepreneurs, and to build digital applications, but, comparatively, but when we talk about access entrepreneurs, they are locked out.  They are locked out because of spectrum, policy.  The opportunity to innovate in the space is huge, but we need to unlock it both in terms of unlocking regulation and unlocking investment, initial investment in all kinds of access layers start‑ups.  It's the unloved child of the 7 layers.
>> SONIA JORGE: There are so many things, it's hard to focus.  But a few points.
First of all, on the issue of competition, and many of you mentioned, let me put the plug, check our affordability report online which has answers to your questions.  But briefly, competition is definitely not the silver bullet.  It is important.  It is necessary.  But governments do have a role in interfering in the market, when the market fails, as Salam was saying.
There is good ways of doing that while you promote development, while you promote innovation.  There is many best practices on how to do that.  That will be a whole other section, how to do that.  We do work with countries that we are engaging with in that area.
The other thing, as a reality check, we are talking about Internet access.  We are talking about mobile broadband.  I was looking at some numbers that I happen to have in front of me, because of something recent.  To give you a few examples, when we are talking about broadband access, this is why access for us and affordability remains the most critical issue to tackle.
You look at countries like Indonesia that in the standard indicators has 121 penetration of mobiles.  Well, only 24 percent of those mobile subscriptions are actually 3G.  This is according to the GSMA.  If you think about the ability of using broadband and Internet, even on the mobile, it is very small, and a very large spectrum of access in a country like Indonesia.  The same thing is true in India; 70 percent penetration of mobile, only 7 percent of all subscriptions are 3G subscriptions.
It goes on and on like that on most developing countries, which are the ones that we work on that we try to have an impact.
The questions, the real priority questions from the policy and regulatory perspective in these countries are very basic.  To me, the take‑away is that they are important, but they need to be answered now because it's not just about the people that live under the poverty line, but one or two billion that live just above and still cannot afford, even if they have a mobile connection, because the prices are still so so high.  And there is just no, it makes no sense that policy or regulation are not doing the right thing to change that.
To me, the take‑away is that we need to work together, all the stakeholders, just as we are doing in many of the countries that A4AI is working, but many of you here, that is where the real impact lies.  So instead of having 24 percent 3G connections in Indonesia or 70 in India or one in Nigeria, we have many many more of those mobile users to adopt Internet and to adopt broadband access much more easily and affordably.
>> DAVID REED: We have two questions with the mics here.
>> AUDIENCE: Mike Hensen, Association for Progressive Communications, APC.  I want to focus on the role of the state in operating networks.  We seem to see a range of models, from complete inactivity in that area, through to competing with the private sector such as in Brazil and Australia, through to monopoly Government networks.
I wondered if the panel might have views on those three different models, given the history of the inability of the state to actually operate these networks effectively.
Then to zero down on the role of municipalities as well, we hear a lot about municipal backbones and municipal networks as metropolitan area networks.  I wonder if the panel has any reactions to the effectiveness of the city‑based backbone networks, and perhaps in particular, what the role, for example, is Google stepping in in Kampala because the Kampala municipality wasn't able to do it, or is the municipal fiber‑optic backbone a not good idea?
>> AUDIENCE: I think we also need to consider digital literacy.  We are talking about to increase the connectivity.  Everybody can use the mobile phones.  But everybody, the people cannot use without digital literacy the smart phones or the computers.  Therefore, it's very important to develop the digital literacy programme; also, usage of ICT in education.  Thank you.
>> DAVID REED: Any reactions?
>> VINAY KESARI:  I'd like to respond to the gentleman from APC.  The ideal role for the Government should be to design a regulatory system with the right set of incentives, and the right commercial imperatives, and then let the private sector develop the right kind of solution.
Unfortunately, that works only in a perfect system.  Many countries around the world do not have the type of perfect market.  Therefore, I think that there is no one‑size‑fits‑all solution.  It really is something that needs to be customized to the history and economic realities and political system of every country.
