• Let’s get started. Welcome to the administration, to the Executive Yuan. I am Audrey Tang, Taiwan’s Digital Minister. If there is anything I can help answering or you would like to know, I’m happy to share. As usual, visitors, journalists, and actually internal meetings, too, that are held in the Executive Yuan, we make a record, but we don’t publish the record.

  • We make a textual transcript, and all of you will receive the full transcript of our conversation. It’s a editable... It’s like a Google Doc, but it’s hosted within the Taiwan government. Anyone can edit anything, anything that you say, of course. After 10 days, we publish it to the Wide Web. I’ve been doing this for one year and a half since I become the minister.

  • If you prefer to be known as a particular speaker, please say your name before saying something. Otherwise, it would just be a journalist asking a question.

  • (laughter)

  • Just to let you know how it will look like, it usually looks like this, so people who say their name, basically. If you want a photo or something, then you can post it there, too. Otherwise, it’s just audience member. [laughs] I hope it’s OK with you.

  • My work in the administration as the Digital Minister, I’m mostly in charge of the digital transformation of our public service. Basically, how to introduce new technologies such as so-called artificial intelligence and so-called big data, so-called cloud services in our everyday work, and to improve the service delivery of our digital service.

  • Additionally, I’m also in charge of the open government plan, which is about making the civil society and the government trust each other more through a series of accountability and participatory measures.

  • I’m also in charge of social entrepreneurship, which is a new way of combining the commercial world’s forces and the social world’s forces, and to solve common problems and develop into these sustainable development goals. That’s my primary mandate. I’m happy to explain anything if anyone starts with a question.

  • May I begin with a general question?

  • Obviously, you are a member of the younger generation in Taiwan. What is, in your view, Taiwan’s identity? What kind of society should Taiwan be? What should be the place of the country in the so-called multi-polar world.

  • That’s a great question. Taiwan, as many of you know, had a generational gap that we experienced. My parents’ generation, my grandparents’ generation, they fought for democracy, but their education and the way they were brought up was very much in a more authoritarian society.

  • In that society, the government is much like the rope here. It’s tasked with organizing, representing the people, and imposing order, and also to arbitrate between the different stakeholders in the society. That’s no longer the case. I’m 37 now, so people who are younger than me, they were brought up after the martial law is lifted. To us, it is improbable to see the government in this way.

  • In our generation with social media and things like that, we don’t need political representatives to organize among ourselves. We just need a hashtag. We also don’t need particular ways to arbitrate. We’re more or less looking for win-win solutions, that is to say values that works for everyone and solutions that works for everyone.

  • I work on, for example, social innovation. We all emphasize the government’s role as providing a space rather than just being an arbiter. The space is because of in a sense that it is composed of people’s intentions and stakeholder stakes. We try to find the values that people share and find a solution that work for everyone but without the government imposing it.

  • Instead, we see that the stakeholders themselves need to discover those values together. To use a very concrete example, when Uber first got into Taiwan, we deployed this AI mediated conversation pol.is where people can use their phone, their iPad, whatever to see how they look within their stakeholders’ groups, their physical friends, their Twitter friends on the opinions, for example, passenger liability insurance.

  • When people press agree or disagree on people’s ideas, gradually we see that people’s consensus emerges, whereas in many authoritarian countries without as absolute freedom of speech, freedom of assembly, people would actually focus on the most divisive statements as a way to mobilize among themselves.

  • We, instead, provide a space as a safe space so people can focus on the consensus statements that are surprisingly many.

  • Just by replacing the public forum from a private sector waste-as-much-time-as-possible design to a consensus-based design, we were able to crowdsource, essentially, a series of legislation viewpoints that we can check one by one with stakeholders in a live stream fashion, and collectively make sure that Uber can be incorporated into the Taiwan eco environment.

  • We’ve been doing this systematically, and ratifying such crowdsourced legislation. To answer your question directly, Taiwan is uniquely positioned in working on the ’17’s Sustainable Development Goals, which is partnership across sectors, to further these goals without sacrificing any other.

  • Whereas the authoritarian governments may be better suited to do some top-down experiments that is to the benefit of one stakeholder to the detriment of many, in Taiwan we’re tasked with to find these kind of common good solutions.

  • We’re pretty unique in Asia in having our value proposition and have the Reporters Without Borders and other international NGOs to be headquartered here, and, essentially, to work in their Asian context in a way that can make full use of the different focus that Taiwan people has on these Sustainable Development Goals.

  • My answers were a regional hub for collaboration, powered by a vibrant civil society, and the government’s absolute respect for freedom of speech and assembly.

  • Regarding this slide, consultation, as far as I understood, do you use this data to form your policies in any way?

  • Don’t you think it’s a little bit, even dangerous to put some questions in such a binary fashion -- yes or no, or I agree or disagree -- because sometimes the right policy’s in the middle?

  • No, this is very much a brainstorming phase, so that if you think that you’re mostly agreeing with it, but you have a better suggestion, then you propose something else for other people to vote on, which is why it looks the shape it is, because people converge on these bunch of suggestions or sentiments that resonates with everyone.

  • While each dot is, as you said, a yes or no sentiment, over time, over a month, or three weeks, or so, it gradually converge on a shape that people refine on each other, compete on the resonance that one can propose for each other’s sentiments. That’s how it works.

  • Of course, in design thinking terms this is the initial brainstorming and a little bit of convergence, like here. The second problem definition we always use a face-to-face and town hall style. We go to the place where this consultation happens.

  • Sorry, for that second phase you need people really engaged, because it’s easy to press yes or no, but then you need more engagement. The people are engaging, then...?

  • Once we make it aware. The first phase is to get stakeholders aware of this consultation, but once they are engaged into this, they are now much more incentivized to view the live stream consultation through YouTube or through other ways, or to, even, attend the in-person consultations.

  • At the in-person consultations, again, they’re extension of the face-to-face meeting format. They are not a substitute of the face-to-face format. Often, after a consultation meeting, we also schedule more workshops so that we can work out the fine details. For example, redesign the tax filing system together through this kind of co-creation workshop.

  • The initial, this digital consultation, is just the beginning. You can almost think of gamifying the consultation to make people identify with the fun of the consultation, but then people get much more serious and is willing to consider in a holistic, systemic perspective, which is the prerequisite of entering a face-to-face deliberation.

  • This seems like a very bottom-up way to make policy, where people crowdsource, see what their sentiments are, and then that gets filtered throughout these different systems. We also heard a lot, since we’ve been here, about areas that need a bit more of expertise.

