This is now a legacy site and could be not up to date. Please move to the new IGF Website at

You are here

IGF 2020 - Day 5 - DC Lessons Learnt from Evidence-Based Research

The following are the outputs of the real-time captioning taken during the virtual Fifteenth Annual Meeting of the Internet Governance Forum (IGF), from 2 to 17 November 2020. 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. 




>> CHRISTOPHER YOO:  As the dictator of this panel, I was thinking you could share with us the order.

>> FRANK ANATI:  Starting the streaming now.  So we can begin in 10 seconds.  We are live.

>> SHARADA SRINI:  Thank you, everyone who are here and who will be watching.  This welcome to the Dynamic Coalition on innovative approaches to Connecting the unconnected.  This is a Dynamic Coalition that has been part of the IGF since 2016 onwards.  And we have had sessions so far on how we Connect underserved communities, profiling case studies of innovative approaches that have been tried all over the world.  Today's session's title are overall database has had more than 1,000 case studies and we have covered in depth case studies of over 125 cases, which are all available on a website called 1 world connected.  We here today we'll talk about some of the lessons that we have learned from studying these case studies from all over the world and we have Amidst us a very distinguished panel.  I will introduce them briefly at this time and then I will open the floor to them.  I wanted to have some house rules in terms of how this discussion will go.

First, we will have introductions of the panelists by myself and then we will have presenting as along three dimensions.  Digital skills, community networks and health care.  Once for each segment of the session, we will have a presentation by a member of the coalition and a commentator by someone who is an expert in that space to make sure that we have and reflect perspectives about what we know about these things in terms of Connecting underserved communities.

After that, we will have a dedicated segment for Q&A as well as any comments that commentators have for this session.  Please post them in the Q&A section if you are using the chat, please change your default to all panelists and attendees so everyone is aware of the questions that you're asking.  As a rule, we would like to do all the questions and comments at the end.  Now, and if there are clarification questions, the panelists can type answers as we go along.  Thank you very much.

So today on behalf of the Dynamic Coalition's panel, we have Christopher Yoo.  We will be presenting our Dynamic Coalitions work on digital skills.  Commenting on that is Clair.  She will be commenting on the additional skills segment and the lessons we have learned.

On community networks, we have Leon Gwaka who is a post doctoral fellow who willing talking about his work and learnings from the community networks segment of the session.  And commenting on that, we will have Dr. Leandro Navarro from Catalonia who has been involved in one of the, largest and successful networks.

Then we have Dr. Haseki.  And commenting on that will be Dr. Sarbani is working on part of her new role.  And I would therefore now request for the first segment to kick off Professor Yoo.  Over to you.

>> CHRISTOPHER YOO:  Well, thank you very much for introduction and for launching us off.  I will note in the chat there is a reminder that this is being recorded and the event is being held under the code of conduct under the IGF.  As one of the co‑conveners of this Dynamic Coalition, I am delighted to have the opportunity to share what we have learned and to really focus my part will be on the digital skills programs.  Just so everyone ‑‑ just to frame it more broadly since we've at the beginning, the whole Dynamic Coalition and the work we're doing is framed by the idea that we need evidence‑based decision making.  We find that there are many innovative efforts going on around the world.  Many of them have no measurement whatever.  So we have no idea what is working and what is not which forces every project to reinvent the wheel and make its impossible for us to learn from prior projects and to take that learning forward. 

Secondly, we discovered the places that did have the information such as certain international organizations couldn't really do the evaluative work because of the nature of their role as an international organization.  They couldn't take the position that certain projects were more successful than others because that would jeopardize their political standing.  So they encouraged us to do this work.  And you have seen three different aspects of our desire to push this forward so people are not just giving you their opinion or even sometimes people are pushing the business model or product they have available to you to convince somebody.  Policy make was, regulators and everyone will use the information we pull together to solve the problem we're all here trying to solve, which is to insure that every global citizen enjoys the benefits of internet Connectivity.  So specifically, we generated a database of over 1100 Connectivity initiatives.  We contacted all of them and we have done over 125 case studies ‑‑ today my segment is about digital skills.  Just to let you know, this is one of the largest segments in the overall database consisting of 339 entries and 44 case studies therefore well over a third of the total amount of who we discovered.

So we discovered in writing this up that this very little good information.  A lot of opinion about what makes for digital skills, what are they and what makes a good initiative.  So I'm going to present to you what we found.

First is the regional distribution and the scale.  As should be probably unsurprising, a large percentage of these are concentrated in Asia and Africa.  There are additional skills initiatives in Latin America and North America, Europe and Oceana at lower levels.  We are always reminded and we'll talk about this in terms of vulnerable populations.  There are aspects of this in terms of low‑income and gender in terms of senior citizens and disabled ability access where we need some sort of digital skills initiatives often in every part of the world no matter what the socioeconomic status is.  And we see the majority of them.  75% in Asia and Africa.  They occur in different scales.  Many are good 70% of them are larger than local either on a regional BASIS or national BASIS.  And this is an important decision when launching such an initiative to try to evaluate the scale at which you would like to operate it.

Target populations very widely.  43% the largest proportion focus at students often through in school training or specifically after school training.  20% to 6% are target on the BASIS of gender to try to increase and empower female populations.  13% are done youth out of school and as we mentioned earlier, there are other lesser represented groups including seniors and disabled which are occasionally targeted, but generally not a primary target in our database.

