This is now a legacy site and could be not up to date. Please move to the new IGF Website at https://www.intgovforum.org

You are here

IGF 2016 - Day 4 - Room 3 - WS161: Public Wi-Fi/Open access models in developing countries

 

The following are the outputs of the real-time captioning taken during the Eleventh Annual Meeting of the Internet Governance Forum (IGF) in Jalisco, Mexico, from 5 to 9 December 2016. 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. 

***

>> STEVE SONG:  Welcome everyone to the session on public Wi-Fi and open access models.  My name is Steve song, I'm from the network start‑up resource center and I'm going to open the discussion by ‑‑ well, perhaps with a provocation.  In 2005 the Economist published an article that sort of declared a revolution that actually mobile networks were going to connect, especially Africa, rather than sort of traditional copper wire networks or other means.  And that has proven to be a bigger initiative ‑‑ a bigger movement, a bigger change than anyone I think could have dreamed.  And it's become the kind of common understand that simply if we wait long enough, mobile networks will connect everyone. 

For a number of years it was regarded as kind of problem‑solved.  But what we see now is that mobile networks are actually slowing in their growth and are now experiencing challenges in extending their business models into sparsely populated rural areas.  And we also see challenges with ability to pay and making access to communication affordable for all. 

So with that in mind, what we've seen in the last 12 to 24 months are some new things happening.  So I just wanted to relate to you some news from the last 12 months in sub‑Saharan Africa.  So in ‑‑ and this is all with the exception of one or two literally within the last 12 months.  In Dakar, prepublic Wi-Fi services, low may, Togo, free public Wi-Fi.  Kenya, poor Internet have just launched commercial Wi-Fi services.  Muingu, which is a slightly over initiative, operates a rural Wi-Fi service.  Liquid telecom have launched over 350 hot spots in nine counties in Kenya and public libraries in Kenya are also offering free public Wi-Fi access.  In Tanzania, just two weeks ago the government announced a free public Wi-Fi service following the news from Uganda where the government is offering free public Wi-Fi after hours in the greater Kampala region.  A major fiber network have also launched a whole sale Wi-Fi New York through the Kampala region which has led to ISPs there offering 24 hours of access for 1,000 shillings or $0.30. 

I mean, the list goes on and on and you're actually going to hear about it from my colleagues here.  But not just Wi-Fi in urban areas, but in townships, on buses, petrol stations, total petrol stations in South Africa offer free public Wi-Fi and most of you will have heard of initiatives.  So the question I hope we can discuss today is, is this a blip, is this a sort of marginal add‑on or what the role that these complementary technologies like Wi-Fi can play, and if they are relevant, what can we do to noble them.  So, and I don't think ‑‑ I want to emphasize that we're not limited to Wi-Fi in these discussions.  There are many pilots going on with TV wide space technology which is a Wi-Fi like technology but operating in unused television frequencies. 

Community fiber projects also springing up where communities, obviously wealthier communities are banding together to build their own access.  And I haven't yet touched on bottom‑up initiatives such as community Wi-Fi initiatives to provide access in underserved areas.  So there's a lot going on but it's almost like they succeed in spite of the obstacles that they face.  Obstacles such as regulatory frameworks that allow them to operate or access to high speed back hall.  So one of the key constraints around these networks is the cost of access to backhaul and it has been moved by many governments towards open access networks around fiber optic backhaul but has that actually brought the price down in order to enable the further spread of these initiatives.  These are all questions that I hope we can dig into today.  And with that I would like to turn over to our first speaker, Christopher Geerdts from South Africa who has recently conducted a study on behalf of research ICT Africa on public Wi-Fi in South Africa. 

>> CHRISTOPHER GEERDTS:  Thank you, Steve.  There's a lot of questions I could answer.  But in terms of this introduction I'm going to focus on the study we did, when we started in most of the desk top and the interviews we done just about a year ago.  The time in between has been spent trying to get some sort of dough manned research done with the providers.  But I looked at nine main free public Wi-Fi projects.  So what I focused on was not all the initiatives with commercial ventures offering free Wi-Fi, which is a number, but specifically where the government has provided.  And why we did this was to inform policy. 

In other words, if government says we need to create a more inclusive access environment using Wi-Fi, how could they do it?  This would be the beginning of a discussion on how it could be done.  If you look at the different ‑‑ to see what extent they are relevant in different contexts and what good practice we could tease out. 

On the face public Wi-Fi looks really exciting because more and more people are getting smart phones, the price is coming down quite dramatically but what you end up is having people with a fairly powerful computer by the standards of a few years ago walking around in their pockets.  And now with Wi-Fi those can be broadband enabled.  Can one deliver on that vision of providing access in such an easy way which could make such a difference?  And that was really the argument of the first of three case studies I just wanted to touch on.  Which is an organization called project ‑‑ they developed South Africa's first large free Wi-Fi project in the capital city and they argue that access is basically a human right and because it's a human right it should be provided by the government.  And they said the cheapest way to deliver is by public Wi-Fi. 

If you use a smart phone or a tablet, you don't have to put in the money of the device, because that's provided by the consumer.  And if you use government buildings, they already have physical security, they already have connectivity and they already have electrical power.  So you can bring down the cost of providing the service quite dramatically.  And then the service would be provided really around government buildings within 100‑200‑meter radius.  So the project is a non‑profit organization and they provide a turnkey solution.  They take local authority from the entire process from installing, planning, managing, organizing, running, assisting with the branding and they're doing the content.  It's basically a one‑stop shop.  All the local authority has to do, and it's a big "all," is to pay.  And payment is really on the high sites and the megabytes that go through.  You would have to have at least a 4‑5 year or even longer commitment to those payments. 

So with the first project just ending its third year, what has worked well?  I've tasted the technology itself and it works really well.  For those of you who have had experience with public Wi-Fi which is probably most of you at airports and shopping centers and restaurants, this is actually an exceptional service.  It uses a better broadband service than many fairly rich areas have in their own houses.  So the first sites chosen in their project were mainly on high education exam puss.  And that gave them a bit of a publicity head start because students are an ideal target.  They buy smart phones even if they can't always afford them.  They learn quickly.  They're social creatures.  And they need the Internet for their studies.  Reports of impact was good, and the gender disparity in that particular group is less significant in the broader population. 

