• Let’s get started. Hello and welcome to all of you, wherever you are, in Europe or Asia, in your office or at home. We’re glad you tuned in. I welcome you on behalf of the Open Society Foundation, King Baudouin Foundation, European Movement International, and Bertelsmann Stiftung. My name is Dominik Hierlemann.

  • I will be moderating today’s event, Digital Democracy – What Europe Can Learn From Taiwan. For this topic, we’ve got someone with us who I could easily introduce for the next hour just by quoting from international newspapers and magazines, restating all her accomplishments, or I could just say, which I will do, welcome, Minister Audrey Tang. Great to have you with us.

  • Hello, good localtime, everyone.

  • Hello. Audrey Tang is all in on transparency. That’s why we run this event as a open video call where you all can see and look at each other. This WebEx call is limited to 200 people, so we’re also live-streaming our event on YouTube. Welcome to all of you.

  • Our discussion will be divided into three parts. First, we have a one-to-one conversation with Audrey Tang. We’ll then bring in four more panelists. Eva Maydell, Member of the European Parliament and President of the European Movement International. Eva, are you with us?

  • Good morning. I’m with you. Again, the video will not be enabled, so I’m afraid you can’t see me.

  • But we can hear you perfectly well. Welcome, Eva. Welcome to Anna Piperal from Estonia.

  • Hello, everybody. I’m really excited to be here.

  • Yves Mathieu from France. Hello, Yves.

  • Bonjour. Also, bonjour to Graham Smith, who is in London this morning and who got up very early.

  • It’s not that early. Welcome, everybody. Hello.

  • [laughs] Hello. Hello, Graham. Later, you will have a chance to comment and to ask your questions in our chat. The chat is hosted by my colleague Stefan Roch. Stefan is with us too. I’d like to express, before we start, our gratitude to Professor Shieh, Taiwan’s representative to Germany, in helping to set up our today’s discussion and for being present today. Hello, Professor.

  • With us today is also Brigitte Mohn, member of the board of Bertelsmann Stiftung. What is our conversation all about? It is about Taiwan. We want to understand its model of digital participation but also its concept of democracy. It is about new forms and techniques of citizen participation. Together, we’d like to reflect on what Europe and especially the European Union can learn from it.

  • Minister Tang, in my preparation for our conversation today, I did a bit of reading. Somewhere, it was mentioned that you like to end your presentations by quoting from the songwriter Leonard Cohen. I think that’s a great habit, but today we’re doing it the other way around. I’d like to kick off our discussion with a quote from Cohen.

  • The well-known quote is “Your faith was strong, but you needed proof.” Of course, my faith in participation and digital participation is strong, but we can need some proof. Can you tell us what is it exactly that you do that made Taiwan a leader in digital citizens’ participation? Where’s the proof?

  • Certainly. That’s a really good quote, by the way. I will go like this, the fourth, the fifth, the minor fall, the major lift, and [laughs] the baffled king composing Hallelujah. In Taiwan, we are collectively composing Hallelujah in the time of the dual threats of the pandemic and also the infodemic. The proof is very clear in that we managed to fight off entirely the pandemic.

  • We have very large concerts, live performance of tens of thousands of people, for months now – it’s not just a recent thing – and with no lockdown. Our GDP is actually recording a growth, especially in the day-to-day street markets, convenience stores, and so on, for the last month. Obviously, we’re somewhat post-pandemic now. There was no lockdowns.

  • Also, as considered by experts – we have Estonian representative here, so I’m saying it’s not my personal consideration – one of the most cyber-attacked [laughs] places in the world, including disinformation, but we managed to fight off disinformation in our previous election, which takes in January, around the same time as the news of the SARS 2.0.

  • We managed to have a fair election that is largely free of disinformation, unaffected by the information warfare that has plagued us just as short as two years ago. We retain the CIVICUS Monitor human rights associations rating of the only country in Asia, and one of the only two if you count Asia Pacific, that will be New Zealand, to be completely open in terms of freedom of press, speech, and so on.

  • Counter-pandemic was no lockdown. Counter-infodemic was no take-down. These are proof enough. It’s not quite hallelujah yet, because we don’t have a vaccine yet. We have something that is second best, which is a physical vaccine that has now reached more than 95 percent of the Taiwanese population. The other five percent perhaps already have plenty of these at home.

  • They don’t bother getting them from the government.

  • In one of your tweets I read, how did democratic Taiwan combat coronavirus, I’ve got it on my mobile here. Soap, hand sanitizer, civic tech, and plenty of trust. Can you tell us a little bit more about civic tech and trust? How do you bring in digital participation in order to tackle the pandemic, and also the infodemic as you’re saying?