  • You have a technocratic approach to industrial policy, for instance, setting an agenda. "This is what we’re going to pursue, is we’re going to prioritize for economic development." What is the balance that the Taiwanese government strikes between sourcing policy goals from the people and imposing this expert vision?

  • That’s a great question. Actually, the first cases of the laws that I worked on, personally, when I entered the cabinet is called so-called fintech sandbox law. The fintech sandbox kind of epitomizes the balance that you were referring to.

  • In fintech sandbox law, I’m going to introduce very briefly, anyone who has an idea, like AI-based banking, cryptocurrency, and B2B lending, whatever, that is not currently allowed by the financial ministry, basically gets to submit a application saying that I have a new fintech innovation and I’m going to violate the law for the next 6 months to 12 months, but for the public good.

  • They have to write a justification of why this is good for digital inclusion, for financial inclusion, for any kind of public good. The financial people, it’s not like the ministry stays completely out of the loop, but they assemble a multi-stakeholder panel to look at this innovation and see if they can partner with any local city government or whoever to realize it.

  • Once it’s in the experiment, we also look at it very closely from a data-sharing viewpoint, from a privacy enhancing viewpoint, and things like that, to make sure that the experiment goes well. After 6 months or 12 months the society get to decide collectively whether the people who have actually used the service, instead of just seeing PowerPoints, feel it’s a good idea, whether this is good for society, after all.

  • If it’s not, then we thank the investors for paying the tuition for everybody else afterwards, but otherwise our regulation, our laws, even, get changed. If it’s just regulation, it gets changed quickly. If it’s law, then the legislator need another debate, which is, again, democratic, but the experiment can extend to up to three years to accommodate for the time needed for the deliberation in the legislation.

  • As you see, this is a co-creative environment. We let the private sector, who has the incentive to break the law, and to work with the legislator who need this real-world data to convince the constituents. We let the constituents’ voice inform the private sector people whether this is a good idea or not.

  • This is all the three sectors working together. We’re extending this fintech sandbox into a couple weeks from now we will push a AI mobility sandbox into the legislation where we will experiment with un-crewed vehicles that are not just autonomous cars, but also freight-carrying drones, ships, and some hybrid vehicles.

  • Also, working with the local governments with an eye on improving the life of their citizens with new technology, and then we will move onto other general purpose sandboxes. In each cases, we need expertise, but we especially need a accountable public way for people from different expertise to have a real conversation among each other, and for the whole society to learn. That’s the main idea.

  • Thank you. What sparked the open government movement? What were the conditions in Taiwan for the open government movement to grow? Why is it so difficult in places like even the US to get anything like this done?

  • There’s not a lot of cooperation between industry and government. In fact, industry doesn’t want to work with government because there’s so much regulation. I’m wondering why has it taken root in Taiwan and not the US?

  • Can I take another question?

  • I was going to ask, basically, what percentage of young generation age group are involved in politics in a decision-making that really makes difference in the government? Because it seems you have a lot of innovative ways to come up with statistic information or get their feedback of the new generation.

  • That would be fine, but as long as, the question was risen already, how do you input that into the people that sometime, like in the United States, you get 20, 30, 40, 50-year-old people that are in positions that are not really, in my opinion, contributing the way it should be done with the new generation?

  • How to work inter-generationally in this political environment?

  • What percentage of your age group, for example, are in the government in a decision role making that can contribute changes?

  • These two questions are very much closely related, so I’ll try to answer it. In many city-level governments, such as the Hsinchu City, the entire city cabinet, the average age is something like 30 years old, or something. In many local city level governments in Taiwan we see a lot of young people in charge.

  • I was just visiting the Taoyuan City. They have a youth affair department. The director is 20-something. The idea is that the local level, the young people are very much in charge now. Many mayors, they’re, themselves, pretty young, and they have their city-level cabinets pretty young.

  • It is true that in the central government, because of the stability, the requirement of being stable, the top senior executives are usually professional civil service servants. There is no easy way for young people to theoretically jump into the top, senior executive level.

  • We don’t think this is a big problem, though, because in many cases where the innovation happens it doesn’t have to be in the public sector. All we need is systematic ways, such as the new civic Referendum Act, for example. We’re going to have Swiss-style referendums starting end of this year working on a lot of interesting issues.

  • There the agenda setting is all done by the civil society. It’s not done by senior executives. The senior executives are, of course, as you’ve mentioned, there might be some reluctance. There might be fear, uncertainty and doubt on this kind of direct participative democracy, which goes back to your question.

  • One of the reasons why people are putting a lot of faith on open government is the occupy in 2014, the Sunflower Movement. Back then we occupied the parliament for 22 days as a demonstration.

  • The demonstration, so-called Sunflower Movement, is a demonstration in the sense that we demonstrated that it is possible, using digital tools, the tools you just saw there, to get half a million people on the street, and more people online, to actually converge during the course of an occupy.

  • In many other occupies around the world, the end of the occupy you don’t even know what they’re asking anymore. There is a general sense of directionlessness in the occupy.

  • Using this kind of, what we call focused conversation method and related technologies, it is possible for the various different factions in the occupy to handle one aspect of that particular trade service agreement every day, and then have some consensus, have some part that still needs discussion, and then move onto the next day, working on the days from the previous day.

  • After 22 days the consensus was so strong that it was a coherent set of five demands. Head of the legislation eventually agreed, and so the occupy was a success. It’s a nationwide demonstration that this method actually does work.

  • Of course, it requires a lot of time. It doesn’t really scale very well. It requires a lot of human power to maintain it, but we’re in Taiwan, we can automate many of these parts. It really requires digital innovations. I think after the occupy, the general population really started demanding for radical transparency, for accountability, for things like that.

  • That’s when the new mayors starting end of 2014 pretty much all used open government as their campaign ideas. Of course, primary among them during their campaign was the Taipei City Mayor, Ko Wen-je, and also the Tainan City Mayor, Lai Ching-te, that both campaigned using open government as their campaign, and now Lai is our premier.

  • I just wondered to what extent the concerns about cyber privacy have resonance in Taiwan. Are there specific concerns to Taiwan? Also, what you describe seems relatively benign, but I wonder if you have any concerns that all this could be misused by a future government, as it has been used in other countries in Asia, for instance.

  • In Thailand, they actively monitor all social media by the government using probably the same tools.

