IGF 2020 - Day 9 - WS247 ICTs, SDGs, and Existing Data Gaps for Measuring Progress

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:  Vincent, are we ready to begin?

>> Yes.  Go on.

   >> CHRISTOPHER YOO:  Well, thank you, everyone, for joining us on our workshop today at the Internet Governance Forum on the role of ICT and achieving Sustainable Development Goals and existing data gaps for measuring progress.  My name is Christopher Yoo from University ofpencil vaina and I lead a initiative called One World Collected and has a goal of collecting as much data as possible as innovative ways to bring more people online and to try to make ‑‑ to try to build a data ‑‑ a central point of data to be collected in a way that allows comparisons across projects so we can learn from what has been done in the past to improve what we've done in the future.

Obviously, our interests in learning from what is going on is the inspiration for this panel and we are pleased and deleted to have three experts who are acutely aware of the wonderful things that we can do with data in terms of closing the digital divide, but also are very well aware of the limits of the current data and what more needs to be done.  And we are very grateful to have them here to share their expertise with us today.  In no particular order, alphabetical speakers are Anne Delaporte of Connected Women Program for GSMA.  Ain tona Garcia gab he will yoas works for and Lorrayne Parciuncula, my apologize, who is director of the Data and Jurisdiction Program at Internet and Jurisdiction and for those of you who know her, Lorrayne, you may have known her and she recently left the position as economist and policy analyst on communications, infrastructure, and services at the Digital Economy and Policy Division of the OECD.

So we have three very distinguished experts working with data trying to understand and make recommendations and investments and guiding how we're actually making positive action to close the digital divide.

So we would like to have 60 minutes with panelists and open the floor to questions and answers at the end.

I'll introduce you so let's just start by the way, Anne, let's start with the basic question of why is data so important?  Why does everyone always talk about this as a key need to closing the digital divide?  You're still on mute, Anne.

   >> ANNE DELAPORTE:  I will say three things.  First like to measure success, to measure success and impact of what achieves the Sustainable Development Goals, I think that's what I'm here for.  But also like to inform decisions because without data you can't set objective, you can't make a decision, and I think that sets of what we're trying to do is helping to track data, monitor data to then to be able to make a decision and as well as policymakers.  And the third one is to reduce inequalities, so using data to improve the life of people in working in mobile industries and using data to close the digital divide to reduce inequalities in gender, so, yeah, to me that's the three reasons why we need data.

>> You're on mute.

   >> CHRISTOPHER YOO:  I made the same ‑‑ I followed the same pattern from Anne.  I learn from the best.  Antonio, how about you?  You're interested in data so why is it important for the decision‑making of your work?

   >> ANTONIO GARCIA ZABELLOS:  Thank you very much, Christopher, and thank you very much for the opportunity of being here with you all.  Do you know, whenever we are designing as Anne was mentioning before, whenever we're designing a public policy on a specific intervention, in our particular country, we have to understand where is the market failure and it's very difficult to understand the type of market failure or type of interventions we need to do unless we have a clear understanding of what is it and what are the main challenges that the country is facing.

So having access to that data, to that information, will not only provide us with very useful information that allows us to establish a status quo of what is the starting point, but also we should be the specific policies and regulations that we need to define either in terms of regulation, either in terms of increasing the level playing field competition, or moreover, even to identify which is the infrastructure gap and what could be the potential financing of the potential policy reforms that are needed in that particular country or region in order for them to catch up with the rest of the country and the rest of the regions around the world.

   >> CHRISTOPHER YOO:  So I find both of your answers fascinating in one sense.  It's figuring out what to do but then also to measure it.  I know you have to scope the actual intervention but I know from my previous work with you, Antonio, after interventions in one country you often discuss with other countries to show what you learn from the previous efforts to show how we can benefit and leverage the benefits into the future and that's a critical role that that plays, and that's again the first thing Anne said, we need to learn how to measure success and not just targeted interventions but figure out how to do things and do things better.

Lorrayne, over to you.

   >> LORRAYNE PARCIUNCULA:  It's a pleasure to be here with this group of friends and experted, so why data is important?  So as humans we like using analogies, right.  So one could be that data is a new sunshine or new air.  And for me because of my background in infrastructure regulation, I like saying that data is an underlying infrastructure for so many things.

But mostly, it is a key asset everywhere for social and economic development.  For business it's about value creation, to improve business decisions, new services, product, and entirely new business models.  For governments it's to inform policymaking, and in academia it's to produce knowledge.  Data is different than knowledge in itself, right.  And for Civil Society, data is important for empowerment, either individual or collective.  So in a nutshell data is very important.

   >> CHRISTOPHER YOO:  So I think you all made a powerful case about why it's so critical.  When you come to meetings you constantly here we need more data, there is not enough data.  So let's go back the other way, Lorrayne, having said it so wonderfully, if it's so important and you all agree to this, so why is it so hard to get and why do we have not enough?

   >> LORRAYNE PARCIUNCULA:  I think, so starting from a more granular perspective, it's that there is a lack of standardization of several indicators, and in my previous role as an economist in the OECD, I would see that especially in countries that were not part of the OECD membership, the actually comparing different datasets, and so just we end up using very generic indicators to compare countries, such as Internet users or Internet usage in that case, but then the experience of the Internet is much more granular than just if someone is using the Internet or not, and a lot of the data that we have is just very basic in terms of being online or offline, and as Antonio was saying, a lot of the market failures, I mean, you need more information to understand what are the bottlenecks there for access and usage of technology.

