IGF 2018 - Day 2 - Salle VIII - DC Connecting the Unconnected: Policies for emerging practices

The following are the outputs of the real-time captioning taken during the Thirteenth Annual Meeting of the Internet Governance Forum (IGF) in Paris, France, from 12 to 14 November 2018. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the event, but should not be treated as an authoritative record. 



>> MODERATOR:  This is a little after 9:00 and we only have one hour, and we have remote participation.  Out of courtesy to you for being here early, I would like to start the program and to get as much of the information we can.  It's our pleasure to welcome you here to the DC Connecting the Unconnected: Policies for Emerging Practices.  What is working and to catalog all the innovative ways people are connecting people to the internet.  And also, to find out what is working and what is not to try to put decision making on a sound basis instead of marketing pitches or different attractive ideas where people fall in love with a particular approach or technology regardless if anyone knows whether it works.  I'm delighted to be joined by Michael Kende, who is one of the co‑conveners.  The other two are Rajan Mathews of Cellular Operators Association of India and Helani Garrity of LIRNEAsia. 

As often happens, Steph’s (sp) scheduled for another panel in direct conflict so she will not make it.  But she is here at the IGF, and I'm sure will be delighted to speak with you. 

So we don't have very much time.  I would like to tell you briefly, if you would, about what the project is for those of you who don't know.  We are identifying innovative ways to connect people to the internet.  We've created a database of all of those initiatives looking at both supply side that is creating infrastructure and lowering prices and also the demand side addressing problems including digital literacy and gender inclusion and other aspects that remain as obstacles to deploying the internet. 

Second, in addition to all 1,000, we reached out and conducted case studies, interviews to get a deeper understanding and to pull information we can turn into financial assessments. 

Third thing we are doing is we are going to analyze them in a systematic way to try to identify trends and patterns and to make recommendations both in a context‑sensitive way.  Fourth, somewhat of a surprise to us.  As we talked with ministers, we discovered connectivity is wonderful, but the health minister wants to know what impact is on healthcare and education.  The finance minister wants to know what impact is on economic growth.  If you look at the academic literature, there's never been a true study done of what those impacts are at the level of social ‑‑ the gold standard of social science which are controlled -- controlled trials. 

We're, actually, then entering the phase where we'll be consolidating information and presenting it to key decision makers in governments and the international finance community and we're making all of the data collect available through a web site which makes it a resource open to everybody.  Cheryl, if you would. 

So as I said, the case studies, we're doing the interviews, control trials and lastly as part of the analysis, the analysis of cost.  Different domains, education, healthcare, gender, financial inclusion.  We're also doing regional breakouts, Africa, Asia, Europe, North America to get ideas how those things have in common. 

So if you would, again, Cheryl. 

Looking at the database, those of you who heard this presentation last year, we identified 750.  We're now at over 1,000 all of which are in the database.  I will show you briefly the information is in there to see if it will be useful.  We have 115 of the case studies done including some focusing on zero rating, health.  And we now have three controlled trials going.  One in Rwanda, one in Nepal and one in Vanuatu.  And we've presented in 12 countries over the last three years. 

So to give you the flavor of 1,000 entries, this is the first page of it.  You have the project name, the country in which it is, the region, stakeholder group of who is deploying it whether it's commercial, government or civil society organization.  Whether it's on the supply side, the demand side, the technology being used and the domain.  Healthcare, education and the like. 

And then beyond that, sometimes there's more than one domain.  There's information about web sites that document the project, where they are available, when it was founded, where it is located, how it is funded, and then a brief description of what the nature of the project is.  And, actually, some people we know were doing natural language processing to try to identify trends in the overall projects to find out where things are going.  If you are interested, if you want to identify projects in your country or identify a particular kind of technology, this is something that's easily searchable in a conventional database form. 

