 I'm just visiting Taipei for a few months, been here about six months, I work as a data scientist in San Francisco and my partner and I were planning on traveling to Taiwan for a little while and then like COVID struck and we ended up getting it coming here on the go-kart visa so we've been spending, we took some time off and now I'm working remotely. And I was just like reading a lot about the digital democracy efforts that you've been meeting in Taiwan and I wanted to ask more about that and understand it. So in particular I was curious about V-Taiwan. Which is weekly meeting right here. Okay cool. Yeah I guess like one question that I had for starters is, I looked at the website and it seemed like a lot of topics hadn't been updated recently is it still active, still going on? Yeah it's still active. Recently they've worked with the open parliament people on getting the national action plan from the parliament side of open government out. There's also a recent discussion about social enterprise related acts and so on. I think there's going to be one about like shared, I don't know, U-Bike or shared E-School terms or something, but I don't run V-Taiwan day to day anymore because I've become the additional minister in 2016, October and then I just handed the governance, the leadership of V-Taiwan to other people in the GoPierre community. But we've taken the V-Taiwan inspired model and installed a roughly equivalent structure within the central government called the Participation Officer Network which primarily oversees for example cross-ministerial repetitions on the joint platform. Yeah I might take those. Sure of course go ahead. Cool. We're going to be on the record anyway. I'll take notes anyway so it's like. Are the meetings open to anyone? Sure of course. Okay cool. I just dropped by. You too can't drop V-Taiwan. Okay cool awesome. And then what are you, so I think there was a paper published in 2018 about the process and I think at that point the major initiative that had gone through the process was the Uber regulation. Yeah. It had to be like Marines. Yeah. On my liquor sale and so on. Okay yeah. I guess maybe you're not the best person to ask but have you had any like learnings about the process since then? Like things that have been successful and things that have been most successful. So V-Taiwan had a very narrow mandate in that only emerging digital economy related issues could be deliberated. And this is fundamentally because it's a self-selecting, self-electing process right. People who find time to contribute set agenda through crowdsourcing. But if we apply this to item of marriage equality or some other domestic issues that are unrelated to the digital then the self-electing nature of the V-Taiwan community could be a problem because there really is no balance in including the stakeholders on the less digitally connected parts of the society. So it was a pilot, it was an experiment. And so it was quite successful in tackling the issues where there's no or almost no traditional representation. The first few cases for example there was no unions of tail workers, there was no unions of gig economy. Nowadays of course there are but back in 2015 there were not. There's no business association of all the start-up of the registrar came in islands. So there's no traditional councils or associations or heads of representative trade councils and so on to talk to you. But we need to regulate that anyway. So it's easier because there was no better alternative. But in more traditional politics where there are like the existing entrenched representatives, this methodology wouldn't quite work as is and it needs improvement specifically taking into the account the face-to-face component of deliberation. Yeah, that makes sense. Do you have plans to expand it to other areas like an increased participation? Yeah, it's already done. But the joint platform as I mentioned accounts more than 10 million visitors, I think consistently millions of visitors every month. So it's almost half or probably already over half of adult population in Taiwan, which is actually not the whole story because one quarter, more than one quarter of petitions there were started by people who were not even 18. So now adults also play a big part there too. And I think that's fundamentally where we're going because instead of having the 16-year old, 15-year old thinking that they have to wait a number of years before they participate in the democracy, why not just have them set the agenda through petitions. So we've already expanded that quite considerably using the joint platform. Yeah, that makes sense. Does the joint platform, is it based on bolsters or is it your own platform? Yeah, each joint deliberation can choose a different set of technologies, but the petition side is our own platform. But the design is from Iceland, from Betere Kavík, and where each petition, once they collect 5,000 signatures, get a point-by-point response by the ministries. And there's a pro column, there's a contra column, but there's no replicaten. So you can