Another thing they have done really well is they have a superb content portal.  The content relates to education, health, employment, and it's offered on an unlimited basis.  In other words when you're accessing that content from that service, there's no limit to how many you can consume.  There's also a focus on the local video creation, which was the part that won the award.  Video information is interesting because it obviously doesn't require a high level of literacy to consume.  So in summary on the good side, they offer a really good turnkey model.  In other words if a local authority doesn't have any capacity or any background in Wi-Fi, they can soon be offering access for return for their dollars in their public buildings. 

So what was limiting in the model?  Well, smart phone access on the street corner is good for information consumption and even video.  But of course it's fairly limiting if you want to do software coding or fairly complex operations which would require a PC.  You had about 8.5 access by PCs on their project.  The rest is mainly smart phones and tablets.  It's not really suitable for serious content generation.  So for a small business if you want to go out to the street every time you want to do a transaction, that's fine.  They come in as experts and they run their project.  So there's really minimal community participation except perhaps on the content side.  Once they've gone into an area and provided their service, pretty much shuts out competition, and they got a huge advantage now.  They've got this high sites introduce cost, guaranteed income and once a Wi-Fi network is entrenched, that network is providing radio interference if anyone else tries to come up and create a second Wi-Fi network.  Customers are allocated a daily bundle of 500 megabytes which is quite generous. 

But there is no way a customer can purchase additional.  Once their 500 is finished, that's pretty much it.  And that was an intentional design which I see to be a mistake.  I think that they're actually reviewing that plan now and seeing the potential to offer people migration to a commercial service if they want to consume more data. 

It is also telling when one looks at South Africa, that even though they've had public high profile success on their project, have been unable to replicate it with any reasonable amount of scale around the rest of the country and recently unfortunately they had to lay off staff.  Also they save money because they're knot non‑profit but they have a significant operator underneath which is actually not non‑profit and we've seen them grow on and now become an operator in its own right.  So one has to look very carefully at governance issues, even when a company or an organization says that they're non‑profit.  Everything has to be free was said very strongly in the beginning.  That was a tenet of their launch but now they're looking at other models, they're experimenting with Facebooks, the free basics model. 

The second one I looked at was in Cape Town run by the city itself which had a lot more capacity for doing its own project, having implemented its own fiber network already.  This focused as a pilot to underserved areas specifically and on residents in general rather than on students.  The city with U.S. grant funding, they looked at two areas, and they provide an open network.  In other words for those Wi-Fi, it allows commercial Internet service providers to provide service on top of that.  They actually have three providers.  As long as those three providers fits in with the proviso that they offer a certain amount which is 250 megabytes of free access per day to each user.  In fact they said it can't be less, it can't be more.  They have to make the provision 250 megabytes.  And those Internet providers can then commercialize the service whatever way they please over and above that. 

So what works with that model?  Well, I believe that model fosters competition much more effectively and it enables comparison of different business models.  And indeed when we looked at the stats coming in you can see how they were learning through the process because each of the three ISPs was approaching it in a different way.  So we got very interesting comparative data, which is available from research ICT Africa in a publication they've done.  And it informed what future models would look like, because this is the beginning of a long journey of understanding.  And if anyone asks, I can elaborate on some of that later. 

What's limiting in their model?  Well, technically they add a very quick amount of feedback that Wi-Fi doesn't penetrate very well into houses built of zinc.  There is no competition at the physical layer, which is now an issue because the project expansion has been so slow, it's still sitting at the pilot stage about a year after they intended to expand.  What I've noticed is once the city owns a network, it becomes a bit preoccupied with the network relative to fostering of competition in general.  And the equipment to manage needed and Wi-Fi to manage multiple providers has been a lot more expensive than if you just provided a single Wi-Fi network.  In fact civil times event ‑‑ several times entry cost.  And there's always a limit of ISPs.  You can't have any commercial because of the limitations of Wi-Fi at the moment.  So those are the constraints. 

And the third case study, which is actually the most interesting to me, and it's in fairly early days of the roll‑out, is the way the prevention government.  The province doesn't ‑‑ provincial government.  So up until today very few people on the private sector had ventured into the rural areas because of the high risk and low returns.  So what the government did was they actually reduced the risk via PPP, a public‑private partnership to roll out finer and they put it out to tender.  And what they could do is they could not buy the fiber but actually use the fiber under this agreement to connect all government buildings.  And by doing the public‑private partnership it could bring down the price to less than half of what it would have cost just to go out to tender. 

In return for guaranteeing their own usage, their own demand as the anchor tenant for all the government buildings for a number of certain years.  Not only that model but the idea of reducing a risk which I see as the most promising going forward with Wi-Fi.  Rather than just providing a free service.  Free Wi-Fi service was added as extra on this project.  So government would pay for the Wi-Fi usage on a per megabyte basis. 

So what works in that model?  I think what works is to leverage the resources of the private sector.  The way to get their interest in moving into these areas is to reduce risk.  And I think the future of public Wi-Fi is finding innovative ways to reduce the risk.  And government's main question should be how can we reduce the risk to make these areas more interesting.  Are there other ways of getting information as you know from public Wi-Fi or a return including advertising and including in fact what's considered the most valuable analytical information.  And by that I'm referring to legitimate analytical information.  The province they were able to deply much more quickly.  Rather than themselves ‑‑ forcing themselves to roll‑out extremely quickly.  So the up turn was extremely quick.  And they're still able to innovate.  And the government can stipulate that those sites are open access.  So they don't have to restrict competition at the service provision level. 

What doesn't work?  I'm just ‑‑ I'm not saying this doesn't work, but potentially as restrictive in the future and that's the monopoly element that near tell now has a huge provincial network and it's very difficult for another operator to compete and emulate that without having been given that head start.  Now, unfortunately with that project, because it was a fiber project, the Wi-Fi actually took second seat, although the project itself took social benefits of Wi-Fi.