  • I think my question is similar, but is there any anxiety that these tools could be taken advantage of by foreign powers? As we learned in the US, any time you have a big adversary bent on disrupting your society, the exuberance that all these new tech, digital tools produces can turn ugly.

  • Very much so. The whole idea of an e-petition system, or a now e-referendum system going online next year, they have to be resistant to foreign tampering, otherwise there is no point in making it a part of democracies, just like the voting ballot system.

  • It’s not particular to the digital. Paper ballots have their ways of being hijacked in Asian countries as well. It is not something that is particular to online. I’ll answer it in three very brief statements.

  • The first one about privacy, I think Taiwanese people have a strong pro-privacy stance. The government, we can’t even begin considering censorship. The political will of the people is so much so that for example, when there’s an epidemic of so-called fake news... I always use the term misinformation, because I think fake news is affront to the journalists.

  • In any case, the misinformation epidemic, so to speak, which results from something technical, it’s easier to share a story than to read a story. It makes it easier for ideas to go viral that are very poorly informed. Our solution with solution space is very much constrained by the strong anti-censorship political will.

  • The ideas that we came up with, such as a rapid response system of, for example, misinformation clarification by all the ministries on their own websites. Then visible as a public listing, as a chat bot, and so on from the civil society, or to add media literacy and critical thinking in our national curriculum from K to 12 education and things like that, they have to be benign.

  • If it’s state-controlled, the people will have none of it. This is a very strong political atmosphere here. For misuse, we had a lot deliberation with stakeholders on our e-petition platform. Second to countersign a petition, one has to provide one’s mobile phone number, but not real name or any other identification number.

  • This is a compromise, because otherwise, if it’s just email account, anyone can get 5,000 of those. With a mobile phone number, but without revealing it to other countersignatures, we provide a middle ground for people to not abuse the system too much.

  • Again, it protects people’s identity. If they go to some length, they can still get anonymized mobile phone. This applies not just to Taiwan citizens. Anyone who is a resident certificate, they are also eligible to participate in our e-democracy platform.

  • We’re trying to be inclusive, but there is a constant struggle, as in anywhere else. We face this challenge by having open, multi-stakeholder conversations on these. Finally, yeah, in the cyber security area, of course, there is a constant pressure of the foreign hackers taking over.

  • Speaking as a civic hacker myself, the system that I built that I introduced to the administration, we always hire the top notch white hat hackers to do penetration testing to try to attack the system that we built. We only build on this solid cyber security foundation.

  • We’re going to have a cyber security law that requires personnels in critical infrastructures in all the different ministries, just to make sure that there is a culture of being cyber security aware. Then using only systems that are penetration tested, open source, and all these by the community, instead of trying to keep it to ourselves.

  • Again, these are challenges, but we face it by facing with the people, not for the people, so to speak. Yes?

  • I have a question. You just mentioned e-democracy platform. We see in China something like an e-dictatorship taking place, with the social rating system and all of this control system.

  • Yes. It is a very, very interesting direction.

  • One of the concerns that we have in the West is that we see the Chinese government doing these kind of projects. We see in the West that big companies like Google and Facebook, etc., are doing the same thing on another basis, on a business basis. We are afraid of the convergence of these projects. How do you see yourself and your e-democracy platform in this?

  • That’s a great question, but there was another question?

  • Maybe just a follow up to that. One of the issues that is in a lot of countries is the collaboration, or the demand by governments, for records of users of tech companies. What happens in Taiwan in regards to requests by police and security agencies for cooperation among tech companies? Are those demands made, and if so, what are the components?

  • There’s the transparency published by a volunteer secretary of the civil society, the Taiwan Association of Human Rights, that basically looks at the requests sent by the security and police force to those international companies, and how they work with each other. The nitty-gritty details are in the transparency report.

  • The overview of this question is that the government, we try to make sure that we work at least as much with the civil society actors as with the private sector actors. The movement that I am part of is called the g0v movement. It’s spelled G-0-V. The movement originates in Taiwan, but it’s everywhere now.

  • There is an introduction to it, if you just google for g0v, or g0v.Asia, or .us, or .uk, dot-New Zealand. There’s chapters everywhere, but it starts in Taiwan.

  • The g0v movement basically is a civil society re-imagination of the government for all the government websites in Taiwan that ends with .gov, .tw, as long as there’s someone from the civil society who wants to improve on this public service, they can register exactly the same website. They change the O to a zero.

  • Getting to the shadow government by just changing one word in the browser. Personally, I was in charge of forking. That is, just I take all the existing data, but moving into a different direction, forking the National Dictionary Project.

  • I personally maintained, with a lot of volunteers, Taiwanese Hakka, Holo, Mandarin, and also English, German, French dictionaries, which then people then forked our project to indigenous Armies and so on, all based on the Minister of Education’s data, but presented in a way that is much more multicultural.

  • We do this in abandon, relinquish our copyright, so on the next procurement cycle, the ministry can incorporate our contributions back to the procurement cycle. This is an idea of forking and merging the government. The very first g0v project was the visualization of the national budget. They show how each ministry and each agency in their projects, how they allocate their budgets.

  • There is a discussion forum where people can ask any questions and talk among themselves. This is merged back to the Taipei City participation budget platform, so that anyone can have a conversation with people in the city government directly without going through representatives by this visualization platform.

  • Just last month, we also introduced this on a national scale, so that over 1,000 ministry-controlled projects, all the KPIs, how they’re spending their budget, how much they’re spending on such things every month or every quarter, with our service delivery percentages and so.

  • These become social object that people can have a real discussion with the people in charge of these projects, again, without going through representatives, directly online. For formulating such platforms, we make not just a result or the process open, but the actual agenda setting, like which things to develop, which issues to talk about, and so on.

  • We do it in our weekly hackathons. Every Wednesday from 10:00 AM to 10:00 PM, I am in the social innovation lab. Anyone can come to talk to me, as long as they agree of the transcript being published online. Every Wednesday night at the same place, there is the g0v hackathon, where people from the civil society can just collaborate on more projects.

  • We try to keep a very close relationship, not just with the civil tech sector, but also through biweekly tours across the island to bring the people, like 12 different ministries, through teleconference. They’re in Taipei. I go to the various rural areas, indigenous areas, and so on, and then talk to social innovators there by using teleconference to bring the rooms together.

  • I go back to the same place every two months so that the civil servants have a real connection with the people locally who are trying to do social innovation to solve local problems. These are our links to the social innovation space. While you are here, you will of course also hear about the 5+2 industrial innovation. Our premier is doing the same tour, but with the industrial innovators.

