IGF 2017 - Day 2 - Room XXIII - OF75 Data for the Humanitarian Field


The following are the outputs of the real-time captioning taken during the Twelfth Annual Meeting of the Internet Governance Forum (IGF) in Geneva, Switzerland, from 17 to 21 December 2017. 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. 



>> HEATHER LESON:  Hi, everyone.  We are going to get started in about five minutes.  We will give our colleagues a chance to depart.  Thank you.  Five minutes.  Thanks. 

(Standing by)

>> HEATHER LESON:  Hi, again.  We are just having some technical difficulties.  We may have to do interpretative dance.  One moment, please. 

(Standing by)

>> HEATHER LESON:  Good morning.  Thanks very much for your patience.  My name is Heather Leson.  I'm very excited to be here at the Internet Governance Forum.  It's really important that we tie these two things together and while humanitarians are not the usual suspects you would expect at an Internet Governance Forum, what happens here the Internet affects how we reach local audiences. 

Today we will have a conversation around data and humanitarian action.  We will touch on topics everything from data protection to data sharing to interoperability.  It's sounds like a five‑day conference but we will do our best in our one hour.  Every presenter will speak for about five minutes about the different kinds of ways they are using software and the Internet to reach audiences.  Then we will have a Q&A discussion which we hope will be a bigger discussion amongst all of ourselves. 

Now, IGF is all about the Civil Society, the businesses, the tech communities and the governments all trying to find common ground.  When talking about data and data action some of the topics that are timely across the board here at IGF are related to what our work is, too.  You will see there's a whole section in the report and Barbara can give that to you shortly.  There's a whole section about privacy and how we can actually work.  It's online, if you check Twitter, if you would like a non‑dead‑tree version. 

We will have about 15 to 20 minutes about sharing information and then move on.  I'll start with the far left ‑‑ far right.  We have Emir Hartato.  They just won an award for creating software to engage local communities.  He'll share more about that.  And next to him is Rania Alerksoussi and she is the coordinator at the international federation of Red Cross Red Crescent.  Then we have CJ Hendrix from the humanitarian data exchange.  And CJ is going to talk about that tool and some of the practices of how we are globally trying to share humanitarian data. 

So just to say we're very thankful you are here, and we will give you lots of space to talk and asks questions.  We will start off with Emir with a couple comments about PetaBencana.

>> EMIR HARTATO:  Good morning.  My name is Emir.  I'm from PetaBencana.id.  A little bit about PetaBencana.  PetaBencana is powered by software.  It's a free web based platform using both crowd source reporting and government agency foundations in real time. 

The platform is used in emergency events to confirm situational updates and removes the need of expensive and time‑consuming data processing.  And these verified user reports are integrated alongside represented data.  PetaBencana.id has been used by millions about safety navigations. 

If you would like to know more about PetaBencana, you can visit me later this afternoon in the booth.  But I would like to inform you why PetaBencana has been so successful in Indonesia.  This is Indonesia. 

Next slide, please.  So this is Indonesia.  It's predicted that Internet users will be increased by 30% in 2022.  We realize that social media and instant imaging has the potential to be a powerful tool.  Information is the most important resource in disasters.  At PetaBencana.id we tried to open access of information and make it available to everyone.  However, PetaBencana.id works because of community participation. 

Next.  Can you press one more? 

However, again, crowd sourcing tools heavily rely on the use of mobile devices and the Internet.  Particular demographics, like disabled people may have problems using the familiarity of mobile devices.  Okay.  But today I'm going to share from my experience to improve on crowd sourcing.  I will tell you a story based on my experience.  But, however, mostly I'm taking examples from my work at PetaBencana.id.  So please enter.

First is localization of the materials.  When I was working in Indonesia translating materials into local languages is important.  As most tools were still in English.  When our team started translating tools and materials into Indonesians, for your information there are more than a thousand local languages in Indonesia but at PetaBencana.id we are trying to localize the information as well and actively working in localization of oral materials not just by words but use other materials as well. 

Community engagement is very important as well.  At PetaBencana.id we are working with a range of different communities.  Through our activities like workshops and seminars we value their input to improve our tools adaptive to the users.  There are many ways and I will show you examples later.  It's also useful to have a local community hub.  Trying to have a local champion can be very helpful especially if you live in the country with more than 16,000 islands.  Say someone from far away wanted to have a community engagement program they don't need to contact people far away, they can contact their local hub.  And again also it's more importantly for better crowd sourcing tools. 

