• For the record, I guess, then, first of all, on behalf of the BBC, thank you so much for having us in Taiwan, in your offices, and to allow us to follow you around for a bit.

  • Let’s begin at the beginning. You dropped out of school at the age of 14. Could you talk about your early life and your first interactions with Internet governance and the values that you went by…

  • I first got an Internet account through the NTU network by way of a friend when I was 12. That was 1993. There was no graphic browsers then. As I remember, those days we had Archie and Gopher and probably it makes no sense to use those words now, but those were pre-world-wide-web technologies that were attempts to make a hyper-linked world.

  • My first foray, if we can call it, into Internet governance is just to set up USENET groups for what we call Perl-mongers, which are people who are trying to spread the Perl language. I also participated in the transition between Perl 4 to Perl 5, the forming of CPAN, and things like that.

  • All this is basically people building a library of reusable software components and trying to make sure, in a way, that people can trust each other, even complete strangers, to reuse parts of our code in your code and vice versa.

  • That is my first foray into Internet governance, just by way of participating in CPAN.

  • What struck you with how different Internet governance was from normal politics?

  • You cannot punch anyone over the Internet. Even though there were latitudes and longitude and we jokingly call them ICBM addresses, but there was no real ICBMs ever flown because of those addresses.

  • Well… Recently, there was one . At that time, there’s no ICBMs flying toward those IP addresses.

  • In any case, the difference was, as I said, the absence of coercive power. You really could not coerce somebody to not speak over the Internet like that.

  • The governance works in a purely voluntary fashion. It doesn’t have the army, it doesn’t have a navy to maintain some kind of sovereign of Internet governance. We just earn legitimately through other means. People earn it through radical transparency, poly-participation, and the emphasis on rough consensus.

  • Audrey, could you just look at Carl when you’re talking, not…

  • OK. Do I have to say that again?

  • No, you don’t have to start over.

  • What made you become interested in moving into conventional politics?

  • From the very beginning, when I was four or five years old, is when Taiwan started relaxing its one-party rule. The Democratic Progressive Party was just born at that time. I remembered all the conversations in my household, plus my parents are journalists, one studying politics, the other law.

  • They all very much cared about the democratization and, indeed, joined and founded many social-sector organizations. It’s in the family. Everybody really cares about the empowerment of social sector. Remember that was years, a decade or more before we even have our first presidential election.

  • It’s all social-sector governance or local-level governance, instead of just electing the president, which took place in 1996.

  • Talk to me about the Economic Power-Up Plan.

  • The Economic Power-Up Plan was a interesting plan that tried to motivate the economy, but it’s “too complicated to explain” at the time. The government at the time, in 2012, tried to film something that basically have a voice-over that says, “The Power Plan is great but too complicated for citizens to understand. So, just follow it along and make it happen.”

  • It’s the first advertisement on the Internet that’s flagged as spam by citizens, from the central administration, so a complete flop. What it represents is the kind of elitist mindset that thinks the public, the “commoners,” have no way of understanding something as comprehensive as a multi-year budget.

  • The civic hackers at the time, four of them, imagined that it’s not because we’re dumb, but it’s because it’s not made in accessible way. They went up and did a interactive visualization of the national budget just to show that people can actually collectively understand budget very well.

  • The sun keeps coming in and out. It appears on your face quite a lot.

  • Yes, please. I think if you can sit.

  • We need to redo anything?

  • I think those are good warm-ups.

  • That’s good. Now we’re good.

  • It must have been a huge contrast to you then. You’d come from a world which is all about participation, radical transparency, consensus. Here was the Economic Power-Up Plan, which was essentially just telling people to trust the government.

  • Did that inspire you to become more involved in g0v?

  • No, it inspired my friends. I wasn’t part of the g0v movement till 2013. That’s three months after the movement started. What they decide to do essentially is just to register a domain name “g0v.tw” and systemically look at all the bad-taste, authoritarian government services and websites and redo them in a participatory way, in a more interactive way.

  • I joined the movement in 2013 working not on a particular government project but on a dictionary project that combines all the different government offerings of dictionaries – because Taiwan had many different national languages – together into a shared dictionary that can share every national languages as well as, of course, German, French, and English. It’s called MoeDict .

  • These were almost like civic remixes. Would that be how you’d describe it?

  • At the time, civil disobedience as well. Because the dictionary that we downloaded said “all rights reserved.” Fortunately, Taiwan has Fair Use clause that says it should be used in proportion. The government publications can be used in a free way if your use is in proportion.

  • This is actually controversial because we download each and every entry. What’s proportional about it? What we’re saying is we’re adopting a device that was just invented back then. It’s called Creative Commons Zero, meaning that we wave all our copyright and any nearby right to our remixes. We were like “All times zero is zero.”

  • We’re making zero profit out of this. We say we’re just a format converter for the Minister of Education. After two years of legal debate, they eventually agreed.

  • Next was the Sunflower Revolution?

  • Basically, after that, between 2013 and the May of Sunflower, there’s also another demonstration, the so-called White Cross demonstration against the attempted covering of a murder case, or accidental killing case, depending, in the military.

  • Random people, out of nowhere, on PTT — which is like a local Reddit except it’s not-for-profit — just assembled in the flash mob. The flash mob became a quarter-million people and exerted a huge pressure for the military to try the military cases with civil code in non-warring periods.

  • It’s a successful movement, in many cases, a prototype of what’s to come in Sunflower. At the time, it coincides with the COSCUP, which is the largest open-source conference here. People went on the street and discovered they cannot do anything because the Internet is saturated.

  • There’s no reliable communication. There’s no nervous system for that flash mob. It is a flash mob, after all. After that, there’s a Google Doc, people just learning from the occupy movements worldwide of how we can make ICT more useful in occupied state like this. After half a year or so of discussion, that is then deployed to the Sunflower Movement.

  • You say by the time Sunflower Movement arrives or happens, you’ve already built a civic online architecture for what would happen.

  • It’s very fresh access, like 10 days before the Sunflower Occupy, in the anti-4th-nuclear-plant protest. It was a typhoon-like weather, so not many people came, but people really want to watch live stream. Live stream was just introduced at the time, so we have much more people watching from afar than people directly on the stage or besides the stage.

  • Our equipment was still hot when the Sunflower Occupy happened.

  • That’s when you rigged up live streamed, big screens outside of the Parliament?

  • Were you there when they burst in?

  • I was there the night before they burst in, when they were just about to burst in. I was there providing Internet connectivity. I actually left my phone there to keep providing Internet while I go home. Two other people from g0v movement came, and they captured the entire breaking-in sequence.

  • After that, many professionals did join. The CPR team came, the “Cable, Radio & Power” team that supported pretty much all the open-source conferences.

  • We all agreed to use the banner of g0v because it’s easier to identify. Many different movements, the Mozilla movement, all the different movements, kind of re-branded under g0v during the Sunflower Movement.

  • How would you best characterize what happened in the wake of Sunflower? Was it a political crisis over democracy itself? Was that what happened, the conversation became…Why was it that the Sunflower, that this could have happened in the first place?

  • The core was that the administration, indeed the legislation at the time, considered the Cross-Strait Services Trade Agreement a regulatory-level not a law level deal, so it could be passed just by the administration itself without a substantial debate or oversight from the legislative branch.

  • The regulations at the time, they have a pre-announcement period that is just one week or two. You can call the public servant… That is it. There’s no meaningful public deliberation to any of the regulations. That so-called “black box” prompted people to move to the street to demand a more democratic process.

  • I wouldn’t say that it’s against democracy; it’s against the institutional implementation of democracy that people feel is not up to times. Nowadays, it’s 60 days. The announcement is online. No exception, all regulations is up for public debate and things like that. That came directly from the demand of the Sunflower movement.

  • Then the government came to you, came to g0v, and brought you into government.

  • Jaclyn Tsai came to g0v. Jaclyn Tsai, I wouldn’t say that she’s particularly representing the government. She was from IBM Asia anyway. That was her second year, third year working in the government. She’s as much a private sector lawyer as a government minister.

  • She naturally sought support from the external network after Sunflower happened because she really want to help Taiwan in building the digital transformation. I remember that she came with two questions. First, how can we regulate teleworkers when they have no unions or there’s no association of people who employ teleworkers?

  • Second, how can we hear about people who start up their start-up in Cayman Islands? Why don’t they set up in Taiwan? Again, they don’t have any associations like that, so how can we think past traditional representativeness when making regulations?

  • So you became Digital Minister, eventually?

  • Things happened after that. There’s many people who participate in the vTaiwan project initiated by Jaclyn. I was mostly just the technical. Then I learned facilitation from the facilitators, then became also a facilitator, and finally an architect on what is permissible, what is not permissible to talk like this.

  • For example, this need to occur at the agenda-setting stage where the government has no idea. If the government has a lot of idea and just want a quick poll, vTaiwan is not a good place to have this kind of conversation.