If you want to know basic statistics about the case studies, the largest number are from Africa and then by Asia.  Latin America has fewer projects.  Their connectivity is relatively good by global standards.  By domain.  The largest percentage is education.  Education includes both digital literacy training and support for school education.  Most are digital literacy.  The other large categories are community networks, commercial deployments and health.  And a number of categories going down in smaller numbers.  Looking at the 116 case interviews we've done.  Again, mostly in Africa and Asia largely following the same pattern as before. 

Cheryl, if you would. 

By domain, again, education, mostly digital literacy.  Community networks, commercial deployments, very similar in terms of what we saw in the original database.  What's interesting is a lot of these are being done by the private sector.  I think that is something that is overlooked and important role that we play.  Civil society and community networking groups are doing very important things.  And the volunteer‑led programs, I think, are critical as well.  What is interesting to me is what many people think of these as competitors or having some sort of adverse relationship.  To me, we all are working towards the same goal which is connecting as many people as we can to the internet as quickly as possible.  And in a world of limited resources, I think places like governments do not need to be spending money on places where commercial companies will deploy.  To understand how to use to target the interventions where they will get at the areas, where they will do the most good is turning out to be a critical aspect.  And we think the volunteer‑led community organizations have also proven critical in this space.  We have different scales.  Some people were wondering.  Some of them were national, local and regional. 

And we have a predominantly more rural than urban.  And then we have some that are doing both. 

And if you can relieve Cheryl of the responsibilities of relieving slides.  She is one of the graduate students working on the project as well.  She just arrived from the United States just now. 

And you have to do the little arrow under the number on the left side for WebEx.  Other side.  There.  The other left side. 

All right.  More about two thirds of the project as opposed to the supply side with a small number being both.  And so that's the basics about the overall database.  Now, if you would, the control trials are some of the most exciting ones to us.  It makes the connection between the actual bill out of connectivity and the goals of healthcare, education and economic development which are needed to mobilize governments.  Even health ministers who are sympathetic to the idea of increasing connectivity will ask how much of my budget should I commit to this?  How much support should there be?

They have many obligations and many things they need to deliver.  They need some information about how this actually works, and the impact it has on the key indicators.  So this is called control trials and called difference in difference methodology which is the best methodology ‑‑ closest for attributing causation that we have in social sciences. 

So we have an on‑going project looking at economic activity and education.  We have two new projects.  One going on in Nepal.  To see if it improves healthcare outcomes pre‑natal and post‑natal outcomes.  And this is being done with partners.  Vanuatu is using connectivity for remote diagnosis.  You have to begin by porting over a mountain to get to the shore.  And then you have to decide which of the two available hospitals you can go to.  And both of them don't have full service.  So knowing which disorder you have you have to take one‑day journey by boat. 

So this is a two‑day commitment.  It's actually very interesting.  They think it's helped at least 70 people in six months improve their outcomes.  And more importantly, it's served as a crowd sourcing fund‑raising platform to allow people who could not afford to make the travel to make it possible for them to visit the island.  We've done the preliminary data on both of these and go back and do the back-end study in the coming year. 

In terms of cost analysis which is a big part of what we're doing, I think a critical contribution, what we're discovering is 62% of the projects we've studied have no revenue model whatsoever.  Some of them are technical trials to validate a technology, in which case I understand it.  But some of them are not.  Without a revenue model, there's no way they are sustainable.  Unless we can find a way to take a free service and transition it to a paid service.  The people who have tried to do that have found it to be relatively challenging.  We have recommendations coming from that.  Interestingly, the demand side interventions which are more of them have more of a problem in revenue than the supply side.  About two thirds of the supply side interventions have some form of revenue.  Two thirds of the demand side ones do not.  The problem is if that's the case, they will only survive as long as the grant‑making authority supplies them.  Even if it's digital literacy platform, if you are going to expect it to be sustainable, it needs to have a way to capture revenue from the benefits it provides to try to create some sort of synergy and longevity.  And unfortunately, that hasn't happened yet. 