ask vote and downvote, but there's no way for troll to grow. And that's very balanced. But if it's cross-ministrial issue, like if it's a white disagreement even within the government about how to define the same, well, we usually use polis in a way to crowdsource the common values out of those, they're very different positions. That's right. Yeah. How do you authenticate using it on a platform? The people registered it like they already do. The joint platform uses SMS. Okay. Yeah. Okay. Oh, interesting. So just you have to have a time when you use number. Well, if you get like 50,000 or just 50 SIM card numbers, the anti-money laundering people will be out here anyway. Yeah. It is actually quite expensive to get a 50 SIM card without getting notified. But some people have two or three phones, that's fine. Yeah, that's it. Okay, cool. Yeah. And then let's see, so have you noticed like as you scale this like any pain points or any issues with, it sounds like access is really good. Yeah, sure. Broadband is human, right? Yeah. Yeah. And are there any other learnings that you have from like the, from the past like two years or like four years, like since you terminated the joint platform? Yeah. I think people really, really like to set the agenda by themselves without the government setting the agenda to discuss. The only exception seems to be like really early stage as a general setting. For example, when we ask people what do we need to do when we're opening up mountains for mountaineering, for hiking and things like that, there are of course competing values at play about natural preservation, indigenous nation preservation and things like that. But also better access also translates to more people care, right? And that's one part. There's also a part about opening up the oceans like professional fishing people and amateur fishing people. And these are like different profiles. And sometimes they see things as zero sum, but actually with education, many people can understand the oceans more and so on. So for these like really broad, like there's no fixed set of solutions, issues, then it's a little bit like the digital issues that we have used to tackle in that the crowd sourced feelings and the crowd sourced like general narratives from personal perspective start to become very valuable. But if this is just one predefined trade off between existing value like one value versus another, then there's less chance for the Polish or any crowdsource approach to work, which is usually a later part of the stage. So citizen initiatives or early enough stage conversation where it's about feelings and values that you seem to draw the most number of people. Yeah. And do you find like special interests like tend to be more or less represented? Because I think like, so one problem we have in the United States is that a lot of policy that's not at a community level is done by like in person meetings. And like just to get to these meetings, like oftentimes I might be held like during the day or like at an inconvenient time or convenient place. And so people who are working can't attend them and like maybe people who are retired or represented or something. Do you find that like having the online platform improve? Like does it increase the presence or decrease the presence of special interests in the decision making process? Yeah, back when V-Taiwan first used Polish, I think the first UberX case really gets a lot of people loving this way or another. But once we resolved out and saw the common values, the next one we tackled was Airbnb. And interestingly, Airbnb said they set an email to all its Taiwanese members asking them to go to Poland and advocate the Airbnb position. So I guess with asynchronous modes of communication, there's more opportunity for this kind of mobilization to go on simply because online mobilization takes less effort, right? But much to I guess Airbnb's surprise, only about one third of the members they mobilized actually agreeing with the Airbnb position because Poland is very high bit rate. It's a crowdsourced like weekly survey. So if this is just a yes or no question, maybe Airbnb would have succeeded in mobilizing the people. But when people go to Poland, they see very nuanced eclectic positions. So that's even if they are an Airbnb member, they would still prefer like, I don't know, better firefighting inspections, like better insurance policies, like better bargaining power from the individuals and as opposed to like people who actually own a chain of, you know, units that pretends that these are just my spare houses. Of course, that's not the case. Right. So so people still have like legitimate improvements that they really want to see, even if they have already lived in Airbnb for a few days before. Right. So I think more than two third actually offered constructive criticism to the Airbnb position, even if they actually given that they're mobilized by Airbnb newsletter. I don't think special interest is that big a problem because people are given much more bit rate to express. Yeah, that makes