  • Governance mechanisms start to emerge, but that’s all with a year or more of experimentation on many, many different cases. When I become Digital Minister, people said that I just get promoted from the understudy. That is kind of the case. Jaclyn Tsai then went to g0v and started working entirely on the social sector. We swapped positions.

  • Let’s talk about vTaiwan. This is so difficult, and I’ve consistently failed. I try and explain vTaiwan to people all the time and I just can’t. Could you try and describe it in as a compact way as possible?

  • There was a paper on SocArxiv , that’s our attempt at just having a condensed version.

  • It’s described as a recursive public that meets regularly among the different stakeholders to talk about emergent issues that the government doesn’t quite know what to do, and have the social sector collaboratively design a participation agenda for each of those cases.

  • What’s the process? This is managed by g0v?

  • The process is just open-space technology. People meet up every Wednesday night, share dinner, share stories, have fun, and occasionally discover some interesting issues to talk about and then talk about it. Anybody who shows up is the right people. Anything that gets discussed is the right things to discuss, and it continuously evolves every day, so pure open-space technology.

  • Does it still go through the ORID basis?

  • They do. On the other hand, other spaces may have different technological components. It may change its process. That’s based on whether the stakeholders are mainly all residing in the same space or they are very distributed.

  • Whether those things are more on the reflective stage, people are still getting the feelings about it or whether people already have very solid consensus in terms of feelings and just want ideas, to start brainstorming. There’s no fixed process. It’s collaboratively determined by open-space technology.

  • People just have to read through the Hackpad and now HackMD and the transcripts of every week’s meet-ups in order to participate in the context of policy-making. Anyone who shows up is the right people. The government sponsors the venue or maybe sometimes some food, but otherwise, there’s entirely in the social sector.

  • Did it begin with almost this kind of activist idea of presenting an alternative to government policy when it wasn’t listening? Its lifetime, it’s gone from government not listening to government listening. Is that true?

  • No, vTaiwan came up with a minister’s questions about teleworking and about start-ups registering in Cayman Islands. Each time, the minister with a portfolio makes sure that the agencies that are related need to show up when a community decide it’s time for them to show up.

  • They participate also in the meet-ups to exchange, before going on live stream, their takes on these matters. From the very beginning, this is a partnership. This is not about protesting.

  • The idea is not to necessarily engage the whole of Taiwan on a particular issue. Is the idea to basically allow people if they really care about…

  • It’s a rolling wave of stakeholders, people who have something to gain or lose if a regulation passes in some particular way. This is not about referendum, which we want everybody to participate.

  • This is more about agenda-setting, agenda-finding, fact-discovery, feeling-discovery, which the plurality, the diversity, the solution landscape mapping is much more important than counting the headcounts.

  • Could you talk about the significance of pol.is? It seems to me the importance of almost like platform engineering informing the nature of the discussion space, which could either be consensus-seeking or polarizing.

  • After operating mostly on Discourse, which is a forum software, for half a year or so and getting responses in the hundreds, people can still very easily facilitate when it’s just a hundred people participating.

  • When we handled the Uber case, we found the interest is so high that there was no human way possible to read through all the comments and respond to them had we still used Discourse.

  • At the time, Chia Kao, another contributor and one of the founders of g0v movement made acquaintance with Colin of pol.is, a Seattle start-up. The problem was that the pol.is wasn’t very useful on mobile. We just launch it as part of vTaiwan Uber while they changed the code in real time based on feedback so it works better on mobile.

  • Most drivers probably won’t bring a laptop while they’re driving. Of course, they still have to park the car to use the mobile phone to prevent accident.

  • In any case, the idea is that we want to make this friction-less as possible and have people crowd moderate and see the consensus that, despite their apparent differences of preferring Uber or preferring taxi, they actually all care about registration, about taxation, about insurance. All those things are rough consensus no matter what your take or what your feeling about Uber is.

  • How did pol.is help allow the consensus to bubble up? It must have began with real polarization.

  • Basically, it’s a three-week or four-week process. We have found that as long as two months is actually preferred, as long as possible. What we are seeing is that after people establish their factions so to speak, there’s bound to be people who are competing for in-group resonance.

  • They want to convince everybody in that group that they have some rationality to get abandoned like this. Invariably, if you abstract this out enough, it’s also going to appeal to the other group as well. Pol.is automatically clusters people to let people see their in-group commonalities, their differences from the other groups.

  • Most importantly, there’s a “majority opinion” tab that shows across the different groups, no matter how small, what their things that they can live with are. Usually, it starts with something that has maybe just 60 percent appeal over the population.

  • Then people discover, “Oh, you can compete on that as well.” Then people start to bring more eclectic, more nuanced statements, try to win other side’s support. You literally see people converging. The lack of reply button is key because if you have a reply button, then people over-focus on the earlier, more polarizing topics. Because there’s no reply button, there’s nowhere for troll to grab.

  • It sounds like it almost gamifies consensus. People compete to draft the most consensual statements.

  • That’s right, and it’s a rough consensus. We ask people whether you can live with that. It’s not like you can sign your name on it. Basically, people re-realize that we are a polity after all of people caring about very similar things.

  • Was Uber the first real test for vTaiwan, the first really serious piece of…

  • I wouldn’t say so. The company law did get changed because of vTaiwan, its very first case. Teleworking, again, is a regulation. It’s not a law, but it’s also passed in due time because of vTaiwan. vTaiwan had a lot of early successes.

  • We say that Uber is the first case that has such a volume that it’s simply not scalable using traditional forum-like technologies, which we learned from Cornell. Using RegulationRoom methodology, the scale of Uber participants would not work. We were forced to innovate, to find something that truly scales.

  • How many people were involved in the Uber vTaiwan…

  • People who have voted sufficiently to be mapped on the visualization, that is to say, maybe seven or nine votes are more, numbers in the few thousands. People who merely have read, interacted, or voted casually is much more than that. People also watch the live stream knowing that we would talk explicitly only the agenda that is set by the consensus.

  • It began with, “I love Uber. I hate Uber.” You had Uber drivers on one side and conventional taxis, vans, on the other, I presume. What were the consensus items which emerged?

  • For example, people generally agree that safety is very important. It is important to have a legal framework where people can buy insurance. That’s a very high consensus. The other consensus is that regulation should change with time. Any social innovation need to drive regulatory change. It’s not like the law, as written, rules them as criminals, so we shouldn’t negotiate. It’s not like that. It’s also consensus.

  • People also care a lot about the taxation issue. It should be taxed. If it’s registered as taxi, you should enjoy the same tax benefits. If it is registered as a rental car for one hour or more, it should be taxed exactly the same as a rental car.

  • Finally, people care about professional license. This is actually a surprise even to Uber themselves. The earliest recruitment they did was Uber Black, and they’re all people with professional drivers license. For a time, they also allow people with no license that’s professional, just amateur drivers license, to operate as drivers.

  • Even the Uber drivers themselves are against that. They really want their fellow drivers to get professional licenses. That’s why the Uber representative at the live stream say that they will work with their drivers to all get professional drivers licenses. That is a very important commitment. Once it’s on the transcript, you cannot really backtrack from it.

  • How many laws or regulations have emerged from vTaiwan?

  • Maybe two dozen or so.

  • It sounds like some of them have been quite significant.

  • Not only the company law and Uber but also…

  • The company law two times, actually, a smaller change back in 2015 and another change around social entrepreneurship around 2017. Then, of course, teleworking, as I mentioned, remote education, telemedicine, and basically digital transformation that concerns all the different ministries.

  • Also, Taiwan is one of the first jurisdictions in Asia that has a good VAT/OECD compatible taxing regime for multinationals. That’s also one of the vTaiwan cases.

  • What’s the moral weight? Say vTaiwan produces a series of consensus items. How is it then coded into law? What’s the process? Is it your job to take that to Parliament?

  • vTaiwan is more regulation than law. Only 10 percent or so of vTaiwan ends up in the Parliamentary floor. Mostly, it’s about regulations. That is the say, the law didn’t anticipate this. It doesn’t mean that the law doesn’t permit this, but how to make it work is the level of regulations?

  • Teleworking is a great example. It doesn’t require changing of the Labor Act. It just need a regulation-level policy as guidelines for people to consider. All these levels is easy because the Parliament doesn’t need to really debate substantially unless they feel like.

  • They can always turn a regulatory pre-announcement into a parliamentary debate. They don’t usually do that if we can show that there is consensus already of quotas. Where it really helps is that it makes the public servants relax most of the fear, uncertainty, and doubt.

  • When they think of Internet participation, they think of pure noise and no signal. What vTaiwan showed convincingly is that with a good social design, we can always get the cream from the crop. We can always get the signal, not the noise.

  • It will actually be much more help to the public service because we extend the invitation to anyone who raise any good points into our face to face deliberations. They also learn of emerging stakeholders as well. It’s as much about changing the mindset of public service as it is about delivering regulations.

  • And as much about choosing the technologies which are going to host the online discussions?

  • It’s completely the opposite of Facebook and Twitter.

