• …so yes, please go on.

  • Actually, if you share the transcript with me, that’ll be great. I’m taking notes, but I can make sure I’ve got all your quotes right.

  • Maybe we can back way up. I’d like to get some of your own history, and how you came to be interested in these technologies, how you got involved with the political side of things, with Occupy, etc. Just give me your background, and then we’ll go from there.

  • My background’s pretty transparent by now. [laughs] I dropped out of junior high when I discovered the Wide Web during a science fair. That was in 1995.

  • My principals, all the teachers blessed me and said, “You don’t have to go to school anymore since you’re co-creating on the arxiv.org pre-print website anyway.” They covered for me, which is why I believe the career bureaucrats are the most innovative people ever since.

  • (laughter)

  • Actually I would agree with that. [laughs]

  • That’s right, that’s right. I work with Internet governance since then, so for five years. That’s before I even get the right to vote.

  • To me, this multistakeholderism really is my native political system that I work with. I see representative democracy as a pale version of it. Then I started quite a few web startups also, and eventually joined this g0v, spelled G-0-V, movement around 2012.

  • The g0v system is, very simply put, forking the government, where all the government website and services that ends .gov.tw, you can change the O to a zero and get into the shadow government, which offer the same information, but in a much more accessible and interactive way.

  • It’s basically the social sector’s way to show the government a better path. I did a, what we call, crowd lexicography, people making their own dictionaries, in various different national languages in Taiwan.

  • Taiwan used to have only one official language, Mandarin. Thanks to the effort of the social sector, starting from last year, we’re now having 20 or so national languages, and each with their own maintained dictionaries.

  • That’s just one example. There’s many around campaign financing, transparency, and just very recently about mask distribution systems, about disinformation flagging and countering narratives, and so on.

  • I think it’s this more than 5,000 people strong, collectively working together to fork the government that really make Taiwan civil society much stronger than any of its particular NGOs or NPOs, because they were connected in a way through multistakeholderism.

  • That’s my background. I studied social interaction design by prototyping it with the early Socialtext folks, so the wiki thinking. We’re basically taking micro-blogging, wiki, speadsheets, whatever, that’s running on the consumer Internet and making into productivity tools.

  • I also worked with the Siri team for six years as a independent contractor on computational linguistics also. There’s a lot of it in communicating poetic meaning [laughs] through very limited communication modalities and so on. That’s a very short background.

  • Tell me when did you start as digital minister, and if you can maybe give me one or two examples of the way in which this sort of more representative ideas around democracy merged with the technology that actually enabled that to happen in a real way?

  • Sure. Actually the GovLab just published a CrowdLaw for Congress, which is a way for collective policymaking for Congress members supported by the Democracy Fund. The problem definition part in the Playbook is exactly the Taiwan example.

  • I would strongly encourage you – because it’s really short – to go through it, which lists the peer-reviewed studies of these things. I’ll also paste you a link that talks about another concrete example which is how we deal with disinformation while protecting speech.

  • These are the two flagship examples, the first one dealing with emerging economic models such as Uber, and making sure that a law corresponds to social expectations and setting social norms.

  • While Uber is more of a commercial entity, the other deals with a more political entity, the PRC propaganda unit to be very specific, and how do we actually strengthen democracy through social innovation, making sure that everybody can participate in journalism rather than over-relying on any state representative. That’s our two flagship examples.

  • I started in 2016, October. At that time, I’d already worked with the Cabinet for two years before then, since late 2014, as a young reverse mentor to a previous ministry.

  • Can you tell me a little bit more, or is there another link where I can read about the way in which you get people together to contribute online news and use that crowd model?

  • Yes, of course. Actually, the fact checker system that we set up during the presidential election is a really good example, but it’s a lot of different initiatives that, taken together, forms what we call this collaborative fact-checking ecosystem.

  • There is really no single link that I can give you, but there is three links that, taken together, takes care of most of the g0v’s or related ideas. One around fact-checking every single thing that every presidential candidate says.

  • Another one around the legislative members and the final one around all the rumors that’s spread in the end-to-end encrypted systems in Taiwan – it’s called Line, which is very similar to WhatsApp. We basically fight it the same way the Spamhaus fights spam, which is basically people voluntarily reporting things.

