 Okay, let me start with or let me explain the basic background here, like I was listening to you on Thursday, local time on Prague City Data Congress. So this is basically a follow-up interview, I would say. But I will start with something like I was kind of forced to do in my podcast to cover also the COVID related topics. And this is a great opportunity for me because I'm really amazed by the numbers. You, I wanted to say scored, but that's not a good expression. No, it's a very Macabre humor. But the numbers are extremely low, like only I am using the Google numbers right now. So it's 442 confirmed cases and only about seven deaths. And it's been a month, it's been a month with no local confirmed cases. So it's largely over for us. Yeah, how did you manage that? Because what was interesting for me that you said on Thursday, that you're not using any location based application? We're not using app level tracing, as you mentioned. We have had no lockdowns. Our business and schools are never closed. And so this is because we acted very fast. We started our response last year. Many jurisdictions started only this year. That's the crucial difference. And also the fairness of the supply of masks and so on. That is also essential. We see the mask for all campaign and that is great. But we started that I think in February, if not January. And also I think the fun of everybody communicating in a way that sees the humor being used so actively, not only by the government official and their spokes dog, but also by every scientist involved. They get into this co-learning culture with the entire population. That I think is the often missed part of the social innovation, because that makes everybody feel, oh, I can contribute to the counter coronavirus. It's not just a government thing. Yeah. What was interesting, you mentioned this kind of like data related way to deal with the crisis, when you basically followed the information on social media and then reacted based on some, let's say, checked data. Well, it's basically agenda setting by the crowd, right? So if you have sufficient amount of people who are informed or at least have the amateur level of knowledge, just by the algorithm of upvoting, you can see what's trending, what's the important topics on a given area. This is exactly as how Reddit works. Yeah. So basically, because you reacted like based on the Taiwanese version of Reddit information, you basically installed the health checks on the airport, is it right? Yeah, that's right. And we set up the health checks for flights coming from Wuhan January 1st. So that's the first day of this year. So way earlier than pretty much everybody else. And also our central epidemic command center was set up even before we had a local confirmed case. Yeah. So basically, this is somehow a part of the very open approach of Taiwanese government because it's not only about sharing the data with society, with your people, with your citizens, but also let's say, yeah, using their data basically. Yes. So do you think that's something we, and by we, I mean the rest of the world probably, is that something we need to learn from this situation? Definitely. And Taiwan did not start to be a perfect score according to your words in stars, right? Back in 2003, our communication channel is very chaotic. Our municipality government and our central government says very different things. We had to barricade an entire hospital unannounced and there was no fixed state of termination. And so it's all very traumatic. And I think we learned from it, knowing that it's only through collective intelligence, it's only through everybody becoming kind of like an amateur epidemiologist, can we fight a COVID together? In the beginning, we didn't know that the novel coronavirus is different from SARS. So we just treat it as if SARS has happened all over again. But it turns out it's not SARS. It's even more difficult with even higher Arnold value. And so we had to improvise on a lot of things, including but not limited to the digital fence, which is the alternative to the quarantine hotel. If you return to Taiwan, you want to go to quarantine hotel, we do that and we pay you a daily stipend. But many people, if it's for 10 days, they prefer to stay in their home if they're not living with people in the vulnerable groups. And if they do so, then we use their phone and use the cell phone tower signal to make sure that they do not escape the digital fence. And that is, of course, compromising the privacy and the freedom of movement, but compared to, for example, a wristband or any other invasive technology or GPS based technology is actually much more coarse grained because using cell phone tower triangulation, we only know about the 50 meters radius. And so it's not as invasive as GPS. And also it's time defined, right? It's just 14 days after which there's no constitutional basis for us to track the whereabouts and to send SMS to the household managers or to the police. So my point is that people know the epidemiologic reason of why doing so. So they can improve the process, for example, writing chatbots that automates the checks and so on without compromising the value because the value according to the constitution is to protect the public health. And we never declare a emergency situation. So everything we do need to be Congress approved. It need to be accountable to the parliament and it need to work in non-coronavirus days as well. So that makes our innovations much more transmissible to the rest of the world because you don't need to make a constitutional exception to use the Taiwan model. Yeah. And also, like when you're talking about the, let's say, bridging into your privacy with this digital fence, I guess there's the line. That's exactly the point when you're breaking the privacy, but with, let's say, voluntarily. I can choose that you will or will not do