  • We use Facebook also to attract people. You can leave comments here, but you will not have binding power. If you go to this platform, your comments are going to have binding power. If you leave comments here, you will not get an invitation to the executive hearing. If you leave comments here, you will. It’s good as a portal.

  • Are you trying to reinvent democracy?

  • We are reinventing democracy as a byproduct.

  • What we’re really getting at is a call for the social sector to own democracy. It doesn’t at all works like the state intermediate representatives. Whether you look at vTaiwan or other g0v projects, it basically lets people amplify their voice and amplify their own hearing as well, but they’re still themselves.

  • They are only speaking for themselves. They are not representatives that pertains to many people. They’re only people who re-present themselves, their arguments. Pol.is is one of the visualizations of people’s general argument, but everybody just speak for themselves. This is democracy more like the early Greek democracy than the later representative democracy.

  • The best thing in Taiwan is that the representatives, the MPs, now sees this process as a complement to their work. When they actually look at the laws or bills brought to the legislative floor, they already know who the stakeholders are, who to contact, what to make of this static document.

  • It’s very important that we operate in this neutral zone in the administration. In Taiwan, the administration propose draft bills, not individual MPs. The MPs can also do it through their parties, but then the administration propose the vast majority of draft bills.

  • It sounds like so much of this, including vTaiwan, it’s basically a story of you trying to find ways of using technology to cause power to flow out of buildings like this actually and onto strange new processes involving many more people. Is that a fair description?

  • There’s power and there’s power. What we are doing is — in the words of Manuel Castells — to bring network-making power into the social sector. Within the hierarchy, within the bureaucracy, of course, there is still vertical power. We don’t call vTaiwan or any projects bottom-up because there’s no bottom in the hierarchy. They are firmly in the social sector.

  • We make sure that I’m at a Lagrange point where both sides can feel the gravity of the other side and also see that there are things that the social sector can commit to that may actually work better compared to a law.

  • In many cases, like cyber-bullying, what ends up is a self-regulation of a social sector oversight from the social sector to the private sector, rather than having the government have direct control over the private sector. The government empowers the social sector to have oversight of the private sector. As one of the possible outcomes of vTaiwan, it doesn’t have to end up with regulations or laws.

  • It sounds like, implicitly, you think that classical, conventional, representative, elective democracy doesn’t work anymore.

  • I wouldn’t say so. I say it doesn’t scale well.

  • It doesn’t scale? What do you mean?

  • Basically, it has a limited transaction speed, rather like Bitcoin. There’s only so much bills that can be processed through this main chain of legislators. There’s only so much regulation that can be processed in each ministry using the traditional methods. When there’s an emergent issue that nobody really has firsthand experience of, it doesn’t scale well. That’s the only observation that I’m saying.

  • Much as Bitcoin must have a kind of slight chain underlining their work, we’re basically offering a way to synthesize the rough consensus in the agenda setting stage before the government even consider how to respond. We have this collaborative governance infrastructure that plugs in into the democracy.

  • But once it becomes a government affair, it becomes a referendum topic in Taiwan. Of course, the traditional methods takes over. But it always was kind of an informed consensus as an input to that institution, so that people don’t feel accidentally polarized.

  • What do you think is going to happen to democracy, I mean, if you had to be honest, of course? If democracies elsewhere, outside of Taiwan, don’t begin to go through the same kind of process, kind of rethinking how they can move faster, how they can use technology to engage more people. What do you think’s going to happen, like in a place like the UK, for instance?

  • I would say that Taiwan is just a larger municipality, geography-wise. From the northmost to southmost is one hour and a half by high speed rails, and we have gotten this “broadband as human right” policy so we can feel very close to each other using high bandwidth live streaming.

  • All this makes this polity feel very close to each other. That’s why our main partners in foreign countries like Reykjavik or Madrid or New York City is the one, they are all municipals of comparable size and sometime population as Taiwan.

  • I think that is currently the limit of this methodology, beyond which we don’t yet have a ready-on-the-shelf solution yet.

  • For example, for the transition of justice with indigenous nations, the east side of Taiwan, this calls for a very different design. But even if we bring the town hall meetings to the tribal elders, the elders still have their own culture. Their idea of sustainability and democracy may actually lead us, instead of the other way around.

  • That takes almost foreign affairs ambassadorship rather than democracy per se. Right? It’s not a classical case of evolution, it’s more like international relations.

  • All this is important because we don’t presume to scale this methodology over the size or the population of Taiwan. If the countries want to adopt it on the federal level, it will have to take a very different shape. Which is why we almost always see people taking the vTaiwan or g0v technologies on a municipal or city level. I think that’s because of this.

  • What do you think it will look like nationally? Do you think it has to be rolled out, something has to be rolled out nationally?

  • Right, I think nationally, one of the key components is, as you said, pol.is. Because pol.is, I think, Canada started rolling out with automated translation between English and French, so that you can see almost through a cultural translation lens different viewpoints from different people.

  • That is also why we experiment with virtual reality in our deliberations, because then people can feel that they’re in the same space, even though they’re very far apart. But it’s more about psychological distance. Any way to shorten the psychological distance is a prerequisite before this kind of methodology can work.

  • Do you see yourself and your colleagues in Iceland and…

  • Madrid and France…

  • Madrid, yeah, France and so on, as a new kind of generation of politician that is trying to actually change what government is as what it does.?

  • Well, I call myself a poetician.

  • Yeah, I write poems most of the time. That’s my main work. I do have a counterpart, Birgitta Jónsdóttir , the Pirate Party of Iceland. In that sense, we’re very much alike. We’re about giving people a glimpse into alternate futures than anything that’s more come under control.

  • Once people inspired in this way, poets doesn’t have control. People just take whatever ideas they want, they run with it. The poet that doesn’t publish radically is just a bad poet, right, people don’t learn about our work.

  • This kind of poetical work in politics, I would say, it has its historical precedents. But it is the first time, because of the Internet, that we can scale this kind of just performance and listening throughout the globe and feel that we’re a part of a kind of kinship. Previously, without Internet, it’s simply not possible for this to happen.

  • Yeah. But even by example, what you’re doing does feel very different from any other minister in almost any other country.

  • Well, it’s very similar to ministers of religions, spiritual ministries. Just preaching to the choir. [laughs]

  • As a kind of moral example.

  • But do you think that, in a way, the democratic crises of legitimacy that we are seeing in lots of different places around the world at the moment, plummeting levels of trust in politicians, apathy.

  • Also between social groups, not just with politicians.

  • Between social groups, of course, yeah. Do you think that has to do with the inability of democratic systems to redesign themselves?

  • I think it’s about the apparatus that we’re using, like radio and television, pre-Internet technologies, it amplifies people’s speech to reach millions of people. But there’s no way to listen to millions of people. Crucially, there’s no way for millions of people to listen to one another. Because of that, it’s much easier to incite world wars, for example, amplified with that technology.

  • Now Internet, it used to be that people want Internet to be truly symmetrical. But now because of AI profiling and all the post-facts and things like that, it has a danger of wherever you see, you see your own projections. You see in the people who disagree with you all your worst fears and anxieties, that’s actually psychological projection.

  • In your so-called friends, that may be actually just sharing a hashtag, you have swift trust, but there, that’s just ended.

  • All this, I think, amplifies the instincts of tribalism in people rather than the more empathy or listening skill tendencies in people. But it’s actually the same technologies, it’s just how we deploy it, and if we deploy it in a way that prompts resonance just like in pol.is, then actually people’s better side eventually surfaces.

  • If there’s a quick poll, [snaps fingers] just like this, a polarizing poll, a binary poll — I’m not talking about any particular referendums in any particular society…

  • No, no particular referendum.

  • Nothing in mind. Where do you see this going, both in Taiwan and globally? Let’s start in Taiwan. In five years’ time, what would be your vision for…It’s difficult, because I know we haven’t spoken about so many other different subjects, which PDIS is doing.

  • Would you see kind of vTaiwan style procedures being used much more often on many more issues? Would you see the kind of cultural government change? What would be the big changes which you think will happen or should happen over the next five years?

  • Now that the public service is well aware of this technology, and indeed, started its own vTaiwan-like network within the government, the Participation Office or the PO network, the culture is growing both within the public service and now, on some municipality like Tainan as well, as it is on the social sector.

  • Because its protocol is the same, that is to say, collaborative governance and multi-stakeholderism, it’s very easy for people to just build ad hoc collisions, because people have very similar ideas and design thinking and things like that. The new curriculum also helps.

  • In five years’ time, I think the participation that we see that is a surge of participations to this methodologies, especially after we wed this process into the e-petition, we see a flow of people are around 15 or 16 years old, participating very actively.

  • You want to join the e-petitions with the vTaiwan-style process?

  • Yeah, exactly, the PO network process. We have a lot of main stakeholders that are just 15 years old. The greater thing about them is that there’s no other democratic things.

  • When I was introduced into collaborative governance when I was 15, I don’t even have voting rights and no referendum rights as well. For me to meaningfully participate in democracy, this is like the only option, it’s like one of the several options, this is the only option.