sense. And just in general, like, what do you focus on these days if it's not like running in Taiwan? Like, what do you mainly worry about like in the past like month? Well, counter-infodemic encounter pandemic has been a focus for the for the past year. It's been quite a bit of time with good car holders, actually. Simplify in the process. So are you a economic affairs or science and technology? Economic affairs. Economic affairs, right. So, yeah, the science and technology ministry just relaxed massively last Christmas so that anyone with the potential to contribute to science or technology are eligible, which is pretty much everyone. Yeah, because I applied on the science and technology track and had to wait for my application to be rejected, which took a month because I guess the past year. Yeah, I know, I know. Yeah, and most of the most gold card applications end up finding that MOEA is easier. So and most change that last Christmas. So it's a bit of a friendly competition. So I think, yeah, what we're seeing now is that Taiwan is becoming a much more transcultural place so that people are based in Taiwan but working on international projects. So the presidential hackathon, for example, we've seen record number of people who were not based in Taiwan, but nevertheless participate in the presidential hackathon. So I spend a lot of time working on the presidential hackathon and also similar hackathons like the Kohak and so on, which also use Polish to crowdsource norms that's acceptable around data for fighting Covid for the five countries participating. And we really get some pretty good like the contact tracing apps that generates one time link, but that's not actually report or you're aware about anywhere outside your phone or things like that. Yeah, cool. That's really exciting. So have you found success in like introducing like I think I think also like like you mentioned, you're working on rebuilding some government websites and properties with with more modern technologies and also introducing some crowdsourcing. Oh, yeah. Yeah, that's right. So it's like, I guess what it would have, your findings been in that area? It's kind of an open-ended question. But sure. Yeah, we run the RAID project, RAID being rescue action by youths. So RAID.pdls.tw, the RAID project for fourth year now. We're about to begin the fifth year where we ask for interns usually like younger graduate level, but also higher undergrad level to dedicate two months of their time, three people at a team to look at any government website that they don't like and change it for the better. And the first couple of years is all about the charts. The first year, the interns did nothing but fixing websites so it works not in an IE only way, which is good because the Microsoft deprecated IE right afterwards. And the next year all we did was to ensure responsive web design. So it works on phones of all different widths, right? But after those two, like very infrastructure work, the third year see, for example, the hiking portal where if you go to hike in Taiwan, depending on your trail, you will have to apply to like up to four different agencies, each with their own website. So we use the open API spec to make sure that existing websites are turned into API endpoints. And then the interns build this single portal Hike.Taiwan that can just plan your trip. And because it's API based, so an ecosystem of apps and so on. Additional services build upon those APIs. Later on, we'll repeat that for Ocean.Taiwan. And then last year they started working with municipal government. So the main lesson is that if the career public service can see the young people as, you know, once bringing gifts, almost like reverse mentors instead of just criticizing them, right? So they can see this partnership as really good in terms of time saver. They don't have to like debug their website anymore, right? Anyone who see something wrong also suggests something better. It decreased the risk and it also improves mutual trust. So much so, I think we don't have the allotted quota of interns anymore for this year's rate because many other municipalities and government ministries start to think about interns as teleworking interns. And so all of them actually want more quota because they see that this is actually a great way to improve the digital service. Yeah, what technology stack do you use in the government? Is it kind of all over the place, like depending on what website it is? We strongly prefer OpenAPI, like the Linux Foundation standard, the speedy open source manifest. And I think that the point is that as long as you speak OpenAPI, that all human interactable parts, whether it's mission writable or mission readable, has an API equivalent. We don't quite care, which technology is that you use because then it's trip like a Lego block that other services can apply into. And so the OpenAPI stack is actually piggybacking on the universal accessibility track, where if people contract a government website and a vendor says, oh, I can do this for people with side, but not with