So those are my three models.  And then just in conclusion if I can add three more conclusions because all together I actually looked at nine projects.  The one as I have said I support a sector model to drive growth.  Not exclusively, but across most of the areas where there is a benefit finding ways to turn it into a real benefit for the private sector.  We have, for example, in South Africa a wholesale Wi-Fi operator and they're more than willing to expedite roll‑outs to huge expanses of the country.  If they can find a model that works for them and government should be engaging them to see how to do that. 

The other thing I noticed when I looked at all my nine projects was they were focused out of the wealthiest two provinces out of the nine.  The other seven really weren't touched and when you look at the projects that Steve was talking about, they tend to be in the capital cities, I think with the exception mainly of Kenya and even where it is nine provinces.  And again, they were only in the wealthiest two metropols.  Even the clusters in South Africa are all connected.  So the rest of the country is pretty much untouched at this stage, even then a few of the ‑‑ in fairly rich areas.  So we're still a long, long way from talking equitable or inclusive access.  It's very hard for the smaller cities to justify Wi-Fi spend at the moment in the face of abject need and extreme poverty. 

I know that many of you would disagree and say Wi-Fi is still important, but we're preaching here to the converted.  I'm talking about when you're facing an audience out in the real world that is faced with potholes, roads being nonexistent, sanitation issues, how do you put Wi-Fi on to the agenda.  That seems to be why Wi-Fi has ended up being restricted to the big cities.  A good project requires good planning and this is my final point.  So that would involve initially an integrative start city type of initiative, good governance, low corruption, adequate capacity to deploy.  And when you look at a Wi-Fi project and evaluate it, it's quite a strong link to the underlying fiber projects that are in the area.  And that's actually my quick way of evaluating them.  If you see a successful fiber project, for example, in your case I would say liquid telecom have deployed successfully, you're likely to see a successful Wi-Fi project very close behind.  So that's almost like a proxy of, and very useful in the debate to link the two in your understanding.  So that's my final point and I'm happy to answer questions on the other seven projects or anything about how Wi-Fi relates to technologies.

>> STEVE SONG:  Thank you, Chris.  That point about fiber and Wi-Fi is tremendously important.  One, from the point of view that the marginal cost of adding Wi-Fi, if you own fiber, is like .0001%.  Wi-Fi is so cheap in comparison that it's just an easy win and because of the capacity of fiber it doesn't have an impact on other services.  The other thing is, and this comes back to your point, I think that the reason for me, the reason why Wi-Fi hasn't extended out of the major metropolitan areas is because of the cost of national backhaul.  Any country you travel to in sub‑Saharan Africa, they'll tell you that it costs more to get data to the cost or the point of exit to the country than it does to get at the rest of the way to Europe.  So lowering the cost of fiber backhaul is deeply implicated in the success of urban Wi-Fi beyond urban centers.  With that I would like to turn over to Moctar Yedaly to get a perspective from the African Union.

>> MOCTAR YEDALY:  I would like to apologize.  I will speak quickly and step out so you will not have a chance to ask me any questions.  I have a matter to take care of.  But I will take it from the point where we just finish the link between the big pipes, the big transmission channels and with the development of Wi-Fi.  This is where we are actually ‑‑ the African Union, our business mainly is about considering the macro big projects integrating the continent itself rather than going into the nitty‑gritty national small project in general.  Sometimes because of those actors that did not fulfill their missions, we step in and we try to see how we can supplement that deficit in terms of providing connectivities. 

While we are looking at big projecst in interconnecting the continent, we start lately to worry about the connected.  We believe that in most of the countries, the universal access strategy policies have failed and even sometimes have been abused for specific matter that are not really in the interest of the cities themselves.  And we start -- and we are just starting.  But thinking the way to provide connectivities to those unconnected to linking it with electrical or power project we do have on the continent. 

Indeed, the African Union have a big project for development in Africa, which has four major component.  Water, electricity, transport, and specifically ICT.  Mind you that many years ago we have designed an African satellite which is now launched that had a small solution about a small antenna that doesn't cost too much has specific links and can provide Internet access be it through Wi-Fi or the technology surrounding that.  But the big pipe was provided because at that point of time we thought that private sector would not go into places where the return on investment would be so low. 

Number two, although the coverage is there, the access is not provided either because the devices are expensive or are not put into place to make sure that those surrounding that access point are not connected. 

So from that point of view, we are emulating what we were preaching in terms of empowering communities to provide electricities through digital platform which is a new way of conceiving economy horizontally rather than vertically.  And allowing people to produce electricity and sell to others around would be a policy that would be much, much encouraging, not on the big cities but in the remote and specific areas. 

This is one of the biggest concerns. 

The second concern for us is once we provide the access to the population, without being somebody who is dictating what they have to do or not to do, wouldn't like to find ourselves in a position whereby the Wi-Fi or infrastructure or access is being used for something that is not, quote unquote, useful.  And we have noted for instance in India recently one of the biggest hub station has been, because it has a Wi-Fi become the hub.  Will that be something appropriate for Africa or Africans in general without again, and it's very important.  Dictating what the behavior should be or the contents should be.  We think it is very ‑‑ it is our responsibility to make sure that once the access is provided, is used for the well‑being and for the socioeconomic development. 

Once again we provide the access by no matter how scheme, PPP, public, private, the issue comes again here.  Why would we provide access or infrastructure while somebody was riding on those without actually paying anything for the services they're providing.  So this is the measure points from the African point of view.  I am sorry I won't be able to answer to those questions.  Some of them might be controversial.  But I will apologize and just step out. 

>> STEVE SONG:  I'm not sure we can let you go after making those provocative statements.  That you should determine what useful content is and the issue of free riding ‑‑ I feel like we have a whole session right there.  If you must go, thank you. 

>> MOCTAR YEDALY:  Thank you very much.  Again, apology. 

>> STEVE SONG:  Well, that's a lot of food for thought. 

With that, I would like to turn over to Carlos Rey‑Moreno who's based at the university of the western cape in South Africa and is one of the instigators of a community wireless networks. 