  • We’re not trying to be partial to the private or to the voluntarily civil, the civil society innovators. We’re trying to be very balanced in the way that we make the innovation happen.

  • Sorry, but you didn’t answer my question about what’s going on in China. For a time, when Internet was not in the government’s eyes, it was very free, and there were many experiments on the Chinese Internet.

  • Then suddenly, when the government took interest in the Internet, it ended up controlling everything. How your system is immune to a future more authoritarian government? Is it immune?

  • I can say as a fact the Internet is designed to treat censorship as damage and route around it. Even today, in the regimes that you mentioned, there are still dark net, underground innovations happening.

  • It is not a coincidence that the Reporters Without Borders chose Taiwan as their HQ, exactly because we can receive the clandestine networks’ data and so on, and publish here without a fear of retaliation.

  • Personally, back in the early 2000, I worked on the Freenet project, which is way before the Tor project, but serves the same need, which is to make the human rights people, whistle blowers and so on, to have a safe space to conduct their work in a way that it is not easily harassed or blocked by a suppressive regime.

  • Now, of course, there are limits to any technological breakthrough. On the other hand, there is the wider international community working on how much we tolerate such behaviors on the Internet without the Internet being fragmented. These are big questions. Of course, it is conceivable that at the end, people all see the light, and see that end-to-end principle is a good idea.

  • For the time being, there are roughly speaking, three very different Internet governance philosophies being battled out. Original end-to-end innovation that starts from the US, and the human right-based thinking that is the GDPR and the current EU working groups. Of course, there is the Wuzhen paradigm that basically sees the cyberspace as a power projection of the realspace government system.

  • These three are frankly speaking, not very compatible. Taiwan is caught in the middle of all three. We try to harmonize the laws so that, thorough open multi-stakeholder consultation, we allow innovators who work on any innovation regimes to still thrive in Taiwan.

  • Today, our government itself must always be held accountable by the people. However, there is no promise that in the future, hundreds of years in the future, that we will still remain a democracy. I think it’s for every generation to ensure that for the next generation, it still says true to the original democratic goals. Seven more generations, and I can’t promise anything.

  • Regarding the practice that you mentioned at the beginning, to record meetings, and to be transparent, show to the people how discussions are displayed in government, can you tell us how much resistance do you find from any other colleagues?

  • Can you mention the Minister of Defense, not very willing to discuss some of the meetings in public, or some of the...? Did you find resistance among your colleagues, and how did you overcome all that?

  • Yes, and you had a follow-up?

  • Something similar, I guess. All of this reminds me of a particular company in the United States, which is a financial institution that records all of their meetings. They have people answer questions and show these maps.

  • I used to work at this place, actually. I found that it was good and bad. People at the top of the company would preach this idea of radical transparency. Then at the same time, sometimes, it was a tool to exert control.

  • There were subtle ways that you could tip the scales in favor of the powerful, like the way you ask a question, for instance. I guess on the same note, has there been any internal pushback in reform to the tools that you guys...?

  • There’s a lot of back and forth. Basically, we have to start by saying, in the Parliament and in the court system, they already do that. It is just the administration which have some meetings that are actually binding, to a degree, but the where and hows are not visible in the open.

  • There are good reasons for that, but the outcome is that without these kind of accountability records, the public trust, or the perceived distance to the administration, is larger, is longer than when the parliamentarians.

  • Of course, we are not parliamentarians. By law, we are not required to disclose the internal transcripts, the full transcript of the preparatory meetings. I made it very clear, two things. First is that what I am doing is an experiment I am not trying to inflict it to my fellow ministers.

  • They sometimes use the tools that we use, but especially around the controversial issues, where the stakeholders have no reason to trust one another, [laughs] they capture the whole meeting just so that the stakeholder can know that there’s no backroom deals. They use it selectively. They’re not saying, "All my ministry’s meetings are to be this way."

  • The second thing is that there are still rules, as I said, there’s 10 days for folks to edit. If there is typos in the transcription, if there are particular exchanges that you think will be prone to be taken out of context, you can always change it. This is not live streaming. There is very little pushback, actually, because they still have the room to edit.

  • During editing, people actually become much more receptive. In some meetings, people just speak but forget to listen. That fix some of that problem. Also, being published also means for those career public servants who are before anonymous, if there is anything that goes right, the credit usually goes to the minister. If things go wrong, the minister has a way to blame the public servant.

  • There is little incentive in the existing system for public servants to propose too risky ideas. Usually, they’ll have to find a political alliance, again, PE to propose on their behalf. On this kind of meetings, because there’s a full context, people know that it was proposed in a way that are known to be risky, or it’s just innovation. We try, and it may fail. It has a better understanding, a better padding.

  • If it fails, because it’s just me doing this experiment, I absorb most of the risk, anyway. If it goes well, then the journalists actually go back to the transcript and find the actual career public servant that proposed this idea in the first place. I make sure to share the credit in my presentations and so on.

  • I think this is a win-win solution, personally speaking, but the prerequisite is that I don’t ever attend national security meetings. [laughs] I am not party to any confidential information. When there is a military drill, I take one day off. I still don’t know where our underground base is. I don’t know where the bunkers are. This is a deliberate experiment is what I am saying.

  • Related to this, is there access to information legislation in Taiwan?

  • Oh, yeah, of course, the freedom of information, yes.

  • Where you can request government decision making information?

  • Yes. It’s up to par with any modern democracy. What we’re doing is above and beyond the FOI requests, because we’re voluntarily providing the full context, whereas by law, we only have to provide when, where, and who, a summary.

  • Is there any danger in replacing journalists, or is it just giving journalists more information to do their job? Does it testify to governments, or, "Well, it’s all open source, then we’re not going to respond for 10 days."?

  • I find the people who love this are investigative journalists, journalists who need time to do a lot of original research, a lot of perspective, a lot of stakeholder interview. By including the full transcript, they automatically know which stakeholders are in the room, and where their ideas come from.

  • Whereas before, if you just interview people after a meeting, if you interview three people, you get three different worlds. It’s very difficult to do an investigative job. The investigative journalists love it.

  • I think the real-time media, who are under pressure to release a short snippet within five minutes or something, they also like the fact that for such time critical things -- especially if it’s live streamed in the first place, like our press conferences -- we also provide them with the full transcript. They can copy and paste, and be done with their five-minute job.

  • These two groups of journalists are particularly positive. I haven’t heard from the other kind of journalist yet.

