• Thank you all for joining the Institute for Public Knowledge and the GovLab at NYU tonight for this inaugural event, in a series that is part of the new Future of Democracy working group here at IPK.

  • For those of you who are new to IPK, we are a social science institute at NYU that supports communication between researchers and broader publics around major public issues. Our working groups consist of graduate students and professors from within NYU as well as members of business, non-profit and academic arenas beyond the university.

  • Members collaborate to write papers, host conferences and meet regularly to discuss their individual projects. If anyone here tonight is interested in joining or learning more about our Future of Democracy event, please come see me, Jessica Coffey, the Associate Director of IPK, after the event, at the reception.

  • The next event in this Future of Democracy series will take place on Wednesday, September 6th, this coming Wednesday, at 12:00 PM in this very room. We will have a conversation with Geoff Mulgan, Chief Executive of Nesta, on Collective Intelligence and Democracy. Please visit IPK’s website for additional details and to RSVP.

  • Now I’d like to introduce the leader of the Future of Democracy working group, Beth Simone Noveck, our humble leader.

  • (laughter)

  • Beth is Director of the GovLab at NYU and its MacArthur Research Network on Opening Governance. She is a Professor in Technology, Culture, and Society at NYU’s Tandon School of Engineering and a Senior Fellow here at IPK.

  • Beth was recently appointed as New Jersey’s first Chief Innovation Officer and previously served in the White House as the first United States Deputy Chief Technology Officer and Director of the Open Government Initiative under President Obama. UK Prime Minister David Cameron appointed her Senior Advisor for Open Government.

  • In conversation with Beth tonight will be our guest, Audrey Tang, Taiwan’s Digital Minister in charge of social innovation. Audrey is known for [laughs] revitalizing the computer languages Perl and Haskell, as well as building the online spreadsheet system EtherCalc in collaboration with Dan Bricklin.

  • In the public sector, Audrey served on Taiwan’s National Development Council’s Open Data Committee and K-12 curriculum committee, and led the country’s first e-rulemaking project. In the private sector, Audrey worked as a consultant with Apple, Oxford University Press, and Socialtext.

  • In the social sector, Audrey has actively contributed to g0v, which we’ll talk about tonight, a vibrant community focusing on creating tools for the civil society and a call to fork the government. I now pass the mic to Audrey and to Beth to get us started. Thank you and enjoy.

  • (applause)

  • Thank you very much, Jessica. Thank you to IPK for hosting us. For those of you in the back, we have seats up here in the front. We promise A, we won’t call on you...

  • (laughter)

  • ...often. If you have to leave early, please don’t worry. Neither Audrey nor I will be insulted. Please feel free to come up and take a seat. We’d love to fill in the room and have you join us.

  • I’m really thrilled and grateful to Audrey for being here for our first inaugural Future of Democracy lecture conversation, as we’re going to have it. I’m hopefully not the fearless leader. I am just the modest hostess of a conversation that I think we’re all eager to have and eager to talk about.

  • It’s very hard to live in the United States right now and not to be in the world right now, I think, and not to be deeply fearful and worried about the state of democracy.

  • We started, let me just give a quick word of this is why we started this conversation. We’re living right now in a time, in a country, in which every day is election day, in which party politics are coming ever before problem-solving and public interest.

  • I think we’re very deeply concerned that this is not simply another cycle in the swing back between one party and another but potentially something deeper and more dangerous that’s going on. There are no tanks in the streets. If you’ve read Ziblatt and Levitsky, the two Harvard professors who wrote one of the many books on the death of democracy that have come out this year, it’s a particularly good one.

  • They very chillingly paint the picture of how the dangers that we face today are not tanks in the streets. It’s not people with guns. It’s not people being thrown in prison. It is in fact what we’re seeing, which is Putin who is permanently in office by swapping the role of President and Prime Minister. It’s Erdoğan in Turkey. It’s Orbán in Hungary. It’s Maduro in Venezuela. It’s Modi in India.

  • Of course, it’s Trump here in the United states. People who deride their opponents as criminals, who maintain a constant sense of threat and support and endorse violence in our political culture, who show contempt for their critics and for the media, and who, frankly, stoke conspiracy theories galore, again to reinforce their power, and in ways that actually make us, I think, deeply fearful about our democracy.

  • Add on top of that the fact that for a long time now, trust in government even before these current spate of authoritarian and populist leaders, that trust in government has been declining over many, many years. In 1958, 73 percent of people said that they could trust the federal government in the United States to do the right thing just about always.

  • In 2013, that number was 28 percent. Right now, in most recent surveys, the number of people who say the federal government at least does an excellent job is now at two percent.

  • We’re going to hear tonight though not about the United States, but about Taiwan. Hopefully, not only about the challenges that are faced, but hopefully some very hopeful developments that Audrey and her colleagues have been able to bring about.

  • We would like to find out the ways in which you are helping to make government and democracy stronger, to do a better job at delivering services, what the vision is and for creating more effective and legitimate policies. Really to understand what it is, what this revolution is that’s happening in Taiwan that may hopefully give us all some hope...

  • (laughter)

  • ...for what’s happening here. Welcome very much. Let me start. We’re going to make this as much of a conversation. Knowing Audrey, she will hack whatever it is that I ask her and talk about, I hope, what she wants to.

  • I will start things off by really just asking you, how did you get started? What was the problem? Was it this problem that we’re talking about now? What was it that motivated you? What was the challenge really that you were setting out to solve? Was it a problem of democracy or of governance or of both?

  • Hello. This mic distorts a voice like any other intermediary, but I’ll use it anyway. Maybe not.

  • Maybe we can switch to the moderator’s one?

  • (background conversations)

  • Still not much better. Actually, can people hear me without...

  • ...without using the mic? This is good because we have this as well. Maybe just at this distance. Is that OK with you?

  • Very glad to be here. Actually, last week I was in London and had a hour-and-a-half conversation with Nesta folks and Geoff Mulgan as well.

  • (laughter)

  • We’re very much in line in value and as well as in the various innovations to democracy that we’re working on. This is my office in Taipei City. It’s called the Social Innovation Lab. It’s very playful and peaceful. It resembles my post as the Digital Minister.

  • As an anarchist, I don’t have a contract with the Taiwan government. I had a compact or a covenant, which is three basic points that I would elaborate further. The three points are in direct answer to the problem of legitimacy that we’re trying to innovate and to solve.

  • The three pillars of the compact are first, a voluntary association. I work with the government, not for the government. I don’t give nor take command. All I do is facilitate, make suggestions, receive fear, uncertainty, and doubts, and do some facilitated work.

  • The second thing is radical transparency. Literally everything that I’m a chair of in a meeting in a cabinet, we publish the entire transcript 10 working days after each convention. It’s the same for lobbying and for journalists and for anyone who come to me during my office hour. Every Wednesday here, I’m in that place from 10:00 AM to 10:00 PM.

  • Anyone can come to talk with me long as they agree for the radical transparency.

  • The third thing is location independence. Anywhere on Earth, I’m still doing my job. This enables, of course, to have creative offices like this, but also enables a lot of regional innovation where I just go to a place and do some ethnographic thing. [laughs]

  • Like investigative reporter kind of thing in a population and in very rural or indigenous places, but meanwhile having the 12 or so ministries in the Social Innovation Lab to see through my eyes how is it like in the field out there and for the people in the field to have a real-time conversation with the 12 ministries involved in a national social innovation plan.

  • Those three taken together I think form a legitimacy system that was initially prototyped during the occupy. The occupy, for people who don’t know about it, was four years ago now, in 2014, where people occupied the Parliament in Taiwan because the MPs were on strike.

  • They did not wish to deliberate substantially a Cross-Trait Service in Trade Agreement or a CSSTA, and because the MPs were on strike, they went to the parliament and did the MPs’ job for them, namely deliberating the trade service agreement.

  • What they do as a demo, in a demo in the sense of a demo seeing not a demonstration of purely protest. For 22 days or so people were just there, and also around the different street corners there was about 20 NGOs deliberating each and every aspect of the CSSTA.