  • We have people who petitioned to ban, I don’t know, plastic straws, to ban also the plastic waste bags and things like that. She initially used a pseudonym, so when we see like 5,000 e-petitioners joining them and therefore kicking off the voting process to start a vTaiwan-style collaboration meeting, everybody in the EPA thought they must be a very senior activist leader, to get such following. Right?

  • But it turns out she was 15 years old and this is her civics class, her teacher just said, “Hey, there’s a new e-petition platform. Refer something interesting.” Then, she was able to gather thousands of people supporting her.

  • Then of course, she, and it’s actually very erudite. Instead of going to strike on Fridays, we just meet on Fridays in a collaboration meeting and then figure out the pathway of banning plastic straws. Which actually takes effect in July for take-out drinks. She got what she wanted.

  • She got what she wanted.

  • That’s right. Taiwan became one of the premier places to have a very firm ban on plastics in the coming months.

  • The idea will be to use e-petitions to see what people care about in issues.

  • Right, to set agenda.

  • Set agenda from the very beginning.

  • From the very beginning.

  • The government even suggests the…

  • Right, so five years after that, when the 15 years old become of legal age to vote, I think they will demand a collaborative governance, no matter which level of participation they are.

  • Then it’s joined into a vTaiwan process to generate consensus items?

  • Would you then have government…

  • Government, by regulation, need to respond within two months in the point by point, of which consensus item that we accept, which have to wait a little bit after people get a new norm around plastic straws, and which is not possible, like changing Taiwan’s time zone to +9.

  • They do have some moral weight to the defense item.

  • They do, very much so.

  • The kind of government is kind of feeling that we have to have a very good reason to say we’re not going to do this.

  • That’s what agenda setting means. It means that government cannot escape from having to provide and account for that.

  • But in collaboration meetings, as I mentioned, the private and the social sector can also commit things. The government can also use their radically transparent transcript and lash them to hold the private sector and the social sector accountable.

  • There’s this neutral accountability. This is not the old style accountability where everything is absorbed by the state, but rather, the stakeholders get the places where they feel they are at an advantage of implementing and everybody commit to that in public.

  • Could you ever see the final stage also just become automated as well? It goes from agenda setting to norm formation, straight away through to coding into law?

  • Yeah, I think it could be automated. Like this year we have a Presidential Hackathon team that basically took law, like drunk driving, which is one of the cases that’s deliberated by the PO network, and is very mechanical. The end result of that law is basically higher penalties, and also, alcohol locks and things like that.

  • If you want to pass sentences, there’s really nothing in it for value judgment to enter, usually, it’s very mechanical. They basically write an AI that analyze the prosecutor’s document and to suggest a sentencing, thereby relieving the chore of the judges.

  • Of course, there still need to be human review to be GDPR compliant, right? It’s not automated decisions, but it’s very mechanical and so all the chores are automated by the machine before the human considers the facts and writes their own opinion on it, which is not unlike how mission learning clusters people and let people discover that their true resonances.

  • That vision, that would mean that, genuinely, some sovereignty would sit within that process as well as in a parliament and as well as in an executive. It’s almost like that’s another branch of government.

  • I wouldn’t say so. I would say that it is a way to assist any decision making process that is compatible with this protocol. You can use it, of course, in the legislation branch, in the judicial branch, actually the corrective branch use that as well. We have five branches here.

  • Yes. It’s a Sun Yat-sen invention.

  • But in any case, all the branches can use this methodology of better legitimacy through collaboration. But it doesn’t mean that the sovereignty is entirely spread out. It means that people can participate in a polity, in the political process in a much more shorter interval. It means that it’s more real time, it’s more in the moment rather than fixed, like four year periods.

  • I think it’s mostly about the time scale that’s being changed, it’s not about any particular configuration, like we’re not just making a constitutional change because of this process.

  • What about constitutional change? Do you want to change the constitution in Taiwan?

  • It’s interesting, right, Taiwan’s constitution, which is like a kernel of the legal system, was originally designed for a place that is much larger than Taiwan, the entire Chinese continent. By necessity, the constitution has been amended to kind of retrofit it into, it’s like retrofitting Unix to run on a microcomputer. It has its own inefficiencies.

  • But then the kernel is very difficult to change, because it requires three-quarters of legislation, referendums, and so on, and it’s kind of the legacy kernel that we have to maintain. A lot of work that we’re doing is kind of just working the periphery to add to what the kernel have already implied, but kind of reinterpreting it.

  • At the moment, I think the kernel still works to a degree, being Unix compatible has its benefits. But there are many people who really want a re-conversation around the constitution.

  • There’s one part of the constitution, if I can change one part, I would change the legal voting age, which is written in the constitution, not any particular law, from 20 years old into, at most, 18 or even younger. But because that’s part of the constitution, it’s very hard to change. Maybe one kind of pinpoint change to a constitution may actually win referendum support.

  • OK, can we just get an answer again, but this time, our audience won’t necessarily understand what Unix is, so…

  • Oh, yeah, we’re geeking out, isn’t it?

  • We’re geeking out, but I was actually just thinking that that was a reference that even “Click” listeners probably won’t know.

  • Yes, that’s what I mean. I get it, but yeah.

  • OK, that is too esoteric, yes, OK. Yes. Let’s try again.

  • All right. Are you trying to change the constitution?

  • Tell us, to the third eye. Yes. Right here. The pineal gland, right?

  • Physically, what we’re looking at is a constitution that’s original designed for a much larger territory than Taiwan. It was original designed for the Chinese continent. Of course, after a series of amendments, it now kind of works in the scale of Taiwan.

  • But there are places are hard-coded, meaning that it’s written into the constitution, like the voting age is 20 years old. Because it’s not a law, it’s very difficult to change, it takes three-quarters of Parliament, it takes a referendum.

  • I think it’s much easy that we focus on special cases like this, like dialing down the voting age that people really do have a broad consensus on, than any very large scale change.

  • Because then you serve an existential proof that constitution can be changed and that people can have a much more fruitful deliberations about which particular aspects of the constitution especially support for the newer generations of human rights and maybe even natural personhood, that’s my favorite, but in any case, can be written into it.

  • But none of this have very wide consensus, or indeed, is being actively deliberated. The 18-year-old voting issue, actually, we do have proof that people really want it, other than, actually, 20 years old. Somehow, they feel that they just earned the voting right…

  • But in any case, in other age groups, people do support for 18 years old to vote, so maybe we start with that.

  • Would you also put vTaiwan itself into the constitution, to have it recognized and protected?

  • I wouldn’t think so, because vTaiwan, the label, it belongs thoroughly to the social sector. I think if we say something that’s vTaiwan-inspired, then usually we’re saying that is collaborative governance or pro-gov for short.

  • It’s good that we write multi-stakeholderism or collective governance first into law. Right? Currently, we’re at a level of regulation. There is a draft law, the Digital Communication Act, in the Parliament right now, that writes multi-stakeholderism into it. It would enable the state to have constant support for vTaiwan-like processes without over fixating on vTaiwan.

  • Because there are many cases like this, right, there is the international Internet Governance Forum, the IGF, and there’s a Taiwan chapter of it as well.

  • There’s many other co-gov forms like that. We don’t want to privilege vTaiwan. But the state support in terms of the law will ensure that, no matter who the minister is, it will survive the minister, because regulation is easy to change. Right? Maybe the law level, and form the norm a little bit before we consider constitution level.

  • What would you say to people who said, well, this is fine, but actually, it hasn’t had…

  • It was fine, but rare?

  • Yes, well, there’s two points, yes, one is fine but rare, yes. That’s one question.

  • You’re now going to ask the critical question that I hadn’t thought of.

  • Yes, it’s fine but rare, in other words, it’s restricted to a small sector, and therefore, not that relevant to most people. OK, that’s one thing.

  • The second thing is…Yeah, do you want to get that one first and then I’ll come in with the second?

  • Yeah, so how many people has vTaiwan actually affected or influenced? How meaningful is it on the streets of Taipei?

  • Right, so I think vTaiwan, even at its highest participation, like Uber and Airbnb, is around tens of thousands of people in each case, right?

  • Our e-petition platform is about 5 million users out of 23 million in Taiwan, which is a little bit more, like a quarter of the population, but still, there are three-quarters who have never heard of the e-participation platform.

  • That all pales in comparison to referendums; everybody have heard of that.

  • OK. What does that say about vTaiwan, then? I mean, if it’s been going for several years now and there are a lot of people who have not heard of it?

  • Right. Why we keep saying something is vTaiwan-inspired, is that vTaiwan just showed that it is possible to listen at scale. But then the tool kits, everything is open source. We find particular parties adopting this same process, like the Social Democratic Party. We find people in the administration, like the PO network, adopting this.

  • The Legislative, the Judicial, and the Control Yuan all took elements of vTaiwan and started implementing it. In a case, this is basically a lineage that began this conversation.