that. So there's probably some kind of border between being, I don't know what to say, well, voluntarily in these meetings and in these use cases. And I'm kind of like going to the question of keeping your data private because that's something you mentioned on Thursday too. That's why you didn't use the location-based app. But this is kind of like a little bit same, but not that invasive. That's exactly right. So the point of the cell tower triangulation is that it's just for home quarantine. It's a defined part of the population. And for people who we don't know that they are not infectious, who are suspectly infectious, which is what quarantine means. That's the constitutional principle defines because of the act about communicable diseases for these people, we have the constitutional basis to do a time limited 14 days in this case, tracking of their whereabouts. And the constitutional court says, even though it may be constitutional to barricade with no other choice, like just to confine them in a quarantine hotel, we need to find more humane alternatives if it's at all possible. So that's why the home quarantine instead of everybody going into the quarantine hotels. But we still do provide quarantine hotel if you don't like the idea of digital fence. And also we still provide a quarantine hotel if you don't have your own apartment or if you would be forced to live with people in the vulnerable groups, which put even more people in danger. And so the idea is that if the human right of this entire process is guaranteed by the accountability to the parliament and to everybody who care to ask, then we're staying good on the constitutional side of things. If we collect this data for other purposes, other than ensuring the quarantine, then we extend outside the original purpose of this data collection, which is why we prefer not to collect new data. This data triangulation is chosen precisely because the telecom providers are collecting that data anyway to improve the service. So it doesn't require a new review of data and a breach of data. If that happens, what happens? We rely on the existing mechanisms to collect, to process. We just apply the data in a different way and only in a time defined window. Okay. The reason why I'm asking, here in Czech Republic, where they installed the app, it's called eMask, basically. Oh ours is also called eMask. But we use it as to pre-order mask. So I guess it's not the same. No, this is different. In Czech, it's Eeroška, which Eeroška is Czech expression for mask. And basically the principle is very easy. It's based on Bluetooth technology. So you won't really turn the, let's say, Bluetooth recorder on and you're collecting just this data. Like, okay, I've met somebody, some other person with this app turns on and I'm not collecting any other data. So it's only collecting the data of people who are who willingly installed and turned on this Bluetooth app. And it's kept in your phones or is it transmitted to a central registry? It's in our phones and when something happens, you're diagnosed with coronavirus, you can share the data. But if you choose not to, if you choose to remain at risk to the society, there's no way that the health authority knows about this. Yeah, I think so. Okay, then that's good. It fits our principle. When we send those SMS to the people who overlap with potential high-risk areas, this is just like sending out the SMS to potential earthquake warnings or heavy rainfall or whatever. If the people choose to do nothing, there's no constitutional basis for us to track them. If the people choose to, of course, do something, like go to a clinic that's far more preferred, then of course we know that they're returning because they say that they've received this warning. But if they don't, we don't force them to. And I think we agree with the Czech Republic on this one. Good. That's a good news for creators of the app, I guess. I won't tell them. Yeah. And if it's open source, if everybody can see that it is actually working the way it's intended, because if it's not open source, even with the best intention, maybe it could actually do something more, but the original designer did not foresee that. So if it's open source at least have its source code published, then of course it's even better. In Taiwan, when we hold the coronavirus hackathon or co-hack.tw, we ask all the entries to use the MIT license, a permissive open source license, so that everybody can apply it to different platforms, but also to make sure that the developers are doing what the designers intended. Yeah. So generally, would you say that this situation we're in right now has changed something about the approach to that open source slash private data? Is it good idea to, let's say, push our boundaries in a good way, with a good will to help prevent the spread of coronavirus? Well, I mean, even though there are some data that are private, strictly speaking, there's very few individual private data, like if we're having a video conference now, right? And so this data, this context belongs to both of us. And if you go to a diagnosis, of course, the person who run the diagnosis and you both know the result. And if you even as simple as taking a photo of you in a party or something, that basically is a shared fact for everybody who are in the same room. So data by its nature, because it's seldom the same person collecting one's own data processing and applying it just for that so person's purpose. This is very rare. So data is social, intersectional by nature. Now, it's human nature that I will want to, for example, I take a photo of us, I will want to share with everybody in the same room, but I do not want any random person, much less the government have access to the fact that I'm taking this photo with a group of people together. And the old problem was that if any of us post it to social media, then we basically