At PetaBencana.id we are not making redundant reporting apps.  We design our user interface.  It's so intuitive, easy to understand and more usable for everyone.  Next.  This is one of the examples of community engagements that we did a couple weeks ago.  We used the funding from the works to do this mural thing.  So it's very useful.  We are not just telling about the PetaBencana.id but tried to educate people by sharing platforms by social media, we are reducing the risks. 

Next.  And this is our user interface so basically when they for example from telegram they just typed slash and got reports, locations, and more information and the report will show up on the map.  Next. 

And because our tool works intuitively we receive reports from people.  2016 we received more 6,000 reports and February 2017 we received more than a thousand reports in 24 hours and hour website as 256,000 users in 24 hours.  However we still have some challenges like limitations, for example expansions for people that don't have Internet connections and applications in rural area.  But yeah, that's all for me.  Thank you, very much. 

>> HEATHER LESON:  Thanks, Emir.  16,000 languages, how do we make decisions with all these different stakeholders and make sure there's continuity to be able to help during emergency response.  I enjoy this point about how do we collaborate around different languages and locations.  I think that's the question all of us have in humanitarian action and that's one country snapshot.  I'll turn to Rania to give us a multi country snapshot. 

>> RANIA ALERKSOUSSI:  I am Rania.  And I'm the coordinator for federation‑wide data bank and reporting system which is a big platform for annual reporting for all Red Cross Red Crescent Societies and there are 190 at the moment.  Next. 

So FDRS, the acronym, it's system for transparency and accountability and performance reporting for Red Cross and red crescent societies.  It was a decision made by all national societies at the general assembly in 2009.  And it was introduced in the IFRC constitution. 

Red Cross Red Crescent Societies do report to that system.  It's self‑reported data.  We disseminate all results through different channels, so we do issue every couple of years a report called the accounts report which we are planning to issue one next year in March.  March or April 2018 there will be a new report that will analyze the data for all the past five years.  And there will be a focus on the volunteering for the Red Cross Red Crescent Societies and the challenges that we have seen within the post disasters as well. 

And we do have a web application that is continuously being updated with the data that we collect from national societies.  So it's basically a knowledge sharing network about the national Red Cross Red Crescent Societies and we have it in four languages, English, French, Spanish and Arabic.  Next.  Thank you.  So the data, what is it?  We have key proxy indicators because they do reflect the capacity and resources of the national Red Cross Red Crescent Societies and also how many people they reach. 

Main indicators are about the numbers of volunteers, how many people are volunteering their time, how many staff are paid, how many people donate blood, the number of local units.  We call them local units because Red Cross Red Crescent Societies are quite different.  Some call them branches, some call them chapters.  But it's really the different levels of offices across the country so the offices that work closely with all the communities.  We try to count all these offices to see where we are within each country. 

Also, the total income and the total expenditure from each society.  We have subsets of indicators.  These are the main indicators, the high-level ones but there are subsets of indicators like one about how many people are trained in first aid because this is one of the biggest activities of the Red Cross Red Crescent as well.  All the data we try to get desegregated by sex and age.  Age was introduced only lately in 2016.  The people reach data is desegregated by type of programs so whether it was disaster response or a long‑term development program or by type of service.  We have an overview of each society and what they have done in their country in the past years.

Of course reporting is not 100% complete so national societies do report the minimum seven main indicators but then the desegregation sometimes is more complex, and we can look at the challenges later about how this is happening.  Next. 

So this is a snapshot from the website.  And that's a map of on the website where we actually publish our data.  We have a map with a drop-down menu where you can select the indicator and see per country where we are on the map.  The size of the bubble reflects the number.  We can have another view also as a table where you can actually see the trend over the past five years as well.  Next.

We do have also profiles for each and every national Red Cross Red Crescent Societies.  This is also a snapshot, so we cannot scroll down but for each indicator as you can see on the screen the trend over the past years whether they reported or not.  With some secondary data that we get from the World Bank indicators.  Next. 