>> CARLOS REY‑MORENO:  Yeah, actually, as Steve has said I'm a post doc fellow of the western cape in South Africa and as part of my research I've been involved in the creation of networks, that are community owned among inter service providers.  And probably the other side of the world in South Africa from the big capital cities that Chris was talking about.  Access to electricity and other services is cars.  Only 3% are connected to the grid and 5% of them has resources to collect rain water so they don't have to go to a river or communal water points that they really work.  That is because the income per person are 15% of the national average.  So people can't afford private access to those resources.  In terms of communications, it's covered by operators.  And nearly everyone owns a mobile phone.  The fact that many family members emigrate to other parts of South Africa to work in the mines or in the farms, or access to location makes communication a priority for many of them.  As they want to be in touch with the people that they care.  Such a priority makes that people spend 22% of their disposable income in communications, many times are going food to do so.  So valid proof of the mobile communications might not be working in some places. 

Sadly, it is very common to hear that many people having to make the choice in between making a call or eating.  And although it's a percentage of disposable income, it only allows them to access a very ‑‑ basket of services with less than 20 calling minutes a week. 

In this context, since the network which was with the assistance of my university is providing services to try to reduce this cost of communications.  To do so they install operating our wireless network that easily provided free voice calls among 10 publicly accessible analog phones.  In addition the community used energy priced to create a mobile phone charging station. 

With that money, they get from the charging those phones, they are able to cover the monthly cost of the networks and those services were later upgraded to provide calls to international numbers for a fraction of the prices offered by incumbent operators.  Even those networks is able to offer voice over IP calls, they might not be able to reduce this cost for everyone as only a third of the people own smart phones and very few decide to use phones. 

The cost of voice communications would be straightforward in the communities had access to the spectrum as probably very is good tell you in a second.  From the two main providers, from the four main providers in the country, only two are providing national wide coverage.  One only cover 25% of the national territory and one of the other two providers and the rest and it doesn't seem that this situation is going to change with the slowdown of GSM roll‑out in the country.  Communities spectrum considering that the same number of calls is currently made.  The cost of communications could reduce from 22% to 3%.  Organizations and individuals that are present in this room at the moment presented this proposal to parliament three months ago and we are waiting for feedback although it seems very positive and we are making strides into that direction. 

In terms of that provision, they are about to make it happen, thanks to a grant for structure to connect to the fiber in a town that is 60 kilometers away.  And provide hot spots in the community.  To tackle the lack of Wi-Fi enabled devices, two laptops will be made available in those hot spots so community members can benefit from accessing that data.  They are also working with the school to use the computer rooms.  The department of education set up those rooms in 2013 you about they have been barely used since then.  In both cases they are working with them to understand what is preventing them from using those computer rooms, provided them with ICT training to overcome them, including among other things, security, security online, local content creation. 

Something similar to this public access to the Internet came very strongly in the first community networks in Africa that took place two weeks ago in Nairobi where public access appear as a common feature across the networks.  In all cases, the cost of the service is decided by the community, but that always as cheap as possible so they can cover the operational cost.  So we are talking about moving beyond affordable access.  We can see with example of the free computers in the school.  Nothing has happened until the community has taken the matter into their hands.  So we are talking about a creating a local economy, creating or providing digital skills, generating local content, and actually giving the communities the opportunity to have a agency about the developmental process and affairs. 

And to answer the question that was said in this panel, if we talk about universal access and the context of South Africa, this model is complementary to GSM networks.  But if we talk about universal affordable access, this model complemented with more changes in the environment, could actually rival with other models that are out there.  But if we look beyond affordability and the goal is to take communities outside the cycle of poverty and disempowerment that they are trapped into.  Community network seems to me the only solution.  Thank you so much. 

>> STEVE SONG:  Thank you, Carlos.  I think that is fundamentally essential point that there appears to be spectrum scarcity, but in fact there is a lot of spectrum that is unused.  And the current model we have for assigning spectrum is not allowing communities in rural areas to solve their own problems.  And you know, what we've seen in the explosion of Wi-Fi services across sub‑Saharan Africa is not because Wi-Fi is such an amazing valuable piece of spectrum.  Just the opposite.  But because it's a very narrow piece of spectrum.  But it's because it doesn't require permission to use.  It allows for self‑determination.  Permissionless innovation.  Which doesn't mean there are no rules.  It just means that the rules are imbedded in the technology so the devices have to play nicely with each other.  And this is a new paradigm we are facing in terms of how spectrum can be managed.  And with that I would like to turn over to Erick Heurta who actually I think has taken one of the biggest innovations forward in spectrum regulation and I won't steal his thunder and let him tell you about that. 

>> ERICK HEURTA:  Can we move a bit, attempt to use my hands and I don't want to disturb that computer. 

Sorry because of my voice it seems that the party was hard yesterday. 

[ Laughter ]

I talk very much. 

Actually, I will talk ‑‑ invite me to this panel, I say why did they invite me to this panel.  But then I quickly accept the invitation because on this experience of developing GSM networks in rural communities, and also before like in looking at the policy of social coverage in telecommunications, I face myself a lot with different programs along this years of information society.  And now it seems like kind of branding programs of many governments, open access Wi-Fi.  We have like thousands of public access places and many governments advertise that as one of the greatest achievement of its policy.  But I think we can look at it carefully.  And I will talk about three experiences just in this way I have faced with.  One is from Mexico, the other is from Colombia and the other is from Nicaragua.  And I will focus on rural areas and public access in rural areas. 

In the first case in Mexico, some of the rural areas, the only connectivity that could be at some point given to those areas are satellite connectivity.  So you arrive there and the government has be careful, I don't know how efficient, someone might hear there's statistics on that.  We can tell you better.  But it's around 30,000 along Mexico.  What I have found is on this rural places are some of these antennas that connect a school, for instance, then the school have to give public Wi-Fi as well.  So if you need to use the Wi-Fi, you get close to the school and then you can use the public Wi-Fi.  But in those cases, really the connectivity of the Wi-Fi connectivity is really poor, you can hardly do anything.  You can hardly download a document.  It might take you some minutes or 10 minutes to send an e‑mail.  And if you're lucky you can send some message on your what's app.  But still allows some communication where physical therapy satellite connectivity cannot be afford.  And there you have some other cases like Colombia where they have almost all the municipalities connected to fiber.  And then they also have a program of public access within that fiber and it seems that works really well. 