The primary source of funding has been grants.  Government funding is a large number.  It's based on the type of funding that's different representation.  The grant‑based models have much bigger problem doing this.  And what I would suggest to a grant‑making authority is instead of throwing grants at projects that seem like nice ideas, the grant making authority should insist they have sustainable projectable revenue models so they can validate a business idea that could actually survive when they stop providing support but could allow it to be produced and ruled out to other places once it's proven to be successful.  That would be the best possible use of that grant money. 

In a different size of scale also vary but by half in half in terms of that. 

All right.  So I'm not going to go through these slides in great detail.  But to show you some of the ideas of the domains we're in, this is one of two slides of educational interventions.  We have age numbers on the right and cost numbers.  They are varied widely.  It's hard to know what to make of these on the demand side.  To say something is a cheaper education program you have to also understand what the education program's goals are and how impact is likely to be.  That's hard to measure in a consistent way.  Not the least of which is if you talk to different countries, sometimes they have different measurements of what their own success is.  Another country measure student attendance.  And teacher attendance has become a major success. 

And one of my favorite stories, you'll learn anecdotes and the most entertaining parts, there's one deployment in India where they ask the students to take a picture of the teacher at the beginning of the class and send it to the ministry of education.  Unless the teachers there at the beginning and end of the day they don't get paid, it has radically increased teacher attendance in a typical smart phone sort of way that I think is increasing the accountability of the teachers. 

There are gender case studies and we've been supporting the best practices for them here.  We've been supporting on gender and the equals initiative co‑led by GSMA, and there are a number of interesting gender initiatives.  Some of which are more interesting than others.  Healthcare.  And agriculture which food security is critical for many countries.  Often, this is about educating, better ways to do things.  Occasionally but more like primary education. 

And lastly, a number of very interesting community networking.  There's other programs here, but this has been one of the great finds here.  There is a niche for the actors where they are providing a critical role community‑based activity building networks in ways that are tremendously beneficial. 

Project costs.  They range widely.  And we have average numbers.  They are quite misleading in terms of how much variance in there.  This is on the demand side.  On the supply side, what we're discovering is there's a wide variation in capital expenditure and operating expenditure.  One of the lessons we've taken from this, although we worry about capital, operating expenditure is probably more important.  If the deployment is losing money every year, if it's operating in the red, I don't care how much capital you threw at it at the beginning, it will die without additional funding.  What we're discovering, there's different things such as back haul.  Not just the local connectivity but to their ‑‑ not funding employments but providing the critical back haul infrastructure needed.  Many are relying on satellite not because it's a great way to do back haul but because it's the only back haul available in a general way. 

The cost data we have is based on interviews and requires a number of assumptions.  Four deployments have given us full access to financial data.  We can do good financial modeling on these four.  Using very different technologies but that's actually a good thing.  We can learn more about how they are doing.  Different power agreements, approaches, different back haul approaches.  Different ways they fund it.  And what we'll see is very different cost structures in terms of what they are doing.  How many people they are benefiting and we're hoping to turn this into a full-fledged financial analysis and what's sustainable and scalable and identify the critical factors each of them for success whether it's access to spectrums.  A number are doing back haul in spectrum to understand what the critical success factors are. 

So what we've learned is the cost varies widely.  Capital expenditures is quite high and that is a real issue we have to deal with.  Operating costs explained are critical and backhaul costs.  And a number of innovative ways people are pursuing to bring down operating costs partnering with local governments.  What's interesting too is there's essential part which is capacity building which is you have to give technical training to the community about how to continue to maintain the equipment.  Otherwise, it will operate beautifully until the next round of updates or the next patches come through.  If you don't train the people to do that, it won't sustain.  So it's not just about money.  And, in fact, many of these community networking groups, they have a lot more time than they have money and willing to invest in it.  I think there's a wonderful name they give it in one of the projects we studied.  They called it barefoot engineers.  They do the training to the trainers.  The people who will become responsible for developing and training additional generations of people in a scalable way that will maintain the project. 