compromised the negotiation bargaining power of everybody else, because then that social media company will have not have to ask the other people in the same room for that selfie. They have that selfie already. And so this coronavirus, I think because the data involved is not as simple as a photo that's taken in the party. It is actually health data, the more sensitive ones. You still share it, but you likely only share it with people who you trust to act in your best interest. You share with your doctor and nurse. If the contact tracer, the medical officer come to you, of course you share it with them. You'll probably share it with your family and probably not nobody else then. So the challenge is how we build a data collaborative that only allows your preferred, your comfortable amount of sharing. And for everybody else, it shares in a way that is aggregate or statistics, they still improve the public health, but without compromising your privacy. That is the challenge. I'm happy to see the Czech Republic rising to the challenge by designing an app like that. Yeah. The thing is that I was surprised a little bit that the share on market of this app is really small. It's like, okay, Czech Republic has about 10 million people living here. And the last number I know is about 300,000 users. So that's pretty low for me. I would expect that when it's right now, we're in the talks like this app being included in some governmental, let's say, program, anti-covid program. Probably there's a more specific expression, but I cannot find it right now. So what I was a little bit surprised that even the techie people, the young generation, was still afraid to install the app and to share the data or basically collect the data because the sharing is the last step. And that's the only way. Yeah. However, I think that's healthy suspicion. I wouldn't say that we should blindly trust the apps. If it were me, I would insist on seeing an independent audit on the cybersecurity side. And also because I know how to code, I will also look at the source code and how it was developed before choosing to install. And in a sense, this is a good practice both in cybersecurity and digital competence. It's just that with the time pressure, of course, people will want to have a wider dissemination of this app. And so I think one of the point of using the digital fence is again, because we use it in a way that doesn't require collecting new data. And we use it in a way that's very clearly marked in its purposes. So I think for contact tracing, if people can see that this somehow benefits them. For example, it's a app that traces my whereabouts, but I don't share it. I don't broadcast any Bluetooth. This is strictly my diary. And then when the contact tracer come to me, I can generate a one-time use website to them, a one-time URL to them, and it's containing only contact tracing relevant information. And without sacrificing any other families or friends, privacy information, then that's I will be willing to install because it serves primary myself interest. Because nobody wants to divulge unnecessary privacy information of their friends and families if the contact tracers come to find them. And so this helps me and also incidentally helps other people. However, if you build an app as something that primarily helped the other people and of negligible use of that person, then it's a harder sell. And this is why for medical masks, we say this protects you because it doesn't make you want to touch your face all the time. It reminds you to wash your hands properly. And that's why we wear a mask. And we don't say you wear a mask to protect other people. That would be a harder sell. You've mentioned it before, I guess, the open source principle that you would check or at least you would want to know how the app was called and what so ever. So for future events, and unfortunately, we can expect something will happen again, something like that, something like this situation. So maybe in the future we should I would say like to not to force, but we or at least the government should want to be open source in these these apps and these let's say let's say precautions to be open about what we do or what they do to protect our health and how they do that. That's exactly right. If people already use that sort of apps in non coronavirus pandemic situations, for example, I can easily think of the kind of personal log as something that people with chronic diseases, they will probably also want to use it because they want to their phone to remind themselves to change their habits for the better for precision health purposes and so on. And so if you add those functionalities, they become something that people are willing to work with every day. And also they trust more because there are more data they feed it. They know that nobody can do a security breach or something that compromise their security. So when pandemic happens, you just say, Oh, by the way, here is a new field that you report whether you have a fever, for example, and then people will be much more willing to use that app compared to a green field application. You mentioned that basically at the beginning you reacted as SARS was happening again and then that's right. And then we discovered that's SARS 2.0. Yeah. And that's the last time SARS actually happened. The Taiwanese government was not that prepared. No, not at all. Yeah, so I guess you had the process of learning in between these two situations, but what have you learned right now? Like within this situation? I guess there still was some space for some new experiences. Yeah. Well, first of all, everybody should learn epidemiology, learning why soap works, why the hand sprays work. These remain the most important technology. Everything else is just adding to that. But if you look at the R0 value and really soap and hand sanitation in general and physical