And the last one I would like to talk about, and this is more like a different experience.  I went to because we have been collaborating in one of these GSM networks in Nicaragua in a very isolated place.  It's like the English‑speaking part near the Caribbean and naturally has been a very difficult place to access.  Both, I don't know why, probably because it has been easier it is close to a sea cable, sea cable comes through that area of blue fields.  So I went to a community I have to go by boat within the river, or lagoon.  This community and it's really, really in the deep, probably like a two‑hour boat.  And then you get there.  And then I went to the park in the city and I have found that I could have video, send video, download photos really quickly and it was public Wi-Fi in the part actually came to me to my mind like some other philosophical issues of Internet.  Like I mean because in the park like everybody instead of playing like that with this mobile.  Well, that's for another chat.  But the connectivity was great.  And then when we, when we have a meeting with the people of the GSM network, one of the problems was the connectivity. 

So I said how does that happen?  I mean, it's not a problem of connectivity.  It seems that these guys have fiber here in the public access area.  But they couldn't get connectivity to connect their networks, this could send calls outside of their own network.  And so I think that is like a very good point of reflection when we talk about public access Wi-Fi.  Shall we be looking at public access Wi-Fi or should we ‑‑ do we have to look further, do we have to look connectivity not to public access.  Because well I think there are like levels of users in a network and I think because he mentioned at the beginning.  You have user probably like myself that just use that to send e‑mails and doesn't do much stuff with Internet.  But if you are like software coding on that, then you need another type of connectivity.  You need probably some connectivity, more secure, more capacity, have to be to create, to program, or to do programming.  And then there is another layer of user that is the one that can expand the network and build a network.  So for instance, in Colombia, I have the chance to talk with someone from the ministry and she was telling me this wonderful things about this program.  And then I ask her how does community can connect to that fiber in the municipalities to take that connectivity to other cities that are farther ‑‑ that are not ‑‑ that are far from the municipality.  And she said ah, I don't know.  But she was the one in charge of that program. 

So I think that you have to look this thing of look at this thing carefully, no.  There are some places that the only thing you can do is to have some public access because there is no other way to do connectivity and then with a satellite connection and that's a best effort.  It probably doesn't work very well but it's a best effort in connecting the people and giving the basic service.  But when you have fiber there, when you have fiber, why you stop the community of doing the best use of the Internet or scaling to other layers of user, artificially you're creating even though there you have fiber.  So and we said that the governments, by doing this, are accomplishing really with Internet access, were actually blocking Internet access to communities.  So we have to be very careful with those policies from the perspective of a researcher and present the perspective of a policy maker.  We don't have to say okay. 

And then the question comes to me is that is that something intentional or is that just ignorance.  And I think this sort of the mixture of both.  Sometimes it is intentional because we think that we shouldn't compete.  That sometimes some of the things these are state networks, we shouldn't compete with the operators.  But the problem here thinks, and we'll talk specially for the case that I saw in Nicaragua.  They cannot get access to the government fiber.  But they can neither their access to the submarine cable there.  They said you have to buy like 5 gigabits and they say we only have 1 or 2 megabytes or something.  And then it comes like within the intention of the mixture of ignorances.  There is a market so you have to enable those other operators, the conditions too in this community operators, the conditions to expand the network.  So it won't happen anything to the big operator, you won't bother him if you allow or give these people access to two or three that they need to operate. 

And also the thing is that it's sometimes government doesn't care because it is popular to say we have a thousand public access points.  But probably the public access points, even though they are giving good services to their users and the users are happy to send e‑mails and images, while this is happening, they are stopping other people to make better use of the ‑‑ not better use.  Another use of the Internet in developing programming or in expanding the networks. 

So I think that is what we should be looking at is connectivity, not public Wi-Fi.  What we should be looking at if we are really granting access to the Internet to people, or we just trying to show something that is not really efficiently done.  Thank you. 

>> STEVE SONG:  Thank you, Erick.  I think that point about fiber is fundamental.  You know, people talk about fiber as if it were just one of a range of technologies.  Fiber, wireless, copper, et cetera.  But fiber is a different category of technology, because its capacity is relatively infinite.  It is the deepwater port of access.  So when governments think about deploying fiber infrastructure, they often mimic the private sector model, which is to extract as much value as possible from the network.  But that's not the strategy you have for roads.  Right?  When you deploy roads, you want as much traffic going on that road as possible to enable the economic development of your country.  And if you were to actually cost out fiber, according to its lifespan of 15‑20 years and the terabits of capacity that it represents, a gigabit connection across your country, the actual infrastructure costs maybe 1 dollar, $50 or something like that, that were there a more strategic approach to fiber, that we would then see an unleashing not just of public Wi-Fi but of community efforts and private sector. 

Anyway, I see the moderator ‑‑ I'm the stand‑in moderator today standing in for Alison Gillwald and I wonder if you want to offer any remarks. 

>> ALISON GILLWALD:  Thank you very much.  I wanted to respond to what Erick said but make a comment about how this fits into our efforts within global governance more generally and the regional impacts of our regional bodies on our policy outcomes.  So we're saying all these wonderful things here, but they're not translating into our environments.  I wanted to say I really don't think it's fiber or public Wi-Fi or whatever.  I think it's just about using leveraging the different resources that are out there as optimally as we can in a resource‑constrained environment.  And I think that's really the mind set that we need to try and get bind to.  Because it's no use us coming here every year and talking amongst ourselves, or talking at our governments or people who quite only despite the multistakeholderism don't feel engage, don't feel there is no ownership of these processes.  So we speak about them and you get the kind of pushback that you saw today.  So I was responding specifically for those of you who weren't here to Moctar Yedaly's input from the African Union which I'm perfectly aware of, we work with him a lot on a lot of initiatives.  And there have been a number of really devastating or retrogressive decisions taken by the Africa caucus around OTTs for example and this kind of conventional wisdom that OTTs are a bad thing that they are actually taking away revenues from operators.  In the face of extensive evidence that OTTs in a changing environment, mobile operators that adopt these platforms and drive the data revenues and abandon the attempts to hold on the voice revenues, et cetera, are very successful.  And yet you're still getting there have been a number of workshops that I've been in this week with governments, African governments that are still being engaged in this forum saying these things.  S

o I just wanted to say that, because I think he's not stated to defend himself.  I think we do need to address it and it's really important.  There are a whole lot of things that are happening globally where we're all drinking our own Kool‑Aid and we get surprised when things don't happen like we expected them to.  OTT is as much of an issue here in terms of trying to get governments to not make decisions that will actually have these devastating unintended consequences for this project that we're involved in, this connectivity project we're involved in at the global level.