So as I've said, we're entering the presentation phase.  The first three years we made 40 presentations on this work.  And that wasn't even our main dissemination mode.  We think in the next 12 months we'll be more aggressive.  I won't bore you with this.  Done it in a lot of different countries.  Previous IGFs in other places.  Been a really exciting project and the reception we get is positive.  At IGF we have eight sessions here as participants of different members of the team.  And two of them are really boring.  They are part of the dynamic coalition main sessions where we each talk for three minutes.  That's just the way it is.  Many of the sessions are excellent.  Participating in projects on gender, disability access, hybrid business models and overcoming barriers.  I will show you a slide at the end that will encourage to you come to all these.  What do we do from here?

Analysis on the cost models to try to give advice to people trying to decide how to connect more people in their country to the internet to find out what are the critical success factors.  We're going to do the breakout reports by continent, by domain, that is education, healthcare and agriculture for people interested in particular aspects.  We're going to continue to do the field work and a great deal of dissemination. 

Then lastly, these are three of the main sessions and the AI disability one isn't on here.  But we encourage you there will be a main session on development innovation and economic issues at 3:00.  Overcoming barriers at the same time unfortunately.  Hybrid business models is today at 4:10. The gender one is going on in another room.  You cannot attend that one as well.  And the disability access one is what time?

11:50 in what room?  Four. 

And, if you are interested, all of this data is available on 1worldconnected.org.  Only 71 of the case studies are currently posted.  We will get all 116 up as soon as possible and the database is available there as well.  I hope you found this presentation interesting and I hope you think that the work we're doing is going to be helpful to you in the important work you are doing in your home countries. 

So in addition to presenting the main results, it is my pleasure to serve as the moderator for this panel.  I would like to pass the microphone over to Michael Kende.  He is one of the co‑conveners of this coalition and a tremendous support of the project. 

     >> MICHAEL KENDE:  Thank you.  I certainly do see the value of this project and the work that you are doing.  I just want to say there's a quote by Lord Kelvin that says if you cannot measure it, you cannot improve it which captures this work.  He's most famous for the work in thermodynamics.  He calculated the bit rate for the first telegraphic cable.  Nobody knew what would happen over distances.  And he didn't stop there.  He also calculated the impact of different designs for the cable on profitability and revenues for the cable and invited on the board and was on the ships that laid the first cable. 

The feasibility of putting it down and the bit rates would be.  And I think that's really captured by this project.  We all have scarce resources.  If the technology will serve people.  But if it uses the money wisely.  If it's profitable, if they have revenues and sustainability and to be able to share this data is important.  So everyone can learn from what works and what doesn't work in these types of projects.  So congratulations for this insight and continuing this kind of work looking not just at feasibility but profitability and sustainable. 

     >> CHRISTOPHER YOO:  Thank you very much.  We are consistent in the spirit of IGF, we are insistent on recognizing the multi‑Stakeholder approach.  Michael was part of internet society and the technical community.  We have a person from business and another person from simple society. 

And I'd like to recognize Cheryl Miller who works for Verizon as another part of the business community.  Recently served her sentence and was paroled into being an attendee.  More importantly, deserves the credit for getting this project into my head.  This project began as a suggestion she made to me many years ago to which I said no.  And you've discovered that Cheryl Miller is a very difficult person to say no to.  So I will say that she has been very persuasive and I thank her for getting involved in one of the most stimulating projects in my career. 

     >> CHERYL MILLER:   Thank you.  That was a kind introduction.  I'm not sure I deserved all that. 

It's a pleasure to be here with all of you.  I really apologize.  I hate speaking with my back to people.  The setup of this room doesn't allow for anything other.  But the connectivity issue is something I've always been passionate about on a personal level.  I was one of the last students in my school to ever have internet access.  I grew up in a very humble environment with few resources and so when my grandparents finally got the money to actually buy a computer, I didn't have the heart to tell them it doesn't just stop there.  You have to then be able to be connected to the source. 