But then beyond the issue of standards, and I think one of the reasons of those issues is that the lack of the required institutional framework in terms of the mandate that regulators may have to collect certain kinds of data, the clear plan and action to use that data, and what for, and an institutional framework to actually have a plan to do informed and evidence‑based policymaking.

In terms of traditional providers, especially when we're talking about Internet and ICTs, I think at that the issue of the mandate is resolved in many developed countries, especially, but for example when we're going in items of trying to understand what is the issue of Internet usage and the data that Internet companies may have and OTTs, that's not really the case.  There isn't a mandate of regulators and policymakers to collect that kind of data.

And on another level, beyond the issue of standards and the issue of institutional framework, there is also the lack of capacity and human capacity both from the technical side of dealing with this data and also in terms of the policy side because you actually need policy people to be able to bridge these two worlds in terms of understanding what kind of information and data knowledge is needed to inform policy.

Finally in the last dimension, there is ‑‑ there is a lack of cooperation between stakeholders, and we are in the IGF, in a multistakeholder forum, so it's always important to talk about that in terms of the cooperation between international and regional organizations and governments and service providers, Civil Society and academia for you actually to be able to bridge those issues in terms of standards, institutions, capacity, et cetera, so that's a first thought on that.

   >> CHRISTOPHER YOO:  Antonio, your thoughts, please.

   >> ANTONIO GARCIA ZABELLOS:  Yeah, thank you, Christopher.  I totally agree with Lorrayne.  I would say this is probably a mix of capabilities and incentives.  Let me just explain and elaborate a little bit more of this.

When it's about the institutional level, one of the things that we have observed is that, for instance, regulators, even those sometimes don't have the mandate and they're lacking the capabilities to collect and gather those datas.  Moreover, they're lacking capabilities to even analyze that data.

So let's say that there is one specific problem as related to the capabilities and the institutionalty framework to somehow endorse the mandate of regulators and institutions to request that information and eventually analyze and provide some conclusions out of that.

So from that end, we have identified two major concerns.  The first one is as Lorrayne was mentioning is precisely the institutional framework of how to make the institutions really capable of requesting that information and having the strong mandate to at times actually introduce some sort of fines or on the industry as a whole whenever they're not fulfilling with their mandate.

On the other side, we have another issue that is related to the mandate and the capabilities to analyze and take advantage of that information with the design and implementation of particular policies, so from that end there is quite a lot of, let's say, quite a lot of areas of actions that we may consider.

But also, there is a mix, and I was mentioning before as related to incentives.  Sometimes it happens that the private sector as such, they are very comfortable in the area in which they are providing businesses, and for them to provide, let's say, data and information on a particular area of the business, it's not right im, they understand that they don't get anything out of that beyond more regulation, beyond more implementation of particular actions that are going against the P and L and so on and so forth.  And so probably here there is a lack of incentives in a sense that probably there is a missing link between the national strategies and that the government as a whole are having and the objectives that the private sector does have in order for them to precisely establish this sort of matching between what the government wants to do and what the private sector may get out of that particular strategy, so there is also an issue related to incentives that we have observed that makes the gathering of that information a bit complicated.

   >> CHRISTOPHER YOO:  Anne?

   >> ANNE DELAPORTE:  I agree with Lorrayne and Antonio have said.  I think on that I would add the idea that, you know, we talk about data, more data, more data, more data, but I think it's also part of like what ‑‑ it's not only about collecting data but it's collecting the right data for actually the right purpose, and often we collect too many data and we've come back to capabilities as well of like private sector, government, et cetera, how you get that together to actually understand which indicators are important, what data I will need to measure progress, to track whatever indicator we want.

And a lot of times it's also like a very top‑down approach where it's more like I would say institutional organizations who want data but local governments don't necessarily see the value of it, and getting back to them like and making them understand why it's needed and often that seems like coming of this data and often we don't have the capacity to understand why they need to do it and don't see the value of it.  And in the same way with the private sector, you know, I've been working a lot on understanding the gender gap in mobile Internet, mobile news in general and it's really working with operators to make them understand the value of it, commercial perspective and social perspective, and so really bringing back like why do I need to collect the data is something that I would add to the rest of it so far of why it's difficult to produce data.

   >> CHRISTOPHER YOO:  So what's also fascinating to me in the work that we're doing, the way international organizations are often constructed, they often can gather a great deal of data, but they can't analyze it in a very hard‑knowed way.  They can't pick winners and losers, if you will, and identify successes and failures as easily.

You often see samples as related to systematic analysis because they have constituencies and members with sensitives that don't want to be told.  No one wants to say it's better than this, this is better than that.  We found that many international organizations we worked with were delighted to have someone with some distance from that to be able to make those kind of analyses and claims.

So this has been great in principle.  I want to ask you both, all three of you a question.  Which is, do you ‑‑ we've set the stage of the need, we have the problems, so can you think of examples that were particularly good examples of when you got the data you needed?  Is there an innovative way or ways to solve this problem that you've seen that obviously we have the reporting by governments, we ask companies for data, man tree reporting to regulators, but is there ‑‑ can you help us with thinking through solutions of not just the problem stories but possibly some success stories or some positive stories about when we actually asked for more data or tried to get more data and succeeded?

   >> ANTONIO GARCIA ZABELLOS:  Well, let me just say if I may, and Lorrayne, let me give you a real example that is happening these days.  At this moment of time we are precisely having conversations with the Government of Jamaica, and they are very keen on improving the level of digitalization that the country is facing, so in response to your question, I would say that ‑‑ I would say like kind of a systematic methodology that in order for it to really be successful, we need to align the interest of the public and private sector.