  • Certainly. The three pillars of digital and social innovation in Taiwan is known as fast, fair, fun. Very easy to remember. The fast part is the collective intelligence system that is run by this not-for-profit, actually run by students, place called the PTT, the Taiwanese equivalent of Reddit. That is the first organization to sign on this very effective counter-disinformation plan.

  • Last December, when Dr. Li Wenliang posted on their social media that there’s seven new SARS cases in Wuhan, it not only immediately get reposted to PTT, but up-voted enough so our medical officers started health inspections for incoming flights from Wuhan the very next day.

  • That’s the first of January, 10 days before pretty much everybody else in the world, including in Wuhan because at that time, Dr. Li Wenliang would get “harmonized.” While he did save a lot of Taiwanese people, we, of course, wish he could save more people.

  • Anyway, Dr. Li Wenliang’s message, because of the absolute freedom of speech and collective intelligence, basically made sure that we have this daily life press conference of the Central Epidemic Command Center that is set up even before we have the first locally confirmed case.

  • Digitally, anyone can just call this toll free number, 1922, or subscribe to any of our chat bots and get timely updates on the latest science, Ask the Scientists were a robotic avatar of the scientists. Also, if there’s new ideas from the society, it escalates very quickly into the CECC daily press conference.

  • Like back in April, there was a young boy who caught saying, “I don’t want to go to school. You’re rationing mask. I get this pink medical mask and people may laugh at me,” and things like that.

  • The very next day, everybody in the CECC press conference, including our Minister for Health, Chen Shih-chung, wore pink medical mask to show not only solidarity, but also make the boy the most hip boy in his class because only he had the color of the mask that medical officers, the heroes, wear. That’s radically trusting the citizens. The fair part is, for example, on distributing of the masks.

  • In the beginning when we’re ramping up the production from 2 million a day to 20 million a day with a country of 23 million people, we worked with the pharmacists across the island, more than 6,000 of them. Not only they have professional credentials, but were trusted in their neighborhoods.

  • That’s not the whole story. The civic technologists volunteer to create this very simple mask map that can show people where in Taiwan do you still have masks and where not. You see the real-time tally after each procurement.

  • For example, when I go to the pharmacy, I show up my National Health Insurance card, which covers more than 99.99 percent of not just citizens, but also residents, I can see in real-time that this simple triangle that represents the pharmacy that I go to actually have their mask storage decreased by nine.

  • The people queuing after me will be able to confirm that this system is indeed working as intended, and then if anything goes wrong, like if I buy nine masks but the people queuing after me refresh the map and actually see it actually grows, they will call 1922 right there. This is how we make sure that people feel calm and collected.

  • I think what becomes pretty clear is that you’ve got the digital tools in place. You’ve got the culture where citizens are able to deal with these tools, but you’ve got also a culture where politicians can deal with these tools and respond accordingly. Let’s talk a little bit about the nitty-gritty, about the tools that you’ve got in place.

  • Can you tell us a little bit more about the platform vTaiwan, about Join, about Pol.is, and how these all developed?

  • Yeah, certainly. But before I talk about vTaiwan and Join, I need to talk about Presidential Hackathon, which is really where the social sector teams meet their counterparts in the ministerial teams and build this common, what we call data collaboratives, which in GDPR terms would be joint controllership.

  • People in different sectors co-producing and co-using the data so that people, for example, can detect earthquakes, floods, predict air quality, and things like that, in a very collaborative environment, so that anyone really, in their primary school and so on, can set up air boxes, very inexpensive – less than US$100 each or €100 each – stations that contributes to our environmental understanding.

  • The Presidential Hackathon is a enabling annual thing where the president, Dr. Tsai Ing-wen, gives an award to five of the winning teams out of 200 or so. Each need to correspond to one of the sustainable goals. The five teams, the trophy is a micro-projector.

  • If you turn it on, it projects the president handing you the trophy – the trophy describes itself – promising whatever you did in the past three months will become national policy in the next 12 months. We do all the regulatory, personnel, budget, and so on as needed. Of course, such Presidential Hackathon teams doesn’t come out of nowhere.

  • That is where vTaiwan and other g0v communities come in. For example, when Uber first came to Taiwan, they work only with professionally licensed drivers, but very quickly, they introduced UberX, which X, I think, stands for no professional driver license, maybe. [laughs]

  • Then it creates a tension in Taiwan. We used a technology called Pol.is, which is artificial intelligence tool, or as I prefer to call it, an assistive intelligence tool, that clusters people according to their feelings about UberX. The feelings are there’s no right or wrong about feelings.

  • For three weeks, people just resonate with one another, and you can see one fellow citizen saying maybe that insurance for private vehicles is important for the passengers. If you agree with me, you will move closer to me. If you disagree with me, you will move further from me. There is no reply button, so there is no place for the troll to grow.