And so I'm very proud to work for Verizon, a company that's been very committed to improving access and also focusing on something that I think is as important, if not more important, building the skills and the knowledge around the resource once you have it.  So once you lay the pipes and once you have connectivity, it's really teaching people and making sure that digital literacy improves, I think, globally, sometimes we forget everyone in this room is privileged.  We have a real problem with digital literacy but literacy in many parts of the world. 

I just wanted to talk a little about Verizon's innovative learning program.  It's something that I'm very proud of that we've been involved in.  It's very focused on making sure that kids that have fewer resources have access to technology and skills.  Those skills that are really needed to be able to succeed in today's digital world.  Very different than when I grew up.  Free technology, free internet access and hands‑on learning.  Those are kind of what we're offering to these children. 

And it started in 2012.  We committed $400 million to this project.  To date, we've helped over a million students.  Our goal is to help 2 million more by 2021.  You can go on Verizon's web site.  You can learn more about the program.  And you might be asking how do you know it works?  How do you measure success?

That's tough.  But we've looked at standardized testing.  Students that come through our digital learning programs they score three times better at math and two times better in reading than students that have not gone through the program.  So we are, sort of, seeing results that are very positive.  So we're happy about that.  We also have a number of innovative learning labs.  In these labs we're teaching students coding, robotics, virtual reality, all kinds of interesting and fun stuff.  And we're also collaborating with a number of different partners.  And so we offer a number of grants for different projects as well.  So if you or your organization has a project that you'd like to apply for, go align and look up the information.  We're always looking for new ideas and this project is one where we're really proud of.  Christopher and the whole team have done amazing work.  When he talks about the multi‑stakeholder spirit, this is one of the few projects that really has embodied that.  I've watched them reach out to all different corners of the community to involve everyone in it.  I think that's why it feels so special.  It's not just, you know, one world connected project.  They also have a great booth in the village.  Please do stop by.  They have fun stuff they handout.  Throughout the week, please stop by and meet the rest of the team.  And follow them online to see the rest of the progress with their work.  So thank you very much. 

     >> CHRISTOPHER YOO:  Thank you for the support you've given just by your presence here and your support of the mag and insisting that I not say no.  So thank you very much. 

The other stakeholder group that is under represented at the IGF is the government.  We have two representatives of government to speak here.  Mr. Fehli?

We had him from the legal division in Georgia who I do not see here.  In that case, our last speaker will be Mohammed Azizi (sp) who will offer his thoughts from the government perspective.

>> Good morning, everybody.  Such a pleasure to be speaking here once again.  I think it is very important for the government.  After realizing the importance of connectivity, what to do next.  A lot of the time, accessibility gives us the impression it has different sorts of meanings in regard of the action lines of what we have to do.  I think it is the connectivity we would like to put as the top priority for the government. 

Now, it is the deployment of the physical infrastructure to make sure we do have the effects while the infrastructure is deployed.  It is important at the second layer, let's say, we have the fiber optics, what to do next.  The local government bodies play an instrumental role.  Since they are mainly located in rural areas.  The government service deliveries ‑‑ then what's the advantage of having it?

Now, we have utilized the universal services fund vastly and we have made sure that we undertake a number of initiatives starting from subsidizing the deployment of connectivity across the country and particular to look into the issues of the marginalized groups, gender gap and communities with disabilities and then, above all, it is the rural areas because of the difficult security situation and everything. 

So the use of the universal services fund is of immense importance for us.  We have been very pragmatic and made sure that we take advantage of the experiences of the countries within the region and then even beyond the region.  Whenever I get an opportunity to attend one of these events, I do speak a lot on how to improve the usage of the universal services fund. 