  • These two, taken together, I think, gives a pretty good picture.

  • Our audience is pretty non-technical. If I were going to try and explain to them what were the breakthrough moments when doing what you do – which is using technology to basically facilitate liberal democracy – what were the turning points that allowed that to happen, would you say?

  • It’s probably occupying the parliament for three weeks and demonstrating – in a demo sense, not in a protesting sense – that this is actually a viable way to listen to half a million people on the street and many more online.

  • Eventually, settling on the four demand – not my list – for the head of the parliament and with zero casualty, with zero people missing, and that actually gets ratified by the head of the parliament. The Sunflower Movement is definitely the watershed moment.

  • Before that, if you asked anyone on the street in Taiwan whether this kind of massive listening scale is possible, they would say it’s not possible. After the three weeks of occupy, everybody see it’s not only possible, it’s actually much more effective than traditional representative politics.

  • To what extent do you think that what you do is replicable in a country, say, as large as the US, in European countries? Where are the synergies, and where are the challenges, do you think?

  • First of all, I think it is very replicable. We’ve held workshops in the NOIC, and the same system that we use have been prototyped in the Bowling Green, Kentucky. Our bilingual system, where we deploy with the AIT, the de facto US embassy on four diplomatic conversations at scale.

  • These bilingual translations were contributed by the federal Canadian government, where everything also needs to be bilingual, so on and so forth. There’s many different success stories already on the Polis platform in particular.

  • What’s important, really, is to choose an emerging topic, where the government genuinely has no idea what to do. If they choose something that require a lot of expertise, that’s already well into the development and delivery stage, that would not work.

  • Then people would have to pay a lot more opportunity cost and time to understand what’s involved. At the very beginning, there’s no self-driving vehicles and we know nothing about them. At that time, I think it’s the best to use.

  • There’s a knowledge base maintained by the creator of Polis, Colin Megill, that talks about what to do and what not to do about deploying Polis, which I just pasted to you.

  • That’s interesting. Having worked in the US around these issues, what do you think that topic would be?

  • Any specific example would be good. For example, in Bowling Green, civic assembly. They just asked a very simple thing. This is a virtual town hall. “What do you think is the highest priority that the town should be spending its budget and energy on?”

  • The top suggestion, apparently, is to put art into STEM education, which is very easily overlooked, if you only look at institutional, and indeed, social media. Actually, they talk about a lot of things, like traffic, zoning, cable, accountability, groceries, liquor laws, homelessness, metro, so on, and so forth.

  • Just making it very specific, very everyday-related, instead of anything abstract, that enable what we call an overlapping consensus among the various different positions or ideologies. The report is here.

  • Let’s talk a little bit about what you’re doing in the context of what’s going on in China with the coronavirus. What do you think? If these tools were actually in effect in mainland China, what do you think would have happened? What would have been done differently?

  • In Taiwan, we are actually using more or less the same civic participation tools, also with the g0v community, to hold the mask distribution system accountable. You can actually find a lot of information based on, by now, hundreds of civic tools around the mask distribution.

  • The idea is that, when people enable a sense of shared reality when it comes to technology, when we have the same facts – backed by open government data, but also, open citizen data – there really is no way for moral panic to happen, because people know what really is happening.

  • This, of course, relies on a couple of things. One is the freedom of speech and assembly. For you to listen, really, at scale, you have to listen to various different positions that initially sound extreme.

  • It’s only after acknowledging those divisiveness and extremism can we then say, “Oh, but people who are extreme is not actually random trolls. They actually agree on a lot of things that we also agree,” and so on, and so find the common point.

  • First, you really have to have that kind of important distinction between freedom of speech that threatens the harmony of the nation and the freedom of speech that actually enable people to see emerging issues much earlier than they would otherwise have done.

  • This includes whistleblowing, but it’s also just sharing everyday experiences that are not otherwise accessible. I think that’s the one enabling condition, is freedom of speech and assembly. The other one I think is equally important is that people need to have a certain faith in the government’s trust in the citizens.