So first point I want to stress in here is that we don't have a clear strategy, a clear idea or where do we go as a country, then it's going to be very difficult to convince the private operators, for instance if we're talking about the telecom sectors to provide you with that data.  You know, so the first thing is that we need to have that strategy of where we are, and what are the targets that we want to achieve.

The second thing is that we have to understand as well the market dynamics, in a sense that know what it is and we are serving and the number of users or the private sector is going down, and the level of requirements in terms of investment for new technology, but even investment for operational maintenance remains pretty high, so it implies also that the telecom operators are facing some additional difficulties, and you know there is a public role that we have to understand.

So in this particular situation of Jamaica, what we have done is come up with a new strategy that is defining with layers and layers of data, and first by identifying which is the infrastructure, not only telecom but also infrastructure coming from utility, and second identify the location of different institutions, schools, hospitals, police stations, and certain public services that know what it is, they need to account the situations that we are facing because of the COVID, you know it's very important first to understand where are they located and second how far they are from those particular infrastructures, in such a way that even in situations of emergency or in situations in which we need to provide particular services or connectivity, we can identify which is the most cost‑efficient way in which we can provide those services, so that would be like the second layer.

And the third layer, it would be very much related to what are the technical specifications or what are the requirements in terms of technology that those particular locations are going to request, and so in order for us to do that, there is a piece of information that we need to collect from the government and there is a piece of information which is critical that only telecom operators and other utility companies to provide.  And only by doing that we'll be able to come up with a sound problem that is identifying the infrastructure gap on the policies that could be fully in place in order for those public locations to be connected and overcome pandemic situations like the one that we are facing.

   >> CHRISTOPHER YOO:  Fascinating.  Anne, please?

   >> ANNE DELAPORTE:  So to add to what Antonio is saying, I think you asked for an example of where the data is collected and works show and I think what we're doing in terms of infrastructure, we develop coverage maps, and so understanding and having really granular data that will help operators roll out cost‑effective networks in other areas, and in order to do that it's really getting, in each country, it's getting operators to give us their data so that we can do these coverage maps and that requires other stakeholders and regulations and governments ‑‑ ICT, et cetera, to get to that.  And I think we started with like two countries and with coverage gaps, eight last year and now we have I think 15 countries that have been mapped in Africa, and so this really helps like not only to ‑‑ it's kind of like a circle where we need coverage data to be able to understand where we need to get more coverage and that is also like helping get us better data because obviously of all the ICT way of like connecting ‑‑ or using ICT to survey, so that is kind of like we need to know what is covered, what is not covered, and then like able to collect your data.

So, anyway, so a way to say that it's just not only for operators to what is covered but also NGOs and government to better target the activities and that's I think has been a very good example of like a good collaboration across different stakeholders to get to that level.

   >> CHRISTOPHER YOO:  Lorrayne, do you have any thoughts?

   >> LORRAYNE PARCIUNCULA:  I like that we're among peers here talking about telecom, so it's just easy to bridge the different discussions.  I'm going to echo what Antonio eloquently put.  I think that the issue ‑‑ I like the point that Antonio made about incentives, and I think I would add to that a dimension of trust.  I think it's very important, in addition to this alignment of objectives, is to foster trust between these different actors in a framework for cooperation.  And you asked for examples, and I can think about countries views that I worked on which is Mexico and Brazil, and a while ago operators in Mexico would give the data that they had on a printed sheet in font 8 for the operator, and they would have to type in the computer, and I'm not talking that this happened 10 years ago.  It happened 6 years ago.  So I mean, it's just ‑‑ it's very recent this change, and it took, okay, a constitutional reform in Mexico to change the institutional framework there and to give more power to the regulator so it wouldn't be questioned in the courts every single time that it decided to collect more data or to implement its decisions, but it also took this building of trust between stakeholders.

And I think it's important to have this alignment, especially so the private sector knows what the data is being used for and that it's being used in a responsible way, and that the decisions can be accounted for.  So, I mean, in the example of Brazil, for example, small ISPs are a regulated by a different sort of service framework.  What happens is the regulator did until very recently and it's just finishing compiling this basic coverage map on Internet connectivity in the country because they weren't the traditional teleco operators so that meant there was a lot of patches in the method they had of connectivity, and ultimately, that had negative effects for ISPs and I'm going to explain why.  The regulator was promoting remedies for microfailures in specific regions where in their ‑‑ in their view, there was no Internet connectivity.

So what happened is that those particular regions and municipalities were actually receiving sort of subsidies to ‑‑ so that big telecos would come and invest in their networks in those places that already had pretty descent and good coverage, but from small ISPs where the regulators didn't have any data from.

So, actually, by not providing data, they were actually having or being ‑‑ having to compete with big players that were receiving subsidies from the government, and they thought that there was no service there.  So, I mean, now that you're actually building these connections and this trust and this understanding that that's why we need the data for, and if we actually know that you're there and you're providing the service, you're not going to be among ‑‑ you're not going to be placed among the regions of Brazil without any connectivity.

So, I mean, it takes, I think awareness and the building of trust in identifying the players and identifying what the policy objectives are there, and so I just wanted to add this dimension to this discussion on telecoms, which is that, yes, creating incentives, creating institutional framework, but making sure that this is done in a way to foster trust among players.

   >> CHRISTOPHER YOO:  So you're talking about how to foster trust among these small ISPs, and explaining to them why it's important that they support providing data.  Is that correct, Lorrayne?

   >> LORRAYNE PARCIUNCULA:  Yes.