So how do we actually do the data collection and how does it work?  It starts at national level.  Actually at community level with the national society because it's self‑reported data they're the ones that provide the data to us.  They're collected from the communities for each program or project in different ways.  So they use mobiles for mobile data connection.  They use surveys with communities and do trainings and that's how they collect their data and this data is consolidated to headquarters, submitted to headquarters in each country and then consolidated and submitted to us in Geneva.  The data FDRS is public so it's available for anybody to visit the website and download it.

>> (Speaker off microphone)


>> PARTICIPANT:  Why is it not open? 

>> RANIA ALERKSOUSSI:  It can happen.  Because we are starting, we would like to engage with people and know what would they like to do with the data.  But that's it, really.  And then when it comes to the global level and how we do that, so it has a global scale because we try to compile all the information.  We do follow a rigorous standardization and harmonization process and rigorous validation process for quality control and quality assessment.

So in the data collection for example we have a big team of data analysts based in Geneva and different regional offices who are in continuous communication with all Red Cross Red Crescent Societies. 

We have a standardized set of KPI's with standard definition as cross regions and so we use standard definitions to work around all the diversity of the different organizations because Red Cross Red Crescent Societies are independent organizations who have their own ways of collecting data, their own systems, so it's quite complex to find a standard system that would fulfill everybody's basic needs.  We allow them also to either enter the data themselves online or offline to cater for the different Internet and technology access and restrictions. 

We do some capacity analysis of our Red Cross Red Crescent Societies and identify gaps.  And according to their data literacy levels we can customize our technical support.  With my colleague Heather some wider ‑‑ to improve the quality of reporting.  In our team in FDRS we also follow up on a weekly basis with all of them to keep the communication and make sure that the data they have submitted is actually accurate.  We make sure that they understand the definitions.  And we triangulate the data obviously with different key documents.  Their strategic plans which are also available to download.  We look at the reporting periods, we look at different sources of data from the different disasters that happened at operations.  So there's quite a long validation process that goes on before we actually publish the data on the website. 

It's very important for us because this is first and foremost for transparency and accountability towards the people we serve.  For disaster response it has proved to be quite useful because when a disaster hits in a country the FDRS data is available and our teams, operational teams who are coming from Geneva regional offices or other national societies who would be responding would have readily available data about a national Red Cross Red Crescent society in the country. 

It's also very helpful in terms of movement, cooperation and coordination so how do we work together as a network of Red Cross Red Crescent.  And for research projects and further data analysis also as I said earlier data sets are downloadable and can be used by everyone for more analysis.  What we analyze and what we have on the website is little.  What we analyze in our reports is a sample, but I think much more can be done with the data that we have.  Thank you, very much. 

>> HEATHER LESON:  Thanks, Rania.  So open‑ish.  This is a gradation, right, to be able to get to more open.  You can download the data sets and use it for a different action.  I think what is really exciting about this particular project is we are trying to standardize data work flows and that's a change.  Just to think through how you go from one national society from one branch like northern Nepal was in a branch and they had a paper document.  This is a way for us to make sure it's online, so people can find it.  If you have the Nepal data, you can take it with our friends from HDX which CJ can tell you about. 

>> CJ HENDRIX:  I want to tell you about the work we do but tell a story to illustrate the point.  First, I want to show you a couple ugly but useful maps.  This is a map we made when working for an organization that dealt with logistics.  This is the 2005 Pakistan earthquake.  This map shows which roads are opened and which are closed as well as information about security concerns.  We produced these maps every couple of days based on information we got called in from all or several little towns.  The roads were opening and closing due to landslides constantly.  We could do that and produce that map quickly. 

Every couple of days we produced I forget how many over the course of the response because we can standardize, and we can roll out forms to our team also.  We can give them GPS devices.  This was 2005 so no one had smart phones with GPS devices in them at the time.  It was slower at the beginning but over the course of the first couple weeks we nailed down procedures and we could do that because we had some degree of command of control.  That got pretty standardized.

Because that information flow was working very well we can bring information from other organizations, in this case OCHA around bringing in their data and joining it to make more analytical products.  The size of the circle shows how many people are running to that place and returning to their homes and how accessible.  This is the analytical stuff we can invest in because the data flow was working smoothly.  That was just looking at logistics.  We had a lot of control there because we were focused on a fairly narrow range of actors doing logistics work and focusing on a narrow bit of content. 