  • vTaiwan at the moment is still very much about the digital economy and regulating the unexpected, because that’s the best shape to get people’s interest when they’re stakeholders but it’s not a popular thing to talk about. vTaiwan wanted to talk about Uber and Airbnb long before they became a global issue.

  • For more cases, like referendum cases, which concerns everybody, it’s not necessary that vTaiwan is the best methodology to talk about it.

  • While elements of it may inform the deliberation leading to a referendum, I wouldn’t say that vTaiwan is a process that people must submit their referendum to. Let us just say the CEC, the Central Election Committee, knows about vTaiwan. It knows of our collaboration meetings.

  • At the end of it, they must design a process that’s more befitting to the referendum, which is every two years now, to have sufficient time for a deliberation that concerns everybody and not just people who are rolled into the stakeholders’ survey.

  • The referenda are every two years?

  • That referendum is a national election?

  • It is, for national referendums, but it’s not tied to the day of election anymore. We found that it leads to a polarizing referendum rather than a deliberative referendum. At the voting booth, people are polarized.

  • Now we’re saying voting for national level is in four years, and mayoral election is every four years, and they’re at two-year intervals. We insert referendum days every August in the mid-years. There’s a year of election and then there’s a year of referendum, year of voting and year of referendum.

  • Is the idea to use a vTaiwan-inspired process or co-gov thinking before the referendum?

  • Yes, but the hard thing is that if we tie it with the election day, then no matter how much deliberation has been done, people on the election day become polarized. It kind of undoes the effect, which is why we now changed it to alternating years, so that deliberations like this will take more hold into people’s minds because it’s not interfered by party politics.

  • Has this happened yet, or is this the next round?

  • The law is now passed, it’s just passed. We haven’t had a new-style referendum yet.

  • I wasn’t aware that this had happened at all. What’s the idea then? You use co-gov…Actually, you just tell me.

  • The idea, very simply, is that we use e-petition as proving ground for referendum. This already happened many, many times. All the major referendum questions, you can find its precedent in the e-petition platform. They want to gauge how much public support they can get.

  • Now the referendum’s countersignature will be very much like e-petition in that you can also do online signatures. Because it’s ready – we’re just waiting for the final cybersecurity audit – you can transfer seamlessly from the people who join e-petition to people who joined the e-signature for a referendum very easily.

  • We use PO network and the Join platform, the e-petition platform, to serve as proving ground for people who are on ostensibly binary different sides of a referendum.

  • Before the referendum probably has ever been set, we meet face to face in a vTaiwan-inspired, co-gov way to make sure that people can roughly agree on what is the actual position that needs to be decided by referendum, rather than have people decide based on the title alone, without understanding the repercussions for it.

  • It’s e-petition, agenda-setting…

  • It could be e-petition. It’s usually e-petition, but the CEC, the Central Election Committee, can also bring their own ongoing referendum topics for the co-gov collaboration meeting. Each ministry that’s going to be affected, they are stakeholders, too, by the referendums, can also bring up those things for collaboration meetings.

  • These are regular referenda, which are often held in Taiwan. At some point, before them, there are a series of questions which are set by the CEC. Is that correct? This is very different. We don’t have referenda at all.

  • We didn’t have a meaningful referendum because the old Referendum Act basically ensures that over 50 percent of people must vote for a referendum to be binding. That means that people who don’t agree, that just don’t show up so it fizzles. All the previous referendum fizzled.

  • Then we changed our Referendum Act so that now it’s much easier for referenda to take hold. There was two mistakes. One is that there’s no campaigning. It’s a silence period during the election day for voting for mayors, but referendum people can still campaign. There are mayoral candidates that are also leaders of referenda, and so…

  • So they can campaign?

  • They can campaign, but their opponent cannot. It’s somewhat unfortunate. Also, we find that no matter how much deliberation has been done before the referendum, when people walk into the voting booth that doubles as the referendum booth, their mindset has become polarized already just by party politics.

  • We already changed the law recently so that the referendum and election take on alternating years. We don’t have a new-style referendum taking place yet. We’re doing training wheels and all the different preparations for the public service to be much more collaborative before the referenda, to have substantial deliberations. All this is made possible because it’s not on election day.

  • The deliberative aspect of this, is it government internal process or not?

  • It’s a transparent government internal process. Everybody see the whole transcript.

  • It’s the participatory officer?

  • Our PO meet every month. Every single issue that’s raised by the POs or e-petitions, you can find our entire debate, just like Parliament, online.

  • That sets the questioning and the topic which will go into the referendum?

  • Will go into the collaborative meeting. After the collaborative meeting, people may actually change their mind to reshape their questions and then decide what they want to propose as a referendum. That’s the basic case.

  • I also mentioned that it doesn’t have to go through e-petition. The Central Election Committee, who is also in charge of referenda, once we train to electronic signature, they can see a trending referendum topic and just decide to bring it to our PO meeting. Each ministry that are also stakeholders may also initiate that, so it’s not just by e-petition.

  • Just back to basics with me. What about the whole thing about the fact that relatively few laws have gone through the vTaiwan process to become law? I’m playing devil’s advocate here. Does this, in a way, not mean that people are less likely to participate because it doesn’t have the legislative teeth?

  • It’s not like that.

  • OK, so what? That’s just a question that people may be thinking.

  • What we did was that as soon as I become the Digital Minister, there was already a extension of the regulatory pre-announcement period as negotiated by then Secretary General Dr. Chen Mei-ling. It used to be one week or two, and people can only call a single person. Nobody knows 5,000 people have already called that poor person before, so it doesn’t work. In short, it doesn’t work.

  • After Dr. Chen Mei-ling arranged so that it becomes 60 days, I joined as Digital Minister and say, “Hey, why don’t we take part of vTaiwan and make the regulatory pre-announcement period, now that it’s 60 days, public, so that people can offer their commentary in public?”

  • Just like in vTaiwan, we have ministers that respond in real-time. People ask for clarifications. Also just like vTaiwan, the government must respond point by point for all the regulatory opinions that are up-voted to a degree that is not purely trolling, in the end as a regulatory impact assessment report.

  • If there is strong controversy, just like in vTaiwan, they just re-announce the regulation again in a modified version, not unlike vTaiwan where there may be many drafts. In this sense, while there’s no part about pol.is, the non-pol.is parts are all part of the regulatory pre-announcement period now, and there is no exceptions.

  • Every single regulation goes through this public debate, point-by-point response, real-time response, clarification, and all this process in Taiwan.

  • Every single regulatory announcement?

  • Every single regulation now. There’s no exception. In a sense, it now applies to everybody. For the bills that we’re preparing for the Parliament, if it concerns trade, it concerns international foreign policy, foreign trade policy, then it has also to go through the same process.

  • People, like the Chambers of Commerce, can have sufficient time to translate that so that their lawyers can go and ask questions publicly and receive public answers before it hits the Parliament. There is no exceptions. In a sense, the inspired by vTaiwan is very amplified, it’s literally every regulation and a lot of bills.

  • Is there an objective, reflective, interpretive and decisional stage in that pre-regulatory announcement?

  • It is, it is. But I would concede that the reflective stage is dependent on the bandwidth of the ministry. If the ministry only has limited bandwidth, sometimes they…In vTaiwan, we say, you must respond within seven days. But sometimes on the joint platform, there’s simply no sufficient jar to respond within seven days. It’s not 100 percent on that.

  • But what’s great is that sometimes they ask for help. They will bring this to the PO meeting and say, we hold a live stream face to face conversation, radically transparent transcript, just to get everybody in the same place. People who left comments on the regulatory commentary board gets invited into those face to face meetings, exactly like vTaiwan.

  • I would say that it duplicates most of the process of the pre-pol.is vTaiwan, of the first year or so of vTaiwan.

  • Does that include like a kind of wiki-fied laying out of all the different facts which are disputed and objectives?

  • Yeah, tabled responses, and things like that.

  • You can get Audrey to explain that. That would be quite good to hear her talking about Wikipedia.

  • Audrey, just the other thing is that when you talk about PO, if you can talk about something else, basically because people may not understand what Participatory Officers are.

  • Sure. OK. The POs, or Participation Officers, are a little bit like media officers that talk with journalists or parliamentary officers that talk with the MPs.

  • But the POs talk to people who are emerging as kind of social leaders out of nowhere. It was a hashtag, about to go to the street. It’s a much more engaging role, because you have to face strangers all the time, whereas the journalistic and MPs kind of stay relatively stable. It’s a more challenging work.

  • In each and every ministry, there’s now a team of POs, by regulation, and in many third level agencies and departments as well. We meet every month to collectively determine which two cases of the month we are using voting to discuss, in a cross ministry way, open to the public with a full transcript and everything.

  • The idea of PO is a vTaiwan-like process where they pull their mechanism into the administration, because the 32 ministries are like 32 different values. If it’s the same value, they might as well become the same ministry. They can counterbalance the jargon and to provide different viewpoints on the same issue. It’s very much like meet ups, but made entirely out of career public service.

  • What was the other question?

  • I’m so, so, what’s the role of facts in all this?

