The following are the outputs of the captioning taken during an IGF intervention. 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, but should not be treated as an authoritative record.
***
>> Kait, can you hear me?
>> Make sure you use the mic.
>> KAIT HOLM: Yes, Kat, I can hear you.
>> KAT TOWNSEND: Yes. If in Zoom you can just introduce yourself, that would be great in the chat so that you all can get a chance to meet each other. And I'd love to do the same for the people who are in the room. So I'll start. My name is Kat Townsend, I spent eight years in the U.S. Federal Government and have been advising governments in Ghana, South Africa, UAE, local domestic around the world on open data and open data policies.
And I work with a group of others who ‑‑ we have various day jobs, but because we believe so strongly in government transparency, this collective is called Open Data Collaboratives. If you would like to follow on with any of these slides or the people, that is the link. It's case sensitive. So, pull it directly.
And you'll see I'm here by myself. Others did not get their visas. It's in the middle of the night. They may join, but I do have a colleague who is on and supporting. So, thank you all for joining today.
I think what I'd like to do is just take a moment to take you all through, but it would be helpful for me to understand what roles you're in. This is a networking event. The point is for us to see and meet each other. Do you mind if we go around and speak about ‑‑ just to say either what role you're in or why you chose to come to this session, what you're looking for and then I can tailor on what would be most useful.
And I ask those on Zoom to do the same, please. Thank you.
>> AUDIENCE: Hello. Good morning, everyone. My name is Kosai, from Kuwait, chairman of a company entrusted in how to sustain ‑‑ oh. We are mostly entrusted in how to govern and sustain open data to government entities, because they are our client. Thank you.
>> AUDIENCE: My name is Varun, from the government of Sri Lanka, a civil servant for 23 years. I have is also served in New York as a diplomat, the Sustainable Development Goals were composed, I was part of the working group. So I found it quite interesting, Sri Lanka has many personal data collection initiatives and it is very essential for data governance models. It applies to all stakeholders. Thank you very much.
>> AUDIENCE: Good morning, everyone. My name is Nisha from the Maldives. I'm one of the cofounders of Women in Tech Maldives, a nonprofit organization which was established. Currently I think we are the only functioning IT nonprofit in the Maldives. We do advocate a lot for open data, which is why I choose to be in this session because I'm looking forward to learning from you and maybe going back home and trying to get our government and other offices to implement this. Thank you.
>> AUDIENCE: Good morning, everyone. My name is Shopa, University of Melbourne researcher. I am an academic. I used to work as an assistant professor at a university. My research on data governance. I am trying to look data not from an economic perspective, and hence I really wanted to see the capacity of how open government data can protect or guarantee values, specifically constitutional values is what I'm here for.
>> AUDIENCE: Hi. I'm from Maldives, government, National Center for Information Technology, working in IT service management.
>> AUDIENCE: Hi. I'm Aisha, also working in the Center for Information Technology, assigned to the Minister of Information Technology. So we kind of manage the government data. We have the data. And we are developing systems to take and use this data. So this session is really good for me to know more about it. Maldives. Yeah. Thank you.
>> (Off microphone)
(Laughter)
>> KAT TOWNSEND: Australia, Kuwait, Sri Lanka, beautiful. Thank you all for taking that time. I don't know if I said it, but I'm from the U.S. Very evident. And just to know who is online, if you all want to introduce yourself a bit on the chat, I'm happy to read out so that others in the room know.
Okay. So I just have three examples that I wanted to run you all through of different ways that the people I've worked with and the people on our team have approached government data. So, to know about our group, who we're coming from. So, we have ‑‑ thank you so much. So, we've worked ‑‑ founding members worked in the U.S. State Department, USAID, White House, on the U.S. Open Government Data Policy.
And from that we built out a platform that was shared around the world and really took this to other countries and, you know, what we have been working with is with youth and startups, academics, to try to build these coalitions and have about 4,000 people across our network that we engage in fairly regularly.