   >> CHRISTOPHER YOO:  So that's an interesting idea, and especially these smaller deployments, if they've got a lot of work to do and the idea of providing data to the government is not ‑‑ they have to see the clear cost‑benefit analysis to make that go, is what I take you're saying.  If they understand that, that makes it work better.

What about the larger telecos, the larger providers, the incumbents?  You know, Antonio, Lorrayne, and Anne you've all talked about aligning incentives, so in the real ‑‑ I mean, let's get down to it.  How do you align the incentives for an incumbent?  Because I know even when a service provider is trying to do some experiment of having supporting innovative ways to connect more people whether it's zero‑rating or something else, they're providing the support and they can't get information out of the actual partner provider that they're using, so how do you overcome the reluctance, the understandable reluctance of private companies, larger providers, to engage in this?

Because an Antonio said, they have to look at their responsibilities to look at the request for information and say, how will this help my organization, and that's the most profound incentive alignment problem that we have, for those people.  Can you give me an example of how people have overcome these kind of barriers?

   >> ANTONIO GARCIA ZABELLOS:  Well, I will take the lead here.  Do you know, I think what Lorrayne was mentioning, it makes a lot of sense to me in a sense that incentives at the end of the day comes out of a situation in which you are going to win something.  Let's put it in a very simple way.  So, telecom operators, the way in which any other private company is working, they decide to deploy, attend, and provide particular areas whenever that they are going to get a particular region.  So it is not about social return, in the case of the private sector, it's about financial return in most of the cases.

So, the incentives in there is that, again, whenever that you are designing a strategy to, for instance, allocating particular subsidies, it's very important that the way in which you decide and design that particular option, for instance, to provide services in those particular locations, you focus on giving precisely those subsidies in those areas that at this moment in time, let's say the financial return is clearly negative.

So unless that is not happening, the operators will not be interested in going, and we do have some experiences in situations where, you know, either the telecom regulators are defining building processes associated to location of frequency bands that eventually remain, you know, not attended by the private sector because of the conditions that they're put in, or even we have situations when they are launching particular auctions to allocate those subsidies that, again, the way in which they were designed is not precisely aligned with what the expectation of the private sector is.  That's why I said at the very beginning that it's very important to understand what is exactly what you want to achieve.

And in doing that, you will be able to design those bidding processes and auctions in such a way that you take advantage of that.  This is what is about subsidies, but even if it's about creating a special vehicle with a participation of a public and private sector, the same thing is going to happen.  You have to create, let's say, this telecom operator that is going to be formed with contributions from the public and contributions from the private, and the governance model and incentives that you're going to give to each of the stakeholder has to be aligned with those objectives that they have in terms of financial and economic return, especially taking into account, and I want toinsist very much the situations that the private telecom operators are facing these days.

   >> CHRISTOPHER YOO:  So, Anne, if I can draw you in.  GSMA has made coverage maps, so I imagine you have similar problems.  So how do you convince people to contribute the information to the coverage maps, and are there challenges, how do you align the incentives to get people to cooperate?

   >> ANNE DELAPORTE:  That's a very good question.  And I am not the one who is dealing with that, so I can only talk to you about my personal experience, and again what Antonio was saying really about the commercial opportunity and potentially, you know, like most of the people that are not connected are in rural, remote areas.  I mean, when you look at the figures, you know, like in Africa, like 25% of the population is not covered by any network and they're in rural areas.

So what is commercial for operators, and so in terms of our like showing them that there is value to invest of these areas comes to trust as what Lorrayne was saying early, like to share data that's quite sensitive, so it's really about like we aggregate all the data that we have and it's a lot of like building this trust relationship with your operators so that they are willing to share it with us because it's, yeah.

So but then on the microlevel, I wouldn't be able to tell you in details of like how, you know, what's happening.  But there is definitely like also role for governance to play and in talking about operators of the importance of doing that to social benefit as well, but yeah.

   >> CHRISTOPHER YOO:  So think, I think, a perfect segue.  Someone mentioned before about the role of the government.  How do you work with governments to ‑‑ what kinds of challenges do you need to do to get them to provide data?  Or as someone said, you talk about to change the data they collect or how do you work with them to get them to see the information that they have, and you acknowledge as a critical component of the information that decision‑makers need?  How do we work with governments to get them to support providing data?  I know in my experience, governments are always worried about privacy concerns, data protection concerns, and that it's a tough one because we don't actually need individual‑level data.  We need much higher levels of aggregation.  Are this ways that we can work with governments to make it easier to get to yes or to change their systematic ways they're going about things to try to improve decision‑making that you need to make or help us connect more of the world?

   >> LORRAYNE PARCIUNCULA:  Maybe I can jump in there because I think it's exactly the point that I wanted to make in the previous session as well in terms of how can we get the government to help the private sector with data that they have as well, and I think there are interesting examples around the world where data has been responsible for reducing the cost of deployment for the private sector, and so I'll give you an example from Mexico as well.  They have a pilot project that works pretty much like AirBnb of rooftops, so what they did is they were able to compile an inventory of rooftops of government buildings so that operators could actually deploy antennas, and we know the bottleneck that it is antenna deployment around the entire region of Latin American and Caribbean and many other countries.  In Brazil it's a very big challenge as well.

But at the same time, a lot of governments don't even have an inventory of all of their official buildings that they have, so I mean, it goes again to the problem of capability that we're talking about in the previous question, and also about what was commented in the chat about the cost of doing that.  And sometimes, I mean, we have a lot of possible applications of AI and neurnetwork and satellite images and all of that trying to understand where the unconnected are, trying to make sure that you have models that improve and optimize transport networks and are also a big cause for deployment, but at the end of the day a lot of this bottleneck especially in developing countries, comes in terms of the fact that you simply don't have lists of what is an inventory of a government.