  • Because of that, every time we had a Pol.is conversation – we just concluded one about our ocean proposals – we always see that, of course, there are some ideological differences. People have maybe five things that they fight about, but actually, most people agree with most other people on most of the things most of the time.

  • It’s just, if you look into a very antisocial social media, you will forget about this.

  • It is a small thing, but why is there no reply button? From a participation perspective, I would be thinking there’s no real deliberation taking place. People are not talking to each other. You just press the button and you respond. Then at the end of the day, you see if you agree with other people or not. Is that, in your perspective, real deliberation?

  • Can you explain us why you’re so happy that there is no replay button on the screen?

  • I’m also very happy that it participate pseudonymously. This is definitely not your ordinary habemus conversation deliberation. [laughs] There’s a very simple example here in the Uber case, is that for people who are driving for Uber at the time, they are actually doing very questionably legal things.

  • If they are required to show up with their real name and/or have a reply button where they can question and even interrogate essentially each other, very quickly, you will get people who get pulled into those divisive statements, like what is sharing economy, is Uber even sharing economy, is it sharing economy if you are not carpooling, and things like that, which is great from a debate point of view.

  • What we want is actually design thinking, just to explore people’s feelings. The facts and interpretations are what the traditional deliberations focus on. What Pol.is does is that it adds a step called reflection of feelings between the facts and interpretations.

  • When people can see that we are a polity after all, we have very similar feelings, for example, around registration, around insurance, and things like that, then people can then expand to the kind of face-to-face, habemus-style deliberation that you just alluded to, and we live-stream them, too.

  • What we are doing is essentially crowdsourced agenda setting that focus on the things that people have broad rough consensus about, instead of the debate that leads to polarization, and maybe find consensus, but most of the time just enmity.

  • This is our contribution from the Internet governance world with the idea of a rough consensus or harmony.

  • Let’s talk a little bit about rough consensus. Unanimity, if I understand you correctly, can never be achieved or is definitely not the goal here in what you are doing. What is the difference between rough consensus and lowest common denominator? Why is rough consensus so good in your point of view? Why does it help us in order to develop policies?

  • In online communities or in face-to-face committees after COVID, it’s the same thing anyway, in committees and communities, if you seek only fine consensus, then people with the most time wins by default.

  • Basically, on Facebook, for example, people who have more time to respond to each other’s face instead of the book part will dominate the discussion simply by dwarfing out, in terms of noise-to-signal ratio, other people’s points.

  • The Internet Engineering Task Force have long since figured out if we seek not the fine consensus, but rather, something we can live with, something that is more like consent, a common understanding that we hone to show that we can OK live with this, then it actually creates a more productive significant inclusion into the special interest groups that will form to explore that possibility.

  • In design thinking terms, if your exploration deviates, then you cannot actually do the defined path and agree on a common understanding. That is why a fielding checking part is always important so you know what other people’s arguments come from. Maybe while you seem very different in your positions, you actually share a very similar value and feeling.

  • We discovered that automatically using assistive intelligence on Pol.is, which is first used in vTaiwan and now also used in a national platform called a Join, which has more than half of the population as visitors and deliberate very large polices, far beyond what is originally the vTaiwan purview which is about digital economy.

  • Half of Taiwan on that platform. I make a note. [laughs] We’ve got a lot of citizen participation experts here in our midst. When we talk about participation and meaningful participation, three things are key – transparency, impact, and inclusiveness. Doing digital participation and being inclusive is a challenge. How do you do that?

  • Very carefully. [laughs] For example, this is my office, literally my office. It’s called a Social Innovation Lab. It’s in the heart of Taipei, and we tore down all the walls. You can see very clearly that there are no walls. Just this soccer field contribute by people with trisomy differences with Down syndrome so everybody who walk in feel very creative. I get random people visiting me.

  • The only thing that I ask is that the conversation of our transcript or video is published online. When they lobby, they always lobby for the public interest. In this part, we make sure that people who have anything to say can get listened to by the 12 ministries.

  • I seek out the least connected places, the remote islands, the rural places, the indigenous lands and so on, and work with our young reverse mentors, people who are under 35 years old, and advice each of the cabinet member so that the people in those indigenous or rural lands have a regular town hall, and maybe with cultural translators.

  • Then connect through video conference to the municipality and the Social Innovation Lab where the 12 ministerial people, section chief, or above, listen to what is happening around Taiwan. By viewing connected spaces for people who are going to be included in this discussion, for them it’s just another town hall. It’s just I arrived a couple of days earlier and do an ethnography or hanging out.

  • I hang out with them for a couple of days and make sure that I can understand where they are coming from. Then we empower the people closest to the paint by essentially broadcasting, amplifying their message to the people in municipalities and so on. It always augments face to face conversation. It doesn’t replace face to face deliberation.