Geographically, roughly speaking it is 50% in rural areas, 25% urban and, of course, many initiatives of the national ones will stand both.  So we do so a disproportionate emphasis of the training appropriately so in the rural areas.  In terms of how it is delivered, this is an area where very little study has taken place.  The question is:  There's a great question about whether do you it through a computer as depicted in this photograph through a mobile device or through a handheld device or tablet.  And there's an obvious question about reach and in terms of cost.  If you can do it through mobility based devices, you can reach a lot more people quickly and do it more cost effectively.  We do discover just under 2/3rds of the training is computer base.  And second, do you do the training in person or online through some sort of streaming situation or through an app and we discovered that 84%, the vast majority of the training is delivered through in‑person training.  The content varies widely.  They can be everything from basics training literacy to employment and work related training, education, entrepreneurship, social skills, financial literacy, eGovernment, health and agriculture.  This is in decending order of importance.  Basic I‑training being the most common all the way through employment education.  And this raises an interesting question.  We talked about digital skills as if everyone knows what it is and it's a unit.  There are important choices about how to focus them and here it's not just a question of one versus the other.  You probably need some mix of a wide range of subjects.  It's a question of:  How do we marshal scarce resources and attention to create a portfolio of content and approaches that serves the goals of society.

In terms of how large the programs are, there are ‑‑ they're different whether they're focused on individuals or focused on institutions.  There are very small number on a very large scale that's over a million programs, a million programs targeted over a million people.  But what you see is a significant number of medium sized programs targeting over 10,000 to a million over 1,000 to 10,000.  So we see very large scale projects or medium scale projects that are very much part of this.  And in addition, there are some target the institutions, a large number of targeted and very large number of institutions and other at a smaller scale.

One of the most difficult problems that we face is actually how do you make the programs sustainable.  Many of the times 2/3rds of the time they're funded by grants.  Some by governments and some by subscriptions.  And with the ones that are not funded by subscriptions, their sustainability depends on the support of the funding agency N. that sense, we are encouraging funding agencies to do that on a conscious BASIS.  Know if you need to keep providing funding in order to sustain, you need to know that in advance.  Most by NGOs and academia, one of our panelists SAR bane is one of the people who is doing it from an academic stand point and is a wonderful person who is also doing community networking.  Different programs use different impact assessments and different indicators.  Most use of qualitative measures.  Some use quantitative measures.  The qualitative ones tend to be surveys and other tend to find metrics and smaller do both.  Impact indicators with employment, academic performance, empowerment, financial wellness and health.

So there are a number of challenges which we asked them to talk about.  The biggest one is lack of sustained funding.  Others are electricity access, lack of man power to keep systems operational, access to devices, Internet access, resistance to the community and the need for local relevant content.  So in short, we are hoping that on a work we're doing we'll lay a platform for success.  And what's critical to us about doing evidence‑based decision making is develop the tools to evaluate impact to find out what is working and what's not so we can learn from initiative to initiative and improve.  We have to include to make sure people are deploying new projects if they include resources and time and effort to evaluate their success.  Otherwise we can't progress and we can't actually validate what's working.  You have to identify the types of training whether it's as I said basic skills or other types, which populations to train, determine the best mode for delivering it in person, through apps in the like or versus classroom, computer based, mobile based, identify the curriculum needed for each type of training, look at sustainability, analyze ways to overcome challenges and curate a repository resources that others can draw on so every person doesn't have to create them from scratch.  If you are interested, here is our website and thank you very much for your time.  I turn the floor over to Clair.

>> SHARADA SRINI:  Thank you for that presentation.  I now invite Clair to present any comments.

>> CLAIR SIBTHORPE:  Great.  It was great to hear that presentation.  Really important to have data on this.  Such an important issue.

I will make start with a couple comments about the broader on why we USS on it and what we learned in terms of responses and points that you make.  Right now there has been great strides in delivering Connecting the infect and the broader topic ‑‑ Connecting the topic.  There's a coverage gap of people who are live outside areas not covered by mobile broadband networks, but that's about 600 million people.  But much more importantly, there's a usage gap.  So when we look at ‑‑ so we have a lot of data on mobile Internet and mobile usage.  So there's approximately 3.4 billion people who live outside areas covered by mobile broadband that are not using it.  So basically so there's coverage, but they're not using it and that is six times bigger than the coverage gap.  So when we look at what are the reasons.  Why are people not using Internet or digital included and we do national representative surveys and we have a lot of data on this is that digital skills and literacy is a top barrier to mobile Internet use among people who have a mobile.  So this encompasses both traditional literacy as well as the ability to use a phone and use the Internet.  So clearly tackling digital skills is an important issue and making sure people can effectively access and use services.  I think it was really great that you highlighted gender and disabilities.  We again see these are segments of the population where there's a lot of significant gaps.  So for example, in the gender gap, we see since 2017 it is declining, but women are 20 are less likely to use mobile Internet in low (inaudible) countries.  There's a gap.  Similarly we're doing research on people with disabilities and we're seeing again gaps there, significant gaps there.  It's great and I would encourage anybody who is looking at digital skills to sort of look at who is not, who does needed skills and who has not been reached and try and focus on that.