OCHA is responsible for coordinating across all domains for humanitarian crisis and has a different problem.  You've got all these different national societies but it's a fixed number of them.  Whereas for us with each crisis there's a whole set of many different actors that may not have had contact with the national humanitarian system before.  You're trying to deal with all people who had no pre‑event training and so on.  These are the five pillars of our work and information management is the part that I work in but information management touches all these aspects.  Next slide please.

When I came to OCHA we thought about how do we solve this problem.  The spelling of place names, for example, is always different and even the same actor will give you a new version of a data set each week and the place names will be spelled differently.  It's a nightmare.  So the first thing that we came up with we need an exchange language to let people keep their existing data systems but add a little bit of information to standardize it when they stand it to us.  We started working on that, we built a prototype.  But we saw that wasn't quite enough. 

There were issues around how does it even get shared?  Data gets built in these crises and walks away on someone's lap‑top when they rotate out.  We needed a place for data to be visible and findable.  We sort of expanded out.  We have this data standard, but we need a place for this data to live and advocate for that standard to give a place for people that use tools.  So we started working on the humanitarian data exchange about four years ago. 

So that's essentially a platform.  I'm show you a couple screen shots.  It's basically an open platform or sharing humanitarian data.  Our goals are to make data easy to find and easy to use for analysis which is where the standard comes in.  Easy to find, that's the job of the platform and tagging and search and so on.

Here is a screen shot to give you a concept of what it is.  There are some top line figures there.  We often have graphs at the top.  At the bottom is the beginning of a search result, data sets found by a search result.  And that's really the core of what the platform is.  It's a way to find data sets about a specific find or country.  So that was good, and we made a lot of progress with that and it helped.  It's expands because of the platform.  We are starting to run into issues around privacy of information, security of information, data literacy.  We need people in the field to have the capacity to use this information to understand data and the limitations of data to make decisions. 

Now, we have zoomed out to something much larger which is the center for humanitarian data which is a partnership between OCHA and the government of Netherlands.  That's opening this week.  Now HDX are components of the products of center for humanitarian data.  The four components are data services that includes the platform, the standard, data literacy trainings, we have done trainings recently in Dakar.  Also looking at data policy, how do we come up with policies that make sure we do a good job of protecting people when we share data?  How do we deal with personally identifiable information, how do we deal with disclosure control, the concept of being able to identify people from a data set?  Network engagement.  You need to get a lot of partners on board to have the critical mass for this to succeed.  Thank you. 

>> HEATHER LESON:  We all need a place to be able to use data and understand that, but I think the conversations we have around how do we do this properly is where we spend a lot of our time.  Each of these different platforms, I know that you'll have many questions and comments potentially about our individuals which is great, and their products that they're talking about.  In our conversation before we joined here today we came up with four themes.  We talked about them. 

Just to summarize.  We think that local communities are involved directly in data use and that really ties to the agenda for humanity.  Leave no one behind.  For us most of our actions requires us to think through that.  But we are as well humanitarians which means we need to do no harm and be mindful about our decisions.  How do we work in those work flows?  Those are questions that Rania and I have.  How do we involve local communities and be engaged and do in a crowd sourcing way and use the Internet? 

We need to modernize as humanitarians and inform the response and include them in their journey.  What are some of the implementation tips and learning and technology and how do we standardize?  And the big topic is verification, validation, how do we make sure we are causing no harm.  Those are the themes we came up with.  I know we have about 25 minutes to have conversations, so I want to take a moment to open up the floor and see if you have comments or comments about some of the things we talked about today. 

When you do speak up, speak a little bit louder so we can hear you.  It would be great if you tell us your name and the organization you're with and why you're here.  Thank you.  Questions?  Hi, welcome.

>> AUDIENCE:  Thank you.  I hope my voice is coming clear.  I'm a senior lecturer at Stockholm.  One of the interesting phenomena that we have seen lately is the proliferation of data sources online that journalists can extrapolate and get information from.  One area that's been somewhat causing a lot of issues is emergencies.  And journalists can't get right information from the sources.  And oftentimes they lead to either exaggerations or understatements or under ‑‑ I mean, issues that end up informing ‑‑ misinforming the public. 

One of the new efforts we tried to push was open data for journalists through projects of the sort that you're working on.  And one of the difficulties is that you need to have some sort of standard.  Standard method.  And yes, each project has its own visualization methods.  You can have various graphs and tools but none of them are applicable to an open application interface that the public can reach to such as the API's that many news organizations provide.  And so this is becoming sort of an opportunity that's being missed because data is there, it's in the databases, but then it's not reaching the public because of the lack of the interoperability. 