  • I was part of the people in Sunflower Revolution not as taking any of those 20 or so sides, but rather as part of the movement called g0v, which, as the moderator already introduced, is the community that caused to fork the government which was start in 2012, two years before the occupy.

  • This is a very simple hack. You can do it yourself here. Any government website in Taiwan ends in gov.tw, and the idea very simply put is that if you don’t like, for example, the legislative website, you can just do a alternative by changing the O to a zero on the browser bar. That solves the discoverability problems.

  • Any public service, by changing O to a zero you get into the shadow government, which does the same thing except with more interactive, more open data. People participating in this movement relinquished their copyright so that by the next procurement cycle, we see a lot of g0v innovations.

  • The first one was the budget visualization and having each budget item a conversation board that actually gets merged this year, so all the 1,300 ministerial projects become a social object upon which you can have a real dialog with the public service and things like that.

  • All this is a fork in the government, but it’s also systematically merging it back. G0v was supporting the communication for the Sunflower Occupy throughout the 22 days, and we see something very interesting happening. We see that with this kind of radical transparency, with a location independence, where wherever you are you can be part of the conversation.

  • With this kind of voluntary association, people just cross-pollinate the different points around CSSTA so that over three weeks, people converge rather than diverge as some other occupies do. We converged eventually on a set of very accurate demands, which the head of parliament then accepts, and so the occupy was successful.

  • That was our first demo, but after that, we’ve just been scaling these conversations. It doesn’t take occupy to start a conversation like this, but actually it’s something the public service is comfortable of doing it themselves. That’s the legitimacy crisis, and the innovation that was brought up to solve the legitimacy crisis.

  • [laughs] I want to push you just on this, but this is very interesting that whereas for...Oh, I’m sorry, here, I’ll grab the bad mic. Is that the...?

  • Obviously, Zuccotti Park, the occupy here, the Occupy Movement and the related social movements were not as successful. There’s clearly some vision that you had or that others had or they coalesced that caused you to develop some really concrete proposals to move forward.

  • Would you say it was a particular vision for democracy that was the cause? Why did this work and those didn’t would you say?

  • I think we benefited from seeing the other occupies around, and also from the two years between the g0v has founded to the occupy, so it was two years of civic tech people basically trying to turn collectivism into something that is more a general setting, which we call hacktivism.

  • We do it obviously through sharing of open data, turning data into social objects through having a good social sphere, where people can ask questions without getting burdened by trolls and things like that. There’s a strong deliberative quality to the spaces that g0v has been building since 2012 to 2014.

  • After that, of course, we try to have a real conversation with career public service, and try to get career public service into the design of such things. Basically, by relinquish the copyright, have the public service take over the maintenance after the initial prototype.

  • All this fosters a culture that focus on what we call scalable listening or listening at scale, meaning that we get people into the mood of listening but do it at scale. We did not certainly anticipate the occupy, which is the real agenda-setting power, but we were almost there.

  • When the occupy actually happened, there’s a lot of ready-made systems for crowd transcript, for crowd live streaming, for a lot of different collection of opinions, for visualizing them in clusters. For getting the digital line in the CSSTA, you just have to enter your company number or our company serial number.

  • Then it shows exactly which paragraph of CSSTA affects you in a comic way that everybody understands and things like that. All of these are ready when the occupy happened, and so we plugged in into the Occupy Movement to have half a million of people on the street using the system that previously only maybe 5,000 people was using.

  • I just want to push a little bit further on this, because so many of the other public engagement platforms that have been developed growing out of different social movements, and here I’m thinking about our friends in Spain, who have chosen to go a different path which is what we might think of as radical direct democracy.

  • Their view is they want to be Switzerland. You don’t want to be Switzerland, and I’m curious why not.

  • We’re not Switzerland.

  • (laughter)

  • Neither is Spain, but... [laughs]

  • Right. I don’t know too much about direct democracy because Taiwan just at the end of this year is going to have the first substantial referendum. I think there’s 9 or 10 topics waiting for referenda, but even in our new revised Referendum Act, there’s a strong emphasis on the deliberative quality before the referendum is actually cast.

  • I think part of it is culture, because in Taiwan we can talk about, for example, consensus, and people visualize things with consensus statements, divisive statements. We can say with a straight face that this online discussion after a while in a safe enough space always produces consensus, but that’s because the social norm already values consensus.

  • Maybe just in a rough consensus way, but we are reasonably sure that it always ends up with something like that, with the right space design, and you probably cannot say that in many other places on Earth. [laughs] This is a defining characteristic that we always want to converge on something, and it’s seen as a positive social value even among people who disagree bitterly.

  • You’ve created this whole ecosystem now of tools and culture for digital listening and deliberation. I know that everybody is super eager, at least I think you should be super eager to get to the demo and to see the details here. I’m wondering if you can, before we dive too deeply into the mechanics of any one platform, give us the overview about the different pieces in this vision.

  • How they are connected? Whether they came one at a time or sprung whole cloth from your head and that of your colleagues. What are the components in this kind of ecosystem that have begun to create this new culture?

  • If we consider the occupy as people really going in and thinking really deeply about one singular social issue, the difficulty usually lies in asking the right question, like how may we move forward to find some common value, i.e. "How Might We" questions using standard design thinking terminology. [laughs]

  • We can separate these into four different phases, where the first is the crowd fact-finding, the checking of feelings, the getting to a point where people think. These are the objective facts that we can all agree on. Then, just by listening in to people’s feelings, we have a month or so dedicated in each topic just for people to check in on each other’s feelings.

  • Visualization technology certainly helps a lot here, because if you can see your Facebook or Twitter friends’ relative position in a crowd-sourced map, then people can actually relate much more, because it’s all your friends and family, so you just didn’t talk about this particular issue over dinner.

  • They’re not nameless enemies, they’re people who you can have a real conversation with and discover. All this is not done by a preset questionnaire, but rather by people sharing their authentic feelings and for other people to rate whether they resonate or not with the feelings. This feeling stage I think is very important, the reflective stage.

  • Of course, after we get a set of reflective rough consensus, the ideation begins, but now we have a clear goal, because the best idea are the one that takes care of most people’s feelings. Finally, the Parliament’s system or the administrative systems steps in and make decisions by essentially ratifying the ideas into a coherent decision.

  • This objective, reflective, interpretive, and decisional stages came out of a Canada research called Focused Conversation Method, and it’s broadly speaking the stages that we used for public consultations, public multi-stakeholder discussion panels.

  • Now tell us about the tech, and how all of this gets...

  • So soon. I can’t wait, I have too many... I’m going to come back to all my other democratic theory questions in a moment. Let me just interrupt again to say if folks in the back want to, there are seats over here on the side. Please feel free to come up and join us, but they look comfy back there.

  • We can talk about the tech.

  • Or you can override me. If you want to talk...

  • No, it’s just fine. I think the tech is worth looking about because really the reflective part was the hardest part. How do we get people with no face-to-face experience together to resonate with each other on feelings instead of just polarizing and trolling each other or scam pictures or whatever?

  • That was the missing piece that we were looking for. Fortunately, there’s some startups in Seattle solving exactly that. That’s the pol.is system, which is the first crowd feeling checking system that we tried. We tried dozens of them. It’s the first one that actually worked.

  • Back in 2015 the topic at the time was UberX, and UberX as you know is this meme called sharing economy, but the payload actually means that code dispatch car is better that law, so we need to obey to no laws. That’s more or less payload back in 2015.

  • Just like any virus of the mind, it’s go through the interactions from drivers to passengers, to driver to passengers. It actually is a meme in the pure sense that if a driver after trying it for a couple of weeks found it’s not a very good deal after all, they will have already infected people, and that’s basically the beginning of polarization.

  • Using this kind of consultation, what we think is a vaccination of the mind so that after people confirm each other’s feelings, it became much harder for people to be polarized by one-sided PR messages.