So just to respond to some of the stuff you also said.  You mentioned that a lot of the delivery of digital skills training is computer based.  That's also when we do our landscaping research and our work in this area.  We're seeing that too.  I think that's a challenge and again, when we do basics, we see it according to the IT U for example.  Mobile accounted for 87% of broadband connections in developing countries.  In our surveys again, we found that a high proportion of respondents access the Internet exclusively over the mobile.  They have no way of seasing it particularly in ‑‑ what is fighting and the training is not delivered on the device they are using.  It is sort business, but when it is not delivered on the device, it doesn't give them the skills they need to access and use and safely use digital services they need.  So when people are receive training on, for example, computers and laptops, they are getting stuck on app islands.  They don't know how to have the skills to effectively manage and access the Internet on the phone.  In fact, that's training not giving them the skills they need to use it.  I think an important issue is think about groups you are targeting and trying to reach and what are the platforms and tools they're using to make sure the skills and training is appropriate to that.  So ‑‑ a number of the towns we also see when delivering digital skills programs and not surprise by what you presented in some of the barriers is that again our research and our work in this area shows this cannot be looked at in isolation, but it needs to be looked at in the context of the range of barriers of them trying to use digital services.  I would call again what our research is showing is affordability.  That is a primary barrier to having a phone that is also a second biggest barrier to using the Internet after digital skills.  If people can't afford or access the devices, you know, the digital skills training isn't going to be very effective.  There are barriers around concerns and security issues insuring this relevant content and services are available.  I would say that again in our research though, we're seeing that video consumption has been rapidly growing in the last two years.  It is much higher now.  But that ‑‑ that means that it is providing an accessible way to access content particularly for those who have low literacy and digital skills.  It's also an important thing to think about we can't replace face to face training, but sort of some of this digital training material and content is useful in the context of COVID where people are constrained and we are seeing much higher content to produce video content amongst some of the populations that we work with in low and miss countries ‑‑ middle countries.  We have mobile Internet skills to training toolkit, which is focused on getting people who just have the basic skills to access in youths the Internet.  It's freely available.  It's been (inaudible) by many countries and reached over 2.2 million citizens that we're aware of.  But we developed the initial modules looking at again taking every research evidence based approach research what people wanted and the user center design approach and identified that as people were looking for one of the basics on the Internet, how to keep themselves safe, how to understand cost and reduce costs as well as key modules on apps.  So that's an area we started to develop contempt on, but we're now currently doing more detailed research to understand than digital needs and capabilities of people in Africa and Asia.  Understand the journey from people when they first own a phone to use Internet to use it more widely and we'll be sharing that.  So basically I agree it is critically important we need to focus on this issue.  It would be great to see when you present next year if we can have more increased consideration of mobile given how important it is as a tool for people.  And in fact, because we think it's such an important issue, we are ‑‑ we think that we need to collaborate more on this.  We're in the process of establishing a community of practice.  We'll call it mobile digital skills and alliance to drive collaboration and anybody who is participating here and interest in being part of that, we feel we node more collaboration, research and evidence.  I will say that I agree with that.  A lot of what you're saying I think it is really important and I think about it in the context of the wider barriers if we're going to be success envelope delivering these initiatives.  So thank you very much.

>> SHARADA SRINI:  Thank you so much, Clair, for those comments.  If there are questions on the additional skills with the fining and commentary that you just heard, please do post your questions in the Q&A panel and we will try our best to have discussion at the end, but also if there are clarification questions, we will take them through the course of the presentation.

We have a slight change in order and next we will actually have the health care segment and I would like to invite Dr. Haseki especially in areas that understand the unconnected and we'll have the community network section after that.  Muge?

>> MUGE HASEKI:  May I share my screen now for my presentation?

>> SHARADA SRINI:  Yes.  I think you should be able to.

>> MUGE HASEKI:  All right.  It says sharing screen has failed.  Let me try again.  All right.  It works this time.  I would like to present some of our favoring on health.  So we over the past couple years, we have generated 13 case studies on health, five of which were on maternal health.  And then we decided we wanted to know nose five maternal health services skilled at different levels and also had different outcomes.  So we were curious what made them in what ways they were similar and different than why they scaled at different levels.  They started at similar timeframes and some of them were more successful in terms of reaching their own goals and changing how behavior of pregnant women.  Today I will focus on too often given our 10‑minute timeframe and compare and contrast to maternal health implications in south Asia.  We wanted to know what challenges they face, what challenges woman face in those areas and how maternal mobile applications can address those challenges through design and change how behavior of woman intended ways.

So I'm sure everyone in IGF and in this panel are aware of the increase in digital infrastructure all over the world.  So here you see how digital infrastructure has expanded over the past 10 years.  The two maps from GSMA show those expansions vividly.  And that also led increase in the mobile health initiatives over the past several years.  Here you see maps from 2015.  More than half a million have been implemented over the past 10 years.  But unfortunately, only around 700 of them are currently active.  So this also let us our research questions why mobile health applications are not sustainable or why they're not working as they intended.  And this digital increase in digital infrastructure, there's a failed gender gap still growing fast.  And the second thing we were curious about was why the gender gap is growing despite the increase in digital infrastructure and mobile health services especially those mobile health services aim to improve health of woman.

Here I would like to compare to mobile health services in Bangladesh and Nepal.  They ever both neighbor countries in south Asia.  Those two maternal health applications they had the same goals.  They both provide pregnant and postpartum women and have families with voice and tact messages.  So (inaudible) provides an access to a help desk.  So we wanted to know about those two maternal health applications.  They started around the same time upon (inaudible) in 2010 and (inaudible) in 2011.  But they scaled at different levels when it became national within the 3‑year timeframe and Amakomaya became regional with only 7,000 and (inaudible) reached 2 million subscribers.  So we wanted to know why they scaled it differently within some other timeframe.  And at the same time, in what ways they change behavioral given gender gap is growing and how they address some of the challenges that are associated with woman in those areas.  Sorry.  So we identified some challenges of woman in rural Bangladesh and Napal.  And those influence of gate keepers in how behaviors and decisions low literacy rates and limited leisure time due to overwork load, barriers to health centers and high 99 of local languages and patriarchal norms.  So we wanted to know how those two mobile health applications tried to address some of these challenges through design.  So both of the applications provide SMS and SMS reminders as well as voice messaging.  But different than upon ‑‑ different an Aponjon receives messages twice a week.  In addition, they also provide SMS for gate keepers, timing choices, help line and self‑registration as well as they are available in three different local languages.  So here we kind of look at how they address some of the sociocultural challenges to their affordances.  And Aponjon and Bangladesh provide literacy in Amakomaya.  There are norms by providing weekly SMS messages to gate keepers.  There are messages to address leisure time of woman who are working and spending their time with domestic responsible (inaudible) working in the field.  They address the low level access to health centers and limited health care workers in those areas.  And also provided local content in three out of 38 languages unlike Amakomaya, which is only available in one language to address the high number of local languages in those countries.

So in terms of results, again Aponkon reached 2 million subscribers.  They were also able to change how behavior of woman ‑‑ help behavior of woman 67% compared to Amakomaya that changed behavior of pregnant woman 60%.