Yet the Internet allows you to have this interoperability so how do I find a potential for this to become a reality in that you provide platforms and look at the journalism community and ask them to extract this information.

>> HEATHER LESON:  Thanks.  I'll turn to Emir to get some comments.  Thank you. 

>> EMIR HARTATO:  Thank you for the questions.  It's really good questions.  At PetaBencana we have an open API, so everyone can plug into our real-time information.  One of the examples is that API from PetaBencana is also being integrated with the national systems.  It's called in aware where people like you can check not just information, not just information from PetaBencana but also disasters that have been going on in the regions.  So that's just one example of how we integrate the API into the national level but maybe not yet the international.  Wait.  It's possible, the API is there, but again it's a challenge to find who is entrusted with this information and how they're going to use it, they're still trying to figure out basically.

>> CJ HENDRIX:  So I think we are trying to get to that world and have made a lot of headway.  We have data from IOM, international organization for migration that on the site as hexal (phonetic) tags in it.  There's still a long way to go, especially around other kinds of data like the needs assessments that are done in response to crisis to understand what needs or where, so people can respond to those.  That data is less standardized.  But again, the difficulty comes from the fact there are so many different actors.

>> (Speaker off microphone)

>> CJ HENDRIX:  Sorry?

>> (Speaker off microphone)

>> CJ HENDRIX:  They're not working to a standard, they just have a format that they're using, spreadsheet format often.  We are trying to add a layer in there where we are at the point of sharing where they add semantic knowledge that comes from the hexal tags that makes the data more inoperable.  We have to build systems to require everyone to share their data into this system and it fails because people are in the middle of a crisis, they don't want to learn your system, they want to respond.  So we are trying to very light weight standardization.  I think we are already in the world that you want to be in but there's obviously more still to do. 

>> HEATHER LESON:  There's also an opportunity, if I might, with the center for humanitarian opening givers a chance to say there's partners with universities so maybe we need to partner more with journalists to how we can work with because I think that's the purpose of this center to bring all these different actors together.  When it comes to being able to tell a story, I think meeting journalists halfway is a good idea. 

>> AUDIENCE:  Hi, Brett Solomon for access now.  Thanks for the work.  We have been tracking the issues of Internet shut downs over the last 4 or 5 years and have seen an increase in disruptions to network taking place largely around moments of hardened political activity but also in humanitarian crisis and in isolated places as well.  For instance shut downs in refugee camps, also a couple of cases in Jordan refugee camps.  I was just wondering given data is becoming increasingly important central to humanitarian response, are you seeing disruptions in the network, deliberate intentional disruptions in the network?  And if so or if not, are you thinking more broadly about how to ensure the openness of the network to ensure data flows can be collected and reported to journalists to be published, et cetera. 

>> HEATHER LESON:  Start with Emir.

>> EMIR HARTATO:  Thank you for the questions.  About the Internet shut down at the moment in Indonesia we haven't had that issue.  Instead the government is trying to expand the network to the rural area as well.  And that can be said from the Internet proliferations in Indonesia is quite huge and social media is quite huge. 

At the moment or platform is focused on the mega city or urban area at the moment but we are trying to expand to the rural area we still need to ‑‑ it's still a challenge with rural area because I know some of the rural area it only works in mega cities at the moment because rural area has less people than urban areas.  It's still kind of challenged and undergoing research at the moment.  Thank you. 

>> RANIA ALERKSOUSSI:  For the overseas operations we haven't had that issue yet but since we have been operating in emergencies for some time already we are relatively prepared around not having Internet connections and we have various ways of people on the ground and not on the ground who communicate via earth satellite phones or other types of technologies that may not require Internet. 

Also our data settings usually can be printed and downloaded and used on Excel and in computers.  But this is something that we do need to think about a bit more for the future for sure.  Thank you very much for your question.  

>> CJ HENDRIX:  Our users are the people contributing data at least are the organizations working in the field typically and so they often will come with their own Internet connectivity, so the data sets do find their way to us, so I don't think that's impacting us at our level, but I would imagine those outages are affecting people using mobile devices to collect survey data.  So I would imagine it's an issue, but we don't really see it so much at our level.