If I could share anything about the strategy for how this works well and sustained, there's a lot of people who try to change by shouting from the outside and saying you must open data. And then there's people working really hard on the inside trying to convince their colleagues. The most effective is to have a partnership where you're working together.
So in any way that the government can find people on the outside that are championing the work that they're doing and saying this is what we need and we're supportive, you can use that advocacy to make changes inside government and vice versa. So that would be ‑‑ perhaps it feels very obvious, but finding that collaboration and finding your counterparts is what's going to make it work and sustain.
So, just three countries and I'm just checking my time here. Great. So, the first I'll share. And it is about 3:00 in the morning for Florence, so I'm sharing her slide deck. But this is Florence Toffa. And she has an external organization. So she has a women in tech organization in her country. And built relationships with the government and learned from them that they wanted to build out an open data portal.
So, I think some of the basics about what is open government data. It seems like you all are working on this. You study it very well. But to think about the history of how this came together, a lot of this has been pushed and supported out of the north. But it's presented as just a different kind of platform that can be used for government to be responsive to its citizens, to improve service delivery, and to have better partnerships and also better transparency and accountability.
It will depend on your context of whether you want to share it as an accountability strategy or, as worked very well in the United States, as a way of supporting GDP growth and making a lot of funds. And that was very appealing to people within government to want to help others to open up data. If you were able to say that we would unlock a trillion dollars worth of growth.
Currently there's about 122 countries that have open data portals. There is a group called the Global Data Barometer that takes a measurement each year of the quality of data. If there is gray, that doesn't mean there's nothing, it just means they weren't included in the study.
Just to be clear, what open data means, it is both technically and legally open. So there's a lot of times that we'll have data available and it's just a PDF on a website. You can't search for it. It's very difficult to find. It's buried in. But it's called open.
So the quality of a machine being able to track and identify and pull in so that people can run their own analysis is a baseline requirement for open. I'm sure I'm saying things you know, but it's helpful to reset are we all talking about the same thing.
Legally open ‑‑ so, we have also had many experiences where we have legally, it's open. And what that means is you have to physically travel to an office and then you have to rent out a book with 1600 pages of text and then you can only read it there and they say, it's open. So, that is technically open, or to the letter of the law, legally open. But it's described as being able to use, reuse, and be able to redistribute widely.
Here are some of the arguments that we've made, improved social value, public services, more transparent and more efficient government. So this is really something that we find is that open data really improves government efficiency. If you know what needs to be public and private, you have to organize your own systems very well so that you can make that determination quite quickly.
And these are just some stats that have been used by the European Data Portal to make the case when they were developing their own portal. It's truly dependent on the context that you're working in, whether it's regional, national, or subnational, understanding what is going to ‑‑ what are the cases and what ‑‑ there's so many cases about open data. So what is the story that you need to tell that's going to say that's the kind of growth that I want or that's the kind of social change that I want.
So, this is providing a legal context for open data. So, role of open data policy. So, you can do all the work within your teams and your ‑‑ and the organizations that you're working with to build up these platforms and to build up these relationships and to open information. But unless you really have either a law that says you must do this by default or at least a government policy, the second someone changes their job, or roles and responsibilities shift, all of that work goes away.
So in addition to training up how people are working and really getting them on board, you need to have a document in place that everybody can point to and say, this is what we're requiring and this is what our values are. So this is the portal that they built. This is now about ten years old. It looks about the same. It is very hard to change. So once you do lock something in place, just know it will be there for a while.
And anytime that you do this, it's not just about we've got the law in place and we're requiring them to do this and they have to. That's not going to sustain. People want to be useful and they want to help, for the most part. And so how can you build their capacity so that they feel a sense of ownership themselves. So, working to develop that culture of openness.