So our key aways, access is not sufficient in including and marginal environments.  Taking an approach can help develop more relevant and sustainable programs, but what I mean by that is integrating gate keepers, husbands and opinion leaders in the community.  Single barrier through design may not be effective or cost effective.  Just providing and addressing financial barriers for woman in those areas providing them free mobile health service or free data is not sufficient, but they also need to address other challenges like gatekeepers or lack of time or lack of leisure time.  And all these areas and challenges they need to address them all at the same time rather than addressing just one of them.  If they want to be sustainable or effective.  So thank you.

>> SHARADA SRINI:  Thank you so much, Muge.  And we had some issues, but I'm glad to note it has been resolved.  If possible, Dr. Belur, your comments next.

>> SARBANI BANERJEE:  Thank you.  So yes.  Mobile health applications my take on it is that mobile applications (inaudible) incredible promise to include health care delivery.  And it is becoming an important mechanism to improve maternal (inaudible).  In locations and connectivity, mobile phones (inaudible) dependent to receive messages (inaudible) as we have seen in the two mobile applications that use (inaudible) in Bangladesh and in Nepal.  Mobile technology has solutions that are being utilized for disseminating information, monitoring and tracking; however, the lack of providing training support to community has helped us.  Because that's one of the key issues that women in rural areas often face that because there have (inaudible) levels, they are unable to get the support from the community health workers because that's the go‑to person in the villages.  Mobilizing communities is also something and (inaudible) interpersonal communication with community members because if you have to go in to the villages into the rural areas, there has to be a certain amount of interpersonal communication within community members.  We have to as you have mentioned it, in your last slide that we have to take into account the amendment of the community leaders and others.  So there is a great dependency on that ownership of mobile phones.  You know?  If it is like a health application and (inaudible) applications have to work.  There is a great dependency on that mobile phones.  Especially woman and (inaudible) access information that women need because men often on one hand don't have the mobile phone, the devices, but on the other hand, they have auto mommy to access information because of the (inaudible) you have mentioned.

So some of the ‑‑ these are some of the stumbling blocks for the sickness of (inaudible) in various parts of the world.  And I would like to mention that in sense of social areas in a world of Connectivity of (inaudible) services and innovative ways by which ML service can work hand in hand with national health services.  How can they work well low Connectivity locations.  So (inaudible) deep into these questions have been working together with remote rural areas in India and health information and dissemination if you go to each.  You can go into Connectivity.  And in the community network that's been seated, rural (inaudible) village and (inaudible) by association and communications as here.  This is a project that I'm a part of now.  So in community, (inaudible) community users have local access servers.  We host (inaudible) spoken in (inaudible) on maternal health on the local access server and the offline measure that we have set up in the community, users the government access and access in that local mesh.  At the health information, users (inaudible) that information is culturally relevant.  For the pregnant and new mothers are disseminated into the community.  And there is also not only one way of communication between us and the community, but also that new information based on the experiences of the mothers like the use of particular local grown (inaudible) which can please the weight of the babies or information about community, vegetable garden that can enable the right amount of nutrition pregnant mothers and new mothers should have.  So these are information that women in the community can last about.  And all this information is shared in local language by the community for the user community areas and can be accessed by all through the local access server.  So it is not only for women who are being empowered with this information, but also the husbands are coming into the (inaudible) by understanding that okay.  These are the things they should give to the women in the house and when they are pregnant and after they have given birth to babies.

Training the community health workers is also another important mode which we go through these and continue remains for a long time within the community amongst the women because these information for (inaudible) tutorials that we create and download it and (inaudible) offline mode by the women in the community.  And in order to make the reliance on the mobile phones less, what we do is (inaudible) we use a lot of ‑‑ a lot of this offline ‑‑ a lot of community review content is information that is generated by the community and some of it is like broadcast (inaudible) television by downloading detected offline.

So during the COVID times, what we see, what we have seen is that.  This type of content we have made specific content related to best (inaudible) during COVID because the new mothers, the mothers who give birth to the children, they're asked by companies to not feed their babies with breast milk because that has a great chance of the baby acquiring the virus.  So what happened is that we made special tutorials on breast feeding for new mothers in the COVID times for these women in the (inaudible).  It made them aware of the use of face masks, how to use the face masks, how to meet the face masks and how to clean the face masks.  (inaudible) help to the women in the community in gaining information on COVID as well as their own health awareness.  Regarding to me the take of it is that I want to mention the (inaudible) services should not be (inaudible) each night in the sense that a fire times (inaudible) services of ML mobile applications, but it has to be a period of timer inship between ongoing governments.  So there are government teams and private places and help in the mean time of these services because sometimes if is only people working in silos.  So there has to be a multi‑stakeholder partnership and we need to utilization of Connectivity should be roll the out as an part aspect of important Connectivity.  It is not that we enable Connectivity for the sake of Connecting, but also meaningful utilization of Connectivity such as the reliance and community networks and like enabling work online and offline Connectivity in the sense they don't need to be 24 into 7 online, but certain amount of offline Connectivity that the users deliverance of online things can also be involved.  So this will enable the services reach remote (inaudible).  Overall, these are my commands and ‑‑ comments and thank you.

>> SHARADA SRINI:  Thank you so much for those comments and for your experiences.  I now ‑‑ I am looking for any questions in the Q&A and I don't see any.  So I want to make another request to all attendees and those watching.  If you have any questions or comments, please do use the Q&A box or use the chat with all panelists and attendees.  So everyone can see your question.  Finally, we have Leon Gwaka has just joined the Connective feel at the University of Pennsylvania law school center for technology and innovation.  He will be presenting some of the findings on community networks that we have discovered over the last few years.  Thank you.