  • I’ve got one more question that I’d like you to answer, then I want to bring in our experts. I also want to encourage all of you to use the chat. It’s a fascinating conversation. We don’t have that much space left to write on your own. Use the chat and then we’re going to try to bring in as many comments as possible. I’ve been working in the field of citizen participation for a while.

  • I constantly encounter two challenges. First, politicians where the beginning say, “Well, deliberation, citizen participation is great but at the end of the day they just think it’s a new form, a better form of communication.” Then bureaucrats who are afraid of using these new tools. Have you encountered the same challenges?

  • If so, what are you doing to get the trust of your fellow politicians and of the administration? What can politicians and administration gain of these new approaches?

  • I just replied it on the chat room but will say it loud again. [laughs] The solution is simply to show that for lower level bureaucrats, it saves their time. It’s much more preferred than sifting through endless phone calls. Each phone doesn’t know that 50 people have already made the same questions again.

  • Having a semi-automated system that can generate cute dog pictures that serves frequently answered questions as memes. For example, this is our physical distancing advertisement that says if you are indoor keep three Sheba Inu away, and if your outdoor keep two of those dogs away.

  • These dogs are basically companion animal of our participation officer in charge of deliberations and citizens engagement for the Ministry of Health and Welfare. Whenever there is something that need to be discussed, for example you wear a mask to do what? To protect yourself from your own unwashed hand. You go back home and take a picture of the dog. This is called humor over rumor.

  • If you get the R value of your messages higher than that of conspiracy theorists, then the citizens actually spreads this ideas for you and you do not have to answer their questions one by one. People will basically train themselves to become amateur epidemiologist. It saves their time. Amortized, of course, it saves their time. That’s the first thing. Second, that it reduces risk.

  • If you introduce something without considering some stakeholders, in Taiwan, we’re a very free country and parliament gets occupied, the Ministry of Education gets occupied, people go to the street all the time. [laughs] If you include the people who complain the loudest into co-creation, then chances are that you are going to reduce the risk down the line, again, amortized.

  • Also, the most compelling argument is that in those face to face meetings, in those Presidential Hackathon conversations in those teams and so on, the public servants themselves are recognized as the experts as they are. They are no longer anonymized abstract PowerPoints or Word documents, where we use Open Office, but you get the idea, Liberal Office.

  • The idea is that they engage in their work also more because they get a credit if I amplify their idea. I always say, oh, this is the public servant like 楊金亨 — former participation officer at Ministry of Finance — who decided to work with the petitioner that said our text filling system is “explosively hostile” so they get promoted much more quickly. If there’s anything that is risky, I absorb the risk.

  • By absorbing the risk, spreading the credit, reducing the time spent, and also make sure that there’s less risk of everybody involved, that is the Pareto improvement, we never trade one for the others. We must always make movements that increase one of those access without sacrificing the other field. This may look slow in the beginning, but it grows exponentially.

  • Thank you, Audrey. Bureaucrats are getting ready for their ideas, not just the minsters and politicians. That’s the time to bring in another politician, Eva Maydell, a member of the European Parliament, and President of European Movement International.

  • Eva, I’m not sure whether your video is working right now, but I’m sure you can hear us and we can all hear you. From what you’ve heard, what can the EU learn from Taiwan? Do we need more coverage, more technical skills? What would you love to know from Audrey Tang?

  • Thank you very much for being able to join you today. Despite no image, I think you can all hear me very well.

  • I would like to thank the Minister because it’s very fascinating, his story. This is the sort of story we need to amplify. This is why I’m glad that today together with the European movement and all other partners organizations, we’re putting this event forward because we need to amplify those examples, and ideally, we need to implement them. It’s about time for Europe to walk that talk.

  • In a way, yes, it is about political courage to answer your question. It is also about vision. Here, what we see is an excellent vision presented by the Minister, but it’s also an excellent implementation in the same time. If you do know the implement it then it would just remain a nice plan that never saw the light of day.

  • This is why I have two questions for the Minister that could help us a little bit more in finding those ways of how to implement some of the strategies that we have put in place. I like to say often that we need more digital leaders, or tech leaders. What I mean is we don’t necessarily need every politician in Europe to be able to code or know how to code.

  • We need them to have that vision and be ready to implement it despite sometimes, perhaps, there being certain and certainties whether that should be the way forward. My two questions, Minster, the first one is you might have heard, but here within the European institutions, we would like to, so to say, increase the commitment to reconnect with the European citizens.

  • In the next couple of months ahead of us and years, we would like to set up the so-called conference of the Future of Europe where the people will debate about what Europe they want to see. I believe all organizations that have set up this event today, but particularly we at the European Movement International, which I preside, we saw a big challenge ahead of this project.