>> PARTICIPANT:  I would just flag it because I think that as we become increasingly reliant on you guys becoming increasingly reliant on productive to do your work I think we are seeing it as part of the work but in 5 or 10 years it's going to be the absolute backbone.  And the opponents not necessarily in natural disasters but in political et cetera environments and so on, those are going to be aware that the infrastructure is central to your response and disruptions will very likely take place. 

>> HEATHER LESON:  I've been at many discussions over the past week about cybersecurity issues.  It's really important that we are there and learning.  I think there are examples throughout the humanitarian action about people not having access to the Internet that has caused issues.  I can't name any off the top of my head, but I do know there are issues and we are working on different portions of it.  All of our institutions are trying to find a way to read it and that's why we need to work with the IGF community to make that happen.  So thank you so much.  And Barbara you said you had a question for us? 

>> BARBARA ROSEN JACOBSON:  I do.  I have a question related to the perception of data.  Data is often perceived to be objective and to be a better source of information than estimates, for example, related to journalism as was already indicated.  And then maybe this is a question for the HDX team here.  How do ensure that the data and analysis and the databases that are on HDX are sound and are accurately put together.  Is there transparency on how these data sets come about?  And have you had in issues in the past of data that was misleading or used in misleading ways? 

>> CJ HENDRIX:  My colleague is here as well and he managers the content side.  He focuses very hard to use metadata.  People are doing things quick and dirty very often.  We want them to be honest about how quick and how dirty.  But it's a great point.  And in terms of validation we don't come through databases and try to validate them.  It's beyond what a team of our size can do.  What we do do is we vet the organizations who want to share data on HDX.  We talk to them, ask for samples of the data.  Many of them are very big actors and we know them well but there are a lot of smaller actors.  We have about 330 organizations on HDX sharing data.  That's a little bit of quality control that we do.  We monitor to some agree to the extent we can the metadata that we receive to be sure that the data is usable.  So that's an ongoing battle. 

>> HEATHER LESON:  I think in HDX you have to add a description to the data set to explain what the agency is.  And I think you also do spot‑checks to identify information.  The question about validation verification, I know Rania spent a bit of time on that.  Maybe you can speak to that.  Maybe you can talk about verification and how you're given good data. 

>> RANIA ALERKSOUSSI:  So we triangulate data.  We look at different sources of data and try to triangulate what is coming to us with what is available in other sources and other databases and online we do basic Google searches sometimes to make sure that it does not contradict with whatever we have.  We do some contextual analysis.  We analyze the context of the country, look at what disasters happened in that country and does these number make sense to us when they are sent to us.  So that's something that I think any data analyst does usually when you do receive data and before you need to use it or publish it or use it for analysis. 

>> HEATHER LESON:  Emir, I know you guys also worked on the complexity of how do you validate social media data.

>> EMIR HARTATO:  This is very difficult because there's a lot of hoax or fake news going on.  Until now we are very grateful there's zero false reports so far because we are trying to integrate the data from the government as well so when the government sees the reports on the website they also do validations by calling the local municipality organizations, trying to check, and they also can turn on the polygon regardless the level. 

So there's a verification process.  From social media and local apps because we are gathering a lot of applications and each application has their own verification method.  They have incentive methods and exchanging certain profile and also, for example, in the other applications they have their own verification method.  Most of the data that comes to us ‑‑ not most ‑‑ all of the data that comes to us are verified very well.  Users can verify as well.  If you can see a lot of reports that means that area is flooded.  So yeah. 

>> HEATHER LESON:  Thanks very much and thanks Barbara for that question.  We have about five more minutes before people come in and ask us to depart so I wanted to see who else might have a question or comment around these topics.  So far we talked about perception of data, accuracy, the risk of vulnerabilities and what do we do about shut downs and we talked about how do we reach and be more open with journalists so we can get better and improve.  I was wondering if there's other questions here today or comments.  Brett?  Brett and I need to be on a panel.

>> BRETT:  I'm really, really interested in this topic because as you say bringing together I think the digital rights community with the humanitarian communities actually an increasingly important area given ‑‑ increasing number of conflicts we are seeing around the world and we hopefully are not descending into a period of total conflict, but we need to be prepared ‑‑ I think we need to join forces and it's certainly part of our priority areas.  And I'm here to learn a lot. 