The second thing, which is very important, is affordability.  The cost of access.  Now, it is easy to say we have connected the whole country.  We have fiber optics.  But, at what cost?

I'm talking about the perspectives of the cities and country which is landlocked.  We don't have any connection to submarine cables.  So we have to go through the other countries.  You won't believe what they charge us.  Immensely high.  It used to cost us over $5,000. 

So the other perspective that comes into mind, what sort of innovative policies could we introduce to make internet affordable for the people?

Again, poor economies, raining from 600 to $1,000.  It is not an easy thing to make sure that the access and the use of internet becomes universal in these countries.  Likewise, I would work closely with the private sector and I do encourage them that they have to also realize that the whole technology is changing, and they have to also remodel their businesses.  Like, 80% of their venue generated is from voice.  Heavily on the voice traffic, they will not be interested to be an instrumental role in the promotion of the data services in the country. 

Do I have any time constraint?

Okay.  Thank you. 

The next important thing is which is very much related to my job.  How to improve the environment.  Now, again, it is easy to say policies, regulations, procedures and legislation.  They are very important.  However, the most important is the implementation of these regulations.  And that's why I've been emphasizing a lot the regulators need to be independent entities.  Yes, we are part of the government but at the same time they don't need to be part of the executive branch like the ministries.  And for that, I'm very much thankful that we understand the importance of the independence regulator and he issued ‑‑ fully independent and no longer part of the executive branch as such.  However, it has been only two years we have become independent and still we are surviving or pretty much not achieved the goal to put the private sector in the lead for the development of the sector. 

Now, if we are serious about connecting the unconnected people, it cannot be done by the states.  Under the internet coverage and all these things.  So what I would lastly say that this is mainly from supply side.  From demand side also.  We talk digital literacy.  No, we first talk about literacy.  Then we are talking about 80% penetration of the internet.  So it is very important that we make the platforms available for people in order to receive the government services through them.  And if the e‑services will not be available, it will be difficult.  And likewise, we have to provide the industry opportunities of newer businesses.  We have already started the repayments for the civil services.  They are top priority because they are 50% of the civil services.  And we hope we will bring 100% of the payments through mobile by 2020.  From the regulator perspective and from the policy perspective, we really want this sector to flourish. 

And the last message which I understand is that it is not the job of a single actor.  The states are instrumental.  If they do not show a political willingness, nothing would happen.  

However, they can only create an environment ‑‑ only capable of controlling the enablers.  They have to really -- we as a government, this is the advice, we have to learn how to work with the private sector and how to work closely with the manufacturers and everybody else.  So this multi-stakeholder phenomena may sound something in theory but practice.  I have seen it on projects that we worked with a number of different groups.  It has been a great success.  And tomorrow I will be speaking more on those initiatives.  But, again, I'm very much happy that I've been given this opportunity to share some of my thoughts with you.  Thank you very much. 

     >> CHRISTOPHER YOO:  Thank you.  We often think as governments as a regulator or funding source.  We're understanding they serve many roles as a service provider, an anchor, tenant of connectivity, as a provider of demands driving services and a source of collaboration among different stakeholders, these are very important rules. 

I'm delighted to say in a short hour, we have managed to cover all the stakeholder groups and we still have about 15 minutes for contributions or discussions from the floor.  I will tell you our work is ongoing. 

So if you have other case studies or projects you know of that you would like to see studied and included into our work, we would welcome the opportunity to hear from you.  At this point, I will ‑‑ [ audio from another room ].

     >> CHRISTOPHER YOO:  We're getting crosstalk from another channel.  I don't know. 

Much better. 

So if anyone would have a contribution they would like to make or a question about the project or any of the speakers, please raise your hand. 

     >> AUDIENCE:   My name is Nicholaus.  I just wanted to say thank you for the presentation.  In relation to the business models and the social impact studies, working during this year in a global research.  But it's actually going to be presented today at 1:30 in 10.  The presentation is going to be about this report.  It's a fifty-countries report in addition to the initiatives being done.  And has gender, engagement, cultural implications and all within the studies.  So I encourage you and the group to go there to compliment the outcomes of this decision. 