>> STEVE SONG:  Thank you, Alison.  I'll give the panel an opportunity just to react to their colleagues very briefly and then we'll take input from the floor.  Christopher? 

>> AUDIENCE:  It took us four years to get to where we are in terms of accessing fiber in a university that is 60 kilometers away and I'm providing to access.  We were talking to providers that were in the area, they were offering us.

There are some places where these fiber is happening.  There is fiber 5 kilometers away from us.  It's a political decision not to regulate that fiber. 

>> STEVE SONG:  Can we have the mic over here please? 

>> AUDIENCE:  I think this is a good discussion but there are some problems here.  San Francisco is one of the biggest high‑tech cities in the world and there's been a fight for municipalization of Wi-Fi.  Who's fighting that?  It's the monopolies who control it in the United States.  We have a situation where they're passing legislation to prevent communities from having public Wi-Fi.  Why?  Because they want to protect their revenue stream even if poor rural communities in the United States are not able to get public Wi-Fi.  So that's something we have to confront.  Monopoly capital does not want small poor communities to have Wi-Fi, because it's not profitable for them.  That's what we're really talking about. 

Now the other question is Facebook, and I'm surprised it hasn't been brought up in this panel here.  Because Facebook says it is providing public access.  And we learned at the IGF last year in Brazil that Facebook says they're providing free Internet access, zero‑based, but in reality, it's a scam.  It's a fraud.  They're not providing free access.  They provide you Facebook and if you want to continue to get Internet, you have to pay service fees.  And they're marketing this.  They've got the Internet.org to market this domain, to market their plan for making profits out of the Internet.  Our view is we need public community access, public Wi-Fi that cannot be controlled by the media corporations.  Their drive is not for public access.  Their drive is to develop profit streams and that's counter to what we believe our interests are.  In San Francisco, other communities in the United States, there could be public access.  There could be public Wi-Fi, but that's a threat to these media companies and that's something we have to face on a global level.  Thank you. 

>> STEVE SONG:  Thank you.  In the back there?  The old man with the beard. 

>> AUDIENCE:  Hi I'm from Brazil.  I make a question to reflect and comment.  Why not create a city and advertise it by municipality law that makes possible sustainability in democrat governance this is a question.  And can we agree that community networks could be a part of digital cities?  Two questions.  Thank you. 

>> STEVE SONG:  We'll take a few questions and then ask the panel to respond.  Over to the left in the back there. 

>> AUDIENCE:  Thank you, my name is Tony.  My question and concern is on the public Wi-Fi access.  Looking at the rural communities who may not be having the last point or the last point of the last mile of the fiber connection, how are we going to address this kind of issue at a national level?  And if we address it at national level, how are we going to make them be aware that they are able to access this kind of national concern.  Thank you. 

>> STEVE SONG:  Thank you. 

Mike Jensen there and just in front. 

>> AUDIENCE:  Thanks, association for progressive communications.  I just wanted to react to a couple of points that were made by the panelists.  First on the OTT thing I think we want just blame the governments.  I think a lot of the operators in Africa have really pushed for some sort of regulator.  Classic example is the ITU global symposium for regulators last year which was sponsored by air tell, one of the largest operators in Africa, they sponsored it very nicely giving everyone a tablet and a mobile phone and everything.  Allow the chief executive to stand up at the beginning of the conference and say I hope you regulators will stop these people from eat eating our breakfast every day.  So there's a kind of motivation from that area that needs to be addressed.  The other upon the I want to consider is the ‑‑ point I wanted to consider is the one that was suggested by the representative from Brazil, maybe public Wi-Fi isn't mutually exclusive. 

Thank you very much Erick for highlighting the issue that is present currently and I think as the cost of backhaul goes down, I think it's kind of inevitable that people will see the provision of access, commodity, a basic type of infrastructure that should be provided just like access to water or electricity.  So how can we take that motivation and not undermine community efforts.  And I think that's our particular point. 

And finally on the issue of kind of provision of public Wi-Fi by the private sector, yes, I think we need to develop some kind of metric around how these private operators are motivated, whether it's kind of a free basic model where you get a limited text‑based service.  Some free Wi-Fi operators that provide free Wi-Fi but only to text.  Then you have to pay if you want to get video.  So is there some model there that could be promoted or should we be looking at something more generic like the great idea for a free slow broadband service that could be provided by any operator at minimal cost.  Thank you. 

>> STEVE SONG:  We'll take one more comment just here and then I'll invite the panelists to respond. 

>> AUDIENCE:  Hi, thank you.  I'm from Fiji.  I'll speak in my personal capacity.  I note the remarks made by the Latin American panelists and the South Africa panels very interesting and those made from the floor.  In the Pacific we have a hybrid of, in terms of backhaul, we have both satellite, CNKU bend across the Pacific islands across the fiber.  In terms of I note that the most recent speaker, Mike Jensen, he mentioned that as backhaul costs are decreasing.  We're seeing the reverse in the Pacific; in fact, increasing not decreasing.  And the other thing I wanted to mention is IP transit costs in the developing world, particularly from the Pacific, and in Oceania we have 27 countries.  IP transit costs are increasing, this is passed off to the consumers. 

Now I would like to zoom into one country in particular which is mine, Fiji.  Governments have legislated things like access deficit particularly to provision and encourage the tel‑cos to take in terms of the last mile and to prevent access into underserved regions.  But what we're finding is because of the increasing costs, despite competition, in spite of competition, we're still seeing reneging and sort of tel‑cos and ISPs and operators are sort of reluctant to provide public Wi-Fi or to provide access.  So in terms of I pretty much like the suggestion that you said, can we have like a dedicated community access?