  • What percentage of your eligible population vote actually in the general elections or other? The way you’re talking so much, of course, the youth are very active in the government and very eager, very upbeat, and are doing so much better than someone coming from the United States, around the world, to take part in the Green Party, the Labor Party, for example.

  • What is the percentage? It should be pretty high.

  • For the 2016 presidential election, I think it’s high, but it’s not very high (66%). Everybody knew that President Tsai will win the election. It’s not as strong a turnout compared to previous elections.

  • What we’re trying to do here is that we get people to vote all the time.

  • If it’s just the presidential election, it’s essentially two bits of information every four years. It’s not that much. There is considerably more concerns that a civil society would want their government to know, hence the open government, e-petition, or the publication of the budget, how it’s used, and everything, the regulatory pre-announcement for 60 days.

  • All these become discussion boards for people to have a much more higher bandwidth communication. There is also participatory budgets on the local level, and there is also referendums. All these are essentially, was to vote on policies, votes on things, votes on budget, votes on money, rather than votes on people.

  • In voting on people, I think it requires a lot of cognitive work, working out specific policies that you already know or you’re a stakeholder, that’s much easier. People can do it in their spare time. What we are trying to do is to make democratic process to also work in people’s spare time.

  • You had a high percentage of people in Taiwan, with all due respect, do not really have education, as unique generation gap. How are they involved, knowing all of these referenda, the topics, and effects on them a so forth? How do you get it to them?

  • I think Taiwan has a really unfair advantage in the sense that we have broadband as a human right. Many other countries say it, but Taiwan is one of the very few places that can actually realize it, because our geography. It really makes broadband access affordable.

  • Generationally, what we are seeing, for example in our e-petition website, of the 23 million people in Taiwan, about five million is on the platform. The engagement rate monthly, I think, is around 10 million or so. It’s a lot. It’s a sizable fraction of the population.

  • We find that the retired population, and the very young, like pre-work, the still in education students, these two groups are higher than the average, because they have more time. [laughs] People who are currently working in the workforce are relatively fewer. Again, I think this is very well-distributed, both geographically and intergenerationally.

  • The other point that I want to make is that all this is in supplement to the existing town halls, the existing public hearings, existing tools, as I mentioned so that the generations or people who are not so versed in writing or in reporting online, they can nevertheless go to a meeting place.

  • That meeting place, using the principle of what we call ambient computing, or civic tech, can record this face-to-face meeting into something that is digitized, that is then fused into the next stage of digitized policymaking.

  • If you’re interested, there is an open government partnership report from Taiwan called the social housing distribution case, where the social housing strategy in Taipei is deliberatively co-designed by the very people who are eligible -- single parents, people with HIV, who are homeless, and things like that.

  • You can consider the digital work there is mostly assistive technology. If it is just a website or an app, it is unfathomable that people who are eligible for social housing will lead the discussion. The process was designed with stakeholders so that no matter what kind of disability one has, one can be brought into the discussion by way of assistive technology, but they are still in their elements.

  • They are not being asked of going out of their elements to participate. Digital inclusion, I think, is very big in our digital development agenda. We have quite a few cases that we can share.

  • This puts a very heavy premium on what sounds like an extraordinary degree of civic participation. I wonder if there is a risk of basically consultation exhaustion in all this. You already spoke about this hewing in the population. It seems like a middle part of the population would have a limited amount of time to be involved in all these discussions.

  • That’s exactly right. That’s exactly right. What we are trying to do, two things. If it’s referendum for every policy, then people gets exhausted really quickly, which is why there is still a pretty high referendum threshold. For e-petitions, it is just raising the awareness that this is an important thing to be discussed about. It takes maybe just five seconds to countersign a petition.

  • People can do it much more often. Then they go on and do their business. Then after 60 days, they receive the synthetic document of the consultation that we did with the stakeholder that actually have time. Basically, it’s a public education tool, in a sense, that people just sign on for the things they are interested in, and collectively determine the agenda for discussion.

  • Still, of course, people who actually have time for face-to-face deliberation, these are a fewer fraction of people. We are not worried, as long as there is a ladder of such participation. Anyone can find a comfortable place on the ladder, and explain the more detailed or expert information.

  • Since this is on the upper ladder, fewer people attend to the people down the ladder in a way, the common language that they can understand. This is just like the public awareness of law, or aware of computational thinking, or again, with the use of fire, or whatever.

  • As long as the technology is infused by the technology into the society like this way, we’re not worried about the civil society capacity to absorb it. If it’s just a few representatives, actually, they will suffer exhaustion by trying to reflect the myriad of social changes that still needs an agenda setting on the national policymaking thesis.

  • This is, again, the same idea sandbox. Many people witness, a few participate. Even fewer can deliberate, but people just discuss this knowledge.

  • I wonder which are the theoretical sources of this kind of movement. Do you have thinkers who are some kind of ideological guides or philosophical guides for your movement?

  • Personally, I come from the open culture movement. The leading thinkers are Richard Stallman, obviously, Linus Torvalds, the maker of the Linux operating system, Clay Shirky, the people who built Wikipedia, Ward Cunningham and friends.

  • There are also socialists who worked with such open source movements, such as Manuel Castells, who worked with the Occupy people. Of course, there is also Asian thinkers also.

  • Asian, thinkers in Asia, such as Kōjin Karatani, I think. Let me just quickly make sure that I am speaking of the right person.

  • Kōjin Karatani is a little bit like David Graeber, also an associationist anarchist, [laughs] who tries to use the more open multi-stakeholder way of how we make the Internet together as a way to think about how people can collaborate in other non-Internet things as well.

  • I think we, and people who had an experience in the Internet society, we started from the point where there’s rough consensus. There is a working anarchistic model, but it only worked, admittedly, on Internet protocols and source code.

  • Then eventually, through Creative Commons, on Wikipedia, on music, and on many other things gradually. How it works on government itself, it is very much an open problem. I also see myself as doing a lot of field experiments.

  • We’re here, we still publish, and we work with researchers and socialists to try to make some theoretical understanding of the work that we’re doing by applying essentially the Internet governance model to the real world.

  • In which regards do you see room for improvement of Taiwan democracy, and to what extent is Taiwan democracy a kind of role model for other Asian countries?

  • I think we’re quite different from Estonia, because Estonia’s constitution is written after the Internet. They took a lot of Internet-enabled, paperless identity as their origin. For example, their X-Road project and other projects are often hailed as example of how to do the "digital first" thinking, if you can start from a blank canvas and apply digital thinking.