Really training people outside of government and also civil servants so that they understand what they're doing, how their jobs change day‑to‑day and the social benefits of the work that they're undertaking. Because anytime you implement a new policy, regardless of the topic, but for sure open data, you have to do a lot more work. And it takes time. And if you want a civil servant to add more time to their work, being able to link it to social work is very beneficial.
This is what ‑‑ a methodology that we use. We use hackathons. Hackathons have been around for years. The benefit of hackathons especially with policy‑makers and politicians is it is visceral. They can see people using the data. They show up. You can teach them something small about how people interact. If you have maps, they can see the data, they can see it mapped out. And then they can feel connected and part of the community.
So I think social events are really, really vital. It's also a place where youth are very comfortable. I love that. So I recommend ‑‑ and you'll see throughout here, we have hackathons in each of these and there's a reason for it. So that's Ghana. Let's see.
This is about 350 students on a Saturday. Florence runs these about once a month. You do not have to have that frequency. She's very impressive with the community that she corrals, but I do think knowing there's that consistency, there's a space to go where people can show up and they can contribute and where governments can show up and they can meet members of the community is really helpful.
All right. So, this is the work that I did to come into this space, the Open Data Policy Out of USAID. We started with trying to open up agriculture data. And I will say, why agriculture data. In ‑‑ it is important when you're trying to figure out what data set you could make open is to choose one that people will not find terribly controversial. So, there's many. Any data set could be controversial, for sure. Land use is very controversial.
But at least in the context of crop yields and weather patterns, this was seen as much more neutral. So it was much easier to create a prototype about let's open up this data set and show how the process goes, than if we had chosen data about health, women and children, security, or anything else.
So there was sort of an effort to just rebrand into something called a Datapalooza. So you might see case studies of this. I think we've reverted the language, but there's a lot of this process of not just hosting the hackathon but going through the cycle of opening data, writing a policy on data, bringing the people together, engaging with the data, bringing out some prototypes and using that to iterate on what the process can be continuing and going forward.
So, this is a hack from ‑‑ with about eight different countries joined, which might seem ‑‑ at the time it was sort of before we had a lot of ‑‑ it was all Skype. It was before we had a lot of awareness that this could happen. This was the first hackathon that USAID did, that an arm of the U.S. government had ever run.
And from that, we built out a prototype and we built out a prototype of how to open up data and we built out the open data policy. So, if I were to condense what those learnings were, if you are within government, or if you're working outside and you find your champion, really important to find a catalyst.
What you need is a prototype, a paradigm of what it could look like. Once you have a prototype and you have a story, you can share that around and say that's what I want to do. How do I replicate that. You find your colleague, you find your contributor, you find your person who is right there with you saying how do we open this.
If you don't have a friend in this, it's pretty hard to do this all on your own. And I don't say that to discourage, but it's really important to find a friend. A lot of change, a lot of government change happens because you have a small group of really committed people.
So, try to choose a topic that's not controversial to start. You can go controversial later, but try to find one that is easy to tell the story on. Figure out who you actually need to have on your side. It's not going to be always the people who are first to raise their hand. You often will need to figure out who approves from legal, from privacy, if you have a privacy requirement, who are the subject‑matter experts that collected the data or are in charge of it and you need their approval.
Mapping out the steps is important to build the plan. When I did this, I realized you would have to go through 47 different people's approval, which is a lot. But even just the act of writing out those 47 steps we could say this is too much, so how can we tighten this up, but building out what that workflow would be from collecting the data set to making it open meant that we could put a real process in place.
I always recommend including media and communications people in the beginning, because you need storytellers and often, especially for those of us who have worked in government, you do so much work and then you try to share it with the world and hope that everybody's excited. But they weren't part of the process so they don't own it in the same way. So I would just bring in their storytellers as early as possible.
And then making time for implementation and institutionalization. One of the big flaws that we have for policy‑makers in general is that we will write a policy and then we will think that it's done. And if you don't take time to actually change people's jobs, change people's work plans, then it won't sustain and it won't stick.