And when we go further than that, we can also think about not only interoperability of datasets but also interoperability of regulatory frameworks and incentives, and I mean in the literature of connecting the unconnected, one of the good practices is fostering initiatives in terms of when you deploy, one particular infrastructure, you deploy a connectivity infrastructure as well in the fiber, and that's one of the big costs for that, so if you're actually able to facilitate to the private sector in terms of accessing data that is available around public infrastructure or the infrastructure that is already there, but the ones that are planned for the future, you're actually creating these incentives in terms of sharing ‑‑ sharing more information and knowledge among the different stakeholders, but your question wasn't how do we make sure that governments do that.  And I think it's about, again, increasing ‑‑ I mean, improving the institutional framework and regulatory frameworks around the world which sometimes are still stuck into legacy patterns and they haven't really moved into being more streamlined and more simplified, and sometimes it's just a patchwork or very old and legacy legislation.

But at the same time, making sure that this is a priority in terms of doing evidence‑based policymaking and not only to facilitate to the private sector to their job and to invest, but also to improve accountability processes from other parts of the society and economy.

   >> CHRISTOPHER YOO:  Antonio or Anne do you have any thoughts?

   >> ANNE DELAPORTE:  Yeah.  I think that whenever that we have been requested information to the governments (this is Antonio), they're always requesting some sort of MBAs, these agreements in such a way that they feel protected about the misuse of that information, and I have to admit though in the case of IDB, those governments at the end of the day they're part of us, they're shareholders for the bank, and we already have let's say by definition, MDAs with all of them since they are having participations from the bank, so it makes sometimes that information request easier, but the incentives at the end of the day and it's very much related to DNA of our bank and interventions and we are not working ‑‑ we are not intervening in a particular country just because we want.  We are intervening because there is a demand or request from those countries, and they already understand that in order for us to design those projects, in order for us to support them throughout the implementation and execution, there will have to provide some information and some data.  Right.  So I can tell you that I have observed a change in the pattern in the last 10 months with respect to the previous time, let's say to the pre‑COVID time, and these days the countries that are approaching us, they understand that if they really want to have a sound project, well designed and with a specific indicators that eventually can be tracked throughout the lifetime of the project, they need to provide data and they have to provide information.

And in this regard, I mean, in all the conversations that I've had so far with the different governments in the region, in Latin America, I can tell you that they are very keen on providing that data andI haven't seen so many cases in which they're requesting this nondisclosure agreement, no.

   >> CHRISTOPHER YOO:  And to be clear you think the willingness to provide data has gone up as a result of COVID.

   >> ANTONIO GARCIA ZABELLOS:  Absolutely.  I think so.  And not only that, Christopher, but even the interest from the government on having that data has increased quite a lot, do you know, because of the precisely the situation that we are facing and designing policies to mitigate the impact, to design policies to coordinate different bodies within the government across precisely this emergency situations that we are facing, but moreover, to define everything to overcome the economy ‑‑ I mean, to overcome the situation and make the economy be prepared for a post‑pandemic situation, and more so the governments that we've opinion talking to are already gathering quite a lot of information and they're very keen on working on data analytics.

   >> CHRISTOPHER YOO:  And so picking up on something in the chat, you know, they're saying that about how do you obtain the ministerial support.  It's what you're talking about is aligning incentives, Antonio, by they're getting something in terms of a support for a project and something that they want to do to try to make it in addition.  Anne, you were going to say something as well?

   >> ANNE DELAPORTE:  Just I'm actually reflecting on what Antonio just said on the willingness to provide data that has increased, and I was actually on reflecting on what because we talked about equality so do they have the data and are they able to put that toward that, so these are just questions bouncing back to me.

But on a more practical level like what example ‑‑ like several examples of work I've been doing in terms of practicality of governments to collect data is really to engage several stakeholders and the table and have each of them having a responsibility and ‑‑ the responsibility for gathering data coming to the same stakeholder all the time and really like having this like sharing ‑‑ having this sharing, cost sharing, who is gathering what sort of data, and what timeframe, and again it comes back also to what is the data that needs to be collected and making sure that, you know, like having ‑‑ or making it as cheap in a way, as possible.

And I think that recently like last year when I was in Berlin and we had a roundtable with different stakeholders who understand the digital gender gap, and turned out that then they were like we don't have the information, and the first thing we need to do is actually measure it, and operators are interested in getting that sort of data.  Governments are interested in, I mean, in different statistical office and ICT ministry, et cetera, and everybody like really cooperated to like we need to get a plan, and we need to have a plan in place to collect the data.  It's a collective effort.  It's not coming only from one.  Yeah.

   >> CHRISTOPHER YOO:  So, what's fascinating to me about some of the things you're saying is, you know, what ‑‑ are we collecting the right data?  And there is a question in the chat saying, it's not just about, you know, connectivity but it's about the quality of the connectivity, and there is one world in which a fascination also we know now with download speeds, and first download speeds matter less than throughput because it's actually delivered bits but now we understand there is latency and jitter and all of these other questions of reliability.

But what strikes me, and so there is a natural debate that happens all the time, so are we collecting the right things.  And so what strikes me a lot of the information you're talking about, the data you're needing, is not even about providing the actual delivered service.  There is a lot of data you need to provide service in areas that don't have service.  You know, you're talking about, for example, AirBnb for government locations of buildings, mapping of location of buildings, or mapping of current existing needs, or for example, demographic characteristics, costs, things that affect costs.