  • We want this discussion as wide as possible, so that it not always includes the usual suspects, the people that they reach political buddies easily. They are always there. It actually has a more wide reach. I wanted to know how did you make your civic engagement platform so popular. What is the key to this high uptake? What is the key to reach those that never really are interested in reaching the institutions?

  • My second question is political priorities in the European parliament, mine, and also of my colleagues, to enable the free flow of non-personal data across Europe, industrial data in advancing AI for the benefit of societies and businesses.

  • What could be the key for institutions to lead by example? How did you ensure that the public servants embrace these digital solutions and interacted on the platforms? How did you basically bridge the gap between digitally savvy citizens, and not so digitally savvy institutions? Thank you.

  • Thank you Eva for your two questions. We keep reaching out not only to the usual suspects while changing, of course, the culture. I want to bring in Anna from Estonia as well, and then we’ve got a couple of more questions. Anna, Estonia.

  • We’ve got Taiwan as a world leader, we’ve got Estonia as a world leader in digital participation. How do you connect that what you’ve just heard from Audrey Tang to your experience? What would be your question to the Minister?

  • Thank you very much. For me also, absolutely fascinating to see our collective culture in this very complicated times, which is very different from European, more individualistic culture, is just shaped the way how we’ve dealt and you’ve dealt with it. In Estonian case, it was not about the collective part.

  • It was more the system that was resilient, that proved that e-services, e-governance overall, and e-participation tools, this is not only the way to provide public services in times of isolation, which we had to endure, unfortunately, but also is the basic hygiene factor, and is just a public safety tool that all Europe, as I now see also acknowledges, the super important.

  • We also had a boost of confidence in doing collective things and had a bunch of robots, which is an AI robot to help to bridge the Corona thing. From Estonian perspective, I would very much like to see in policy how we could learn from you how to be so collective and fun when it comes to doing these things.

  • I think also in Europe, the biggest problem is too much focusing on the problem itself, too much focusing on the privacy issues at some point that can block and bridge the creative process. We should more focus on solutions in order for us to be able to move faster and to take action.

  • Maybe fail, but maybe find out what is the way to move forward. You are fascinating for me that you’re moving forward with the situation that it is today, and how we can embrace this, what is going on, instead of creating and ideal solution.

  • My question is very simple. Do you consider that coronavirus, that we in Europe is just very devastating for many countries, but could it be a good thing that will help to boost this collaboration, e-participation around the world, e-inclusion? Do you think it has a positive impact?

  • Thank you, Anna. Minister, the floor is yours.

  • The first question which is about how to make that people participate, the answer is very simple. They set the agenda. They set the time and place.

  • There was someone with a random pseudonym, because we allow pseudonyms, the pseudonym was, I think, "I love elephants and elephants love me", nice pseudonym, and started a petition that very quickly collected more than 5,000 signatures, which was the threshold, and the petition reads, and I quote, “We must gradually ban all the plastic straws and other single-use utensils from our national identity drink, the bubble tea.”

  • It’s, of course, an environmental petition, and it gets people support it. We learned a page from the platform called Better Reykjavik that collects all the supporting arguments and all the counterarguments but never allow them to reply to one another, so we get both of the very good arguments.

  • We very quickly set up this collaborative meeting where the meet the petitioners and the people who collect the signatures, as well as people who make such straws and such utensils in a collaborative meeting. Face-to-face meeting, that is at the agenda of the citizens.

  • Of course, it takes some time to prepare, but will never take more than two months or four months, tops, if it’s exceedingly complex, but usually no more than two months. People are invited to talk to other stakeholders. We meet the petitioner, she was just 16 years old, and we ask, “Why would you start this 5,000-people, actually tens-of-thousands-of-people straw movement?”

  • She was like, “Well, it’s our civics class assignment. Our teacher just tells to start a petition that will resonate with people.” We’re happy that we included her in the conversation, because otherwise, she may go to strike on Friday, and what will happen then, we don’t know. [laughs]

  • Anyway, then they manage to work with the producers of the straws and the straw makers. The utensil makers say that we were social entrepreneurs 30 years ago, because Taiwan had a very bad case of hepatitis B, and we used these things to make sure the virus doesn’t spread. Of course, hep B is now fixed. There’s a vaccine and drug, the drug at least for it.

  • They are also looking to improve their business and have a higher margin, so they collaborated on new marketing campaigns to make a carbon-neutral or even negative capturing devices out of circular economy and so on. The young people used their protesting energy to advocate basically, to market the social entrepreneurs’ products.