So all of these things that I've said, there are guidelines. There are job descriptions. There are case studies at that website. It is a U.S. lense, but it has been forked and taken around the world. So I would definitely recommend, it's written on GitHub, it's easy to copy and share. If you're looking at job descriptions or guidance on the hackathons, that is a location.
And so after that prototype, that development work, we have the Development Data Library, data.USA.gov, it cuts across all different sectors and it does sustain. This was under the Obama administration. We've had a President since then that was not interested in transparency and collaboration. This stayed online because there are groups of civil servants that sustained it because it was part of their work.
It wasn't just something that a political group put in place and left. Okay. So I think I've talked through this. Again, find the data set and write down each step of the process. I created a working group with the approvers. It was really important and beneficial within USAID to take notes and share those out around the world. Often work happens in silos, or it only happens at headquarters and nobody knows what others are working on.
So demonstrating ‑‑ working on open data and working in the open was really important. And then synthesizing and then taking those 47 steps down to 8. And then when you publish you try to get the public on your side. So for sure you have your external person, if you're working externally you are the external person, and as soon as that data goes up, that's when you hit your communications campaign.
Isn't this wonderful, don't we want more of this, look at all these things that we can do. It's not just about that one data set, but the other ones that can follow. And then the last example ‑‑ just because you've heard my voice for a while, I'm going to see if it's possible for Kait. Is it possible to have a virtual speaker join? It might be a little complicated.
All right. We have Togo, Ethiopia, on the line. And there's a few others who want to introduce themselves. Kaitlyn Holm. You have the slides in front of you. It's about five and hopefully we can take the last ten minutes to hear from people in the room. It would be ‑‑ maybe if we have her face. If it's too hard, I can just ‑‑ she's unmuted. Great. Kait, can we hear you?
>> KAIT HOLM: Yes. Can you hear me? Can you hear me? Hello?
>> KAT TOWNSEND: Yes. We can hear you.
>> KAIT HOLM: Okay. Perfect. Okay. So, I will talk to you guys a little bit about some of the work that we did in the UAE opening up their ‑‑ opening up and building their open data portal called Bayanet, done with the Telecommunications Regulatory Authority. This was done as an initiative from the UAE, which is a conglomeration of seven emirates. This was done as an initiative to bring together some of that very decentralized data that they had in order to create a national platform.
And as part of this, we hosted a hackathon for happiness, which helped us to show and establish the fact that there was a need for this data as well as how people intended to use this. Because there's always a little bit of a tricky dance with helping to convince governments to open their data. And illustrating one, that there's a need, but two, specifically what types of data might be most useful or most valuable.
So, in this case the UAE was really interested in innovation as well as the possible economic benefits that might stem from that. So this hackathon for happiness that we created in partnership with universities took place across all seven emirates and really gave us a large swath of data on who might be interacting with these published data sets and how these interactions might produce applications that may have economic benefits, social benefit, or benefit for civil society.
So, like I mentioned, this hackathon took place with ‑‑ across I think about six weeks or so? All seven emirates with universities. But it also included people from industry. It included civil society members. So it gave us a pretty large data set to pull from.
There were some challenges with building this open data portal. In particular, one of the challenges, as you can see here, that we had was converting data between Arabic and English, making sure that both were represented. As you can see on this slide, these are some of the steps that we followed.
The first step was finding the data. This was quite a bit of a challenge because the data that had previously been published had mostly been for internal documents or organizational metrics. So we had to go to each of these different ministries or agencies. Some data was from university libraries. Really, really difficult to find pieces of information at certain points.
Once we did this, though, we were then ‑‑ had to look and explore how the metadata, how that information had been gathered, how it had been cataloged. And this was all part of an initiative to, as I said, federate or nationalize the data. So we were combining data from one emirate that may have, for example, let's say on camel populations, may have take then data and only looked at camel populations around watering sources.