Perhaps one of the ways we can certainly do it is discussions such as this is things that could sort of go up given that you have multi ‑‑ different all sorts of people who global areas of regional areas of influence.  But in terms of on the ground, perhaps one of the things that in cases where government has reneged on deficit, perhaps one of the things we could do is consider texting centers or to consider operators, first‑year operators that are providing a backhaul, whether it's satellite or ‑‑ and I know for education research networks, like R net, there's already discount.  But one of the things we could do is to explore how we could increase access by providing stimulus.  Thank you. 

>> STEVE SONG:  Thank you.  So some difficult questions there.  Are there things we can do with municipal bylaws?  Are there other models we could look at for municipal access.  And how do we address fiber at a national level. 

I want to make a brief comment about national fiber networks and turn over to my colleagues. 

We face a conundrum with national fiber networks in many countries where they're either run ‑‑ they're built to run by the private sector, where a scarcity model prevails and the maximum amount of value is extracted from those fiber networks.  Or they're built by the state.  But the state often doesn't have the capacity to run them well, or more importantly, when it comes to pricing, they're reluctant to do anything but match whatever the private sector is pricing.  And I think one answer to that might be to open up the ownership of those state‑owned assets to smaller operators, not just the large mobile network operators, but allow a consortium to emerge, as they have done undersea cables to allow small ISPs to go from being renters, share croppers on this national fiber backbone to owners.  And the difference once you become an opener can be profound because then you can choose how to amortize your office over what period as opposed to paying whatever the market rate is and I think that would unlock those state assets in a way that would relieve the government of the burden of risk and taking perhaps decisions that they're not ready to take in terms of pricing that are allowing more flex among the different players in the market. 

So who would like to respond to some of those questions?  Erick, I see you.  

>> ERICK HEURTA:  I would like to talk a bit about economy.  We always said this is a very a phrase that always counts for in policy and regulation at telecommunications.  The market failure.  We have to go where there is a market failure.  But the thing is to say if there is really a market failure, which market are we talking about?  A book on architecture of economy.  What it says is there are three layers of economy.  The global economy, the local economy, and the subsistence economy.  And each economy has the capacity.  So it's not about the failure of the market.  The market disintegrated into these three layers of the economy.  But if we want to put some company that works well in a global environment, and make him pay that company and work in the subsistence environment, it's going to fail, and if that company don't want to go to that place, the same way if we want to put a company of systems market to go into a global market. 

So what do we need to do?  If we are from subsistence market to go to the global market, we have to be integrated and aggregated.  For instance, if there is a local like local network, wants to get into the market of fiber, that it's a global market, then he has to aggregate many, many small operators like him and probably as I said where they then are able to build that part of the fiber or mark partners in fiber and that's it.  Both are the same.  But in the same case, if there is a company that want to survey subsistence market, have to break its network, or their administration of the network, so that there is a company, a small thing in this subsistence market that can make that approach. 

So that is very important to stop framing that way the failure of the market.  It's the failure that we have said to some stages of the market to allow participating in their own market.  So their regulations, their regulation that it means the access to fiber, to access to essential elements, the regulation blocks them and that is not a failure of the market.  Because the market is there allowing these small companies to get into that market and bring services.  It's a failure of the regulation. 

>> Amazing point.  I would like to talk about what Tony said and the gentleman from the United States with regards to the fight these type of models receive from monopolies and big companies and how the government could enable what Tony was saying how people would know.  I think there is an important effort that these community‑owned initiatives need to do to try to fit into their regulatory framework.  Although it's important what, was saying about probably different environments for different economies, at the moment they don't exist.  So trying to fit into a framework that only plays into that one single economy no matter they are different is very important.  And I think at least in our case we have managed to fit into the regulatory framework. 

So no matter how many battles we are given by the operators, we have played by the same rules and applied to the same law as they are doing.  And in that sense, I think to Tony's point, I think it's very important to make governments aware that community networks or municipal networks are enabling other types of access and enabling other type of solutions for places where the market ‑‑ because of the economy, so actually they promote these models as well.  They don't see these models as evil.  They don't see that the only way as Steve was saying at beginning of the session, that the only way for provision of services under resource are mobile networks.  So I think there is a lot of on our plate so that we are making the effort to play their game, but that they could play some of our game as well and that would have very important synergy impact and transforming the society into a way to be transformed. 

>> CHRISTOPHER GEERDTS:  I'm going to go a slightly more practical route rather than get into high level global politics and policies.  Luckily in my life I've lived through a couple of liberating markets opening changes.  The first was when markets were actually liberalized and governments used to manage telcos but told they you can no longer.  I used to have to drive to make calls to my family because the government wouldn't put a phone in my house.  Luckily that ended in the '90s.  That was a huge wave of change.  The next wave of change I saw when mobile operators started to compete with the fixed operators.  Especially in Africa that changed the horizon completely.  Not only because it offered mobility, because it offered multiple players.  I see a new wave happening now and it's Wi-Fi driven.  It's smart phone slash Wi-Fi driven.  And I don't want us to go backwards. 

I want us to go forward on it much and why it's based on Wi-Fi is purely as Steve said.  Wi-Fi is an accident of regulation, that there was a small slither of spectrum that was license exempt and everybody drove the innovation.  That's the reason.  And we want to build on it that's why everybody in this conference is walking around using their phones, connecting in many cases to the local Wi-Fi.  My roaming hasn't worked since I've been here.  I've been fighting with my operator for three or four days.  But it only matters when I'm on the bus between the hotel and here.  The rest of the time I control how I access.  And OTT applications are manifestation of that.  I've been the lord of my managing my own telecommunications.  For the first time.  When I worked at a mobile operator for ten years, we used to tell our customers you can take a phone with the package, and it's pretty much tied to them.  This isn't tied to anyone.  I can change any time.  So I don't want to go backwards. 