We can draw from best practices of countries that have faced similar challenges.  But for example, to compare the challenges to providing broadband access in India to say the U.S. as it is today, is really comparing apples and oranges.
I don't think there is a one‑size‑fits‑all solution, although the ideal is one where the Government plays a very hands‑off role.
>> STEVE SONG: As an observer of African markets for a number of years, particularly South Africa, I can say with confidence that it's a dangerous thing to hand your network over to the Government exclusively.
In terms of regulatory policies, I want to echo my colleague here in saying, there is no recipe book.  Right?  There is no one ‑‑ each market is individual.  It is a bit like raising children.  
If someone showed me the rule book on raising children, I'd like to see it, because they are all different and complex.  You need to take what you can learn from other networks and then apply them uniquely, because there are lots of markets.  Kenya, for instance, by rights should not be a competitive market.  The dominant operator has 75 percent of the market.  But it has the cheapest prices in East Africa, but in sub‑Saharan Africa also.  But it is not just the number of competitors.
If I have to pick one thing out that strikes me about where I see change happen, it's leadership, where you see strong leaders, in Kenya, in Nigeria recently, Johnson, and also the industry leaders as well, who basically show the way of how the market can evolve.
Finally, on municipal strategies, yes, I think municipalities are increasingly seeing themselves as economic entities that need to develop progressive strategies to attract businesses and to grow, and clearly, fiber is just like an essential part of that.
  (someone speaking off microphone).
>> SALAM YAMOUT:  My take‑away, it is not an either/or solution.  We can have many kinds of network, private networks, public networks.  The governments could be running and operating school networks, hospitals network, municipalities could run city networks, and private sector could run other networks.
I think the take‑away is there is a space for everybody, and there is a solution on depending on set as the vision that you want to achieve.  The solution exists.  It's the political way.
>> SONIA JORGE:  I thought I already had done my take‑aways.  But if you give me an extra minute I'd love to say we would love to here from you at A4AI, we are doing interesting work.  If you want to learn more, go to, and learn about it, and join us, so that you can also be part of the process of having an impact at the policy and regulatory level, not just within the countries that we work, but through the research that we do, but also at the international level advocating for the right messages for affordability.  So thank you.
>> JANE COFFIN:  I would say build your roadmaps and the community.  You need the communities of interest to facilitate discussion, so your champions emerge, and work on the regulator becoming a business development entity, so you can move forward through that mechanism.
>> STEVE SONG: Healthy markets nurture small players to grow up to challenge big players.  This is something that is signally missing in the wireless markets.  We need regulatory and policy change to enable small access entrepreneurs.
>> JENNIFER HAROON: At risk of repeating myself, and my fellow panelists, I think I'm going to say what I said from the beginning, which is that there is no silver bullet, and as Steve said, there is no one answer for every country, every municipality.
So listening to your constituents, engaging everyone in the process and looking at examples, but then tweaking those examples to fit your situation is really key.
>> VINAY KESARI:  If you are designing a national broadband plan or a policy for the entire country, I would say try to make sure it is technology agnostic as far as possible, so that people are able to use the technology that best fits the problem in a particular region and in a particular terrain.
Make sure that if public money is being invested in a particular type of technology, the policy environment does not unduly favor that particular technology to the exclusion of other technologies.
>> DAVID REED: So my take‑away, having listened to the comments and the audience and on the panel, from folks who are far smarter on this topic than I am, as there is some controversy around the local access, and what is going on, folks saying, hey, it's completely broken, to others saying, it's kind of working here.
So I think for the IGF, maybe that is something they can take note to try to move forward a path to fix it where it's not working.
So, with that, I'd like to thank the audience for fantastic participation here.  I'd like to thank the panel for their participation.  Thank you.
  (end of session at 6:04)
The following is the output of the real‑time captioning taken during the IGF 2014 Istanbul, Turkey, meetings.  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 session, but should not be treated as an authoritative record.