But there might have been another that only measured camel populations in rural or mountainous regions, so understanding that metadata was a real key in order to create more complex and comprehensive data sets that included information about the entirety of the UAE.
Once we were able to understand that, the next step was to clean this data, like I mentioned. Sometimes these internal documents and metrics were not always clear or consistent about how they were documenting the data and what types of units for example they would use.
And then we had to convert the data. This is that slide you see now where ‑‑ or the image you see now. And part of this was sometimes the data would come in Arabic, sometimes in English, but it was really important that we made it accessible in both English and Arabic. And this could be a bit of a challenge because especially with the Arabic font, a lot of programs won't recognize that that is a font or a script.
So there was quite a bit of a challenge in figuring out how to either scrape that data or how to input it in a way that would be machine‑readable across multiple platforms. The next step was to ensure the quality. We had in each data set, over 2300 data sets. We had a language editor, two data auditors, and I personally looked through and viewed every single data set to make sure that this data set matched the initial source material and wasn't duplicative of other information we had published previously.
Last step was sustaining the data, working ministries and making sure they knew how to publish and continue to publish that information. And this links us into the last point here, which was on training and visualization. So part of that was training the ministries on how to continually update their data, why it was important, and how to follow up.
And then visualizations came in the form of trainings at our hackathons, how to help students produce visualizations based on the data and metrics they were working. And excitingly, we can see that this project has gone on to be quite significant as the UAE has expanded upon this and developed Bayanet AI, trained based on this data that we initially put up for work for Bayanet. So here's an example of the ‑‑ what it looks like when ‑‑ the landing page for the open data portal.
And you can search here using any of the terms that are included in the data sets in both Arabic and English as well as tags, which we added as part of the metadata and any of the text in the metadata. Okay. Then this next slide, this is actually a picture of us at the hackathon at the planetarium. You can see Kat in the corner. Hi, Kat. (Laughing)
And this image shows one of the teams presenting their work to one of the sheiks. Kat, I'm going to let you take it from here. If there is anything I missed, please feel free to add it.
>> KAT TOWNSEND: That was awesome. Thank you, Kait. It's not a terribly complex formula that we run through. It just takes a bit of work and some time.
If we were able to start a little bit on time, this is the kind of breakouts I would ask you for. I think probably the third ‑‑ you know, I am happy to help you all and we are happy to help you all run through any of these. And if you're able to answer these three questions, then we will be even more effective in helping you chase down, okay, how do we get that prototype together.
So, what data is possible to make open. Who can you work with to actually do the work to get that data open. And I think it's really important to tell the story and answer the question. If this, then that framework. If you open this data, then that impact will happen. So, for example, if we knew about internet quality and cost, we could build maps of companies that were supposed to have a service and we could allocate funds the way they're intended to to improve connectivity.
But truly any way that you can get it into this framing, it's so much more impactful if you say here's how I will apply it and here's the impact. And then I don't know if the last slide comes up for you? It might need a refresh. Maybe the deck does. But otherwise I'll just share it on the Zoom. It is just the contacts. It's just a slide that has the links that were shared today and contacts if you are interested and if you want to follow up with any of us, because we're at about four minutes.
With that, does anybody have any questions or want to share their experience? Please.
>> AUDIENCE: Yeah. Oh. Thank you. It's really not ‑‑ I don't think it's a question or something, but, you know, it's just I'm researching on internet governance, open data. And I was doing this particular research not comparing models, but focusing on the Korean model of open data, because I found that they have been doing a fantastic work. India and South Korea have been promoting open data, including data related to the pandemic lockdown and people who were inside and outside and all of that, which is why it was easy for the government to track people.