     >> CHRISTOPHER YOO:  Thank you very much.  Please. 

     >> AUDIENCE:   Good morning.  I'm Sean from University of the Western South Africa and representing the community network project.  I know you have covered it before, but lots have happened in the last year.  You said there's government representation.  We have deputy minister for telecoms and postal services from South Africa here.  There's lots of innovation from the policy side so governments driving an open network and policy around that.  We're trying to advocate for spectrum and other issues around that.  And government's very keen to support the community network model through program.  A lot of things happening that supports the community network model and will feed that further into your project. 

     >> CHRISTOPHER YOO:  Thank you very much for also the excellent work you do in making this possible.  We are only studying what you are doing on the ground.  Would you like to make some remarks?

     >> AUDIENCE:   Thank you.  Good morning.  My name is Debra Abrahams.  Just to add to what Sean was talking about, we ‑‑ in order for us to breech that divide that we're talking about.  But in your inputs after the last presentation, you made a point where you say we turn as a regulator which is not the case in South Africa.  Where we say, you know, you can't do it alone.  Where we have a Stakeholder forum.  That's why we rely on the support of the private sector.  After that, introducing incentives that will encourage the private sector both big and small businesses to say what is it that they can do and what is it that we can do as government in order to assist them to enhance the work that needs to be done.  One of those is the issue of the location of the spectrum.  Put a clear determinant that as much as we're going to take, we have to open up space for new entrants.  But to make sure that we're able to cover a large vast of the areas. 

As I'm talking to you in South Africa right now in terms of cover reach, we're not there.  But that does not translate into the demand that will need to be created.  As government, we have responsibility not to just provide connectivity or assist the private sector to provide connectivity but creating this demand.  And one of those is to ensure in the areas where we have covered, we make sure to provide educational training skills to empower small business what is it you can use the internet for and how it can help you achieve what you want to achieve. 

If you are in the production center, how can you make sure that you can ‑‑ to say we share the content and that is an opportunity.  For us, it was a twin task to say provide skills at the same time, connect people, provide skills but also incentives including financial ones.  And, also, talking to the regulator that was mentioned earlier.  Very important for the regulator to be independent.  We have an independent regulator in South Africa.  Our system of operations in terms of budgets and reporting or accountability, it has to be done via an executive to parliament, although we do provide policy directive.  It is within our interest of government to make sure that our people are better served.  Our people get the best.  On top of that, these resources are equally or at least in a qualitative manner.  I will pause there for now.  Thank you. 

     >> CHRISTOPHER YOO:  Thank you, Madam.  It's an important thing we have.  I've seen this in the United States.  Governments used to look at companies as opportunities to get funding.  And now, we're seeing a change all over the world where we understand true partnership means understanding all in the actors who are essential to make this work and understanding what is important to them.  And making sure that we come with solutions that are responsive to the needs of every multi‑stakeholder group.  And only by understanding the other groups and really finding a compromise that works for everybody can we achieve the goal we all seek which is universal connectivity of the internet to the remaining half of the world that has no connectivity.  Cheryl. 

     >> CHERYL MILLER:   Thanks.  Yeah and to just follow up on the really great points that you made.  Also, I think we need to think about the importance of local content, not to lose that.  And, also, as we are moving into this new digital world, to be mindful that not all languages are available online either.  And so there are some languages that are actually fading.  So what is it we can work on together to prevent that from happening?

And how do we make content most relevant, so people can receive what it is we need to receive in order to be a successful small business owner or successful student or successful farmer for that matter. 