  • Taiwan, I think, is interesting because we have an operating system, and its kernel [laughs] — the constitution — was invented for a much wider area than currently, and has a very long tradition of paper, paper keeping, and a highly professional, but also very bureaucratic bureaucracy.

  • Our digital transformation, I think, is of wider applicability to older republics of hundreds of years long that also have a lot of existing inertia, how to handle public opinions, and paper-based thinking. I wouldn’t recommend a newly founded country to use Taiwan’s constitution and legal system as the basis of democracy.

  • For existing democracies trying to do digital transformation, especially on a national scale, I think Taiwan is pretty unique in having our population density, and its so-called central government, but still very accessible within the geography of essentially just a larger city.

  • If you go from Taipei to Kaohsiung, it’s just a couple hours. I think this particular geography and density, Internet connectivity, makes it much easier for us to try new experiments digitally without worrying about the odd pockets of population that are not connected, that are not online, that are not literate, and things like that.

  • We just have fiber optic lines to our offshore islands and call it a victory. I think is a particular atmosphere, but I wouldn’t say that our democratic system itself is the best, or that you need five branches of power instead of three branches. That’s for the political scientists to argue.

  • I will certainly say that we are now being very reflective, in the sense that, through the referendum acts, through the sandbox acts. Also, the sandbox acts, Parliament is essentially also doing co-creation with the people by having the people’s input in the Parliament in a much more equal basis, rather than the representative knows it all idea.

  • I would like to maybe mention this information. We’re also, before talking about how to find, that there is a, let’s say initiative to fact-check, like instant fact-checking. How is government involved in all these, and is the initiative coming from the side of government, or more like the civil society?

  • No. We’re not funding any fact-checkers, otherwise, there will be conflict of interest. We’re trying to build the legal system so that the information providers need to be accountable. They need to have a privacy policy, for example, a policy to settle disputes and so on. We’re not saying which policy is the best one, but each large service provider need to have one and be held accountable by the civil society.

  • For the fact-checkers, I know of three independent groups all working on various kind of fact-checking. Instead of taking sides, the government is providing an automated way that we syndicate, essentially, each agency and ministry’s clarification statements so we respond very quickly to rumors that’s being spread online.

  • We publish our clarifications, but in a way that is very easy for a machine-to-machine protocol to receive. If you work in the television and cable news, for example, before you report a rumor from the Internet, you can very easily check whether on the bulletin the government has already said something to counterbalance it.

  • You don’t have to take the government’s word for granted. You can still say, "The government is probably lying. [laughs] We trust this." It’s within your journalistic discretion, but at least the government is not being unresponsive, like takes 7 days, takes 60 days just to catch up on a accusation.

  • The civil society has been really active. There is a Line bot. Line is like WhatsApp. It’s an instant message platform, and this information there is particularly difficult to track because there’s no search engine for such group messages that went viral.

  • There’s a LINE bot that if you add it as a friend and you share, for example, a piece of news with them, and the bot will say this has been fact-checked or not, and provides additional information for people to read. They serve as kind of inoculation points in the epidemical kind of way.

  • This database that powers this bot, much like how spam email was handled, is by people flagging and reporting, and people who voluntarily check. There’s a huge voluntary user base that maintains this bot collectively.

  • Government, for us, is just one of the message providers. We’re not trying to dictate the relationship of those fact checkers to the media. That’s for them to figure out, but we try to make it such that we respond in an open, rapid, and structured way.

  • Can you give us a few examples of some disinformation that’s being pushed into Taiwan from outside?

  • Sure. I don’t know whether it’s from the inside or outside. [laughs] That’s the nature of disinformation. Particularly, I think even before I become the Digital Minister, there’s a recurring rumor about me, Audrey, is able to track people’s whereabouts using GPS.

  • (laughter)

  • It caused for people on the street to turn off the GPS location function of their devices, otherwise Audrey will track them via the satellites. [laughs] If I view it as something personal, then of course, I may by upset or inclined to use some legal ways to find a person who perpetuated this rumor.

  • In fact, I just very quickly responded saying that I view it as a great discussion to begin a discussion on what exactly can GPS satellites do. What kind of privacy expectations should one who mobilize and assemble on the street be? What is our current law in particular of the police collecting people’s identities when you go on mass protest like this?

  • Basically, when viewed as a kind of invitation to open a conversation, we usually find people who are worried or who share those misinformations, they still may turn into very useful contributors if you just send an invitation for people who complained the loudest to join in the kitchen so to speak. [laughs]

  • Last May, the first of May last year, there was a widely-circulated myth that the income tax filing software was broken on Mac and Linux because of the TradeVan company, the rumor says that they have a corrupt relationship with the Ministry of Finance. It was very widespread at that time.

  • We, again using the e-petition platform, invited people who complained or were worried the most into these co-creation workshops to work with the TradeVan company directly to co-create this year’s tax-filing software, which many people said is the best one they’ve ever used of any government service.

  • Again, just by creatively trans trolls into embracing trolls, hugging trolls, [laughs] to only reward constructive behavior, we were able to harness the attention and energy of people who would spread such disinformation, but without getting lost in the disinformation itself. After all, it’s just a symptom that we don’t have a good enough relationship.

  • If you have a relationship with your friend that you go to dinner every week, if you hear gossip about them, you won’t just believe the gossip or spread the gossip. You’ll check with them the next time you meet. If you meet only once a year or so, of course the gossip has room to spread.

  • I think all these are just symptoms that open government still needs work. [laughs] Instead of seeing it as a problem in itself, I think these are just symptoms.

  • What’s your opinion about the Chinese society credit system experiences? What’s your opinion about that?

  • I don’t know. [laughs] From a system design viewpoint, I think they’re uniquely positioned in doing this experiment. The experiment is unimaginable [laughs] in a democratic country, and also unimplementable without the technological inclinations of the technocrats. You really need a technical prowess and the kind of social atmosphere in order to do this experiment.

  • I’m sure that they will carry it to its logical conclusion... and we will have plenty of think about.

  • Other than that, I can’t really comment on it. First, I don’t have a say in where these experiments go. When I was young, I worked on tools and communities for people who don’t want to be subjected to this coercion and surveillance or from any of this kind of meddling, interference, to ensure they have the free software alternatives. Still now I record webinars and introduce free software tools so that activists anywhere who want a secure communication method, they know they have alternatives that doesn’t depend on off-shore data centers. They can all run it locally using just a box or whatever.