>> LEON GWAKA:  Thank you.  Let me change my screen.  Can you see my slide now?  All right.  So in this presentation, we're going to look at community networks as models to Connect ‑‑ and Connect and underserved communities.  So I think the starting point of this work is we acknowledge that over the years, there's been progress in terms of Connecting the unconnect the communities and in receipt reports, there's been suggestions that mobile Internet is actually surpassed to get 1% of the global population.  But when you think of this 1% on the website, it simply is an indication of how much people remain connected.  So it's still not connected mobile Internet.  And when we dig deeper into the regional ‑‑ into the different regions, we start seeing that there are regional disparities way in Africa, in Asia that even many more people that actually are not connected, not even close to the 51%.  And you start to see the different access rates among different social groups.  To a particular focus on women who continue to have less access as compared to men.  So when you look at the whole idea of why do we have so many people who remain unconnected, we think that the current models that are being used to develop Internet infrastructure, telecommunication infrastructure they have demonstrated over years that they're not reliable.  We cannot rely on them to actually achieve access or provide that further access to especially to rural communities.  What I call the current models, we look at governments that mainly funding the governments that have traditionally been charged with responsibility of developing communication or telecommunication infrastructure.  But in most of the poor regions, those governments simply do not have good resources to provide access to everyone.  And the second part is when you look at the Private Sector, they're much more concerned about financial returns on the investment.  And the ‑‑ there have been claims that in usual communities and restrained communities, there isn't much demand that can warrant them to make those investments.  When you look at community networks, I think that there is possibly an opportunity to provide Connectivity to resource constraint communities.

Now, when you look at community networks, these initiatives that are viewed in the March and mostly we say they're forecast and also community driven.  So they lay an eventual role in developing this network.  We also look at it that require less capital expenditure and operating over time.  You require less financial resource as compared to traditional systems.  I think the idea really addresses out of the challenge that is governments that are not well resourced and also Private Sector that is profit driven.

What we have seen in developing this has sprint across the globe in exchange that we have diverse visions of community networks.  So ‑‑ there was ever since ‑‑ there was a studdie in Africa and that study identified things such as village where there were different initiatives and some qualifiers community networks or even just Internet ‑‑ Internet centers.  We believe in each of those initiatives, unique lessons that we can lend, but also when we compare them especially if we can compare one region from another or those that I understand region there are many lessons that can guide us in terms of developing this community networks.

Now, we have 21 case studies in our database.  They are distribution.  You can see that we have most of the initiatives in Africa followed by Asia and then we have Europe and we have America and then we have Oceania.  What this indicates is that the case studies are distributed across the globe.  It is not only one region where we find the community networks and if we look at the number of years this network has been in existence, we see that community networks that would be 1 in 5 years and others that have been 6 to 10 years and then we have others beyond 11 years.  So we can see that there's a distance spread when you look at the number of years coming to accept the number of years and networks have been in existence.  And we believe that we can lend and produce lessons from those that have been in existence for a long time to also inform the units that have just think about informed.  When you look at the location, we see that there is also other community networks that are actually implemented in urban areas and most of them from our case studies, most of them in rural areas and in terms of scale, we have more community networks that are implemented at village level and we have others that are at district level; however, we'll clarify that in terms of the district, different countries use different ways of classifying whether there's a district.  So I think that will be important to note.

Now, given that we have community networks that are from diverse regions and different sizes, we need to develop a mechanism to understand how best ‑‑ understand community networks and an individual case study or how best we can develop mechanisms to compare them.  So we believe that we can apply the technology organisation development approach and classify or rather we can use this approach to try and characterize either one case study or characterize different case studies and it enables us to draw (inaudible).  Now, from what we have, we found that we have 90% using the unlicensed spectrum.  And also in terms of cost, 28% reported that the initial cost of their projects was less than 10,000.  And they continue innovation and investment which is informing new technology that can be used to do at the community networks.

In terms of organisation, 72% reported they consulted and have been involved in the management of the community networks.  That includes the process of repairing and maintaining.  Then we have in terms of revenue generation, 67%.  (inaudible) revenue generation module; however, the revenue generation model that we are seeing in this community network is such that it does not match what is currently on the market because they are trying to serve the resource constraint communities.  So they're trying not to over charge.  So even though they say they have a revenue generation model, they are enabled to recoop either the capital expenditure or sometimes the operational cost so they continuously rely on funding from government from Private Sector and the main lesson we find on this aspect of organization is that most of the community networks are trying to engage in strategic partnerships with academia, with industry with governments so that they are able to generate or to have sustained funding for their operations.

One of the ‑‑ then we move on to the environment and in terms of the environment, we are looking at where the community networks are operating.  And we found from (inaudible), there were indications on the impact that regular frameworks are having on the community networks.  But from our case studies, we found that community metrics are finding ways to actually overcome this challenge and others have managed to achieve license exemptions from the government and others have managed to demonstrate to governments that the usefulness of this metrics and as a result, they have actually prompted governments to update and review their regulator frameworks, but this does not mean that the governments implements those new regulator frameworks.  And we can also just so maybe from the technology part that 90% of the community metrics I use on a license spectrum and they are overcoming the regulator requirement.  Then the other aspect in the environment is that most of these community networks are set up in areas where there is no electricity and there is the option of using solar option.  Then in terms of anchor institutions, this is where we're looking at where this community metrics can set up.  We see that in Africa only 33 have responded.  33% responded that they managed to have anchor institutions or managed to have facilities where they can set up and operate.  In Latin America, it was only 20% who had site support.  Now the main concern with anchor institutions is the security of the equipment.  So if there is no pro facilities that can lead to the equipment being stolen or destroyed.

The next part we tried to understand when this community metrics is being set up, what drives them ‑‑ what drives those who are setting up the community networks and from our findings as you see that the main or the most reported reason was simply infrastructure development.  So we had 19 ‑‑ 19 who reported that there was (inaudible) studied this community metric was to provide Connectivity.  It was just for infrastructure development and along with that is about developing technical skills in the community or in part in communities to be able to Connect themselves.  And the others we have reported that community metrics are being implemented to support development dimensions and we can see these education, which is with 15 from education and we had 10 from health with 9 from gender and we had 7 from agriculture.