The question that I have actually is about security, data security and cybersecurity.  I think again in the same way as we are seeing disruption of networks we are seeing international isolating and identification of vulnerabilities within databases and collection mechanisms.  I'm just wondering if there is kind of a sector ‑‑ what kind of mechanisms are in place to ensure that the information which you have which is really like life and death information when collected and combined, what sort of protections are in place for that.  It might just be a comment. 

>> HEATHER LESON:  I think it's a really important question.  On this panel we have a lot of field experience, right?  Maybe I'll start with Rania.  How do you work on security and making it better from a technology point of view and also working forward? 

>> RANIA ALERKSOUSSI:  Okay.  Thanks, Heather.  I'm not sure how to answer this question ‑‑ (laughter) ‑‑ to be honest.  But in terms of health data in particular in quite critical and sensitive and we do need to protect it and in the past ‑‑ I mean this discussion was not happening even though on paper when you are collecting data on paper you still do need to make sure that it is safe guarded and protected and not distributed.  So currently at the moment our health times, for example, are conducting data protection needs assessments to look at this issue more in detail because we do when there are emergencies, for example, we do have field hospitals that are deployed even in normal situations in longer term developing projects we do have some field hospitals, so we do have very sensitive data that is currently.  So I know for example that they were working on some research to look at palm vein identification to make sure that any kind of personal identifiers are not distributed. 

So we are working on it.  We respect not there yet.  But we are working on it and it's quite a priority for us at the moment.  Also my colleague Heather is working with the IFRC.  They launched data protection guidelines so that's something we are using and disseminating and rolling out with our teams.  Thanks. 

>> HEATHER LESON:  I think the question around security is something we can work on.  We have security experts within our different organizations but I'm super mindful that we can probably spend a whole day, but I am aware there is another panel coming in.  So just to say that we touched on a lot of different topics, everything from data protection to security to cybersecurity risks to Internet shut downs to accessing information.  These respect just the icing really. 

I think there's a lot more that can be done in terms of how we connect humanitarian action and data flows and digs making to reach the most vulnerable.  Data is only helpful if it can be used for information.  If it can be used for decisions and knowledge and evidence and that's a choice but we need to make good choices on our data flows, so I'll ask my colleagues to give a one sentence take away on what you learned today.  Just one sentence.

>> (Speaker off microphone)

>> HEATHER LESON:  I got to keep it colorful. 

>> EMIR HARTATO:  Basically people have been collecting data.  It's very important.  It's very critical that data is not just about number and is about status but about quantity.  It's about people and we need to take care of it.  That's all. 

>> RANIA ALERKSOUSSI:  I think my point and throughout the past few years we have been working with data, I think what is important is also learning how to use the data, read the data and take decisions accordingly.  What has been happening really is we are very eager, we have all the technologies now and using all kinds of technologies and tools to collect data, but we are not yet at the point ‑‑ we are starting but it's still a long way to go until we use it in our decision making and policy development. 

>> CJ HENDRIX:  For years we built systems that were, you know, paper‑based or they could work offline because connectivity was always so terrible in many, many of these field situations.  And over the years that's gotten better and better.  Over year it's gotten better and better but maybe start thinking about that may be a liability as well because that infrastructure a target.  So that's an interesting thing to think about. 

>> HEATHER LESON:  So have a plan and use it.  Barbara?  And again Barbara is from dip low foundation our founder.

>> PARTICIPANT:  Thank you, Heather.  I wanted to make a quick comment or update on that this session is actually part of a track of different Open Forums from international organizations here in Geneva that are dealing with data so if you want to know more about the others we will have another open form on environmental data, and you can find more in your program. 

We are also doing reporting from the sessions as well.  So it's just an update I wanted to give.  And I also want to real thank the panelists and Heather because I felt that it was a very inciteful discussion.  It touched on so many aspects of data collection and data analysis and data sharing and limitations of data sharing, so I thought it was very inciteful.

>> HEATHER LESON:  Thanks Barbara.  And thanks for helping.  I appreciate that.  Thanks for everybody for participating and thanks to Emir, Rania and CJ. 


You can learn more about our work.  If you go to the schedule you will see their biographies.  And we are here for the rest of the day for the most part.  Thanks.

 (Session concluded at 11:14 a.m. CET)