  • This checking in each other’s feeling is done simply by people looking at one sentiment or another, clicking I resonate with this or not, and see their avatar move among the clusters formed by the similar-minded people. There’s two things that’s worth mentioning. First, we’re not looking at the numbers at all. These numbers mean nothing.

  • These just [laughs] basically measure the diversity of the sentiments at the moment, and anyone can just override by proposing something that’s more new and it’s more eclectic that resonates with more people.

  • The second is that basically there’s no reply button so you cannot really troll someone here. That’s the two-secret sauce, so to speak. [laughs] We sent that link to all the drivers and passengers and unions and whatever, and after three weeks, as I said, they came out with this very strong consensus statement just like we did during the occupy.

  • A thing that had an afterward attack with the live streaming of the meetings between the stakeholders, and we bind ourself to use only the consensus statements as the agenda for the meeting, and checking into consensus resonated feelings one by one with the stakeholders. Do you agree? If you don’t, why not?

  • Just by committing their words in a live stream way, transcribe and so on, people can refer to that transcription as the social object. That leads to the much easier ratification, because people cannot easily go back on the statements they have when they know that thousands of people are watching the live stream.

  • That agenda was done by the same people who formed over the course of three weeks through live stream. That is the UberX case, we did maybe 26 or so cases this way, before I became the Digital Minister, and that’s the vTaiwan process in and on itself.

  • That’s the vTaiwan process. A couple of questions. Tell us something about who’s participating here? vTaiwan over the course of those 26 pieces of legislation they’ve been developed have engaged about 200,000 people, is that number about correct?

  • That’s right. If you count people watch line stream or in pol.is, because it’s a very low threshold. All you have to do is just click a bunch of agree or disagree.

  • For me, your own reflection on it. 200,000 is a pretty big number, but Taiwan has 23 million people, right?

  • It’s still a fairly low number.

  • Who were those people who were participating? Were you satisfied with who has participated? Let’s talk about who those individuals are or how you got them a little bit, and why you think it wasn’t more.

  • vTaiwan, as I said, is mostly about getting this thing, this "How Might We...." thing right. It doesn’t even talk about the ideation and decision making which is like the follow-up statement that each ministry has. What we focus on in this point is very much what we call a diversity in stakeholders.

  • For UberX, of course, we have to talk to unions, to various different taxi companies, to Uber itself, to the various horizontal groups that’s formed around this particular issue. For Airbnb, it’s a very different population. For anything that’s related to digital economy, for example, for the privacy protection, either one is a bunch of very different people and so on.

  • There is no constituency of vTaiwan. It is mostly just people sharing pizza, food and whatever every Wednesday in the Social Innovation Lab, [laughs] and basically over at dinner think about things that they would like the politic to talk about, and also invite the various ministries and agency people to join in on this pizza discussions.

  • Over the course, I think the Uber one, we got three month or so before we even agree on the name of the title of the consultation. We eventually settled on "riding a car driven by someone with no professional license and charging you for it." [laughs]

  • That absolutely neutral sentence was basically formed over many month of stakeholder gatherings which the stakeholders who participated in this eventually went back to their communities and spread this questionnaire and these oral engagement forms.

  • It is very much a stakeholder conversation platform, much like the Internet society or standard-making association rather than trying to do ourselves as something that has a power of a referendum. We’re very much not on this stage.

  • We’re very much just on the fact checking and the feeling checking stage. That was the vTaiwan position, which is why it’s seen as complimentary to hierarchical power or to representative democracy, but not reinforcing it, just complementing it.

  • Just complementing it, but it was a very deliberate decision to start with that component of the problem as opposed to instituting a citizen jury or something that used sortition or a random selection of a representative sample of people.

  • You may know James Fishkin’s work out of Stanford. He’s the granddaddy of this field who said it’s only legitimate we’re going to get a representative sample of the population, 400 people together in a room, and measure their opinion. Why start here?

  • It is not very clear to me that once you get after the sortition, that these people will actually carry out the deliberation back to the communities that share their sampled statistic characteristics. That’s the first one. A second is that even if it is statistically fair, some people are just better orators than other people, better at rhetorics, better at convincing other people, given a limited amount of time.

  • It was not at all clear to us that we should start with a representative sampling. Of course, if you want this decisional part to be informed by such sortition-based sampling, we do have some of that here going on in Taiwan, but it is more of in the ideation and decisional processes and not at all in the first two phases.

  • These two are disconnected. They can be connected by a shared question, but when we started, we were not entirely sure that we should start with a random sortition, mostly because, first, there’s no culture for it. There’s no jury system in Taiwan. We’re just starting to introduce one, so it’s experimental, and people don’t have any prior experience to sortition.

  • The second thing is that we want to make this really lightweight so that any city public servant can run it by themselves, and sortition is expensive compared to this mechanism.

  • How much does this cost, by the way?

  • (laughter)

  • Who pays for the pizza?

  • Yeah, who pays for the pizza? We had a donation box. After each meeting, people just chime in with coins and whatever. It is all very much crowd-funded of the people who show up for the pizza. Really, what government does is essentially two things, first, to agree to appear on the pre-meetings sending their people from the right agency.

  • Finally, after the questions and the reflections are synthesized, the government to reply point by point. That’s the two main commitments, and they don’t cost anything from the government agencies standpoint.

  • For an anarchist, a self-proclaimed anarchist, you are remarkably concerned with government involvement and what government thinks.

  • I’m concerned with public servants.

  • [laughs] Tell us a little bit more about how you have engaged public servants, how you’ve gotten them to participate. Getting them to the table may be even harder than getting the pizza eaters into the room.

  • I’m really curious what how you’ve convinced public servants to participate, and importantly how you really had the idea that that was as crucial that it is. I agree with you completely, but it’s surely not a universally held view that they are key to the equation.

  • You can’t even bring government in. Look, you yourself went into government, so it would be wonderful to figure out why that happened. Did you take a wrong turn? How did you end up in government, and then being so good at convincing government become part of this process?

  • First of all, during the occupy, the legitimacy or people’s approval rating of the central administration was at nine percent. It’s not exactly two percent, but it’s dangerously close.

  • (laughter)

  • That is a time when the entire public administration lost legitimacy. Also, at the end of that year of the occupy, every mayor who did not support occupy lost the mayoral election. The mayors who did support the occupy, they found themselves mayors, was not preparing any inauguration speech.

  • (laughter)

  • This is also something that happened in Spain actually.

  • That is the time, a kind of moment in democracy, where the people in the public service very much did not want another occupy, which is why the new premier at the time just invited the neutrals during the occupy -- the facilitators, the communicators, the people who worked with all the different 20 different NGOs -- essentially as mentors or advisers to the public service.

  • I remember the first few lectures I gave, there were three lectures to 100 people each, and they were ranked 12, almost the highest rank in the career of public service. There was exactly 300 such people in the Taiwan Administration in all the different city, different branches in the government.

  • All the 300 officials have ranked 12, basically, through the three-hour lecture on how vTaiwan works, how the occupy technologies work, and how to communicate with people. I found that most of them are actually very much pro to this kind of conversation, mostly because in career public service, the deal was pretty bad because if things go right, the Minister takes all the credit.

  • (laughter)

  • If things go wrong, they take all the blame.

  • (laughter)

  • Of course, people don’t innovate in such circumstances. they found that there is something in that for them for radical transparency because if there’s good ideas, even in the early stages, people discover about how professional they are, and also how truly concerned they are for the public welfare.

  • You don’t usually see that because you see that only after the administrator takes or rejects their proposals.

  • Second, it really reduced the risk because if you communicate with people who would have been on the street but now is willing to go the Social Innovation Lab, then there’s much less risk for everybody else involved, so more credit, less risk. Also, less work. So why not?

  • That is why after which I trained another a thousand or so public servants of lower ranks in the civil academy doing this.

  • Wow. Let’s stick in with vTaiwan for a minute. I want to come back and ask you about the lab and the rest of this ecosystem. Talk to us a little bit about the impact from your perspective.