  • Nowadays, of course, the plastic straws are actually banned for especially indoor drinking, but increasingly, outtakes as well. I use this simple example to say that we didn’t set a time, we didn’t say the agenda. We make sure that people responded to it in the here and now with the stakeholders that they previously would be very difficult to reach.

  • Audrey, Anna was talking also about failure. Is there some kind of failure culture in Taiwan? I mean, did you fail in one of your projects, and what did you learn out of it if it happened?

  • We fail all the time, and we fail very loudly. For example, the mask distribution map that I just show you that the civic technologists contributed didn’t start that way. On the very first day that we launched, a lot of very angry pharmacists used the feedback form, tells us that the time saver that you discovered with these civic tech people, they actually wasted our time completely.

  • Instead of swiping the NHI card and give out masks as we imagined, they are taking the NHI cards and give them those number plates so that they can walk away and then collected the mask at evening. Then they very slowly process the NHI cards.

  • You will see that the stock numbers in their pharmacy grow very slowly downward, while everybody who show up thinking that there are still some masks available, run into a situation where in very large font, it says on the front door, “Don’t trust the app.” [laughs] It was a tremendous failure.

  • Then very quickly, because at the time, we’re working on weekly iteration, we roll out a new mechanism every Thursday. After three iterations, we work with all the pharmacists so that they can publish their different opening hours and their ways of taking the NHI card. They can press a key to disappear from the map, and things like that.

  • It didn’t stabilize until maybe three weeks in the future, but we did fail maybe 10 times during those iterations.

  • Thank you. There’s no second thought that I go to the UK now when we talk about failure. Graham, I want to bring you in. You’ve worked on citizen participation for a long time. You have theoretical level. You have got practical experience as well, of course. What’s your question to Audrey, and do we have in Europe that sort of failure culture as well?

  • I think clearly, we don’t have that failure culture, which is one of the real problems here. Taiwan is to be highly widely acclaimed if that kind of culture is available. It’s certainly not in the UK at the moment. I have a couple of questions which are related to the same, which are about structural issues, if I may, Audrey.

  • The first is around the governance assistive intelligence, as you call it. I like that term, or algorithmic governance. There’s a lot of concerns particularly in the UK, and particularly because of a couple of policy disasters, about the way in which algorithms can reinforce differentials, and this question about how we ensure oversight.

  • I’m thinking, particularly having used Pol.is myself, it’s not quite clear to the user how these different decisions are being made about who goes in which groups and things like that. I’m really interested in your questions about how do you ensure democratic oversight of these algorithms, of associative intelligence?

  • The related structural question is, the kind of examples you’ve been given, I would talk of, they’re difficult, but they are relatively simple policy issues like the straws, like Uber. What about things like dealing with the climate emergency, dealing with the existential questions of Taiwan and Chinese relationships, or UK military presence?

  • How are you using digital tools to deal with those really fundamental policy issues and really complex policy issues? My sense is that that’s where digital gets strained. I’ll be really interested to hear your response on that.

  • Thank you, Graham. We’ve got also a question on that in our YouTube chat. What about more existential, more significant questions?

  • The US-Taiwan military collaboration is actually one of the Pol.is topics in the AIT@40 digital dialogue. I pasted a link to the chat room. That’s that part of the question. I don’t know about UK military presence. Maybe we need to talk to our Foreign Service. [laughs] We have Ambassador Jhy-Wey Shieh here.

  • Anyway, what I’m trying to get to is that, of course, as you correctly witnessed, that Pol.is is really not machine learning. It’s not deep learning. It’s not AI in the current generation. It’s AI in the previous generation. That is to say, it’s just a very simple dimensional reduction tool, principal component analysis and k-means clustering.

  • These two are kind of easy to explain. These are quite transparent algorithms as algorithms go, as compared to RNNs, and CNNs, and so on. In our Pol.is setting, there is always an open data check box, and we check it. Everybody can download the open data and visualize it however they want. Maybe they visualize it in three dimensions virtual reality, or maybe they visualize it as a song or something.

  • Actually, there are Japanese rap bands that remixes my material as their songs. This is not hypothetical. [laughs] In anyway, they can present it however they want. That answers the part about the transparency of the algorithm, as well as the earlier question about how to make sure that bureaucrats feel that this is not one more inbox, one more thing to check, one more technology to learn.

  • It’s simple. Basically, we build them as time saving tools. We actively seek out the places where they feel that it’s the most time consuming and automate these first, because actually, the vast majority, I would say, of Presidential Hackathon cases started with someone in the trenches, in mid to lower level bureaucracy. It’s just they can’t get their innovations heard, so they work with the social sector.

  • If this case works, they can say, oh, I’m just proposing this. I get all the credit, and actually they get promoted very quickly, but if it doesn’t work after three months or after deliberation, they can say, oh, I am just participating in a hackathon. It’s all the petitioner’s idea, or things like that. We have plenty of these going around, and we actually encourage these kind of things.