  • Beyond this tool-providing activist or hactivist role, I can’t really interfere directly with their domestic policies. [laughs] There is a sense of let’s wait and see how it plays out. I don’t know yet what kind of outcome it will be. I will say that all my friends in social system design are very eager and waiting [laughs] on what kind of results are from this kind of state control.

  • How about the hate speech in the Internet? Can you write anything, or can you be prosecuted for that?

  • The interesting thing about embracing a troll is that, in our spaces for discourse and discussion in e-petition, we have a way, for example, if someone writes 100 words, 97 of which are ad hominem attacks, hate speech, and whatever -- exclamation marks -- but three or four words of which are constructive or could be construed as constructed.

  • We establish a way for the moderator to be able to basically edit out all the 97 words, leaving only three words, and then reply substantively to the three words. Therefore, everyone who are bystanders to learn that the only way to get the ministry response is to be constructive.

  • It is still not censorship, because if you have too much time on your hands, you can click revision history and see the original version. It’s not like we sensor people.

  • It is true that once people get out with the catharsis, so to speak, and find out they haven’t been kicked out, but instead rewarded, but only for the constructive parts, they’re much more willing to be a constructive member of the community.

  • Sometimes, the hate speech is a way to get attention that they cannot otherwise get from real-world interactions. It’s become a kind of proxy, but because interaction across the Internet is shallow, it’s not a long-term relationship, so they wake up feeling very antsy and do another hate speech post.

  • The way we do our safe space and do our convergence, focused-conversation space, is to encourage people to participate, and then Aikido their sentiments into something that’s constructive. It works on roughly half of the other cases. The other half of cases, [laughs] they find some more enjoyable forums to practice their art and leave our forums alone. That’s the limit.

  • Currently, you can’t be persecuted by spreading misinformation, except on a widespread epidemic and during election periods, try to get someone elected or not elected, because these two are social and democratic foundations of the society. Otherwise, you can’t be fined just by spreading misinformation.

  • Sorry, I should have done my research. As a factual matter, are the protections on speech in Taiwan more like in the US than in, say, Europe or Canada?

  • Of course, there’s libel laws. There’s all sorts of laws that if someone feels hurt they can also apply. We’re designing something DMCA style. You can order a take down, but you have to file a lawsuit within 10 days -- I think it’s 10 days -- so there’s a way for the administration to work with the judiciary system in a balanced fashion to handle these things.

  • We also now have a anti-harassment law that one can issue, like some particular person who keeps interfering with my online and offline work, keeps stalking me, one can file some order for them to stay away. If they keep violating it, then it’s punishable. All these are done in a way that doesn’t harm free speech, as I said, somewhat differently from some European contexts.

  • Here in Taiwan, free speech is seen as something that is a value in itself. It’s not an instrumental value. I think mostly because people still remember, people younger than me don’t, but people older than me still remember when there wasn’t free speech at all. [laughs] They finally got free speech after generations of struggle. We might as well try to keep the verbatim version of it.

  • Maybe after three generations, people will decide free speech is just instrumental, and it’s not a core value in itself. But for this generation, it’s still very much a value in itself.

  • When you were talking about regulating hate speech and leaving only the substantive parts, is that the government deciding what is substantive and what is...?

  • No, this is in our e-petition or our vTaiwan discourse platforms. This is in space we manage. We’re not trying to tell PTT, Facebook, or Reddit how to do their moderation business. We leave that to the moderators.

  • So just on the e-petition?

  • On the platforms that are potentially agenda-setting for the government. We’re not passively collecting Twitter sentiment and use that as our input source. It’s as unreliable as it gets.

  • (laughter)

  • If we do passive collection and use that input source, the problem is that, by doing this, we implicitly encourage what we call precision persuasion, so Cambridge Analytica and friends. The idea is that if people can be persuaded to send signals in a way that then affects the outcome of public administration, then it’s a vicious cycle.

  • By saying we only reward voluntary contribution, people become aware that they are taking a part of democratic process instead of being passively collected. Then we have a much higher chance of convergence, because it is a long-term relationship. It’s not transactional, like passive sentiment analysis is. I don’t ever look at sentiment analysis reports, [laughs] lest it affects my judgment.

  • These are just users commenting on the platforms that you manage?

  • Right. Every budget item, every regulatory change, every e-petition, they become a social object that people can have a real discussion both online and offline about.

  • And you can see the revision history if you wanted to?

  • If you only see the three words, you can click, unedited, and see the whole hate comment?

  • Yeah, that’s right. The software that we use for, for example, the vTaiwan process, which is the civil society version of a research platform for this kind of engagement, there’s a very sophisticated algorithm that says if you’re first registering, you can’t post pictures. Or, you have to stay for how long and participate for how long before you can post links and things like that.

  • It’s a self-regulating community, powered by algorithms. Again, it’s all open source, and the process of determining the threshold, the setting itself, is also done in a multi-stakeholder way. The government doesn’t control that particular part of the platform.

  • In vTaiwan, we experimented with a lot of different ways of how the community can be more aware of itself instead of trying to use one-size-fits all solutions to moderate speech.

  • How much the general population is aware of the existence of these sophisticated tools?

  • The e-petition platform, as I said, is close to 5 million participants out of 23 million in population. It’s the second-most visited website that’s offered by the government, the top one being the real price registry of land and house prices. That’s obviously a large demand.

  • You mentioned the labor intensity of it all, but as participation grows, and hopefully it will continue to grow, when does it become too overwhelming as far as monitoring this?

  • For all the parts that doesn’t require human judgment, we are using AI, obviously. For all the parts that does require human judgment, we use crowdsourcing. The idea is that everyone who participate, it’s not just typing their sentiments or idea.

  • They’re also using simple upload and download, as you have said, just as you saw, and there are the way of putting a face on the crowd. People can see, in context, how their ideas and opinions relate to other people’s ideas and opinions through a mind map and things like that.

  • All these are being facilitated in a way that we try to release the burden of the public servants who want this kind of process by introducing tools that can effectively automate some part of it and crowdsource the other parts of it -- crowdsource meaning having participant do most of the work.

  • If you’re a citizen with some sort of problem, like a pothole in front of your apartment, how do you report that to someone?

  • It’s hard to get 5,000 people [laughs] to countersign one pot in the front of a house, but we do have local issues. There was a popular tourist destination in Taiwan called Hengchun. It’s in the south-most part of the Taiwan island.

  • There is 8,000 people who petitioned for us to station helicopters in Hengchun to serve as ambulance cars. Their closest large hospital is 90-minutes drive away. When people have a stroke, have a diving accident or whatever, it takes a very long time to get them into somewhere with equipment.