  • 26 pieces of legislation, 200,000 people participating, how would you describe the impacts both in terms of individuals, in terms of the institutions, in terms of society? Are those laws better laws as a result of this? How do you know?

  • This is a very fair question. Of course...

  • [laughs] To which you have an answer to everything.

  • (laughter)

  • This was the detailed interfaces, more colorful but less legible. I would say that the vTaiwan is a kind of existential proof to everybody involved that this is possible, but at the time, the Uber case in particular, I think there’s three notable omissions.

  • First is that we, the civic hackers, did all of this ourselves. We did not actually involve... Oops, I think it went to sleep or something.

  • Check the input terminal.

  • Just hit the key and it will wake up. We may not...

  • No, that’s fine. It will come back. Here we go. At the time, the civic hacker did everything and did not involve the career public service in the preparation.

  • I think that is why it did not actually scale into city level or municipal level, because while everybody see that this is obviously working, the "how to make it work" is not exactly common knowledge to the public service especially on the municipal level.

  • The second thing was that when we did this experiment, Uber was only operating in Northern Taiwan in Taipei and Taoyuan and so on, so we did not invite taxi drivers from the Southern Taiwan that will go back to the owners because the legitimacy was simply not there.

  • The third thing, finally, is that people keep in this conversation want to broaden the scope to talk about platform economy in general, but not UberX or Airbnb in particular. We should have gone with that and that also came back to haunt us [laughs] because then, we’ll have to do a case by case for each and every cases after that.

  • We remedied some of that after I become the Digital Minister. I think, what prevented vTaiwan from getting into all the municipal places was it is very cutting edge, and the people who do it did not actually do it with the career public service.

  • It is seen as a plug-in or an oracle in computer science speak, where you can just plug in and it gives you a good resonating consensus, but it is very much a black box from the current public service point of view. They may accept that there is no legal risk in doing this, and also, that it also saves them time for most tasks.

  • So I don’t think it became really popular because the public service at that time, still, they don’t know how to upgrade it independent of those civic hackers.

  • Does that explain the...I think someone else has said that 20 percent of the deliberation sort of happened on vTaiwan and have not led to decisive government action. Is that because of the lack of public service or just the participation or is the nature of the issue?

  • Not really. There’s only one case that people reached a consensus and that did not get turned into a law. That’s the online liquor sales case. In every other case, the consensus was respected. Sometimes, it did not lead to action because the collective consensus was, we don’t need a law for it or we don’t need new laws for it.

  • The cyberbullying one, in particular, after various rounds of discussions, people generally think bullying has existing laws to work with it. What we should do is, basically, have a foundational law that treats online behaviors in the same way as offline behaviors.

  • In cases where the metaphor doesn’t hold, provide breaching clauses for it to hold, but not to treat cyberbullying as a different thing as bullying. People generally agreed to not have a special cyberbullying law, which was the original ask.

  • In many cases, people deliberated, and after a few months, decided that, maybe, the best course of action was not governmental action, but action from the social sector, the civil society and private sector to build new norms. That case count for most of the 20 percent where it did not lead to government action.

  • In your wildest dreams, does every piece of government action go through this process?

  • No. [laughs] Mostly because getting people on the same page is really difficult. This checking of reflective feelings rest on the fact that people can look at a description in some crowd-sourced data about private drivers and charging people for it. Everybody who participate have an idea of what is it like.

  • If it is too far in the future, for example, if we’re deliberating about, I don’t know, zero-knowledge proofs and data agency based on mutual ledgers governance systems, then it will require a lot more intuition-building before we actually enter the checking of feelings.

  • Alternatively, if we’re talking about transitional justice of indigenous nations, the past vary too wildly. The same concept don’t even hold the same currency in people’s mind who all came from the 16 different indigenous nations in Taiwan.

  • The transitional justice process, we can’t just say, "Go use the vTaiwan process," because the basic empathy, the basic vocabulary have to be built before even the facts can be considered. If it’s too far in the future, too far in the past, or is a mixture in between, I don’t think this actually is the best focus.

  • Do tell us a little bit, if we can do it without... apologies for the visuals.

  • No, it’s just fine.

  • Although we can try to troubleshoot them in parallel, I want to ask you about the rest of the ecosystem. I’m sure there’ll be lots more questions that I haven’t asked yet or we haven’t the time to cover on the vTaiwan mechanics. I don’t want to run out of time to really talk about the broader, the role that the lab plays, the role that some of the other pieces, which I won’t give away and let you tell us.

  • All that was up to mid-2016 before I got this post of Digital Minister. After I got into this post, there was this compact of three clauses. My main aim in the public digital innovation space that we set up was mostly just to get a career public service the confidence of running these process by themselves.

  • The platform we chose was the Join platform, which is at once, e-petition platform, the regulatory pre-announcement platform and the budgets discussion platform that I just showed the people about. You can fill it in your head.

  • (laughter)

  • Basically, you look at one piece of the budget, you don’t like how it’s being used. You can petition for it. After you collect 5,000 electronic signatures, the government is committed into authoring it, substantially, then hopefully leading to new regulations, which gets pre-announced on the same platform. You can then have a discussion also on that with public service.

  • The e-petition part, when I became the Digital Minister, it’s really pretty good for things that pertains to one single ministry or one single agency. It’s spectacularly bad if it is cross-ministerial. That is because no ministry want to answer for the other ministries.

  • If it concerns three different ministries, each ministry use a lot of words, dutiful reply to explain why this is respectively not their business, their business, and their business. It’s pretty bad. [laughs] The first thing we did after I became Digital Minister is to set up a team of what we call participation officers, or POs, in each and every ministry and have them form a virtual team.

  • It’s just basic game theory. If you’re going to interact with the same bunch of people for the next four years, then you better collaborate. Previously, it’s different people every time.

  • Now, when people do a petition in Taiwan, they know, even if it’s cross-ministerial, even when people petition for, for example, in South Taiwan there’s people petitioning for stationing helicopters to serve as ambulance cars because they are too far away from a major hospital.

  • It could be solved by depositing the helicopter, which would be the Minister of Interior, or by building faster roads, which would be Transportation, or building a large hospital there, which would be Health and Welfare, and/or doing more relocation of population. Actually, there’s National Defense there, as well, because there’s a defense air base there.

  • It could be solved in five different ways. What prevented these ideas from being fully explored was the previous leaders, no virtual network of such participation officers to fully all go to that town hall function and make sure that everyone...

  • (laughter)

  • ...to make sure that everyone is on the same page and also making sure that...

  • It’s almost down. If you can just open QuickTime, start our recording, and then we may actually back to these issues.

  • (laughter)

  • Please don’t do that.

  • (laughter)

  • It’s OK, just start mirroring and then start a new recording.

  • Not sure my QuickTime’s here.

  • (laughter)

  • We’ll first turn on the mirroring, and then we start QuickTime. Once QuickTime has started recording...

  • This gives me a moment. While they’re doing this, we’re getting close to the time in which it’s time for your questions instead of mine. We’re using the hashtag IPK, #IPK. Obviously, you should feel free to just raise your hand, but you can also send questions in via Twitter. We’ll try to take them by both modalities.

  • If you’re watching this live-streaming, by all means send #IPK. For people in the room, you can have your choice, have your cake and eat it, too. Hooray. Thank you.

  • Let me just take a couple minutes. These are the participation officers. Every month we meet, vote, and talk about questions that requires cross-ministerial support. Here are the petitioners. Just by petitioning, they automatically get invitation to such collaboration workshops which are, if they want, live streams, so that people can also participate over the Internet.

  • It’s not just one collaborative meeting. For example, the tax system case, there was four subsequent meetings, where people just collaboratively co-created the tax reporting system. In the Hengchun case I just talk about, it’s literally all the different ministries and all the different local stakeholders.

  • The reason why we can get all the different ministries was that, in the participation officer regulation, all these are national regulations. In the e-participation regulation we said that whenever there’s more than one owning agency, each considering each other owning it, and they only want to support it, everybody owns it. They all have to go to where the people are.