  • Yeah, because it’s very important with many polarized arguments, that facts can get very lost, can’t they?

  • Yes. We found that the best way to share facts is that to continuously share it, even before it become a social topic. In the same platform, Join that you have at TW where you can find regulatory, public commentary, you can find e-petition.

  • There’s also a public accountability utilization of budgets straight from the inaugural g0v project, look exactly the same. I wonder why. That shows almost 2,000 governmental projects, all the KPIs, all the spendings, all the procurements, everything is actually listed there.

  • Anyone can just ask for like long-term healthcare, water and sanitation, disaster recovery etc., and ask their questions, again, publicly.

  • There’s no 60 days until those is entirely continuous, just like the original g0v design. It’s just that government now publishes to each social object that is a budget item, like every quarter or so, saying that this is what we have changed, thanks to your input.

  • Right now, in the presidential hack-a-thon, the national audit office is completing a piece of this puzzle by adding the implementation reviews that they are required to do by law into this dashboard, so that people can track not just the national projects, but actually, the contracts, individual-wise.

  • Is it just budgeting or is it on any particular issue, could I load whatever kind of facts that I think are actually relevant on something?

  • Right. A few things. Budgeting and project planning actually has a higher level strategy, so they must also say which higher level strategy of industrial innovation or additional transformation, social innovation, the president’s promise to the young people or whatever view they come from.

  • Then from that, like presidential promise to the young people, we actually have another like mind map that shows all the presidential promises and all the young people’s ideas about how to explain those promises. Because I’m also co-chair of the National Youth Advisory Council, so we have another website that shows the entire context of policy-making when it relates to the youth.

  • Yeah, there are contexts, but at the moment, Join is mostly about the national project level, so we can complete the puzzle from the strategy level as well as on the contracting level, and this is now being done as a part of this year’s Presidential Hackathon.

  • What about, Audrey, if you tell us about what the whole pol.is platform means in simple terms, in the sense that it can bring people with very different views, who think that they’re enemies, actually find common ground.

  • That’s right. You just put it verywell.

  • If you could just…Yeah, you need to say it.

  • It’s the Digital Minister.

  • Yeah, exactly. It does that by actually, am I right in saying it’s not just the one question? He asks, actually, several questions. It sort of works…

  • Infinite questions.

  • Infinite questions, exactly.

  • You can post statements for other people to chime in.

  • Yes. But by doing that, the algorithm quite cleverly…

  • Highlights those that do divisiveness, anti-consensus.

  • Yeah, something like that. Do you want to ask the question of that type, do you think?

  • Sure. Pol.is is a simple system that is an AI-powered conversation.

  • Just start again, one second. Start with pol.is.

  • Pol.is is an AI-powered conversation platform that very cleverly takes away the chores, but opens up the agenda setting for people.

  • Anyone can propose that I feel that something and something, for other people to resonate or not with. You’re constantly placing one statement for your fellow citizen. You can click agree or disagree or I don’t understand this. Then once you answer, you see another, and then another statement.

  • Once you answer a few, you will start seeing your avatar on the two dimensional map where you’re grouped and clustered with people who share similar ideas. You can find that over the place through your social media friends and families, it’s just you didn’t talk about this over dinner, huh?

  • They’re not nameless enemies on the other tribe, they’re your friends. You just happen to have different views. Then so you would be prompted to also share your more nuanced feelings, to try to win your friends and family across. Those then become social object for other people to vote, agree or disagree, on.

  • The clever bit, not just the visualization of the tribal kind of coming together, but rather, the idea of people do have a majority opinion, regardless of which side they are on, is reflected very early on. People kind of compete to bring up the most nuanced version that they can win everybody across.

  • Invariably, after three weeks or four, we always find a shape where most people agree on most of the statements most of the time, rather than just four or five divisive issues. It shows both the division and the consensus, but much more importantly, it always show that the consensus outweigh the division.

  • It’s a consensus engine, in a sense.

  • It is a consensus engine.

  • If we take the Uber case, for instance, could you just give us in really simple terms what the divisive arguments were and then what they actually found through the common ground in the same way of what pol.is was saying?

  • Actually, the case is not called the Uber case on pol.is. It’s called People Without Professional Driver’s License Driving for Profit. It’s very specific. The theory of overlapping consensus tells us that the more specific it is, the more people are able to transcend ideological boundaries. If we start saying, “Sharing economy – good or not?” it’s lost.

  • Basically, it’s a very specific case. The initial division is essentially people who would prefer convenience over the legality of the service, versus people who would prefer a insurance, a professional license, and the safety of it over convenience. It’s different tribes. Roughly speaking, it’s Uber drivers, Uber passengers, other drivers, other passengers.

  • You can literally see them in the four corners. Over time, people do find consensus. People start agreeing that having insurance is, after all, very important…That in order to have insurance, some kind of professional certification is therefore very important, and that taxation need to be…

  • Sorry. The sun has just come out. So maddening.

  • That’s such a great answer.

  • That’s such a good answer.

  • The question now is, with a concrete example like Uber, we were talking about Uber, how very differently is it?

  • Oh, sorry. You were answering that absolutely brilliantly before the sun came out. Basically…

  • I just do the same?

  • Yeah, just say the same thing.

  • We don’t call it a Uber case. The case is very specifically, what about driving for profit without a professional driver’s license in your own car? The idea is that the more specific the case is, the more likely people can come to consensus. If we ask, what’s your take on sharing economy, it’s anybody’s guess, because it’s too vague as to make any sense.

  • What we were proposing essentially is that we ask all the drivers, it could be taxi or Uber drivers back in 2015…That’s when Uber first started engaging amateur drivers. And so, everybody came in. Very quickly, we can see the different groups or tribes, if you prefer, Uber drivers, taxi drivers, Uber passengers, other passengers.

  • Initially they are broadly in two camps. The first camp prefers convenience and the sleekness of the experience over the fact that it’s not lawful, but the second camp says that safety and lawfulness is more important than convenience. It would look kind of zero-sum.

  • The crazy thing about pol.is is that it shows not only how people are different, but how people are the same. What kind of statements, if you propose for other people to agree or disagree on, actually can win over support from the other group as well.

  • Using machine learning, it shows both the clusters as comprising of your friends and families on social media…It’s just easier to talk about Uber over dinner. As well as people who are completely strangers but are actually on your side, morally speaking, by voting very similarly to you.

  • Basically, it recreates a sense of equality in the sense that people can convince their friends over there, and with the arguments supported by the strangers over here. People then started proposing, for example, insurance is very important.

  • Whichever side you are on, people generally agree that if accidents happen, insurance need to cover it. With insurance, of course, comes registration, and with registration comes the need of professional licenses, actually, because it’s for business.

  • People also say, “If I’m just recovering my electricity or fuel,” then it’s not considered for-profit. It’s just ride-sharing. But if you start earning money much more than that, then it is kind of actually a profession. This distinction is made early on. Everybody agrees on that.

  • At the end, what we’re seeing is that everybody broadly agreed that the regulation need to change with time. That taxation, for example, need to be fair. If Uber operates with a rental car company, it need to be taxed like that. If it operates with a taxi company, you need to be subject to the same rising duties as a taxi driver, and things like that.

  • People have those five consensus items. Then we talk about precisely those items and nothing else in a live stream conversation, inviting people on all the different sides to join, and the Uber representative at the time in 2015 crucially agreed to work with their drivers to help them obtain professional driver’s license.

  • That became the foundation of the Multi-Purpose Taxi Regulation that is now part of the law. The Multi-Purpose Taxi Regulation basically says, you can have surge pricing, you can have all the different way of hailing. It doesn’t need to be painted yellow, which is almost Uber-like.

  • It enables the existing taxi company to also have a fleet that operates somewhat like Uber. They are all registered and insured the same as the existing multi-purpose taxis.

  • Didn’t Uber, though, Audrey, a lot of the problem with Uber in many parts of the world is that they said that, “We’re just a platform. We’re not a company,” and that’s why they wanted not to be treated like any other taxi company. Was it a case of wanting to enter into Taiwan? Do they claim to be a platform?

  • They are a platform here. They work with rental companies to offer e-rental service. They also work with taxi company through Uber taxi to offer e-taxi service.

  • They are a platform company here. We recognize that completely. The partners that they partner with needs to be subject to exactly the same law as rental and taxi companies.

  • Can you talk to us about, almost like a brief snapshot maybe of a few of the other main cases that you are most proud of?

  • The alcohol one surely is…

  • Changing the company law, yeah. One of the things we might try and build would be you basically going from seemingly intractable polarization to consensus.

  • Again, it’s not the alcohol case. It’s about purchasing liquor products over the Internet, so it’s called eLiquor case. The case is very interesting in a sense that it’s the only case in vTaiwan so far that would prepare consensus statements to the parliament. The parliament broadly seemed to agree with it. There was a parliamentary election.

  • The new crop of MPs says, “It’s not something that we have say about.” That’s it. It’s the only case documented that is not actually quite a parliamentary agenda, which is a product of vTaiwan. Everything else has either well entered into parliamentary agenda or already became regulation or a law. This is the only one that went to a set of a MPs and then another set of MPs came in.