So, I guess, the question is are we ‑‑ what is the data that we're not getting that we need the most that is the hardest to come by that influences your decision‑making?  I mean, what is it ‑‑ where is the gap in essence in terms of the data we're collecting?

   >> ANTONIO GARCIA ZABELLOS:  Yeah.  This is even a very interesting question, like the last one posted about powerlines and availability of electricity.  I totally agree that with this comment.  You know, let me give you an example of how we are handling the design of a project from, you know, from the Development Bank approach and I guess that is pretty much similar in the case of other multilateral development banks.  So, first thing that we do is to precisely by means of using georeference, GIS, we're trying to collect information about where the infrastructure is.  Infrastructure means telecom infrastructure, electricity infrastructure, and we're from abroad, railways, and something that could be eventually the vehicle by which we can, you know, provide Internet services and Internet access.

So, let's say challenge is to have inventory of that infrastructure.  I totally agree that sometimes there are governments that don't even have that information, right.  Here we go to the point raised by Lorrayne in a sense that either they don't have it because they don't have the mandate, or even if they have the mandate, they don't have the capabilities to make that request, you know, strong enough that the telecom operators, or in general whatever company of infrastructure is capable of that or giving that information.

So here we have another important issue to keep in mind.  If we're intending to use infrastructure for electricity to provide Internet services, this is another piece of the question that we have talked so far which is the regulation associated to that particular sector, which is different to the ICT and telecom sector, and this could be a major issue because sometimes it could be a situation in which the countries and electricity companies decide simply not to provide those services because as far as they start doing those services, they will have to comply with particular regulations that at this moment in time, I mean you're just pulling in a balance the cost benefit, it makes no sense for them from a financial standpoint.

So there is a whole branch of discussion in this first step that, as I said, it begins with an inventory of that infrastructure, that is a major issue.

Second phase is precisely as you were mentioning where the location of different institutions do intend to connect, institutions or municipalities that you want to connect.  And here there is a number of informations that we request in terms of the quality of services that you mentioned, so what is the bandwidth that they do have, or moreover, which is the bandwidth that they may acquire according to the area and population that they're intending to serve.  Because you know, this may have an impact in terms of the technical solutions that could be put at the disposal at that particular location, so we need to somehow come up with a demand analysis that makes us understand how much usage that infrastructure is going to have, and according to that, you know, to be able to go and move to the Step 3 and identify which are the technical specifications, and those technical specifications are pretty much related to SLAs, you know, Service Level of Agreement, and even issues related to affordability as well as it is another major issue that we have to think of.  Or it could be in particular a aspect associated to estimating what are the technical solutions that makes or not make sense according to the demand or populations that we are going to achieve.  So that would be the third step.  And those are basically the three areas in which we are in particular designing the design of a particular project and in some locations we are getting that information, and in others we are not able to get those informations in 100% basic, and we have to have conversations on how to find alternative varrables that somehow allow us to make a proper design of the project.

   >> CHRISTOPHER YOO:  That's extremely helpful.  I think that's a very useful way to frame things.  Lorrayne?

   >> LORRAYNE PARCIUNCULA:  Yes, no, I want to add to Antonio's point there in terms of what data we are missing, and just to give an idea I worked a lot with OECD countries but also non‑OECD countries, and one of the studies that we did in 2018 was about bridging the rural digital divide, and we carried out an assessment to understand what were the targets around the countries of all OECD countries in terms of bridging the divide, and in EU it was pretty standardized and they usually have the 25 megabytes per second or 50 megabytes per second for X percent of the population, for example, but the problem is that a lot of countries that are non‑EU simply have in targets fast broadband oar ultra fast broadband or super fast broadband, and so there is a prefix to broadband there which is not really standardized among countries.  M and it's interesting because by not even have that stated within their national broadband plans or strategies, there is no way to ‑‑ there is no incentive to measure it.  So, I mean, I mean, the concept of broadband, it's still one that was understood and coined by ITU years ago by 250 megabytes her second when we were talking about connecting everyone, but in most countries it's not satisfactory speed, and in you said it's not only about speed and it's in terms of all the measures about quality, and it's not only about the quality in terms of the moment quality and but also robustness of the network and about the user different profiles of the network.

Households have different needs than public offices than businesses than small and medium big enterprises, and so you have different profiles of need as well, so we just need a better and more granular understanding of what we mean by connectivity there, and that is data that not even OECD countries are often collecting for their own regions, and it's again, it's not only about what percentage of the country is connected, what percentage is not, but it's also trying to understand and geolocate those different speed limits or quality levels of the Internet, and that I got to say, it's not ‑‑ it's not there yet even in the most developed countries.

So when you are looking into developing countries, it's just ‑‑ you're just many steps behind.  And that's very hard in terms of targeting public policy for specific regions to try to understand where the gaps are, and I mean we mentioned issues around basic connectivity maps around the Internet, and in the chat mentioned ‑‑ it was mentioned the issue of just electricity provision as well.  I mean, there is so much knowledge which is not there, and it is needed for you to actually do good public policy and planning.

   >> CHRISTOPHER YOO:  So it's fascinating.  One of the things that we saw in a previous panel we organized there was a person deploying, and I think in Uganda and they wanted to look for anchor institutions, they wanted to diagnosis to this schools and there was no comprehensive list of schools that the country was able to provide.

Interestingly, one of the initiatives I'm working on is the Gig Initiative seeking to connect every school.  They're using satellite maps, and actually are improving data and finding and able to create lists, so there seems to be some opportunities in terms of using advanced tools to try to get some greater data out of this.