When I think free public Wi-Fi, I think let's open the market, let's allow more players, private sector players.  I'm not fully tuned into most of the dynamics of the United States, but the dynamics in the markets where I work are very different.  We want to encourage the private sector to play a much stronger role, because we want more and more people to offer services to compete.  And if one day we get to the point where we have one dominant player, then I will deal with it but not in the meantime.  That's my comment on Wi-Fi in general. 

Looking at rural areas, I think Wi-Fi plays another role.  I've seen networks extend for 300 kilometers providing commerce service over multihub Wi-Fi.  And I chair an organization which had 300 such members.  Not tied into fiber particularly, I was just mentioning it because it's a bit closer to public Wi-Fi.  There's a huge amount that can be done in the countries where it is allowed and providing massive high capacity backhaul.  And for short distances now they're talking about gigabit links.  So the technology is really going to another level and we shouldn't overlook it.  And not only does it allow connectivity in certain areas, but it provides multiple opportunities.  W

hen people think wireless access networks, they think rural areas.  Not so.  Cape Town has five providers making a living for providing alternatives to areas which even have fiber.  Johannesburg, the richest part of Africa, has multiple service providers providing Wi-Fi based links.  We need to think about wireless as a complementary technology at every level. 

Jumping to one of the other comments on the community networks in urban areas, I don't have a lot of experience except the one in South Africa where costs are very prohibitive these networks started popping up and they were very geek‑driven.  You could look on a map and see where people were connecting nearby to you and you could form mesh networks yourself.  You can get software, get a server repurposed from your old home computer and you could start connecting and it was becoming big and it's almost dead now.  It's limited to a couple of people.  And it's because the prices came down so much.  So it seemed to me that people really couldn't be bothered to carry on with those networks if the price in the market is right.  But that's just my own experience.  What it did do, and that was really exciting for us in South Africa, was generate a whole generation of techno‑savvy young people who went into Wi-Fi in different parts of the country and turned it into commercial operations.  But I didn't see a point in it coexisting when the prices were right in the market to go to so much effort.  But that's just an opinion of mine based on one set of experiences in about three different cities. 

In terms of the free basic model, the last one was about free basic and Facebook.  I like the idea which most of our Wi-Fi operators are doing of offering a certain amount, whether it be 250 or 550 megabytes a day per user and making you pay above that.  Innovation if you pay your rates and taxes in time you get a double allocation.  That pays for itself extremely quickly and I like free writing Facebook but a whole set of public interest portals and that should always remain free just like Wikipedia has created that, why not extend it to the wireless access layer.  And that's actually pretty much what these providers have done. 

>> STEVE SONG:  Thanks Chris. 

>> ALISON GILLWALD:  Thank you very much for that. 

I just wanted to comment quickly on a few things.  The smart city and where the community networks shouldn't be or couldn't be part of those, in fact in South Africa the real driving of broadband take‑up from the public point of view and the public policy point of view really happened at the metro level and provincial level national was taking too long.  Ultimately at national we did catch up and we put some sort of requirements for public Wi-Fi, but it's really the provinces particularly the two economically big provinces that have driven it and they're main cities.  Both in the Johannesburg model, smart city model and the Cape Town smart city model are these community networks. 

As I said they're not at the core of it, they didn't drive it or anything else, but they're certainly part of that initiative.  The interesting thing in the western cape broadband initiative which this is very much part of, is that the communities ‑‑ the underserved communities were given priority in terms of getting them connected and also with the public Wi-Fi, putting the public Wi-Fi in their model as Chris has explained, unlike the Johannesburg model, was that they would provide these public‑private interplays and they took different forms at the provincial level and private level but basically people were allowed to connect to the fiber that had gone out but they were given small subsidies in the one model to connect.  And after a period of time that subsidy would go and they would have to run the model. 

And the interesting thing is this conception of community in these modern environments and urban environments.  And this was some of the problems with our early universal service models.  That where you got these slide‑side driven models, at least there was a genuine community to run it or take it over or entrepreneurial model, it never really took off.  I think they face the same thing again.  The one community that's actually been successful in this model is Delft.  Because it's a completely dislocated apartheid bleak community they've got a very strong core and that public Wi-Fi network is the one that's successfully selling on data.  So they've got the 5 megabytes that are free but they are successfully selling on.  A lot of the other commercial models have actually collapsed.  They're not working that well. 

So I think actually that also relates to some of the universal service models and kind of learning from what we had before and the questions you were saying about market failure.  That when all the initial attempt to provide secondary universal access to the incumbent, letting another player come in to offer those services, that the incumbents have said historically were not affordable, when a second player said we'll go in, the incumbents said we can do it.  What we're finding now is something slightly different.  A very changed environment with different layers of complexity and this sort of goes back to the question of governance that we need to be thinking far more opportunistically, and what can be done. 

So at that point in time operators wanted to keep everything in a monopoly, et cetera.  At the moment they're in a competitive environment.  They were required to go and put this spectrum in some remote part of South Africa or something.  It's an obligation for them.  But if you actually said to them now, well, if you are not using that spectrum or you're not running an operation there but you've got that spectrum and you let that go and let communities who want to use it, but we're going to give you this high demand spectrum that you want, there are a lot of operators who are quite happy to do that.  A lot of them are wanting to get off the second networks.  We need to be far more strategic about it. 

And the final point about these parallel systems that we've got.  If we're speaking about in this environment, I do agree with you Mike these telecom interests working through the ITU system and that's where they can get their interest base represented, that's where government is representing them and then you've got to parallel system here and we're talking about innovation and regulatory innovation and using opportunities and they're not talking to each other.  And I think it's really incumbent on us to get us all in the same room more often.

>> STEVE SONG:  Thank you, Alison.  I'll just leave you with one thought.  I'm afraid we're out of time.  It wasn't very long ago that it required a national investment and a national operator to deliver services.  What we see now is the incredible potential of municipalities, of communities, of individuals to deploy communication infrastructure and what we need is for the regulatory and investment frameworks to catch up with that.  Thank you very much for coming. 

[ Applause ]

[ Session ended at 1:02 ]

Contact Information

United Nations
Secretariat of the Internet Governance Forum (IGF)

Villa Le Bocage
Palais des Nations,
CH-1211 Geneva 10
Switzerland

igf [at] un [dot] org
+41 (0) 229 173 411