  • Basically, going to where the pain is, empower people closest to the pain. Don’t introduce new systems for new systems’ sake, but rather, assist people who are already overwhelmed. Just this morning actually, a couple of hours ago, we voted for our, this month’s cases for collaborative meetings as I posted the link earlier, the previous link that I posted before the AIT one.

  • The one from the petition was about the shipping fraud from the Facebook Marketplace that comes all the way from PRC and paid by the receiver. That’s a citizens’ initiative.

  • The other one, as example, is from our environmental protection agency looking to automate the environmental standard checks, on especially motorcycles and things like that, to get people into a more circular economy mindset when it comes to the environmental externalities. That is, of course, a very open question. They propose this because they waste countless time working on this.

  • They want to collaborate with people who are creative people who can turn, for example, the water refill stations into a Pokémon game so people don’t buy plastic bottles any more, and things like that. These are all very small things, I admit.

  • All the large structural issues, like what is platform economy, what is surveillance capitalism, and so on, can be boiled down to maybe like 10 or a dozen typical issues, like the UberX case, and if you resolve them one by one, like the 40 digital dialogue we had with AIT, as a whole, people will get a much more holistic understanding of dialogue structural issues.

  • Yves, the French, Parisian perspective, you’ve recently co-organized a French climate convention with randomly selected citizens, and all participation community looked at that process, because it was a powerful process. President Macron responded to it. How can you connect that experience that you made here with what Audrey has been saying?

  • Thank you. Hello, Minister, and hello, everyone. Yes, it was a great experience at the level of France, that citizens’ convention on climate.

  • It showed that when you put citizens together on an important issue, they create intelligent solutions and they create trust in a world where most of the people believe that trust between citizens and governments is in free fall, then the societies become incredibly divided.

  • We see that when we put randomly selected people in a room, in fact, they create intelligent solution and trust. This is true at local, national level, but it can also be true at global level. I have two points for you, Minister, but it’s also two points for everyone in the audience.

  • On October 10, Missions Publiques and its global partners will deploy a global citizens’ dialogue on the future of the Internet in 80 countries of the world. The idea is to feed the process at the UN level about the future of Internet and its governance. Based on your experience, you know Internet can be a tool to fuel trust in society. It can also be a tool to fuel mistrust in society.

  • If you could add a question to that dialogue that will engage citizens in 80 countries, if you could add a question on that global citizens’ dialogue on the future of the Internet, what would be your question? That’s one of the questions I have for you.

  • The other one is, is there a global political objective that today seems impossible to achieve and that if it were to be achieved in the next 20 years, it would give full meaning to the mission you have set towards yourself and towards society?

  • I didn’t quite get the last sentence. I didn’t quite get the second one. I got the first one about a question to the future of Internet deliberations.

  • The other one is about, is there a global political topic, a global political objective that today seems impossible to achieve globally, and that if that objective were to be achieved in the next 20 years, it would give full meaning to the mission you have set towards yourself and towards society?

  • Yeah, that was the last few words that I missed. For the first question, I think a great question to ask is just to take the RFC 8890, the very recent, just past RFC. The title was The Internet is for End Users.

  • This is important, because this is the first time that IETF really openly say that IETF is a political organization in the traditional sense of the politics, that the end users is the end goal of the Internet’s existence. "Rough consensus and running code" was always defined very roughly... Only people who produce running code are parties to that rough consensus.

  • A multi-stakeholder conversation culture, while it may produce very good consensus, maybe only benefit people who can afford to subscribe to the IETF working group email loops. These are a strict minority of the people. The question that is asked on that RFC, for whom do we go through the pain of gathering the consensus and writing running code? I think this is a great opening question.

  • This would provide a reflex space for each and every participant of that panel. You just ask the simple question that IETF just answered. Then ask the people, how would you answer differently? For whom do we go through the pain of gathering rough consensus and writing running code. Is it future generations. Is it seven generations down the line?

  • Is it rivers and the mountains that are currently disenfranchised? There are many possible answers that is not captured by that RFC. I would love to hear about those answers too in addition to what is in that particular RFC. I hope that answers your question.

  • The global political objective, I think one of the global objective is around the idea of sustainable development itself. As we know, by year 2030, we will meet some SDG targets, and we will miss some. Hopefully, we meet more than what we miss. There are structural issues, such as COVID and such as climate change, that threatens to also undo the progress that we make on the other SDGs.

  • Interestingly, the climate change one may be actually improved by the coronavirus. In any case, we will, by 2030, make another run at something like the SDG and just continuing the trend from the multilateral trend in the MDG to the hybrid multilateral/multi-stakeholder model that is including the major groups in the SDG.