  • This sounds like a lot. This is a circular concept, but this really is important, because if you don’t have this kind of meta trust – the government trusts citizens more than citizens trust the government – then it’s actually fascism.

  • The citizens who trust the government without the government trusting back is essentially falling into ideological trap. Only by demonstrating that the government trusts citizens more, then the citizens can rationally trust somewhat back into the government. This also applies to surveillance capitalism, by the way.

  • Say more about that.

  • Again, in Taiwan, what we’re saying is that, instead of, say, a large IoT company or a large multinational company collecting, say, environmental data, humanitarian data, and so on, and distribute it through charity or…Nowadays, they don’t say charity anymore.

  • Corporate sustainability relationship [laughs] to the citizens, we instead have people democratizing the devices needed to collect such air information. For example, AirBox, which is less than $100 USD, and we have more than 2,000 primary schools and so on participating in.

  • It’s only when even the primary schoolers understand that the idea of data stewardship is that you can actually see what the people demand of your data in terms of quality, in terms of accountability, and so on, can they get into this mindset of a data steward, rather than a data consumer.

  • A data consumer is like, to use Glenn’s favorite analogy, it’s like a labor worker before the invention of labor union, having zero bargaining power. If they band together, forming a data coalition – and, indeed, in Taiwan – negotiate with the environment minister.

  • Saying, “We have 2,000 AirBoxes. People trust us more than they trust your 70 stations. Why don’t we band together?” We allow the government to join our ledger, and we together put pressure to the industrial areas and allow us to install the lamps to measure the air quality there.

  • This way of social sector gaining legitimacy over public sector, and the public sector supporting but not controlling, the social sector’s ledger. These two together pressuring the private companies to conform to the norms already set by the social sector, otherwise facing social sanction.

  • I think it’s a very viable alternative than the other way around, which is the code, the architecture of the multinational companies, setting the norms.

  • One practical question. When I refer to the technologies that you’re using to facilitate the democracy, should I just basically say distributed ledger tech? What else should I throw in the mix?

  • I think distributed ledger is a cheaper accountability layer. We’re using ledgers whenever it makes sense, but we’re not always using ledgers. What I think would work, I usually just call it digital social innovation, which is the broadest possible term.

  • If you want to do it in a more RadicalxChange angle, I would suggest data collaboratives or data coalitions. Ledger is an enabling technology, but it’s neither sufficient nor necessary for data coalitions to form.

  • What is necessary for data coalitions to form, do you think? What’s the needed ingredient?

  • I think there’s three main ideas here.

  • First is that there needs to be a public value, so that people, despite their different positions, different incentives, can agree that this provides public value.

  • Shared public value, I think, by far, is the first thing. You can’t do this by designing financial incentives alone. It needs to correspond to a public value, so that’s the first thing.

  • The second thing is that there need to be an open innovation tradition. If one stakeholder enters and then patents everything else, it actually excludes, like this is too much of an early-mover advantage. You see that in many cryptocurrencies, by the way. [laughs]

  • Basically, the open innovation need to ensure that latecomers have their share of value to be created and captured, too, instead of privileging only the original inventor. I say this with all due respect to Vitalik. That’s the second idea, which is this intergenerational justice or solidarity, however you want to call it.

  • The third thing, I think, again, is the demonstrated impact. You can have the best mission, the best purpose, the best shared value, and the best mechanism design. If at the end this data collaborative mechanism doesn’t show that it actually moves substantially, quantitatively toward this originally shared purpose, then it is all for nothing.

  • This is what we usually call impact assessment, or external accountability, or whatever other social innovation terms you want to use. The idea is that one need to demonstrate, much as a government is accountable to their constituents, that this coalition is really creating value for all the coalition members.

  • Otherwise, they might as well fork off and run something else. I think the mission, the mechanism, and the measurement, these three are…I just randomly throw three labels, but they capture the core ingredients.

  • When you’re thinking about your own demonstrated impact, what would you cite? What would you put on that list?

  • For example, the Join platform, which is our national participation platform, we don’t use the term e-participation anymore, because it’s well above 10 million citizens using it out of 23 million citizens.