A question in the chat about electric power.  It is a critical issue and sometimes the solution is to connect people more, but there are essentially off‑grid solutions whether it's diesel power generators or solar generators or maybe different ways, and it's a huge problem that may not be solved in one particular way.

So, we only have a few minutes left before we switch to question and answers, and if someone has in the audience a specific question, I encourage you to put it into the Q&A section instead of the chat and phrase it in a question.  And if not, I will start asking questions and continue the discussion.

But the last thing I want to talk about is granularity of the data.  So one of the things that I know, for example, Anne your work with connected women, and sometimes we just have data on total usage but we're missing important cuts, and you mentioned earlier in your comments about gender.  There are other communities in terms of urban and rural cuts, you know, disability access, senior access, you know, vulnerable youth.

I'd love to hear more of your thoughts about how we're doing on getting the data we need not just in terms of its amount and the dimensions, but the granularity to get at the policy issues that we all know are important.

   >> ANNE DELAPORTE:  Yeah, so in connecting is really interesting with gender but as you say like getting to what is a person age of women in rural areas that don't have access compared to men men, disabled, employed, not employed, and what we're doing each year is we have a survey that we run across 10‑20 countries which is a sample and so we have these kind of like cuts of, kind of like ‑‑ I'd say the most important for us in terms of like mobile phone development in the GSNA which is around world and right now for digital inclusion looking at disabled people.

What we don't have in terms of data where it's a big, big gap now for us, it's around digital skills.  And Lorrayne talked about connectivity earlier, and you know it's ‑‑ that's on the demand side, it's really understanding the granular ‑‑ like have a level of understanding of like what do people have the digital skills to operate Internet because we know that most of the people in lower‑income countries use Internet on mobile, and currently also data that exists are very much either in developed countries on limited income, but also the skills or ICT skills are usually based on methods and indicators that are computer based and not for low and middle income countries and that's a massive gap that we have and where we try to do more on.

   >> CHRISTOPHER YOO:  Lorrayne?

   >> LORRAYNE PARCIUNCULA:  Yes, I want to echo Anne's point and to say that which need more data and more evidence to understand the differences and how far we're behind in terms of including women in the digital economy, but it's also not only about including a column of binomial male versus female in certain some of these indicators, all those that's also, but that's certainly not enough to capture the spectrum of gender identities.

But, also, while it's a problem in many countries, the lack of this data, but it's also about representation.  And I think it's about the questions that we're not as keen in the surveys as Anne put.  And I mean what about how safe women feel in the Cyberspace or on the Internet or in the entire ecosystem of the digital activities that they carry out?  How productively they can use those tools, and if they have the skills and capabilities to use those tools productively, but also how they use ICT tools to increase awareness on issues that are important for women.  And then at the same time how sensors and wearable devices are compiling data on their own bodies or in health or in reproductive health of women.

So I think it's ‑‑ I think it's around inclusion in general and the questions that we are not asking, and at the same time, what are the architectures or standards and practices that we assume are working for everyone, but that at the same time, excluding either women or trans people, or seniors or youth or people of color, people with disabilities, and so I think the discussion is much broader than simply supply and simply connectivity and simply disaggregating to the level of male and female in certain indicators.

   >> CHRISTOPHER YOO:  Yeah.

   >> ANTONIO GARCIA ZABELLOS:  Yeah.  Christopher if I may jump in here.  I can hear that there is definitely a challenge.  I totally agree with Lorrayne in a sense that, you know, we don't find, even if we want and there is a it can component in almost every operation that is dedicated to gender, even if we try to include specificities related to the gender gap or even indigenous populations because there is also an issue related to diversity, and we find some difficulties to find data and design the program itself.

But, do you know, I think that the issue that Anne mentioned about digital skills, I cannot agree more.  I think that one of the things that I have observed, and this is almost repeating in every country, is that sometimes they are designing programs on digital skills and thinking about the average of the population.  And I totally agree that first this topic needs to be brought into the top of the discussion and it is not just about connectivity, but it's also allowing the people and understanding people and the different users either in terms of gender, in terms of diversity, or in terms of the business in which they're involved.  This is a major issue.

But immediately after is how do we design those training courses on digital skills according to the maturity level.  So here I would like to bring a very good example that is happening these days in Mexico.  Do you know, Mexico is a very large country with about 120 million people divided into 32 states so why not the initiatives in which they are involved has to do very much with the digital skills, and but it is not just digital skills for the entire country, because as you may understand, there is a lot of differences between one area and or another.  There is a lot of differences between or across the different states, but one of the things that they have done is to run social demographic and economic analysis that understands which is the level of maturity, the digital maturity that the area has and what are the benefits that particular sectors or particular areas of population social social strata to benefit from the design of particular digital skill programs.

And so, again, this is an example of how to use particular data to make decisions in these particular cases on data skills which is another very important piece of the equation that we haven't brought at the very beginning and I'm glad that Anne brought it now.

   >> CHRISTOPHER YOO:  So I think it's a great point because what's interesting is we spent most of the time talking about infrastructure and the supply side and providing service.  Digital skills, the comment also about talking about local content and we know that there are other adoption drivers that are part of the equation which were much less developed in terms of collecting data and standardizing.

And so there are two questions in the Q&A and they're related so I'm going to ask them at the same time, and they're talking about, someone asked what is the place of statistical standards in allowing data to be collected more cheaply and comparably.