If people were hiding, how to reach them in case emergency ‑‑ in situations of emergency. You know, we realize, there's a possibility of being used, all these tools used as surveillance, right. And one of the things that ‑‑ I mean, of course that's related to a personal data. But then in South Korea there was also this particular problem that Google for many years very persistent to ask the location of certain kinds of buildings within the geography, which South Korea did not want to reveal for the fact that because of North Korea and the potential damage that ‑‑ the potential attack that they could create in the critical infrastructure.
So now that is nonpersonal data. So maybe the point I would like to raise to everyone who's thinking of the government's perspective, if they want to open data, personal or nonpersonal, you need to see your country's goals or agenda that you want to meet and not just follow the model that is going on in every other world, because they may have different values. They may have different targets to meet.
But what is it that you want to do? The second point, there was this one particular research which happened in India and they were just analyzing what kind of data they have. And the problem was that the quality of data. And it's not just in India but pretty much everywhere. Yes, it's machine‑readable, it looks fantastic how it is there. But then if the data is ‑‑ the juice is actually in the details of the data.
If it is not detailed it is pointless. You are just wasting public resource then. So just having to say that we have an open data initiative but it doesn't do anything because it doesn't give enough information to be used. Which is why before opening data I would say quality standards like having standards and ethics is the first thing that a government should work on or anyone that wants to work on.
That's exactly what happened in India. They opened data, but then they did not have those values and now they have this problem, what do we do with all these things, there's a waste of public resource now. Things are getting on track. They are working on these things. So that's just a lesson I thought I could share with everyone.
>> KAT TOWNSEND: I do think for sure all of this work is a balance between ‑‑ as many of the themes that we have at IGF. Freedom of expression, of information, and security. What looks like personally identifiable information in one context will be different around the world, which is why the U.N. doesn't have a global open data policy. Each country and subnational needs to change.
What seems fine in one government, you change governments, and it is a completely different situation. So trying to start with, again, those more neutral data sets, data sets that are a bit more about what the government does and less about people is sometimes ‑‑ can be helpful there. And for sure on the quality of the data, yeah. You can definitely have insincere actors who check a box.
Look, we did this work, it's open, you're still complaining. They say in civil society all the time. That's where the application is really important. So you don't just have open data for open data's sake. We're opening this data in order to apply it here. So you can see that as a whole package of stories. We just don't have it as a resource, but here's where it's being used in service.
So if they are consistently seeing this is the kind of data that is useful, this is the kind of data that's just fluff, then you can make it possible to have some kind of catalyst and better examples there. They did give us an extra ten minutes.
>> AUDIENCE: The thing is that government data, although it is a structure, but it is not organized. They have the format of a structure of data, but actually the data itself is not organized. Sorry. Oh, sure. In that case, we are talking about government data. So it is public data. That's what we are looking for, economic value, social value. The slide says how to improve public services, transparent government, efficient government.
How you would convince the decision‑maker, what control comes under each line that tells him if this is achieved, if we can format the government data for the sets ‑‑ again, noncontroversial, neutral, available to all. What controls you would apply here or indicators that you would have to the decision‑maker? This is met, you will perform better. Bureaucracy is heavy in the government. You can find departments within the same government entity and they don't share data, right.
And now you're talking about open data to others. So you're talking about the top management who may not know the real data with the middle and lower levels, right. And they need it. And they know that they need it. So how you would convince them that such approach is ‑‑ can be efficient? We do have a law of freedom of information, but to what extent it's in effect, that's a question.
>> KAT TOWNSEND: So, I think your question would benefit more from diagnosing, let's try to figure out what the topic is. But I think what I hear is what ‑‑ who are you trying to convince. So identifying who the stakeholder is, not just government in general, but can you find one team, one leader who has the power to say I can at least ‑‑ I maybe can't open up every ministry but I can open up one section in my ministry's information.
>> AUDIENCE: (Off microphone)
>> KAT TOWNSEND: I hear that. And ‑‑ but it is hard. And so ‑‑ and I will say that just one of the things that anybody gets concerned about is they don't want to look bad. Everything is their reputation. They get very scared of opening up data because they don't want it to be messy and show that they didn't do well. And they're ‑‑ so now you give them an option where instead of being criticized heavily, okay, let us help you.