  • That is why, in every case after the online liquor sales, we’ve made sure that we include the committee members of the parliament into or at least their assistants into the early stage conversation to make sure that these carry over, no matter which elections may come.

  • The core fight basically is how to identify the risk of minors getting access to liquor by posing as a different identity online. We have on one side, people who just don’t want any kind of spread of liquor products and on the other side, people who point out that it is already legal to buy it over Amazon and have it delivered to Taiwan.

  • It only limits Taiwan’s own agricultural products. That is actually easier to track the online trail as opposed to people just randomly sending messages to each other and getting it delivered. These two sides, while seemingly opposites, actually agreed on the same thing, which is a fair protection of minors from the liquor sales.

  • Then people start discussing on pol.is, broadly speaking, that if you recheck identity on the collection side, not just on the buying side, then it is a good way to verify the identity. One of the compromise solution emerged out of pol.is is that it can be delivered to a convenience store, which is everywhere in Taiwan.

  • Just like buying liquor in a convenience store, you still have to present photo ID for that. If you have an addiction to alcohol, it’s far easier to just walk to a convenience store and bring back liquor product than ordering it over the Internet and having to waste six hours.

  • We would think that it is a replacement. It is not an addition to the stock that is already in the convenience store. Even the people most concerned about minor safety do agree with this compromise position, as opposed to have surfacing this very reasonable compromise that tend to get lost in debate if people vilify each other.

  • Have there been any issues it’s just been unable to find consensus on? Have you encountered stuff which you think are genuinely divisive, genuinely intractable?

  • In my time running the vTaiwan community, which is from end of 2014 to end of 2016, around two years, we rejected a lot of cases. For example, marriage equality was proposed by the Minister of Justice to vTaiwan.

  • After the meet ups discussed this a little bit, we decided that there is no stakeholder discovery process over the Internet that has an advantage over the regular way. There really is no obvious connection between people who participate in the LGBT movements and Internet use. They’re totally orthogonal. It wouldn’t really make sense for an online population to have a conversation.

  • Compared to Uber, which is the mobile app, eLiquor, which is e-commerce, telework which is tele, marriage equality has nothing to do with Internet governance, not even by any stretch. We’re like, “Yeah, this is an Internet governance platform and maybe in marriage equality it’s better to deliver it elsewhere. It ended up becoming e-petitions and so on.

  • Basically, what we’re seeing is that there are consensus items that can be reliably produced, if we have some idea of how to reach the stakeholders.

  • If it is beyond the Internet or indeed beyond most mobilization methodologies, you really have to go to the ground and talk with people, social workers, local gatherings, and things like that, and bring the technology to them, rather than asking them to come to technology.

  • That is our main learning. It’s that using a pure Internet way of stakeholder discovery. It has its limits as of how much the stakeholder can be discovered. The more related to Internet governance it is, the more legitimate this consensus-making mechanism is.

  • Presumably for marriage equality, your view on that would depend on your age or would likely depend on your age, and Internet use, or platform.

  • Mostly spirituality, which has nothing to do with Internet use.

  • The other question I was going say to you was what does this mean to you personally, the whole thing about using computers to help decisions?

  • Democracy, yes. What does it mean to you personally, given your very early interest in computers?

  • When I was eight, a few things happened.

  • When I was eight, two things happened. I got my first personal computer. Before that, I was writing programs on paper. So I can actually program and have it run instead of just in my mind. The second thing is my dad went to Beijing to cover a student protest.

  • He waited until June the 1st to come back to Taiwan. 徐宗懋 after him get dispatched and get hurt very badly. That was the Tiananmen Square. People in Taiwan paid a lot of attention to both personal computers – at the time, most of the PC clones are coming out of Taiwan – and the Tiananmen Square, which people see as the way not to go.

  • Right after, there’s a Wild Lily Movement. There’s the sequence that led directly to the presidential election in Taiwan. People didn’t want to repeat the same mistake that people made in Tiananmen Square, just like the Berlin Wall. You know the rest.

  • In a sense, all these are made possible into the people’s consciousness thanks to not just personal computer, but digital photography, but also facsimile machines, and all those digital communication tools without which there really is no way for a journalist in a very restricted environment to get their messages out and reach a global audience.

  • At that time, Internet and democracy are very much both in the mind of the Taiwanese population in my generation. I don’t think it’s personal. It is generational. Then, when we are of the legal age to vote, we’re actually the first generation that have the legal age to vote for our president, along with everybody else.

  • It is a newly democratic country, and people start re-thinking democracy, already informed by the Worldwide Web and already informed by how Internet governance works. Participation is designed right in when we’re beginning to amend our constitutions to make way for more participatory governance.

  • At that time, it is called community building or 社區營造. They have been running for more than a decade before we call it collaborative governance. There’s a grassroots tradition of collaborative governance here as well.

  • For me, personally, I see it as fusion between the three traditions that I want to conserve, the conservative part of the concept of anarchist.

  • One is the people’s hands-on nature of forming co-ops, of forming social-sector organizations to collectively decide our communities for the future. That’s the social sector. We also have civic media that, empowered by the Internet, can reach readerships and viewerships far wider than traditional printing press.

  • We have the free software people who shaped Internet governance from the very beginning.

  • Those three traditions, when flown together, makes a better governance mechanism than any single party can do. Indeed, the ethos of g0v movement is to be maximally inclusive of all those different traditions.

  • If you could give a summary answer, that’d be great because it is quite long, just about, “The principles of g0v, include three things.” Then just say what they are.

  • If you go to g0v.asia, it lists three main groups or activists’ traditions in the g0v community. The three groups of people are the social activists who are very hands-on, the free software people who were part of the Internet governance from the very beginning, and the civic media, who use the Internet to reach viewers and readers far wider than the printed press.

  • The three kinds of people can fill each other’s gaps and, together, form a collaborative system that is much more legitimate than any democracies that we have seen before.

  • Do you feel there’s a real sense of urgency and peril in this work? What happens if democracies don’t remake themselves?

  • I’m going to head over to you about how democracy can become so much better and so much stronger, and seeing examples of it. But if I go back to the UK or much of the States at the moment, the conversation is democracy is dying. Democracy is on the way out. Democracy is on life support.

  • It is true that the institutionalized representative democracy, being that it’s three bits every four years, it’s too slow a bandwidth for people to like. People now mobilize by hashtags, doesn’t like the fact that their ministers takes months or quarters or years to respond to popular hashtags.

  • If ever, right. We also see a new generation of what I call populist democrats, rather than populist tribalists, who make full use of Internet to have real-time conversations, as our conversation is.

  • If you’re reading this in a radically transparent transcript, you’re part of this conversation of context-making and making sure that people can understand where each other are coming from, not just the policies that are the results of policy-making, the sausage machine not being transparent.

  • That is, at the moment, not what all the democracies are doing. Many democracies are, as you said, reverting back into a kind of ritualistic democracy where people feel that they have no real choices. In that environment, it’s very easy for a tribalist populism to grow.

  • That, I think, is indeed a rare chance where people can genuinely see different glimpses of future that is populist but not tribalist. That’s what Taiwan tries to offer not just to this region, but to the entire world.

  • Are there any lessons to the UK?

  • Yeah, or any other. In a way, it’s a reiteration of the last one, but just a simpler explanation, really. What does it say for democracy in the future? What can the Western world learn from Taiwan?

  • Actually, getting a shorter version of something you just said would be good to begin before the sun goes back.

  • All right, let’s go.

  • It sounds like technology is also changing people’s expectations around democracy.

  • Yes. As I mentioned, my first experiences with Internet governance, they’re all real-time. If I send a email, if there’s a consensus from our working group, within months, they become policy. Then they become…Just a second.

  • If I send an email into a working group, and we have consensus, then it becomes a request for comment. That becomes policy. That becomes code. It all feels like a direct engagement with the policy-making process.

  • Democracy, I think, is not getting further away from people. It’s just people have much more interesting diversions from democracy that are maybe on the community level, maybe an online community level that feels much more real-time and much more responsive than your ministers.

  • The two solutions that I can offer to this is that, like my office hours, I meet with people every Wednesday from 10 AM to 10 PM. The only condition is that we publish everything online, whatever we talk about. Many people who spoke out of self-interest need, in that environment, to find public interest in their arguments to talk with me. Direct access to a minister is really the key.

  • The other thing is the horizontalism within each ministry. While every ministry, of course, has its hierarchies. If we send one people who is not afraid of working out loud and let other minister know, then we can actually reveal the processing of the making policy.

  • When people order something from a food delivery or from Uber Eats, of all things, they don’t expect it to arrive at the door instantly. They do expect to know where along the way the position is for their order of food.

  • It’s the same with democracy. If we have people doing e-petition and having to wait indeterminate amount of time for the indeterminate process to happen without their involvement, then it’s not very interesting to participate.