  • We went there, took five hours explore every single options using the Policy Lab methodology of mind-mapping. Basically explored all the different options before setting on the insight that we should actually retain people’s trust on their local hospitals, and so we really should build a larger hospital, after exploring all the different options.

  • Those are every other Friday. Next Monday, after each collaboration meeting, I bring the synthetic document into the premier meeting with the Premier and other ministers, and see whether the Premier is OK with it. Once the Premier is OK with it, of course, after a couple weeks, the hospital just [snaps fingers] gets a budget and it gets built.

  • It’s a very quick turnaround system from people surfacing there is a local or a national issue, to the participation officers crowding in, and to make all the different solutions possible understandable. Then the people getting a consensus, and the Premier actually gives blessing to the cases.

  • Using this new e-petition-driven method, we handle another 40 cases -- this is in parallel with the vTaiwan service -- about half of which has led to a new budget, a new policy, or so on. The other half, again, it’s not because of inaction. It’s because, after thoroughly discussing whether Taiwan should change its time zone to +9, people decided maybe it’s not the best idea after all, so it did not actually get changed to +9.

  • How do you ensure, though, that in any of the processes the information is brought to bear to make that decision in a rational way, based on data, based on information, and not based on irrational preferences?

  • It’s the career public service doing all this preparation. We’re just facilitating and providing the missing proficiencies. Even at that, the facilitative, the recording, the translational proficiencies, we make sure that it’s all transferred back into the public service. Of course, they would insist on that, because it’s them running the show. That’s the first thing.

  • The second is that we also leave room for reflections. In the Hengchun case with helicopters, we used two rooms. A smaller room, about 20 people, just like this, of people doing co-creation, discussion, and so on. A larger room that can fit hundreds of people, actually in a kind of town hall, is where we watch the livestream together.

  • The digital ministry is in the town hall with the people, watching the livestream in the smaller room, and serving as a kind of ESPN anchor to explain what this slide means, what this move means, whatever, and making sure that people who want to vent, who want to protest, they can just come to me.

  • The media is in the town hall, obviously, while the professionally, neutrally, facilitated, high-performance meeting is happening concurrently. People won’t protest for very long, because everybody want to watch the movie, also because it’s not live streamed back. [laughs] All the protests, all the shouting, doesn’t actually affect the deliberation that’s happening in the smaller room.

  • On the other hand, any new point they make that’s in relation with the mind map is tagged and brought back through Slido, through other digital tools, so that people there can also see the outside people’s contributions, but always within the context of the mind map, that they’re doing the mapping.

  • What’s the experiment you next want to try? What haven’t you done yet? Is it scaling what you’ve done, trying something new, or both?

  • There’s a mayoral election coming soon. We’re already getting some interest and commitments, from both existing mayors and mayoral candidates, of basically taking this system, which is national regulation, and introduce it at the municipal level, so that they would want to try this, not just for participatory budgeting, of which there’s many, but also policymaking, at city level.

  • We’re also seeing a lot of interest because Taiwan is now doing a lot of experiment on what we call sandboxes, which is this idea of people experimenting with breaking the law for a year or so. I think this is worth sharing. This is essentially us encouraging people who want to do platform economy.

  • Our final solution is just, for people who want to break the law -- to do platform economy or fintech, or later this year UVs, and the UVs could be hybrids -- to break the law for a year and work with any willing municipalities. They can break the law for a year, basically running with a forked version of the law which the government had set up in place.

  • After running it for a year, we use this kind of consensus-gathering mechanism to make sure that the society really thinks this fork is a good idea. If it’s not a good idea, we thank the investor for paying the tuition for everyone...

  • (laughter)

  • ...and reduce the risk for everybody else afterward. Otherwise, it may be extended to a larger scope, including a business model for another year. If it’s a regulatory change, it would [snaps fingers] just happen if people deemed this a good idea.

  • If it requires a law change, then it’s up to four years when the MPs deliberate on that, in which the people applying for a sandbox just essentially gets a monopoly in municipality, in this kind of business experiments.

  • A lot of the steps that we’re now taking is basically empowering the municipalities to be able to run this kind of process when they deal with a sandbox application and evaluate its applicability to the local social good and social needs.

  • I’ll do one more for me, and then I’ll save the rest of mine until we get a chance for everyone else to come in. I have to ask, because everybody always asks me about the risks.

  • With the introduction of and the moving of so much of democratic life in your vision online, how do we square that with the dangers of, people always ask about self-surveillance, of privacy, the risk as we look at what China’s doing with regard to social credit scoring and potentially rating people, and using that score to determine whether you’re allowed to have a political voice.

  • What do you perceive to be the greatest risks in this process, or is it all upside?

  • We’re all very wary of vendor lock-in. When we first used pol.is for the UberX discussion, there was a lot of flak from the civic tech community, because it was proprietary. Even though that the folks running pol.is said that they would share all the data, we have no way to know whether they were actually sharing all the data. They’re really just a Seattle startup.

  • Basically, we don’t know what is the data policy like. We took the bet, but we also peer-pressure a lot for them to go Affero GPL, which is one of the most libre version of open-source licenses.

  • Also making sure that we can host it locally and basically have everybody can be part of the governance system that makes sure that this mechanism of democracy is itself democratic, in the sense that its code, its data, its operation is community-owned.

  • After I became the Digital Minister, of course, I took this philosophy into my office. Basically, we built this sandstorm.io installation, which is, again, another startup. But it’s all open-source, and we get our separate security people to audit it.

  • What this does is basically it sandboxes the apps in a computer science sense. It boxes the apps that runs on top of it so that they don’t have to worry about cybersecurity authentication, authorization, and whatnot. We can use all open-source technologies for collaborative decision-making, just note taking or whatever on top of this platform.

  • Our career public service can even just learn some JavaScript and write a small application for ordering lunch boxes together or planning trips together -- that actually happened -- and running on this platform without worrying about cybersecurity, about infiltration, about surveillance, and whatever.

  • The code itself, being audited and open-source, is a collective resource managed by the entire worldwide civic tech community, of which Taiwan is one installation of many.

  • I won’t monopolize any longer. You do have an initial question coming on Twitter but I’m going to look for a show of hands in the room. I’ll ask you, if you don’t mind, to also pull up the hashtag because my battery’s about to die. I met you let self-moderate the Twitter questions, but do we have an early volunteer? Please.

  • Thank you so much for your input. I got to ask in the age of fake news, what’s the role of journalism in...

  • ...in your project, and moving forward? What has been the role of journalism and how do you envision the role of journalism moving forward?

  • I don’t use the f-word myself. I prefer to call them disinformation when they’re intentional and misinformation when they’re not. That way it’s not an affront to journalists -- both of my parents were journalists.

  • Anyway. I found that radical transparency is really empowering investigative journalism and just quality journalism in general. A lot of journalists’ work is just to get the scoop from the ministers. When a minister basically publish everything, agree to no exclusive interviews, everybody is on the same ground.

  • When a lobbyist come, like Mr. Plouffe here, talking for Uber at the time, we make sure it’s not just on transcript record but actually on 360 recording, so any journalist can put on VR goggles and relive the moment of the negotiation.

  • I think this is very powerful in the sense that once everybody gets some facts, the journalist with perspective, with time for investigation, with their own life experience to bring to the table can produce much more powerful pieces, but still within the time frame that still attracts popular attention.

  • Without these raw materials’ being made available, their life is much harder, because they have to compete with short-term attention and also try to get some sort of information out there, but it is much higher chance because they’re racing, essentially, with the tabloidic journalism to produce something. Then it will contain more misinformation in it, which will reduce the quality of their reporting.

  • Yeah, the investigative journalists, I think, are really our friend. In PDIS they are also my colleagues who during the occupy ran the e-forum, which is the neutral journalism outlet during the occupy. It’s very much on our mind that what we’re doing is basically investigative work for journalism, but in a minister’s post within the cabinet.