Now, we know that is important to understand what is important to your community.  If a community is manically best on agriculture, then we need to find a mechanism to link the community network intervention to the particular dimension that the community is interested in.  And this approach we believe can assist also in terms of finding of the community metrics.  So the different dimensions, which is health, education, agriculture, they often have some support in terms of addressing challenges which they face.  And we believe that community metrics part of the interventions that can be implemented in agreeing challenges which are across the different dimensions and we need to develop strategy partnerships with those institutions, education institutions or agriculture institutions that you can utilize or we can somehow support the community metrics to in terms of funding.  So this is work in progress.  What we have at the moment insures that community metrics are available Connectivity models and in our case studies, we have documented successful case studies such as the (inaudible) network in Africa.  What we need is to continue producing evidence to demonstrate what works.  And at the moment, most of the studies link towards demonstrating the technical aspect what technology works, but we also need to produce more evidence in terms of the organisation, the nature of the organisation and also the environment in which they're operating.

Then the next part is about integrating this community metrics with development dimensions and we ‑‑ by integrating them to develop or by tieing them to development dimension, it provides an opportunity to also develop other melt wrecks for impact assessment to understand what the impact this community metrics are having.  And also educated when we tied them to the other development dimensions, maybe where there is an argument to persuade some funding from those dimensions to support the community network initiatives.  And in terms of the legal frameworks, why we did not find this particularly to be a major challenge that reported by in the case studies, we believe that there are changes to legal frameworks, but we should insure or provide support to push for implementation of the changes and supporting the different community metrics pushing for those changes in the legal frameworks.  I think that's the end.  Thank you.

>> SHARADA SRINI:  Thank you so much, Leon, for your presentation.  Leandro, comments?

>> LEANDRO NAVARRO:  Yeah.  Thank you.  So I ‑‑ congratulations for the study because it is difficult to gather evidence from many initiatives.  In my experience also, I found it quite hard.  And, um, and I found the draft report very useful to give an overview of the different initiatives.  Evidence depends on communities to report and probably some support from academics to report them in a form that can be used and can be integrated in front of studies.  I found many initiatives don't have any interest or any priority on generating evidence from their activities and it's perfectly respectful, but, um, but it happens.  I suppose that the studies even more relevant because when we talk about connected or unconnected, there are many differences.  So there are those who are connected and I remember in one talk that among the connected, there are those who don't care about monthly bill, which simply they get it and they pay.  Those who when they see the bill coming, they need to find a seat.  They need to take a good moment, take a coffee and then look at it and see if they can afford or they get disconnected in the middle of the month.  So there are different types of connected.  And then from the unconnected, you have the self‑connected, which are the community networks those who find their ways to Connect themselves is like saying, well, we are looking at how people eat and then if you only look at the restaurants, of course, not everyone uses the restaurant model every day.  And the business mods makes a difference and then if you only see that, you don't really understand how people sustain themselves because they cook at home and they cannot afford to go every day to a restaurant.  Nowadays with the pandemic, there's not an option.  Or maybe because they don't have the skills to use a device or because as I said, they prefer to do it at home and, um, also sometimes the unconnected they mismatch is they simply cannot afford.  They don't have money or they don't have enough money because they live in a subassistance economy.  Money is very scarce resource and they need to invest that money into some other services.  It also Connects with the previous parts of the discussion about like digital skills and especially health care and other aspects.  So you may have a limited amount of money and you have to prioritize which one you want.  And also there is remaining category of those who are not self‑connected.  They are simply isolated and we don't know for those isolated, we don't know anything.  Maybe there are no evidence.  Maybe there is evidence that they are not connected.  Maybe they have fragile forms of Connectivity.  Maybe they don't really need it.  Ones ‑‑ who knows.  I saw it also I saw in the draft report that's also connected to the general aspect of infrastructures.  So maybe it has to do with electricity.  Um, and, um, also education in these times.  If you want to continue being educated, you need to be on a school and have home connection.  Otherwise things do not work.  The need and the use of Connectivity and community networks also has to do with if people use it for enjoyment or if now people use it for survival.  So, ah, we can see in the current situation, we found that some families realize that what used to be a tool for getting information now is about a tool to getting education, getting medical services, getting a lot of thins that have to be done work from home.  So I think that has changed a lot meaning of Connectivity and the importance of having ways to self‑connect yourself if you are not like at a target business of the ‑‑ of service providers.  And, um, one thing that is difficult to see, but you also mentioned is how this Connectivity is organized as said in many forms.  The commercial business model is one, but it is not the only one and some communities can have like, um, traditional ways of creating and governing their infrastructures and, um, it is difficult to compare because they are not the ‑‑ that's the way a particular community works and another community works differently and then, um, it's not a choice.  It is simply the only choice is that there are ways to work which is more like formal or informal based on money exchange or based on assistance to Bartering of things.  So it's a challenge.  It's nice to characterize.  It's useful to have to know, but you cannot really do too much about comparisons or rankings or anything which makes the study challenging.  Also, I mean, to what extent is useful to collect evidences is useful to understand scenario and see diversity.  And also as you mentioned in the presentation, it's useful because some communities rely on external contributions from technology knowledge or from investment from subsidies and some others do not.  But I find that the start up resources are important like initial knowledge, initial hardware, initial loans or the money to set up, expand and maintain the networks.  Even I saw reference in one interesting case where a community has its own infrastructure and then they InterConnect with a commercial provider so you can also find this interesting panelships between community initiatives that will ‑‑ that deal with the retail with the local services and the local support, connected to back hold providers or some other commercial providers that certain point where ‑‑ both is a win‑win situation.  The community provides the best possible support and, um, they collect money and might be they pay a fee and commercial providers find a way to make a business in a way that is respectful with the community.  So yeah.  (inaudible) to your work because I think it adds to and brings entry point to other detail studies and is focused on a specific community and you do the hard work of trying to put them together and extract some commonalities and differences and situate the communities with respect to other possible business models and other ways of delivering Connectivity.  It is just a really, really hard problem.  So yeah.  That's my analysis and comments and my encouragement for you to continue this work.