blindness, they could be disqualified for discriminating against people with blindness. So we piggyback on that and say if you deliver a website but you don't speak OpenAPI, well, you discriminate against robots that could be disqualified. So that's the thing we insist on. OK, that makes sense. Do you have like, maybe I could have figured this out already, but do you publish a lot of your government websites as open source and GitHub or GitHub? Yeah, sure, sure, sure. Our GitHub repo is PDIS. PDIS, OK, yeah, I think I've seen that before. That's awesome. And then so I guess you also focus a lot on like fighting disinformation in Taiwan is basically just like fast, like I guess like the strategy is like fast response to disinformation specifically around COVID. And like, I think, I think like using memes and humor to discredit information and disinformation. Yeah, like, has it only been for COVID or you've also been fighting disinformation? It's for general purpose. So the COFACTS project of Dev Zero is one of the first team to use wiki like dynamics so that when people flag something as fun or potentially disinformation on even into any encrypted channels like line, it gets trending on a dashboard. And then the COFACTS people can just wiki crowdsource the most viral ones, the ones that have an R value above one, so that the fact checkers can concentrate their energy on one because each day there's only this much of mental bandwidth of the population, only this many disinformation will trend any particular day, right, because it's competitive space. So if we focus on inoculating only the trending ones, then we don't waste our time working on the ones that will probably not be trending anyway. So the dashboard is also important. The economic sector, for example, Trend Micro, Taiwan's leading antivirus company also have a what they call a farm jada, a disinformation buster, right, that also works on this area where you can invite a bot into your online chat channels and it just scans each incoming message and just post our clarification if it detects scam or disinformation. Another startup called Husqa also offers a very similar offering called Meiyui and so on. So there's an ecosystem about it now. Cool, awesome. And have you extended back to the like some of the work we've done with like introducing more technology government, have you like seen any challenges or pushbacks or like any like kind of stumbling blocks and getting wider, I don't know, like more and more adoption of open technologies in government? Yeah, most of all, I think the fear and certain doubts seems to be centering around cybersecurity, because when you use open source code, you implicitly let random strangers upgrade your system, right? And that's not a very good position to begin. So literally the first thing I did when an institutional minister has to recompile the Linux kernel so we can run secure Linux and run a security product called Sandstorm so that even untrusted applications could be run within a container that deflects or mitigates pretty much all the XSS and other attacks and so on. And we got the Devcore team, which I think second place in Devcom CTF to do penetration testing until they say Sandstorm is secure, right? So that's what we do, just to have a sandboxed security perimeter and also run penetration testing. So the Polis system at polis.gov.tw now is also public digital infrastructure, even though it's open source, we also do penetration testing and things like that. Yeah, yeah, gotcha. And so it's the most of the technology used for kind of web-based running in Sandstorm? Yes, sir. Yeah, and OK, interesting. And we use that for all our transcripts and things like that. Yeah, yeah, like do you use like, is there like a Google Docs equivalent, I forget? Yeah, sure, we use CodeEMD, which is the open source variant of the time you start up Hack time fee. OK, gotcha. And then, I guess like another question I have, are you involved in the digital ID card project? Mm-hmm, yeah, sure. Yeah, and I guess like what are the goals of the Taiwanese digital ID cards? Is there like an eventual ambition to allow people to vote, for example, using their ID cards? Well, at least for the e-collecting, yes. And currently, we already have a PKI card, the Citizens Digital Certificate, or the CDC. The thing is that the CDC only maybe has one quarter of people who have ever used it once. The active users is even smaller, and mostly because the CDC really isn't used much in other than paying taxes and so on, right? So the CDC has a way to authenticate oneself is going on, there's nothing going to change about it. The main idea from the Ministry of Interior is that since we already have a paper-based ID card and a PKI-based card, the CDC, maybe we can ship these two cards in the same form factor so that more people will have access. Of course, that also raises a lot of cybersecurity issues because for the paper-based card, sometimes people, when they go to a parking lot, they just exchange this card for a coupon of the parking ticket. But if this is a