  • This kind of reliable information is also when Taiwan is...There’s a lot of disinformation campaigns, especially now we’re this close to election. The administration’s committing to replying to all the spreading of disinformations in a timely, open, structured fashion so that every ministry can reply within three hours or four hours.

  • Whenever there’s a disinformation campaign, there’s a clarification from the ministry. We’re not censoring speech at all. What we’re doing is ensuring that people get this habit of waiting a couple hours, and then see whether a clarification came out from the administration. That’s all we’re doing -- basically defining a social norm around rapid response and reasonable disclosure from the ministries.

  • More questions. In the back. I see a hand. Oh. Andrew. [laughs]

  • Thank you. My question is about, you mentioned empathy earlier and how this approach doesn’t make a lot of sense for issues that are longstanding, deeply held lack of information on different sides.

  • My question is whether you have explored any other approaches for establishing that kind of empathy between different parties, or if you think the problems are so complex and long-lasting that they frankly just should fall outside of the remit of a digital ministry.

  • There is the Holopolis project, which is now taking place I think in Madrid. They are setting up a lab there. It started in Taiwan and it was initially in my research proposal. I was just working on it when they told me that they want me to be the Digital Minister.

  • In any case, the idea, very simply put, is to use augmented and mixed and virtual reality to make sure people have some lived-in experience of one another before entering a discussion if it’s too far in the past.

  • For things too far in the future, for example, when building a hypothetical, I don’t know, airport or whatever, at least we can take those blueprints from the architects and situate people in future versions of that airport and actually feel how it is like to be maybe a non-human, because for endangered species we can also look at their lived-in experience as well.

  • I did a VR conversation with a bunch of schoolchildren who all very much appreciate that I scaled down my avatar to be the same height as them. They could be at eye level.

  • (laughter)

  • There’s a lot of design that went into this very carefully to make sure that people can be literally in each other’s shoes or avatars, and also that there is meaningful use of chat bots that can go back and forth on this lived-in experience to simulate a kind of conversation you would have if, for example, the spirits lived there, the spirit of a river, the spirit of the endangered species.

  • It’s a little bit animist, even, but we’ve found that it is actually very effective in getting people into the mindset of listening to social objects that cannot speak for themselves. Future generations, things like that.

  • There is a lot of active research, but mostly it’s now done by my other colleagues who are actually interaction designers in PDIS and not me personally, so I’m just talking about their work. If you’re interested, join the Holopolis project.

  • Hi. Thanks. I was wondering, you’re describing how in the case of Taiwan, Occupy Movement -- it’s part of a networked social movement. There was networked social movements all across the world in the last six years. Some have been more ambitious or "successful." I challenge how we read or understand success, because I think of certain kinds of social movements.

  • One question I have is, you didn’t have the opportunity necessarily to embed these civic deliberation methodologies into a political campaign, because it sounds like the ministries were scared and opened up space that, traditionally, social movements would need to amass space and go run for elections, kind of like they did, the Municipalists did in Spain three years ago.

  • I’m just wondering, have you given any thought to how these processes might look in that context? I’m interested to hear any thoughts you have.

  • I think Taiwan’s political system is different though. People elect directly the president, who appoints the premier and appoints the cabinet. That is separate from the legislative function. Actually, a majority of bills passed by the legislation starts as the draft from the administrative function.

  • Because of this system, the administration is remarkably party-free. There’s more independent ministers at the moment in the cabinet than members of any party in the cabinet. We can’t say that of the legislative function at all.

  • What we’re saying is that basically this system protects, creates kind of a buffer zone, for this kind of conversation directly with the population, without threatening the legislative. Because if it requires a law change, eventually the MPs will have a say, party politics will play a role.

  • Just at the feeling-finding, or the fact-finding and consensus-finding stage, nobody objects for the career public service to do a little bit more to prepare the MPs better for what people feel. I think this, the administration doing some preparatory work for the eventual referendum or the eventual MPs, this is something that everybody can get behind.

  • Because of that, it is already a very good idea. We don’t need to campaign especially hard for that. Rather, people who held some reservations, as I said, lost their mayoral elections anyway in 2014. Within our new political landscape, we don’t even need to deal with the people who refuse this kind of thing, because four years ago they were just all gone.

  • More hands. I’m going to ask you a question off of Twitter while people are thinking. From Tamas in somewhere, writes...

  • Oh, you’re here. Ask your question.

  • No, no, please. My phone just died, so now you have to ask the question. You don’t look anything like you do on Twitter.

  • I had to ask the question on Twitter because I Googled join Taiwan, and it was very funny to get the first hit. If you want to see that it’s all on Twitter.

  • It was about the question that Google has as the first hit is "When does Taiwan join China?" something like that. I had to show that. The question that I have is about the risk of public servants. I like how you reframed that.

  • As a public servant myself, I see lots of colleagues who are very afraid to join any participative process. I was wondering, can these people in your radical transparent world, can public servants join anonymously in an online world?

  • Yes. Of course. They can choose any nickname they want and basically anonymize themself. The transcript is published after only 10 working days of collaborative editing. People who don’t want some words to be taken out of context, people who want to add in more supplementary material, people who after reading the transcript actually change their positions, they can all reflect that in the final published version.

  • The upshot of this is that the public sees career public service as something very professional after reading all these transcripts. My work in channeling back the Reddit equivalents is also make the Internet people participation sound very professional, because I remove all the exclamation marks and cat pictures

  • (laughter)

  • ...and actually only delivered the substance back to the meetings. I think it really builds mutual trust after you experience this for a time or two. It really aligns what career public service is like, because it really reinforces the message of what the values that our career public servants hold without exposing them to personal risk.

  • If everything turns out very well, we can always re-identify yourself and take the credit for it.

  • Other comments, questions. Please.

  • Hi. My name’s Nathan Storey.

  • Audrey, it’s nice to see you again. I just want to note that a few months ago we were in this same room for the Fearless Cities conference, and it was this exact room where you were appearing remotely.

  • My question is about the role that you see Taiwan playing in these emerging networks of cities, local governments and movements that are forming networks with each other outside of the nation-state level, and why you were doing that.

  • Why you were on a panel, Fearless Cities, with counterparts from Madrid and Wikipolítica in Mexico, an opposition party, and why you were traveling to London and here. What are you trying to accomplish with all of this travel?

  • It used to be that it takes paragraphs to explain, but now with the excellent technology that is Sustainable Development Goals, all it takes is a few numbers. I’m working on 17.18, 17.17, and the 17.6 of the Sustainable Development Goals.

  • We found this to be really effective, because digital social innovation in the context of the Sustainable Goals is really the glue that holds people caring about society, caring about environment, caring about education, about equality, about whatever, together in a way that is to the benefit of everybody instead of everybody working on different directions and canceling each other off.

  • The methodology in the global goals is just to enhance availability of reliable data, of getting people on the same page, literally using distributed ledgers or whatever as needed to make sure that people trust the evidences that their action is having, impacting, on the respective domains.

  • Encourage effective partnerships by basically building a common vocabulary of, say, #crowdlaw, of a catalog, a system that people can compare what they are doing within their very different narratives but with some metrics that people can amplify their work and compare their work against each other.

  • Finally, to share innovations in technology, so that when we solve this for Taiwan, we know the limitations, but if you’re operating on a different municipality, maybe that limitations don’t apply to you and you can extend our vision and do better. We were happy to learn.

  • This kind of open innovation, I think, is the key to getting some sort of solidarity across the different municipalities who are all working on this very same legitimacy problem, but with very different cultural norms. I think Taiwan’s role as one of the places where really there is no other choice but innovate without leaving anyone behind. Because otherwise the people who get sacrificed just occupy.

  • It is essential [laughs] for us to document and to share our findings, but also work as one of the reliable partners to hold such evidences and such data and such studies, and participate in the global network.

  • You have two, one here and one here.

  • You came out in the room and talked that you’re an anarchist. When I think of anarchy, I think of a lot of passion and protest. We see people here protesting on the streets and at the end of the day, nothing gets down.