>> SHARADA SRINI:  Thank you so much, Dr. Navarro.  Can transpost like a bus be a place for Connecting the unconnected?  I think this comes from some of the newer projects that are trying to have connected buses.  I'm aware of this being piloted in Peru as part of an operation that the World Bank is financing which I now also am a part of, the organization of the project.  But I wanted to understand what thoughts might exist around transient Connectivity more broadly.  Buses are transient.  They're not consistent Connectivity.

>> CHRISTOPHER YOO:  If we can ask Sarbani to join us.

>> SHARADA SRINI:  Absolutely.

>> CHRISTOPHER YOO:  One of the interesting things about bus travel is they tend to be along arteries where there's a fair amount of density, but in addition, there's access to rights of ways where people can actually lay some of the Connectivity that will work that support things.  So you'll find that you'll see bus kiosks in a lot of places and Wi‑Fi hotspots, places where you can get Connectivity because they tend to have certain facilities and resources around them and did is much easier to Connect buses even though they're transit and mobile because they do collect people and they tend to move among the mobile Internet.  Those serious of roads are sometimes the first things Connect and are often among the first connected.  So you can see some potential.  But what I think you're getting at is it's not one technology or one solution to solve everything.  You'll see some solutions work in some areas and we shouldn't say that will not Connect.  That's great.  We solved part of the problem and it may not be one grand solution, but many overlapping solutions that allows us to Connect.

>> SHARADA SRINI:  Any other panelists have any on transient Connectivity?

>> LEANDRO NAVARRO:  Yeah.  I think the advantage of self‑connecting is you can choose how to do it because when you talk to Telecom providers say yeah.  The quality metrics that we need to comply, well, in case there is nothing or do you it yourself is like when you cook.  You like what you cook because it's what you have.  So it's your solution.  And also you can see that when you see a community from distance, it is difficult to imagine what would work, but when you are part of the community, you can realize and you can think a lot and find out how to use the local resources the best.  For me, one nice example is there is computer aid that used to ship computers second hand computers and containers to remote locations.  And then when they did realize the container itself was interesting.  Now they ship containers that are classrooms in which they put computers and then they have a classroom where they also have solar panels.  So they bring their electricity and they only need a satellite connection or cable connection to power up.  Now a rule used to be outside the box to bring the computers.  Imagination is unlimited.  And this is ‑‑ anything that works is fine.Y is yeah.  Imagination to the power.

>> CHRISTOPHER YOO:  I love the idea of imagination to the power, Leandro.  There's a lot of these ideas, but I mentioned there's the box and we have seen in our studies a lot of questions about back haul and the costs sorted with that ‑‑ associated with that.  These are things that need to be studied because satellite a lot of latency, but it is ubiquitous.  There are tradeoffs and one of the things is to study this in more detail.  Something I should have said earlier is really affecting Clair and her leadership through this area through connected societies, connected women, the current leadership of the equals on gender and the new GSMA skills alliance.  It is really pushing out the information we have to make decisions about this on an evidence based way which we think is the most important.  And I appreciate her emphasis on the growing important of mobile, which is absolutely true.  What is interesting is it is true that computer base programs dominate and they tend to be in terms of (inaudible) schools.  Afterschool programs dominate.  What we don't have is great metrics to understand what is the most effective.  That is opportunistic.  I think mobile based education ‑‑ digital skills training has a tremendous amount of promise.  We have to hash it and validate it and figure out how to start creating plans to implement this as well as we can.

>> SHARADA SRINI:  I don't see any other questions, but I will take the moderators prerogative to ask the last one because we have a few minutes left.  To all the panelists today, what do you think are the biggest open questions in terms of understanding how to Connect the unconnected today?  Let's start with the commenters.  Clair?

>> CLAIR SIBTHORPE:  Sure.  I think ‑‑ I think while we have a lot of answers and certainly that's why we do these big national representative surveys to answer of these questions, there's still a lot more to be answered.  So, um, so I will say a few things.  First of all, we don't understand the impact of COVID.  So we're very lucky that we're continuing with our surveys at the moment.  The reporting will start to see what that impact is.  We'll be an impact and I think from a gender perspective, I am concerned it is not going to be a positive one.  I think COVID has changed things and will change key barriers and affordability and such.  I think that's an area we're hoping to fill with more.  And then two others general comments I would make.  I would say that we don't have as much for grasp on the ‑‑ some of the impacts around policies in this space.  So how do they affect digital inclusion and some of those policies ‑‑ the impacts of some of those policies in different areas.  So I think kind of a more detailed understanding of policy impacts around digital inclusion and arrangements of things would be useful.  And we don't ‑‑ we don't have we're working now with the University of Pennsylvania team around and also looking at individual impact on that.  There is a lot of impactability information around household level or national level, but what is the individual impact.  And then I would say on digital skills specifically because that's what I was asked to comment on.  I think there's a lot more learning we needed to do around what mechanisms are ‑‑ we know ‑‑ there's a lot of content, but what other mechanisms needed to be delivered?  So when is it appropriate to deliver it on skills.  So when we do it, we're learning, for example, you can educate people about what is the Internet and some of the basic stuff via video.  But some of the more detailed more advanced skills perhaps need to be more face to face.  What are the delivery mechanisms that will have the impact and how can we ‑‑ how can we ‑‑ and on which type of skills and which type of audience because I think we need to understand that in more detail if we scale up our efforts.  A lot of work we're seeing is done face to face.  That is very resource intensive.  Mobile operators have an advantage they have a network of 8 (inaudible) which can deliver it, which is great.  But not everybody does and I think we need to ‑‑ there's a lot to be understood around delivering mechanisms which is appropriate and which type of skills and which is audience and how effective is it and how can we kind of get it so that these things are more effective and we can get the numbers up scalable.  I said quite a few things, but I think there's a lot and I think on gender, for example, some of the policies around gender, what is their impact (inaudible) to learn from that.  I will hand it over to somebody else now.