  • Basically, participation online is now a norm, like over half of the adults do that. The most active ones are actually 15 or 16 years old. They don’t have voting rights. They can’t even vote in referenda, but they are extremely active there. Second group is around 60, 65 years old. In any case, that’s the thing I would cite.

  • The other thing is to say that we bring tech to people, rather than asking people to tech. We bring these virtual town halls to everywhere in Taiwan, and joined by the municipalities. I would say that this enables a kind of town hall, but virtualized and amplified policymaking model, that is portable everywhere in Taiwan because we have broadband as human right.

  • I would also cite that we have 98 percent or more coverage in rural, indigenous, and remote islands, and that the top place in Taiwan – almost 4,000 meters high – is actually very fast, 10 Megabits per second, €15 per month at unlimited data. If you don’t have broadband, that’s my fault.

  • [laughs] That’s a nice campaign slogan. “If you don’t have broadband, it’s my fault.”

  • That’s right. It’s a human right violation…

  • Wow. That’s interesting. I know you mentioned Occupy as a turning point, but what was it? You talked about 10 million out of 23 participating. Was there a particular watershed moment, and to what do you credit that?

  • The Join platform is special in the sense that it’s actually various systems rolled into a single portal. You can do municipal participatory budgeting there, but you can also oversee the budget KPI and spending of the government there, make comments, and get public responses.

  • It’s also a regulatory preannouncement, much as regulations.gov. It’s also a petition platform. What’s four different websites in the US is rolled into one single website. We also recommend if you petition for this, you might want to review that regulation and things like that.

  • It really is something that accumulates various different CSOs – civil society organizations’ – interest on its various sections. One of its early success stories is where we had a tax filing system that’s working very stably.

  • I think, not long after I enter the cabinet in May of 2017, suddenly, everybody using non-Windows systems is unable to file their taxes online because they used Java applets, and Oracle deprecated Java applets.

  • A lot of designers complained on the petition platform. We invited, through this network of participation officers, everybody who complained into co-creation workshops, face-to-face or online. We co-created a tax filing experience for Mac and Linux systems on the next year that won, I think, 96 percent approval rating.

  • So much so that the Windows users also want to use it. Last year, it’s rolled out for everybody at 98 percent approval rate, which is unheard of in government digital services anywhere. I think the reason is that thousands of people feel they have at least one Post-It Note into it.

  • Instead of the government designing and delivering a design, it is co-designed by everybody who complained about the existing system. That’s one of the early wins, is the co-creation of the tax filing system. I can cite more examples, but you also read the general principle in the participation officer website, which I just pasted in.

  • OK, brilliant. Just to go back to your point about broadband, in the US right now, broadband has become a hugely political issue. 5G has become an incredibly political issue. How could using collaborative technologies fix that in the US, do you think?

  • 5G, are you talking about the component sourcing? I’m just trying to make sure that we’re talking about the same political issue.

  • There’s that, and then there’s also who’s going to get it, who’s going to get it fastest.

  • I think Taiwan also serves as an interesting example, because while we have auctioned the consumer-using spectrums, we also allow sandboxes. If you want to use millimeter wave to empower your local rural community’s drones’ delivery of medical supplies or telecommunication across the specialty doctors in the main island, your local nurses, and whatever.

  • These can all be applied on a sandbox basis, meaning that you get the spectrum that is aside from the existing consumer-facing spectrum. You don’t have to wait for the large four or five telecoms to go to your vicinity.

  • Rather, if you can demonstrate social purpose to your local people, you get your dedicated band to experiment with 5G. I think our main idea is, again, inclusiveness. We want the 5G testing grounds to be near the people who stand to benefit the most.

  • It’s a must, rather than a nice-to-have. We roll out sandbox applications exactly for this kind of use. That’s one thing.

  • The other thing is that we are also saying that, if you have a good application of 5G technology for fintech, or for self-driving vehicle, and so on, you can challenge any regulation in any ministry whatsoever.

  • Aside from money laundering and funding terrorism. We know what happens with those two, but everything else is fair game. This is basically a time-limited way to show the people that your idea works really well.