Let us help you with a good story about how you're being proactive. You're showing your information. You're being transparent. You're bringing in the youth with these hackathons, you're bringing in the community. Thank you so much. And it is important to have top‑down, but you need to give them a story, a few examples of here's how we've changed and there's stories in the newspaper about how great this is.
And I'll say I am sure your context is difficult and I'm sure Sri Lanka is difficult and I'm sure Maldives is difficult. And I can only offer that in the U.S., the offices don't share data with each other. The offices don't organize the data very well, themselves. If you ask a Freedom of Information Act they say we'd have to find that data, it will cost you $20,000, do you want to pay $20,000 for us to answer your question?
So there's not a perfect model. There's only all of us trying to build slightly better models for how we can have better government systems. So just to say that all of these resources that we're sharing is that attempt to try to bring that information‑sharing into these different spaces that are reticent to do so.
So . . . But are you saying that you don't think it's possible without a top‑down approach?
>> AUDIENCE: How you would create a commitment? Governments ‑‑ they are too much in deep of their bureaucracy.
>> KAT TOWNSEND: Yes.
>> AUDIENCE: 20 individuals, is busy to finish their work. Now we are asking them to publish information.
>> KAT TOWNSEND: Yes. For sure.
>> AUDIENCE: Will not realize that publishing the information would make his life easier.
>> KAT TOWNSEND: Yes.
>> AUDIENCE: The beneficiaries of the service or his organization. So, bureaucrats are always busy with their work. They are inside the box. They are not out of the box.
>> KAT TOWNSEND: Yes.
>> AUDIENCE: Now, I'm not saying that this is the right approach, but it is our perception that the top leadership would always see the case because they need information to make the right decision. So they would say yes, we want certain information to be public and certain for us to be in the loop and make the right decision rather than say, what was the case here or there and wait for a week or ten days and he need to make the decision yesterday, because the information is not ready.
Or suddenly there is a crisis. Or a political issue that just came up and suddenly he's exposed, right? So, you know, there is no mindset for open data. There are countries that hold government entities to data and to a specific format. This format is outdated or that mechanism ‑‑ at the time it was done, it was good, but then there was more data needed, more information needed, can update it, so it becomes old that nobody is using it and there is demand to modernize it.
>> KAT TOWNSEND: You're not wrong. I see the hand. I want to make sure that other people can ‑‑ because the point is networking. People can help each other.
>> AUDIENCE: In Sri Lanka, there's an initiative we joined, Open Government Partnership. In 2015 and '16 there's a lot of civil service society activities around it and the government also responded. I think we could submit a second report for the 2019‑2021. After that I'm not sure. But ‑‑ right to information. So if the data sets can be more open, it's not only for individuals to send requests for the paper documents and all of the things.
So it's kind of a win‑win situation. So open, what government can show to the people or the stakeholders, there's a list ‑‑ request in particular.
>> KAT TOWNSEND: I think that's very true. I will say that usually ‑‑ we're just going to close ‑‑ usually if you're looking at the people who submit Freedom of Information Act requests they want a very specific data set. And often it's from a journalist perspective, it's accountability, which is vital. It is hard to get civil servants to convince their bosses to get excited about accountability, because it doesn't feel good to get yelled at.
And if you want to get people to be excited about transparency, you give them a positive story of why it will be in their interest to open it up. And if that is you're helping the youth, you're building businesses, you know. There's a nice story in the newspaper, everybody cares about their reputation. When you build those models you can shift perception and then yes, as happened, it took 15 years for the U.S. to do it but we have a website that is easy to use.
It does take time. Nothing that you build today is going to be modern in 5‑10 years and it is about that consistency. But it's a global community, a lot of people working on this. We want to work together. Appreciate your time. Thank you so much.