>> SHARADA SRINI:  Yes.  Dr. Navarro?

>> LEANDRO NAVARRO:  We need to get out of the trap of Telecom monopolies that we got into several monopolies of large Telecom providers in general.  I think it's like a stretching model too much.  There's a point where customers can not afford the prices and it is not profitable for them.  So no need for them to cover everything.  If we allow the unconnected to count because they are unconnected and how they can raise their voice and how they can find ways or propose ways.  It's difficult for governments to typically (inaudible) in the capital towns to realize and to understand what are the needs for those connected and who maybe do not speak.  Some people say that it's important to enable people to self‑connect because it's like saying enabling people to (inaudible) a home.  If you don't have a kitchen at home, you don't have a choice.  You need to go to the restaurant or starve.  We know for food it is not acceptable and Connectivity is not either.  And this pandemic situation has shown that governments have to face sometimes the trade off of preserving health or preserving their economy.  So I think this is the same thing for Connectivity.  So sometimes I need to preserve the private companies and to empower them, but also at the same time, it has to preserve the rights of those who are connected and those who are unconnected.  And these diversity of models is really a challenge and that Connects the policy definitely.

>> SHARADA SRINI:  Thank you so much, Dr. Navarro.  Dr. Belur?

>> SARBANI BANERJEE:  (inaudible) but now it should be that unselected should be able to Connect themselves.  So given that opportunity to the people that are connected, there is a lot of understanding (inaudible) on how to Connect themselves.  So like for example, starting from particulars and the use of technologies that they can (inaudible) assemble in the villages or how to generate contain in the villages and the communities, how to use the contained in the day to day lives.  Everything is around that and I think because how much we are going to discuss about policy makers, with the Telecom operators to please take connectivity to the last mile the other ones (inaudible) need Connectivity.  You don't have any other means of communication.  And but then (inaudible) Connect themselves in the communities.  They too did a very good job of connecting.  In India, what we have seen is that during the COVID times, the gender enunciation has increased and it's become the digital divide that's widened.  And some of the articles that are there, I have read that the long struggle that we might have control to become independent, to become (inaudible) and has again come back to zero.  Now again, women are back into the houses and then they are (inaudible) who don't have jobs.  They are relegated to the homes taking care of the children.  For all that struggle, I think it has to again perhaps like Clair mentioned next year's report and we look forward to understand what has been the real impact of COVID.

>> SHARADA SRINI:  Any other comments from Dr. GWAKA?  Sorry, Dr. Haseki?

>> LEON GWAKA:  For me the open questions stem from the current situation.  I think after current situation we have seen a lot of momentum in terms of discussing about Connecting them.  Name momentum in terms of initiatives lead from governments and lead from private sector.  People have demonstrated willingness to actually contribute towards overcoming the challenge.  But I think it remains to be seen whether that momentum can be maintained post pandemic and see where the governments are willing to still commit resources as much as they did with a Private Sector that was willing to contribute as much as they did and even the discussions around the Internet.  Then also maybe commitment to using, I think, Connectivity for different initiatives.  I think those are interesting things that I'll be looking at.

>> SHARADA SRINI:  Thank you.  Professor Yoo?

>> CHRISTOPHER YOO:  Before I do anything else, I want to thank everyone for their insights and participation on this panel.  I do think it is interesting we talk about the impact of COVID.  There is evidence in the U.S. that the job losses have been much hard or women than anywhere else.  So I don't think any part of this world is immune to be impact, but there is a gender differential there.

I guess what I would encourage us to think about is to be openly experimental about nie ways.  And to try to do it in a way that allows us to learn from case to case in a much broader context.  I find many people want to latch on to one model and say, oh, this is the solution to everything.  Or fixed wireless or some particular model and the answer has got to be more complicated than that.  What we need is analytical structure to break it down into components and patience.  Learning takes time and you don't generally often get the entire information about the entire system all at once.  You get local knowledge one piece at a time and then as you get multiple pieces, you discover what worked in one context doesn't work in another content and you know the reason why.  It is an incremental long run play.  So what that tells you is someone once told me the best time to plant a tree was 30 years ago.  The second best time is today.  So on some level, we need to start planting some streets and understand that we wish we had a forest already.  But given that we don't, that is going to take a long time that we take a longer term perspective and do what we can to generate that decision and make it going forward.

>> SHARADA SRINI:  Thank you, all.  So we are almost coming to a close on the session and I actually learned a lot from this session.  So I want to thank all of panelists and all of the attendees for being with us today.  But also for the generous amount of insights of where we are today in terms of knowledge, what we know of in terms of different models around the world and also a lot of discussion especially now at the end on what remains to be known.  What are the things we need on focus on?  What are the open areas we need to work on and I think that is really the spirit of the Dynamic Coalition which I have been very, very lucky to be a part of.  The spirit of the coalition has been always to try and go out there and get as much data as possible no matter what shape or form and try to get lessons that synthesize so everyone is not doing the same thing over and over again.  And having a discussion on that every year, the platform of the IGF has been really useful because it brings together so many different stakeholders from different parts of the world to both realize and understand where we have gaps that are still missing, but also what we have learned so far.  So it has been really enlightening and I want to take the time to really thank everyone for sharing their time and being gracious enough to share their insights as well as their knowledge in the session today.  I know that the session will be available online for a long time to come for others to also get to know and access this and I hope the attendees both here as well as in the future benefit from the discussion we had today.  I certainly did and I'm incredibly grateful for it.  Thank you.  We are just coming to a close and we are closing exactly on time.  So I am also very, very happy that we are able to do so on time.  Thank you, everyone.  And very, very grateful for your presence.

>> LEON GWAKA:  Thank you so much.

>> SHARADA SRINI:  Thank you so much.

>> LEANDRO NAVARRO:  Great moderator.

>> MUGE HASEKI:  Thank you. 


Contact Information

United Nations
Secretariat of the Internet Governance Forum (IGF)

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

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