  • We are in the post-SDG era. We will maybe just drop the multilateral part altogether in forming of the groups and in setting of the agenda. How is that even possible? How could a non-multilateral or very limited multilateral process be made when doing this post-SDG SDGs? That is a burning political question at this very moment.

  • That is one that is worth a worldwide conversation, much as the worldwide conversations around the Paris COP.

  • Thank you very much, Minister.

  • I’d like to highlight what’s going on in the chat. We’ve got some people who want to press you a little bit harder on algorithmic governance. Rose Longhurst. Rose, would you like to share your question here?

  • Yeah, I’d be happy to. I’m just wondering about the actual governance of the system. Obviously, we talked about open data and the transparency aspect, but that doesn’t necessarily speak to the governance and the literacy needed to be able to interpret the data.

  • How do we ensure that there’s real engagement rather than just making sure that the information is available? Thanks.

  • I would argue that if you teach about literacy you will not meet the necessary governance criteria in the civil society. That is why in the K-12 curriculum in Taiwan starting last year, which I’m part of the committee to make the new curriculum, we officially dropped the word “literacy” from 10 years ago, used in our curriculum, when it pertains to media literacy, digital literacy and so on, and data literacy.

  • We now replace that with the word “competence”. By media competence, we mean the capability to produce media. Random primary-schoolers nowadays maybe have more Instagram followers than I do. Obviously, they are producers of media.

  • With AirBoxes, with those participatory mechanisms, with people proposing plastic straws need to be banned and so on, obviously they are creative producers. They are policy producers. They are also data producers when they curate the water and air quality measurements and so on.

  • The whole point is that starting from the first grade to make sure that people feel as citizens even long before they have the right to vote, that they can weave their curation, their data stewardship, into the fabric of the decision-making, first in their school, then in their community and in their township, and then in their city or metropolitan area, and then countrywide.

  • That is part of our new curriculum starting in the previous year.

  • Only when they feel as co-creators who can essentially fact-check all our presidential candidates – that actually happened earlier this January. All our presidential debates and platforms and so on are crowdsource fact-checked by random bunch of people, who many of them are just high schoolers or even primary schoolers, who work of course in conjunction with professional journalists – can they feel that they have the competence needed for joint data controllership and for joint media production.

  • Finally, there is a lot going on also on the YouTube chat. There is one question that says, “It’s a fascinating discussion, but how do we apply all of that in Europe?” I’d like to connect it with my final question and reconnect it to Leonard Cohen, one of the most covered artists of all time.

  • As we all know, there are beautiful covers that are really good and some awful covers. Do we need to cover what’s going on in Taiwan, in Europe, or do we need to create our own song? How do we apply your wisdom in Europe?

  • In the words of Leonard Cohen — “Ring the bells that still can ring and forget your perfect offering.”

  • This is important, because there is no perfect offering. In Taiwan, all these mechanisms that I described to you today is under essentially monthly or even weekly iteration.

  • Just today, we changed our rules so that we take at least one petition and one bureaucrat-proposed ideas each month so that each side would not feel neglected by the voting that is done by our participation officers. Last month, all the cases that we handled was initiated by the premier and the ministries.

  • We feel that the petitioners, actually, petitioners tell us they were being sidelined and so on. The Taiwan model is not a model that you simply copy. It is rather a idea, a Buckminster Fuller-ian idea that instead of fighting a old system, make new system that will make the old system feel obsolete. You don’t have to convince everybody that it is, the old system is obsolete, just a few people.

  • Increase the R-value. Make it viral. Make it ideas that’s worth spreading. This is in any sortition, any citizen assembly. The final question is that can they bring back to their community the rough consensus they have reached in a deliberative space. If there’s sufficient amount of cute cats, dogs and memes to do that, then yes they can. The ideas worth spreading will spread.

  • If the R-value is rather under one, then even if they achieve perfect harmony, kumbaya, and all that during the deliberative space, that will not have a impact to the society. Think about fun and optimize for fun. There is no perfect offering.

  • Thank you. That was pretty perfect. Some applause to you, Audrey Tang. Many thanks for sharing your insights, your experience, your wisdom so we can all learn a lot. Thank you to our four panelists, Yves, Graham, Anna, and Eva, for sharing their questions and also their experience.

  • I need to re-watch our talk because there was so much insights in it so that we can all learn from it, I’m sure, so it will be tremendously helpful in creating also the conference on the future of Europe. I’ve certainly learned that you, Audrey, are much more into Leonard Cohen than I expected. That was perfect.

  • Thank you to all of you. Have a good rest of the Monday, and have a good start into the week. See you. Ciao.

  • Thank you. Live long and prosper.