  • What you’re doing in Taiwan, I see a lot of organization there. It takes a long time. How do you maintain that momentum of passion that you always have for what you’re doing and with everyone else who’s involved, how do you maintain that through this process?

  • Thank you. I’ve wrote entire treaties about this [laughs] back when I was working in the free software community. It’s called -Ofun: Optimize for Fun. There’s a whole bunch of methodologies to basically celebrate small successes, make things fun and make the fun contagious.

  • I think that the origin of the anarchist thinking for me, because I learned anarchism first from the classic texts but also from the fables and stories of Laozi and Zhuangzi in the old Daoist tradition. They were very much against hierarchical power as well, but they explained their philosophy in a way that is fun, that appeals to even a five-year-old when I first read it, Daoist texts. It’s basically very intuitively appealing.

  • When you maintain this intuitive appeal without actually throwing bombs, things like that, and destroy people who don’t agree with the anarchist agenda, you’ll get what I call conservative anarchism, meaning that a anarchist tradition, they’ll respect the traditions but don’t reinforce them, who work alongside hierarchical power and shifting them into peer-to-peer power but without usurping it within the old power logic.

  • I think Buckminster Fuller captured it best. When you don’t fix a broken system, you make one, a new one that makes the old obsolete. Then that’s exactly what we’re doing, which is a lot of fun.

  • (laughter)

  • We had a question over here.

  • Hi, I currently study the urban grammatics, so the public health and all the healthcare as well is my interest in directions. For example, I think if people go to the hospital, it’s always expensive and time-consuming.

  • I think that is a problem that people voice concerns to. I wonder, in your opinion, how your ecosystem could address it or help to improve it in and/or does your ecosystem of e-participation already accumulate some solutions about it?

  • There’s a lot of people petitioning and working toward digital health. Taiwan wasn’t that big on telehealth before, but because of the vTaiwan methodology, we found out a lot of people who want to do e-health not because they’re too far away from a clinic but rather, they want a continuous ongoing relationship with their clinicians and so on.

  • It services real needs and cover that with the aging issue in Taiwan. There’s a rapidly declining birth rate.

  • (laughter)

  • We work with elders for whom it is much more difficult to actually go to a clinician. All that resulted in the Additional Health Initiative and Telemedicine Bill, and whatever of this year. This is one of the most popular, actually, topics in both vTaiwan and also the Join platform. The Join platform gets a lot of people passionate about having the clinicians, the large hospitals’ doctors.

  • As well as if they’re in off-shore islands, then the helicopters, the people who actually does the showing on the same page using open standards, using the cutting-edge technologies such as voice assistance to make sure that elders receive the medical care in the accent, in a culture that they understand, in the words that they understand, using metaphors that they understand.

  • And can also capture their non-verbal response to such cues and such dialogues in a way that feels comfortable to them instead of asking them to speak perfect Mandarin or whatever to get into the system. That is one very active research agenda.

  • I’m very happy that vTaiwan played a very small role in opening up the public imagination into the inevitableness but also the urgency of working out additional health.

  • We have two hands over here and in the back. [laughs]

  • Hello, my name is Marco, I’m from Brazil. I read the Chris Horton’s article "MIT Technology Review." I don’t know if you have read it. In general, the article says that you have a very ingenious system for crowdsourced output.

  • Then, they said that there is a statement that the platform has its limits. It needs the real power. I would like to know what we are doing to get the real power? It means what the people institutional reforms you are thinking or trying to propose to, especially keep these changes continuing and doing next administrations in Taiwan?

  • I think the consensus of the vTaiwan community, which I certainly cannot represent, but I can re-present them, is to scale out, to scale up, and also, to scale deep. That’s the three different directions that the various vTaiwan community actors are taking.

  • Scaling out, I already mentioned, meaning the municipalities, city and even smaller communities need to be made comfortable in running this process by themselves without waiting the national government to do it for them, which is why the sandbox experiments and so on is so important.

  • It gives something that’s pertinent to that particular place to deliberate on without waiting for the central administration to do it for them. That’s scaling out.

  • On scaling up, there’s many people who think that vTaiwan should be coupled with the referendum process or some other process that gives final binding power a lot more than what we already have at the moment, which is just really consultative power.

  • At the end of this year, there’s a national, what we call, Digital Communication Act. This is for all the things pertaining to digital communication, to Internet governance, to multi-stakeholder governance around transnational issues.

  • Something like the vTaiwan method must be used, and that’s the first time that this method, what we call open multi-stakeholder consultation process, is written into the law itself. That will give the scaling up the binding power that it needs on a national area, but we are still waiting for the legislators’ green light. It will probably happen in a couple months. We’ll see.

  • Finally, scaling deeply, scaling this idea into the K-12 curriculum system. It’s for the students to co-create curriculum with the teachers, which is just happening now.

  • The new curriculum goes online next September, where we will redesign the capstone and classes to make sure that K-12, junior high, senior high, and college level solve social environmental problems collectively as part of their learning, instead of just waiting until they’re adults, and then participating in this process.

  • Personally, I’m a junior high school dropout. My first foray into democracy is in the Internet Society. That was when I was 14. I just imbued myself into this raw consensus process for six years before I even get my first voting right in representative democracy.

  • I think that kind of formative experience is really important, and we really need to scale deeply into the minds of the junior high school students, who will lead the future direction of the Earth, anyway.

  • I think we had two questions over here, is that right? Should we gather two questions?

  • Then we’ll begin to wrap up. Do you still have a question? No? Kai, was that a hand for you? Did you change your mind?

  • No, but I can ask one.

  • I saw a hand before. Maybe it was a nose scratch. My apology. We have one all the way in the back that’s next.

  • I’m going to focus a little more practically, I guess. My name’s Kai Feder. I work with Beth in New Jersey capacity as of two and a half weeks ago. [laughs]

  • When you’re looking at, with civil servants in particular that are used to very rigid structures, and you’re engaging them in this very different approach to governance, innovation, etc., are there specific types of characteristics and skills that you think really lead to more successful outcomes?

  • That either A, attract the specific civil servant to get involved into something like this, and also, help projects succeed internally, and be able to navigate within their respect agency or role.

  • Do you want to take that, or do you want to take two questions?

  • Sure. Maybe let’s take two questions?

  • [laughs] From the shadows.

  • That question fits, I think, well with mine, which is what do you think of if a governor of a state said, "I love this process. I think we should implement it," and pulled his CIO, "OK, make this happen." A, is it top-down implementation of this type of...?

  • Has the g0v community thought about how to support such an implementation? Is there something antithetical about applying it from an executive level on down without the community self-organizing this? Also, and more concretely, how many apps?

  • You’re using Sandstorm. You’re using all these softwares. What is the minimum viable resource investment necessary to put that infrastructure together to be able to implement this in any type of way that would be respected in the public arena?

  • Like how to boot strap, huh? I think the minimum boot loader is very simply a physical place. The architecture of the physical place, of the public additional innovation space, I think, determines the kind of people who want to go to it and who want to stay and who want to basically make collective decisions and take risks together.

  • I say this because this place was literally co-created by hundreds of social innovators. They asked for a kitchen, and they got a kitchen, a chef. It opens until 11:00 PM every night and so on. Basically, people feel that this place is where they are, where they belong, and where they is willing to come back even after spectacular failures in experimentation.

  • This whole notion of going into a space and see some self-driving tricycles roaming around [laughs] is seen as kind of a social norm. Every time you go into Social Innovation Lab, you will find new ideas and new experiments running around.

  • For me, if there’s one single thing is this recurring, reflective, recursive space that allows for a culture of people authentically sharing their experience and failures, dreams, and whatever, and then still willing to come back to the place. This is also the place we hold all the teleconferences with me touring Taiwan and using video conferencing and so on.

  • The 12 different ministries people, when they go to this place, it feels like play, even if it’s actually work. It feels like play, because they get to see new experiments, new social innovations along the way. They understand that they’re not under any sort of risk of attack, of protest, or whatever.