  • If every time along the way, every couple weeks or so, there’s activity to do, just like vTaiwan with different stages, then people feel constantly engaged in the policy, even when it’s just forming and not solid into a policy itself. The policy-making process itself can be turned into a social object that makes people much closer to the policy-making apparatus.

  • What could the UK learn from this?

  • While I’m not terribly informed of UK politics, I do know that the UK basically defined a lot of what we know as liberal democracy nowadays in the world. Maybe one of the things that people can do is just to look at the existing system that are working pretty well on the floor, in the real world, before Internet arrives.

  • For example, people’s way of getting a hearing from parliamentarians. MPs way of accepting petitions. If you modernize that and empower the people to have real agenda-setting power, while overcoming the noise-to-signal-ratio problem, then you can very easily augment that kind of consensus making into a already pretty working well, well-oiled, institutionalized democracy system.

  • This is what my friend, Indy Johar, says is the dark matter of democracy. It’s the public service looking at newer engagement methods and see only the confusion and noise.

  • If you can somehow introduce parts of the vTaiwan-inspired process so people see more signal than noise, then actually career public service may very well be in full support of this kind of mix of online integration.

  • What happens to democracies in the digital age that can’t reform themselves in any way?

  • Well, they become fossilized or they become kind of ritualistic. People still perform elections. But the turnouts tend to get lower over time. People tend to accredit less legitimacy to public institutions. Sometime, like what happened in Iceland, the establishment lose legitimacy at some point, and people are forced to draft up their own constitution, as an attempt, anyway.

  • I think people basically demand change, nevertheless. It’s just whether it’s an evolution or whether it’s a revolution. But you still get change in the end.

  • Democracy is going to have to change.

  • Hold it, your mic is rubbing against your clothes.

  • OK, sorry. No, it’s fallen, sorry. It needs to change, just a second. Do I need to redo that segment or can we use this soundtrack?

  • Yeah, we could use that.

  • Democracy will have to change, it’ll either be an evolution or revolution, but it can’t stay the same.

  • Democracy is already changing. It’s whether the democratic institutions go with the flow and adapt into the collaborative governance culture that people are already having in their communities, or they remain fossilized and rendered somewhat distance and ultimately kind of irrelevant.

  • Does that mean that if democracy becomes irrelevant, fossilized, doesn’t reform, it could actually collapse before it’s reformed?

  • I wouldn’t say that it’s quite collapsing. Right? The Occupy back in 2014 is not about collapsing the establishment, it’s about demo, demonstrating a way of governance that actually works more in real time for a specific topic that is across straight service and trade agreement.

  • The Occupiers don’t presume to do this for each and every subject, but for that particular subject, with 20 or so NGOs, it’s a viable demonstration that something new is possible. It’s the existential proof.

  • I would say that it is a revolution in a way that it changes people’s mindset. But it’s also evolution in a way that there is no overt violence and there’s no overt changing the constitution or anything like that. It’s just public service collectively realized, “Oh, it can be worked in our benefit as well.”

  • Sorry. Can you just say that again? Just in terms of this. For instance, what about all democracies taking this on? Taiwan’s a new democracy. Do we want to, what happens to old democracies?

  • I just say that, introduce some adapters. Right? That’s what I just said.

  • For democracies that has been running for centuries, obviously, while Taiwan’s research may seem radical because we’re a new democracy, everything’s possible. But there are elements of our experiment that just turns out well that maybe adapted very quickly, like the e-petition system and the pol.is system for getting consensus out of noise.

  • All this are simple click and play technologies that can be introduced as part of the digital counterpart of that analog petition process, the digital counterpart of an analog party member meeting process.

  • It doesn’t have to replace the face to face element. Rather, you can live stream the face to face meetings, have many more people participate in real time in it, and also, asynchronously, through pol.is and other technology.

  • That is to say, you can take an existing already very well-oiled institution and amplify its effect on people’s relevance in their minds, rather than saying, “OK, we scratch this and we do something else entirely.”

  • That’s a really good answer.

  • Yeah, that was great. What about rebuilding trust in e-petitions? Because for instance, the taxi driver on the street probably say. He said that he hadn’t actually heard of vTaiwan, and as a whole, he…

  • Surely he heard of the petition over the e-taxi case.

  • If you ask specifically for the petition, or for 103-1, then he may have heard of it.

  • That’s good, but…

  • No, he had heard it. He’d heard of you and he’d heard of e-petition.

  • That’s right. Because the vTaiwan is kind of a team providing the process…

  • Yes, you would naturally, yeah.

  • It’s like asking people whether they have heard of Nesta, maybe not.

  • All right, but nevertheless, he did say that a lot of people in Taiwan are simply still not that interested in democracy, in politics.

  • Also, they don’t trust politicians, right? Do you think this is all a process to be able to trust politicians, in a way?

  • No. I don’t think so.

  • Yeah, OK. It’s a good question, though, I’ll answer that.

  • All in all, I think this process of co-governance is a way for the public service, including politicians, to trust people. It is not about people trusting the politics, which is what the Economic Power-Up Plan advertisement tried to do, is blind trust.

  • Blind trust is worse than no trust, because it’s asking people to trust their government and the politicians without them trusting back. That essentially leads to fascism and we all know how that went.

  • Basically, what we’re doing now is radical trust from the government to the people by making the state radically transparent to the people. We’re not saying that people have to trust us, rather, people can independently set up their interest groups, their unions, their co-ops or whatever, and advance their own agenda. But based on the same thoughts, based on the same context.

  • But that’s the early important point. If people are starting to talk past each other and living in parallel worlds, that’s something that we don’t want to happen. But if, like, taxi drivers, they have their own line groups, they have their own interest groups and so on and can start a debate based on that e-petition that responds to both sides with the same facts.

  • At least it is a conversation that has a chance of converging. That is our main contribution, is making the state apparatus radically trust people most and then people, some of them, may trust back.

  • Is there anything you can say about, in terms of decision-making, the usual way of decision-making is like the people who are the most articulate, the loudest, the most charismatic personalities, they all win the argument. Is this a way of actually taking that away and actually making more about the argument itself? I don’t know, do you think that’s a fair point, or am I putting words in your mouth?

  • Right. It’s a good question. Like using pol.is, you have to essentially be poetic. It’s a limited number of characters and you need to succinctly make a statement that resonates with people. That’s a poet’s work.

  • Whether that means that people are good at rhetoric, just shift our oral tradition to a written tradition, that’s anyone’s call. Right? But we make sure that along each stage, there are different modalities. If you’re not that good at writing your statements, you can still show up in the virtual town hall meetings and make your case by using memorable language.

  • But I think the balance between the stakeholder groups is very important, which is why I think the root of vTaiwan, that is the weekly meet ups, and the root of PO network, which is in our monthly meetings, is actually more important than any particular issues we talk about.

  • Because this about people who are not comfortable with any modality, have a voice to say, how about we consider that modality? That makes the inclusion the core of the open government process in both cases, rather than just transparency or participation, which are instrumental values, inclusion and trust is the core value that creates the impact.

  • What do you mean by modality?

  • Some people prefer to share their positions through storytelling. Some people prefer to do it through visual drawings. Some people prefer the interactive model that they can show other people, Lego-like. Some people prefer poetry.

  • Basically, people have different communication modalities. The idea of intersectionality is that we’re all weak on some modalities and stronger on other modalities. The idea of transcription or translation between different modalities, by making sure that our dialog here is captured as video to include the memorable gestures.

  • As transcript for people who are more of a textual inclination, to make mind maps of people who are more of a topological, visual inclination or more design space inclination and make sure that everybody can participate into the same arguments by receiving them in their preferred modalities, and that’s inclusion.

  • By modality, you mean like mode of presentation?

  • Yes, or mode of communication.

  • OK, I see. Would you just be able to say, just modes to communicate and say it.

  • Everybody have different modes of communication, like we’re talking face to face, this one mode. This is captured now by this camera, which captures all my memorable gestures.

  • But there are people who prefer textual modes of communication, which is why we’re making a transcript, in case you’re reading this. We’re also making, for example, mind maps for people who are more inclined, who are thinking in a spatial dimension.

  • Each democratic process, just by way of having this multi-stakeholder meet ups, to collectively decide a process, can always include more modalities, that is to say, modes of conversation, into the conversation.

  • People who are good at one but not so good at others don’t get excluded just by the fact that they’re not that comfortable with textual communication or with rhetorics or with any other kind of poetic expression.

  • That’s wonderful. Just one summary thing about your vision and your personal feelings about using the power of the Internet, the power of computers to…

  • When I became the Digital Minister, it’s the first time that Taiwan has a Digital Minister. The HR people asked me to write a job description for a Digital Minister. This is what I wrote. It’s a poem.

  • “When we see the Internet of Things, let’s make it an Internet of beings. When we see virtual reality, let’s make it a shared reality. When we see machine learning, let’s make it collaborative learning. When we see user experience, let’s make it about human experience. And whenever we hear that a singularity is near, let us always remember the plurality is here.”

  • That was brilliant.