     >> CHRISTOPHER YOO:  That's an important point.  Turns out the lack of access to local content is one of the main barriers to adoption approving relevance.  One of the case studies some of the threads that have come out, we talk about capacity building, the need to train engineers.  There needs to be capacity building on the local content side as well.  If people are going to produce local content, critical mass helps there a place where people can come together and share and build as content builds on other content frequently and having access to that as part of a larger dialogue is an often a neglected part of a plan for deploying the internet.  If there doesn't have that sense of community, you don't get the local content.  Without the relevance, you don't get the adoption.  It becomes an interesting growth of the policies we need to intervene.  Please.  Then you, Michael. 

     >> AUDIENCE:   Hello.  I'm Damian from Argentina.  Special interest group of internet society for the community networks.  I want to pay attention to one thing.  When we talk about unconnected areas, there are also areas ‑‑ not only rural areas that are unconnected.  There are also areas in the city unconnected.  We are working those areas.  Work in the capitol city of Argentina.  They are dangerous area.  They are called ‑‑ right next to the next region, that's why we are working there.  What is important is also the local content you mentioned.  We work in that topic very much.  We make work production we want to bring to the internet node not only consumers but producers of contents.  If you want to know a little bit more, I'm available after. 

     >> CHRISTOPHER YOO:  Thank you for reminding us it is not just a rural phenomenon.  The high percentage, the significant percentage of projects that are in unserved and underserved urban areas which is absolutely critical. 

So we have Michael and a remote participant who will speak after that. 

     >> MICHAEL KENDE:  I just want to build on this point about local content and note that most countries do have local content.  We did two studies to show it's not just having local content, but local hosting of the content is critical.  Most web sites that given to a developer who puts it in London or New York or wherever.  It's a little bit cheaper or more reliable.  The result of that is it costs a lot of money to bring it back at $30 per megabyte.  So it increases the cost for the ISPs and it's a lot slower as well.  So a lot of people will just give up because there's that latency, hard to do video or interactive things.  We did the two studies where we showed the theory of this is much better to put it local.  And we could measure the impact of moving the content locally and interviewed the developers and they were all very happy and helped build schools and local hosting skills and local hosting revenues.  Usage shot up because people then get fed up because the web sites weren't loading. 

So when you are thinking about the local content to make sure it's hosted locally as well to help build up that infrastructure and skills development and to make the content more accessible and cheaper to access. 

     >> CHRISTOPHER YOO:  That's wonderful.  It's been the main driver before doing ISPs regionally.  There's another great study in Bolivia where everyone inner‑connected the ISP.  There were no hosting capabilities, it would simply travel and go back internationally again.  It becomes an important part.  The negative side of that is we found what happens when that doesn't happen.  You've already been represented here.  Delighted to have you here.  We have a remote participant. 

     >> AUDIENCE: [ Speaker not on mic ] -- in terms of the region itself.  Most is in Spanish and often ‑‑ so we were glad to have that. 

     >> CHRISTOPHER YOO:  Okay.  Any other comments?

We have time for just one or two more. 

Well, we've reached an appropriate end coming up on the hour.  I would like to thank you for coming today.  As we said, there are other events and many different parts of this to connect people.  As I mentioned, there's a disability access track as well.  We're speaking at one of three in a row which is another part of neglected ways we do not always serve the communities well in terms of connecting more people to the internet.  We encourage you to continue to follow us on the web site. 

We will continue this work into the next year, actually.  We would ask your help.  If you have case studies that you know of that you think we should include in our work, we would be delighted to meet with you. 

If you come by booth number 1 or see us afterwards, we'd be delighted to make those connections. 

And second is we are looking for venues to discuss the results in the coming year in 2019.  And what I will say is you, the community, know best of all where the most effective place is to communicate with real decision makers who would benefit from the kind of information that we're pulling together.  So we would really welcome your input on what you think the most effective locations are.  They are often the people working in region are the people who know the best where those venues are and where you can have the kind of communication that we all need to achieve the goals that we share. 

With that, I will close this session.  Thank you all very much for coming.