distancing, they dominate the factors. Everything else is good, contact tracing and things like that if they're necessary. But if you don't do these two, then everything is doomed. And so I think knowing why the epidemiological model and with playable simulations, I often recommend people to play a simulation by Nicky Case. And that simulation I think is one of the best that I've encountered so far that explains exactly how epidemiology works and what individuals instead of being anxious can do in a day to day way to spread the right message on how to mitigate the coronavirus. And if you get into the habit of sharing what you have learned, then the scientific knowledge has a higher R value than conspiracy theories. And only then can the civil society mobilize into a collective intelligence think tank and do tank to counter the coronavirus instead of waiting for one or two people to make incidentally the right decision. And that's not the Taiwan model. Yeah. We're kind of getting to the end, but I wanted to ask about something which really, really amused me on Thursday. There was your humor over rumor. Oh, yeah. That's the dog in the bottom. On one hand, it's really simple. I totally get the importance and basically the principle. But maybe can you explain like, how did you get there? Like, that's kind of courageous to have your Yeah, because we have no other choice, right? In Taiwan, we pride ourselves in being the only country in Asia, according to Civicus Monitor, that has a complete freedom of speech and press. And a journalist's word is as good as a minister's word. And for many other jurisdictions in our vicinity, that's not the case. When the disinformation crisis took everybody, all the liberal democracies by surprise, there's many jurisdiction that says, okay, then let's just give the administration more power. Maybe they can do a takedown, maybe they can do a lockdown, well, in the info sphere, not the city. And that, of course, is the easy solution. But then it decimates trust from the citizens. And it also shows that the government is not trusting the citizen enough to entertain different ideas. And for the things that are intentional that causes public harm, of course, there should be a response strategy. It's just takedowns actually angers even more people, because those conspiracy theories thrive on outrage already. And so if you take it down, people are even more outrageous about the takedown itself. So how to counter the infodemic, disinformation, even intentional malicious ones without resorting to takedown is kind of the question posed to Taiwan. And we can answer only by innovating, because there's no existing playbook for that, or the easy answers that compromise the privacy and the freedom we cannot choose. And so we have to try all sort of different things. And it turns out that humor is pretty much the only thing that worked. And rapid response, of course. And so a quick humor over rumor turns out to be the most welcomed response from the citizens and from the civil society, because then you work with journalists instead against journalists. And no solution that I know of that work against journalists can win in disinformation space. Plus you're showing this open source principle again. Because you, the part of the thing is not only the humor, but as you mentioned on the example of the issue paper rumor, you actually shown them what is what is produced. Exactly. Where and so on. Yes. So it's about remixes. It's about allowing the comedic professionals to have the full reign. But the payload is still very serious clarification. If you only have what's funny, but there is no point in the fun, then people don't share it voluntarily. They may be just straight for the laugh, but they don't add their own message and their own reading into it. And adding their own message and reading ensure that they internalize that scientific knowledge and can make useful contributions to the community. If this is just something that's funny, followed with something that's clarifying that scientific, these two doesn't connect in the mind. And so the conspiracy theory will then still have the opportunity to sell this court. If you have an adversary or many adversaries at that that tries to sell this court, your only consistent response would be to make each message stand alone by itself so that when people see it before seeing the conspiracy theories, they have a vaccination effect so that the next time they see the rumor, they say, hey, I left about it. It's not like that. And that's when you know you have succeeded in the humor over rumor playbook. Final questions. Aren't you afraid that when you in the future there's going to be a different prime minister that he's not going to be like that funny? He's not going to have that sense of humor? Well, our vice premier is also very funny. But what I'm trying to get at is that this is a culture. When the ministry see that the premier tries this and it works very well, the people who are involved get promoted. The author of that, we only have one pair of bottom each gets promoted from being a assistant to the premier is now our administration spokesperson, our official spokesperson for the government. So everybody and their comedian friends see that this is really the one true way. If you want to get into the communication in the government, you better start making friends with professional comedians. And so I think it has become a culture. It's not a top down thing. It's just people is realizing the science behind it and the kind of mental pathways that we can take from the single kind of thing that makes us upset. You can go to outrage or you can go to humor. And these are mutually exclusive. This is basic science of the mind that people are starting to understand and applying it into public communication. Great. Perfect. Thank you for your time. I'm really glad you connected.