  • You usually get a year or so, but for Presidential Hackathon, you have to demonstrate in three months. If the president likes your idea, she gives out five awards. Those five awards ensure the personnel, the regulation, and the budget required to realize your idea within one year.

  • How do you think these technologies are going to impact Taiwan’s ongoing political evolution in relation to mainland China?

  • I’m at four percent battery. I’m just going to grab a charger, and I’ll be right back.

  • I’m back. The question was actually two questions. One is about the future of Taiwan’s democracy enabling these kind of experiment to continue and shape these experiments. The other thing is about the PRC relationship. Did I get that right?

  • Exactly. One of the things, I’ve argued this more in an economic context. I’ve looked at, say, the way that the US is trying to think about tech regulation, versus how the PRC is structuring its innovation economy, and it seems very top-down in the PRC.

  • I’m arguing that in the US, we should try and think bottom-up. I’m curious if you have thoughts on that, both economically, but politically, how that might help enable you to continue to foster liberal democracy in the shadow of the PRC.

  • I think this is a very pertinent question, and indeed, may be one of the most important questions of our times. Liberal democracies amplified by social media, creating this false sense of us versus them divisiveness actually disables much of the advantages of a liberal democratic system, because you don’t get useful signals anymore.

  • Without useful signals, democracy don’t function. We all know that. I think the Taiwanese experiments, including our referendum that is separate from the voting day, every alternating year – binding, but only for two years – and things like that, these new mechanisms are all designed to let the society try out with their firsthand experience.

  • “What if we designed a mechanism a certain way, different from the legislator’s vision?” Or those local sandboxes that can basically overwrite certain national level regulations and laws, and demonstrating that it’s sometimes a really bad idea that other municipalities shouldn’t follow.

  • It’s about devolution, but it is not devolution in a feudalism sense. It’s about devolution in a kind of swarm-like, open innovation sense, where every municipality learns from every other municipalities by honestly shaping data, again, in a data collaborative way.

  • I would argue, actually, people have much more in common to each other. It’s just that institutional – and now social – media, select the parts that are divisive.

  • If we get into the habit of if we see something wrong, we just send a pull request to a public-governed code base, as the Ethereum community and g0v community both do on a day-to-day basis, even if there are unfairness, injustice in the society, people’s intuitive response is going to be, “Oh, I can fix it.”

  • Or at least, “I can try out a new mechanism that may fix it.” Instead of, “Oh, that’s because the ruling party is my party that I don’t understand, and that these are all non-people, and my party will absolutely do better,” and things like that.

  • A, I wouldn’t say nonpartisan, but rather a transcultural, like being able to view the culture, including the political party which is your own upbringing, of course, through the lens of mechanisms designed by other parties and other cultures.

  • I think that is, on the long term, what our education system with the new curriculum last year is now doing. The teacher no longer hold standardized answers to standardized tests, but rather enables the students to shape their own transcultural identities, which is like everybody’s unique.

  • It’s like choosing their own stars out of existing constellations to form a new constellation, which is my favorite metaphor. That’s the answer to your first question.

  • The second question, I think to the PRC, we serve as a pretty good counterexample. Like an existential proof that it’s OK to have a mechanism design based, a very Georgist design, that enabled market, but without a lot of those externalities caused by the market.

  • Because we do have a Georgist constitution – it’s written in our constitution of the self-assessed license tax – it’s about enabling platform cooperativism, which are all very hip 100 years ago and still in our constitution. [laughs]

  • It serves as what we call a regulative thoughts, because that constitution never really took part in any part of the mainland within the ROC rule of the mainland up to the end of World War II. It did show that it is possible to do so.

  • First in Taiwan, but now, we’re also showing around our vicinity that this kind of mechanism is possible, and you don’t have to, for example, encroach on the freedom of speech and of the press to do counter-disinformation work.

  • We showed that in Chulalongkorn University in Thailand and working with the Thai civic tech community. We worked closely with the Korean and Japanese communities, and indeed, held a joint hackathon at Okinawa, which is the middle point between the three jurisdictions.

  • I think all this serves as counterexamples to PRC, where they say that the only efficient narrative to counter disinformation is to maintain the harmony through automated censorship and a surveillance state.