CDC, if this is a digital certificate, then that opens a host of attack services and so on. So people understandably, and with the more experience about the health ID card, which is universal last year, people demanded that maybe we need Estonia to have a specific act about the norms of what to do, what not to do with a universally issued Citizens Digital Certificate. So we're now working on that act. And if it's passed by the legislature, then there could be real consequences for parking lots or hotels, insisting on having the photocopies of your ID card. Yeah, or explain sports centers, I guess. I know. It's like a rapid or something. That's right, that's right. I mean, if people ask for your photocopy of the backside of your credit card, you'll probably flip it out, right? That's true, yeah. I guess even the Justine ID has a lot of... It has some information, and I don't know what's on there, but it's probably information That's right, exactly, exactly. So we improved that design, but people said that at regulation level, like what to do, what not to do, doesn't really carry any penalties as such. And so people want to do a special act for it, so we're working on that act draft. Yeah, yeah, okay, cool. That's exciting. And then, I have another question here, but I heard you. Sure, sure. Where did you write it down? I guess, I had like... That's fine. That's fine. I think you're also working on promoting digital innovation in Taiwan, and just like trying to bring in more... Like what are you doing in that area? I think it was the goal, especially the gold card, to bring in more entrepreneurs and people to start... People who are here for the, I don't know, food in broadband. Hiking and surfing. Yeah, yeah, yeah. And frankly speaking, we are not saying that the gold card holders need to be forever staying in Taiwan, but we do want to share our experience in both broadband as a human right and also digital competence as a basic education curriculum plan to people in the world, so that when we fix local problems like the infodemic and the pandemic, the Taiwan model can also be shared with the gold card holders and through them to the international community, how to tackle the pandemic with no lockdown and how to tackle the infodemic with no take-down and offering our model as a preferred, at least more democratic model that other liberal or social democracies would like to partner on. So this is a more Taiwan can help approach in international health, I guess. Yeah, that makes sense. And like, what else are you doing aside from the gold card program to promote innovation? I guess like, I don't know a lot about this they have a general purpose inbox. Yeah, I haven't tried to start a company in Taiwan, but I guess like, my friend who started a company, I think in the hardware space, primarily they were saying that, for example, it's harder to incorporate a company in Taiwan, like it's harder to incorporate an LLC than like, in the United States like, you can do it in like half a day or something, but I guess it's like a multi-week process here. Yeah, we've shortened that, I think it used to be a like physical face-to-face process for most people. It could now be done online, but I'm not pretending that something like the Stripes and you know, click here to start a company. A company starting as a service exists in Taiwan. Currently, probably the whole process still takes a few weeks. I think up to four weeks if I'm not mistaken, but at least it's quite streamlined. And we have a like startup help center and actually the northern startup lab center is actually just across the aisle, it's actually that building. So you can go to the startup hub and they'll walk you through the process. Yeah, and a lot of technology in Taiwan is hardware and it feels like the software community is more nascent. There are like some Taiwanese software companies, but it feels like there aren't many big brands that are like kind of well-known outside Taiwan. Well, there are two business accounts, right? I mean, many enterprises are powered by, like you say, appear and things like that, but they don't say on their website, power by appear. Yeah, yeah, yeah, yeah. Yeah, it's like less like consumer brands. They don't say power by trend micro, right? Yeah, right, right, right, right, yeah, yeah. Okay, I was just curious about that. Sure, sure. That's most of what I wanted to ask about. Yeah, I'm just curious about, let's see, with the pull of stuff, because I know the pull of stuff was created by some guys in Seattle. That's right, that's right. They're now pivoted into a foundation, I think it's called math and democracy. Yeah, yeah, yeah. Yeah, I'm not sure if it's actually being used, it's been used in a few other places around the world, but I'm not sure. Yeah, I'm not sure, but have you liked it? A ton of it in Denmark. The Canadian government did a French translation and bilingual interface. Singapore used it for use engagement. There's quite a few users and our contribution in penetration testing this and so on also bolstered the government's