  • People cannot attack them over the monitor anyway, so everybody acts very civilized and so on. Gradually, it lowers the fears, uncertainty, and doubt of public service when it comes to public engagement. That is the only thing that’s missing.

  • The career public service’s perfectly capable, generalists and specialists to deal with these kinds of issues. It was just the silos that prevents their empathy from showing forward and things like that.

  • Just creating this recursive space and culture I think is the boot loader that you just talk about and also what I would recommend when bootstrapping from a new municipality and whatever. We hold training classes, actually, this June in NYC, and also very soon, I think November in Canada, and in many other places.

  • We’re building a English curriculum of the curriculum that we’re offering to the municipalities in Taiwan.

  • Did you get Kai’s question?

  • Just about the general personality and/or technical skills.

  • Sorry, general personality?

  • Personality or technical skills as well, though, that really make civil servants that are engaging on these projects successful, especially given the fact that you’re introducing them to a very new, novel environment and...

  • I didn’t get much time to talk about PDIS, the Public Digital Innovation Space, which is kind of a reincarnation of the g0v culture within the central administration. PDIS is at the moment I think 22 full-timers, about 40 or so interns. It’s like a small internal startup within the central government. It’s like Policy Lab or whatever and everywhere.

  • What sets PDIS apart, though, is that I talk as part of my compact when I joined the cabinet that I can poach at most one person from each ministry to work full-time in the Public Digital Innovation Space. They’re not like the POs. The POs, they grow. They get new POs in the third-level, fourth-level agencies. The PO network just grows. It’s now more than 60 people now.

  • PDIS remains small, because it’s at most one person from each ministry, which means a maximum capacity of 32 people, because there’s 32 ministries in Taiwan, you see. Because of that, it is by definition cross-functional. It is by definition people who care about various many different things.

  • What’s a constant here is that everything joins by voluntary association. I don’t even give them command. I don’t even rate them or score them. Everybody writes their own job description, writes their own scorecard. If they want to do something they have to pitch to the rest of the team, who all came from different ministries anyway, so you better find some common values.

  • It means that first, all the mechanism that came out of PDIS doesn’t sacrifice any other ministry. It is not a place to do ministerial politics, because every other people are there too and we work out loud using the traditional open-source free software mechanism.

  • The second thing is that when we do the recruiting, we find that people who are willing to join naturally are more of a giver. They wish to contribute more than what they can take to the public good. This is not some HR criteria.

  • It’s just in a anarchist workspace you have to be this kind of people in order to have fun. Otherwise it’s just not fun at all, because if you are after petty politics or whatever, you do don’t get much satisfaction from an anarchist minister.

  • (laughter)

  • That’s the second thing, is the fulfillment of giving and contributing. The third thing, I think, really is just the PDIS being a kind of risk-free space for you to do the experiments, because if anything goes wrong, it’s always Audrey’s fault.

  • (laughter)

  • I think that’s what empowers the career public service. We have our PDIS member from the ministry of foreign affairs [laughs] and there’s many other ministries’ people currently in Taiwan.

  • Do we have any final questions? OK, you’ve earned it, cameraman.

  • (laughter)

  • Thank you. I have a small two-part question. The first is related to the minimum viable resource question, which is, is technology crucial to decentralized consensus-making, to what you do?

  • The second part is, if it is, then have there been attempts to game the system, because anything that’s online, whether it’s Facebook or whether WhatsApp, there’s always people who find a way to game the system and have there been instances of people trying to game vTaiwan and how did you work on that?

  • I think we usually say civic tech, but I mostly think in a framework of calm technology, in the sense they’re technologies that allows people to focus more on each other, rather than distract people’s attention from each other.

  • Calm tech, or assistive tech, or ambient tech, however you want to call it, of course doesn’t have to be digital. It could be post-it notes. It could be white-boarding or whatever. It could be sign language used during the occupy. It could be people’s microphone.

  • There’s all sort of technologies that you can deploy without it being a digital technology. We use digital technology mostly because it allows this experience to scale horizontally and for the ideas, thoughts and reflections reached in a face-to-face setting to ripple out without dying down, because there’s just not too many people joining the protest on the street.

  • It is more in a conservation part of the consensus-making process of it than the amplification part of it. We’re not too big on the amplification part of it. Even though, having said that, the Join platform is now 5 million users out of 23 million population, which is not too bad.

  • The gaming part. Back when we did Airbnb case, which is right after UberX, Airbnb sent a email to all its Taiwan members asking them to come to pol.is and support the Airbnb position. What they found out was that had this been a simple yes/no question or a simple questionnaire, maybe people would have behaved as the email told them to do.

  • But because this is a open-ended, reflective space, only one-third of people they recruited this way actually agree with the Airbnb position.

  • (laughter)

  • Many other people have much more nuanced, much more eclectic, much more resonating feelings, because they’re just motivated to press one like, but what it gets from opening the link is actually a larger crowd, a larger system, a more holistic approach, a overview effect, if you will, on the problem space.

  • In that reflective space, people behave very differently than people mobilized just to press one like on things. Of course, we see people trying to use bots and things like that, but it’s not really a big problem.

  • If you write a bot that does exactly the same for 5,000 different entries, it’s just one dot in the principal component app, because we don’t even look at the numbers. What we’re looking at is the diversity, is whether you can propose something that resonate with more people.

  • If you can write a bot that generates a sentiment that resonates with more people, I for one welcome.

  • (laughter)

  • But so far that has not happened.

  • (laughter)

  • The hour is getting late, so let me close this out in the following way. This past Saturday was International Day of Democracy.

  • International Day of Democracy is a UN-created holiday, for those of you who don’t know, which really tries to zero in and focus on one specific aspect of the Universal Declaration of Human Rights that celebrates its 70th anniversary this year. That is namely the idea that democracy, in that 70-year-old version, is the conduct of periodic and genuine elections.

  • For those of us on this side of the table who think and feel that voting once a year is not enough, and that we can do better. That we can do better in the way that you are trying to do in Taiwan and I think have shown us may be possible, I wanted to ask and point to all of you that on your chair, you will find a manifesto on what Audrey referred to a number of times in passing, the idea of #crowdlaw.

  • The idea of crowd-sourcing plus lawmaking, the idea of all of us can play a bigger and better role in engaging in how our governments make law and policy. If you like this idea of doing more things like vTaiwan, PDIS, the participation officers and the digital lab thing, which was digital...

  • Social Innovation Lab.

  • ...Social Innovation Lab, and this whole ecosystem of initiatives that really is a thicker, more active vision of democracy, I would just invite you please to sign the manifesto and leave it behind on your chair and we will collect it.

  • You can also go online to manifesto.crowd.law. Audrey has signed it. It’s really a call to all of us, to our city councils, parliaments and legislatures, to our technologists, and to each of us to play more of a role in the way that you have shown us.

  • I want to just ask you as a final question, then, to give us the big vision. We started with the worried fear about democracy. The future of democracy for you. Are you optimistic, are you pessimistic? How do we realize this big vision? Close us out, then we will drink wine and cheese, and pepper you with more questions. Last word.

  • [laughs] Back when I took the post of Digital Minister, as I said I had a compact, not a contract, but they still want a job description. Instead of a job description, I just wrote the administration a poem or a prayer, [laughs] which would serve as the job description as the Digital Minister. I think that answers the question, which is why I want to read the poem to you now.

  • "When we see internet of things, let’s make it internet of beings.

    "When we see virtual reality, let’s make it 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.

    "Whenever we hear that the singularity is near, let us always remember: the plurality is here."

  • (applause)

  • Thank you to all of you for coming. Geoff Mulgan, the only person I know who’s just as smart as Audrey, and who is also a Buddhist monk, by the way, in case you didn’t know that, and advised three prime ministers in the UK, is coming to talk about his book "Big Mind" and collective intelligence and democracy, 26th, next Wednesday, over lunch.

  • A little bit shorter, but I think we feed you a little bit more. With that said, please come. Mingle, talk. Cheese cubes await. Thank you, and thank you, Audrey, so much.

  • (applause)

  • It was great. Thank you.