  • We’re saying, “Well, that’s very efficient, but you’re paying a lot of social negative externality for that.”

  • Could I ask you just a couple more questions? Is there a good example that you could cite within Taiwan of a social issue, a political issue, in which there seemed to be polarization, but actually, when you used various techniques – ledgers, quadratic voting, whatever it was – that you actually found that there was more common ground than not?

  • Yeah, there’s a lot of examples. We often cite the UberX example, because it’s the oldest one. It’s from 2015. Recently, there’s quite a few also. I would cite the digital dialogs, because it’s significant that even people of various different thoughts around the USA’s role in Taiwanese politics nevertheless agree on what we should do, vis-Ã -vis the AIT, which is the de facto embassy.

  • I would share with you the reports.

  • That would be great. That would be great.

  • The first one, “How To Make Taiwan More Unique in the World,” I think it’s very illustrative. You can literally see the one divisive point and the other four somewhat divisive points, but then a lot of consensus around that.

  • We run the same for security collaboration, economic, and people-to-people ties. I think that also enable everybody participating to see whether you agree with the statement, like, “Wherever PRC closes an international door to Taiwan, the US should open it somewhere else,” which is for a while the most divisive statement.

  • Literally, half of people agree, and half of people didn’t. Some people think this makes us a vassal state or something. Nevertheless, everybody agrees that the US should send people to the Presidential Hackathon. [laughs]

  • There’s, again, I think just the shape itself, of seeing one divisive statement, and a lot more consensus every time reflects the polity back to the people. Always, it’s this shape that convinces people that we have much more in common, actually, than we would otherwise think.

  • The UberX example, is that documented somewhere that I could look at, just for detail?

  • Yes. I think it’s meticulously documented by a BBC series from Carl Miller. I will first paste you the raw data, like when he interviewed me with a lot of background information. It’s a lot of material, but you can get most of the context, including the political context.

  • That’s not in BBC Click because of time reasons, from those three interviews. Then if you also view the BBC Click episode, I think he shapes the narrative very easily. I think that episode is called “Can Taiwan Reboot Democracy?”

  • Do you think distributed technology could basically supplant the need for regulation in the sense that, if you have truth emerging in a decentralized way, you don’t need to regulate big tech, or do you think you still do?

  • That depends on what you mean by regulation. If you mean regulation by something that is imposed by the state to the private sector, then I would say most of the regulations can be co-created through what we call norm shaping or norm designing.

  • If you do norm design well – like in Taiwan, we did a norm design for anti-spam for countering online exploit of children and so on, and whatever – and whenever we have a good, firm norm package, we don’t need a law. Indeed, we passed no laws around spam, but spam is not a problem anymore in Taiwan. That’s one interpretation.

  • The other interpretation is that whether we should abandon that, their state-enforceable rule to the private companies.

  • In truth, when we talk with the multinational companies about the norm package around countering disinformation around election time, we always say that there is a blooming ecosystem with civic journalists and institutional media, social media, and the fact-checkers, and so on.

  • If you work well together, we agree not to pass something like the NetzDG, because we understand that’s the right thing to do. If you refuse to work with the social sector, not only you will face social sanction. We may be forced by the social sector, because they have higher legitimacy than us in the public sector.

  • We may be forced to pass something like that. Your choice. That’s our usual way. I’m saying that, with the right multistakeholder design, the carrot of no regulation needed is always there.

  • At the moment, you need to back it up with a stick. If you don’t show up at the table for two years running, maybe we have passed something just to satisfy the social sector.

  • Interesting. All right, I think I basically got what I needed. How long are you in the US, just so I’ll know what time zone you’re on for any fact-checking?

  • I fly out, I think, Saturday early morning, something like that.

  • We’ll finish up by Friday. I’ll just email you any questions that I have before then. If you could send me a copy of your transcript, that’d be awesome.

  • Of course. I’ll do so tomorrow or so. I still have some time Friday. I’ll be back to New York around 7:00 PM. If you want to do a follow-up call, something like that, we can do that.

  • OK, wonderful. Thank you so much. It was so nice to make a first contact with you, and really appreciate your help.

  • OK, be well. Bye-bye.