The second question is from Susan, which is can we require work with countries for implementation plans to implement SDGs and infrastructure and it would allow them to baseline, track, and report progress and bring together all resources including multilateral and NGOs to work union sen to work on budget and provide them on time.

So to me the common thread is standardizing ways we approach data in ways that are relevant, and there is a tension here, which is there are some variations across countries that things don't always generalize, and there is a problem where a lot of the projects are specific to different things, but they're asking for a ‑‑ in terms of effectiveness and cost reduction, can organizations like the positions, the ones you're in, can they advocate some generalized forms of data collection that will actually make it more useful, more effective, and is there a role to try to encourage or to model, to improve, to standardize, if you will, the way we collect data that would be beneficial?  Or are we missing an opportunity?

   >> LORRAYNE PARCIUNCULA:  I think I can jump on that one, and the answer to your question, is yes, yes, absolutely.  I think there is a role to not only international organizations, but also regional organizations to collaborate in working together to create those standards, and, of course, there are challenges to doing it, especially when it comes to the sector that it does a lot of it which is already embedded in legal frameworks, so just to give you an example on the work I was doing before in the OECD, it wasn't able to ‑‑ we weren't able to find a consensus in terms of defining broadband because a lot of countries had already the definition of broadband within their legal frameworks and that is a big challenge, but I do think that by iteration and by including countries and statistical offices and regulators and ministries into the discussion around evidence‑based policy, and also by fostering this sharing of good practices in the methodological side in terms of compiling, collecting, and also comparing data, we actually build the importance of building that and we move toward capacity building doing that, at least from the side of the government.

But I think it's important to also stress how crucial it is to have different stakeholders around the table helping with a process of understanding what the needs are and understanding what are the questions that are not included in the service in the statistical frameworks as well, so it's very important to include stakeholders around the table when we are negotiating the standards, the statistical standards.  But the standards are therefore that reason to, exactly reduce the costs in terms of having one model or one methodology that other countries can apply to, so I think there is a lot more room in terms of interoperability between those ‑‑ between the different statistical frameworks of Internet organizations and regional organizations alike.

   >> ANTONIO GARCIA ZABELLOS:  Yeah, and if I may, Christopher, I would like to bring here a real example.  I think that the points raised by Lorrayne, I mean, they are mentioning in a sense they are bringing in precisely different view, about but let me give you an example that I had a opportunity to work with her directly three years ago or something like that, Lorrayne, we were drafting toolkit on broadband policies for North America and Lorrayne was precisely in charge of leading this group within the OECD and where the first starting point was precisely to understand what was going on across the region and the different countries.

And she mentioned something at the very beginning that I believe is very important.  One of the problems that we have faced and that we are actually facing whenever we try to make a comparable indicator, is precisely the level of heterogeneity.  I mean, the information is very different.  I mean, it is very heterogeneous, let's put it this way.  It varies very much between country and other countries.  And so having the political will, having the institutional capabilities as she was mentioning and having the coordination across the different institutions, you know, they are going to be involved in information gathering, but more importantly in the analysis, will help us, and will help them eventually to establish comparisons between one and another.

And so in addition to the points that she mentioned, I would like something that was also mentioned at the very beginning and very much related to make the variables comparable.  Which is a tricky thing.

   >> CHRISTOPHER YOO:  Anne, any thoughts for roles on coordinating data and making it more effective.

   >> ANNE DELAPORTE:  I completely echo what Antonio just said.  That's something that we see all the time.  Every institution has different way of measuring something, and it's still don't have a solution on what or how can we make it better.  And talking to international institutions, ITU and mobile, and ITU data, GSMA data, and we all have a different way of collecting information on mobile, gender, Internet access, connectivity in general, and it's ‑‑ it's a lot of discussion and raising of awareness around this that at some point will make it happen and hopefully get to something that everybody will agree on.

   >> CHRISTOPHER YOO:  So, I think that that's something that is an interesting idea.  So, I heard two things.  Among other things that Lorrayne is getting at is the collection of data can actually serve as an educative function in terms of capacity building as well.  And it's interesting because I've seen this not just at the national level but at the grant and application level and when you're actually giving out money, by making people report certain kinds of information to apply, that's often the first indication that they need to be developing a plan for these different dimensions they have to report on the grant application, and it's possible that there is a way of collecting data that is a way to single to different authorities that getting this information is important, and it's something that throughout ‑‑ because obviously, one of the things that you're seeing is until you have the data, you don't know where you are and you can't do anything.

But as you know, just collecting data can actually change people's behavior.  The fact that something becomes visible, automatically increases salience and without regulatory restrictions to change behavior, behavior can start to change and it's an interesting soft‑way to do this and there was a question in the chat about starting with user needs at the rural level and I think this is what you're all saying, almost all of the data collection, the collection you're talking about begins at the rural level and you have a plan what to get but aggregates from the bottom up to make it all work because otherwise it's never going to function properly.

We are out of time.  At this point, I would like to say thank you to our panelists for leading a very stimulating discussion.  It's a challenge and I'm sure that we will continue to have conversations about, or I'd like to say that we solved all the problems about data gaps and data collections in this hour or 75 minutes and we can now proceed and not talk about this again further, but I think we all know that the problems will continue to evolve, and actually a sign of progress is to have new problems to work on because it will always be a challenge of how to bring or marshal the information that we need to bring more of the benefits of the Internet to all of the world.

On that I would like to thank Lorrayne, Antonio, and for being part of this panel and I think it was a fantastic discussion and we really appreciate all of you being here.  Thanks to everyone in the audience for your questions and participation, and we look forward to seeing you at future events, later events at this IGF and future IGFs as well.  Thank you.