support from other parts in the world because they now know that it's more cybersecurity hardened. Yeah, yeah. And then I guess like, I guess ultimately what I'm asking about this question is because I think it'd be interesting to introduce some of the same ideas in the United States and work towards some of the same causes. It's been used in Bowling Green Kentucky, at least. Yeah, oh yeah, yeah, I think I remember hearing about that. Yeah. Yeah, I guess like one of the problems is or like Taiwan is a little more I think dense and homogenous and it's just like smaller in general. The United States is like very big and I think one advantage I guess of Taiwan is that like if, I don't know, like if you're having an argument with someone in Kaohsiung like you, you could, you chance are you've been to Kaohsiung because it's only like a few hours in the train. But you can say that in many of the smaller states like New Jersey. Yeah, I guess like in the United States things tend to get nationalized. Like that's the, for better or for worse like media is like very national now because a lot of smaller media companies can't continue to compete. So they have like big national brands like the New York Times or CNN that have more of a national focus. So a lot of people's news also has a national bent to it. So people tend to get very involved in issues that are actually kind of like not even directly affecting them. Yeah, but from the Bowling Green experience I think the rough consensus of people like putting art into STEM education like diversifying the broadband choices and things like that are decidedly local. So I think it's just a matter of airtime given. As I said, the national media of course focus on national issues. I would argue even the more anti-social corner of social media like on Facebook and so on. Also the viral memes tend to be national or even international in focus. But if you look into the more pro-social corners in Taiwan, of course we have the PDT which is the equivalent of Reddit by it's run by the social sector like literally National Health University student pet project. And because they don't have shareholders or advertisers naturally the local issues like the policy in Kentucky, Bowling Green surfaces up and people discover the more pro-social aspect of things. So I think there's a lot to be done if you use the keyword digital public infrastructure. You'll probably see quite a few communities working on to recreate the experience of town halls and public parks and public libraries and so on. But in the digital space. So we don't have to deliberate about our agenda putting arts into education or diversifying broadband on the digital equivalent of the Nile Life District where there's private bouncer selling addictive drinks which is Facebook. So that we can talk public infrastructure that are also digital. Yeah. I was just like a sign up. You mentioned PDT I spent like a while trying to sign up for an account and I had to go because I think. It's open only to NTU students. Oh, okay. That's why it's literally NTU project. Okay. I thought it was like anyone in Taiwan but then I try to sign up and they like need like it. Yeah. I think you need like NTU. Cool. And so I think that's most of the questions that I had. Is this, I think you're changing the format that the office hours going forward, right? That's right. So starting next month only registered social innovation organizations can book at a time. And mostly it's because the accountability also matters. When people walk in, I want to understand that our conversation is for the betterment of the public good. And if someone is registered as a social innovator then at least once a year they have to publish a public account of them doing that. So it's a threshold that's actually not that hard. So if you complete this 30 days startup creating process and also register on the social innovation platform. Yeah. All right, yeah. I think it covers all the questions that I had. Yeah, it was nice meeting you. Cool. You're still, like I said, we can wrap up here. Unless you have any questions you wanted to ask me. Well, yeah, are you in touch with the timetablecar.com community? A little bit, yeah. I haven't contacted any other meetings yet but I would mainly to go to some of them. But yeah. There's a digital community. Because they put something very much like the rescue action by youth. Because the original go-kart website from the government was so horrible they just created a new one on GitHub. And then they get invited by the national government council. So now go-kart.nat.gov.gov is actually co-created by those forks. So the work gets merged back. And I think we really need more people doing that. You asked about how to foster innovation. That people systemically forking the government I think is one of the best ways. Yeah, yeah. Yeah, that's really cool. Awesome, well, anyway, yeah, it's great to meet you. I'm a Shakespeare fan for, I guess, the pandemic, so. Yeah. Yeah. OK, really good questions. Yeah, thank you. Thank you. Have a great day. Yeah.