  • The petition is not done by just the local people. Instead, before they check in bed-and-breakfast, for example, they would scare people by saying, "You know what would happen if you have the diving accident? Scan this QR code and sign the petition." [laughs]

  • We have a lot of tourists signed that petition, which is why it’s raised on the national agenda. On the surface of it, it’s a Ministry of Interior problem, because the police helicopters are the Ministry of the Interior, but they say, "We don’t really have Blackhawk helicopters to spare," and, "Maybe the Minister of Defense."

  • Defense says, "No, we’re retreating from that base. Maybe it’s a transportation problem. We need a faster highway." The Ministry of Transport is called, and the Minister of Transport said, "Maybe you should just en-biggen the hospital, [laughs] to build a larger hospital there, and it’s a the Ministry of Health and Welfare."

  • I think part of the reason why the Participation Officer Network, the PO Network, is so useful is that we have a national regulation for participation officers that’s just like media officers and the parliamentary officers. In every ministry, there are teams who talk to the stakeholders.

  • This is a stable network, so that when they call each other, they will not push things around. In the regulation, we say, "If A thinks B should take charge, B thinks C should take charge, C thinks A should take charge, then everyone is in charge." [laughs]

  • All those four ministries and all those people, we drove to Hengchun and have a town hall meeting there with the multi-stakeholder mind-mapping. We spent an entire day, and many, many meetings before and after that.

  • We converged on the solution that basically says that the government should spend a lot of money -- I think hundreds of million US dollars -- to make the hospital there of top-notch equipment.

  • Then maybe we can fly the north, larger like Kaohsiung, doctors to train in these hospitals. When things happen, at least we can fly the doctors there, instead of flying patients with stroke to the north one, which doesn’t make sense anyway.

  • This allocation of budget, again, it is not something that’s totally novel by the e-petition. The e-petition gave it legitimacy, gave it visibility, and by the end of the multi-stakeholder meeting, it also gives the feasibility. Everybody can see it is the most feasible of the four potential solutions. When I bring the consultation results to the premier...

  • Every other Friday, we have such meetings, and then the next Monday, I bring it to the premier. The premier could say, "OK, let me take a look." He went there and took a look, and then allocated a budget immediately, knowing that the constituents will all be at least not unhappy, and some of them will be very happy with it.

  • This is how we respond to local issues, maybe not potholes, but the things that have 5,000 or more attention.

  • I wonder, does it go back to the consultation exhaustion issue? Is there a danger that all things are blended together, important issues, less important issues, and they’re all subject to the kind of exhaustive consultation process? I wonder, when your government came to power, whether you did a defense and foreign policy review, for instance, whether that was subjected to this process.

  • Yeah. There’s many national forums. I think the judicial reform process is significantly influenced by the g0v community’s methods before. The civil society, the Judicial Reform Foundation, did a mock national referendum using the methods, but the pension reform forum did not use much of the same methodology.

  • It all depends. It depends on the person who hosts the forum. It depends on whether they think this is something that a wider population should be engaged in or just the stakeholders. In our open, multi-stakeholder methods, anyone who declare themself stakeholder, anyone can declare the stake, becomes a stakeholder.

  • In many other national forum processes, only the heads of the associations or the heads of representative, some council, were considered stakeholders, so, they differ. For prioritization, as I said, every month we take on two cases for cross-ministerial collaboration, so the bandwidth is limited.

  • We have all the 31 ministries’ participation officers vote at the beginning of every month which two topics that we choose for this kind of multi-ministerial workshop. Otherwise, the ministry itself, the PO and their teams in various agencies, still handle it, but it’s mostly single-ministry issues.

  • As the Digital Minister, there’s no digital ministry, so I mostly work on cross-ministry issues. If it’s strictly within one ministry, usually I don’t even see it. That’s also how it’s prioritized, relatively.

  • Any other questions? Are we good?

  • Are you given a yearly budget?

  • A yearly budget for what?

  • Are you given a yearly budget?

  • Not personally. The ministries have budget. We in the Executive Yuan, we work with the ministries, so they get budget. There is, of course, a long-term plan, for example a four-year plan that we’re drafting with various ministries on social innovation, and that plan has some budget.

  • In the so-called IoT for Public Good, which lists all the environmental data and so on into one platform for people to share, that has plenty of budget. The hospital has plenty of budget. Personally, we don’t have. The entire premier and the ministers without portfolio, we share the budgets so that we have some travel budget, perhaps, [laughs] but it is strictly to pay for our necessary daily expenses.

  • How many employees working for you directly?

  • Since you asked, [laughs] it is somewhat an anomaly in my office. Usually, a minister without portfolio have maybe two or at most three staff. That is the normal case. Here, because we say we’re the Public Digital Innovation Space, the PDIS is a space, anyone is welcome to be a volunteer, as long as you’re a public servant.

  • We get volunteers from all various different ministries. At the moment, we’re, roughly speaking, 20 full-timers. We try to not recruit more than one person from a single ministry, [laughs] so we can be much more cross-pollination going on. In addition to the 20 or so full-timers, we also have 35 interns at the moment.

  • Usually the central government maybe use one or two interns, and they’re mostly in Taipei, just because of the location we’re in. But the 35 people who are our interns, they’re all over Taiwan. I work in a telecommunication/telework kind of way, so my interns can also work in any cities. Some of them are actually not in Taiwan, anyway.

  • Yes, they all work in our virtual work space, and collaboratively working on improving digital service. This is like a virtual, small office of sorts, but the salaries of all the various people from various ministries. For example, NCC still pays Yeh Ning’s salary, [laughs] and they join on a voluntary basis.

  • I don’t, for example, rate or rank his work. He provides his own assessments of his work. I just sign on it. This is a kind of peer-to-peer relationship.

  • Are those voluntaries, they are, in exclusivity, working? For instance, voluntary from the Justice Ministry, he is not working with the Justice Ministry? Is he detached?

  • Not exactly detached, though. Anyone who work here still maintain a connection with their original ministries. It’s the idea in software development we call on-site customer. We’re trying to improve the public service experience for all the public servants, so we need real public servants who can assess whether this is a good idea or not.

  • In a sense, they’re also evaluating this methodology in the context of the ministry they work in. Some of them just stay for six months or a year or so, and they go back to the ministry. This is how we work in a hub between all the different ministry who all want digital transformation, but none of them want to be exposed to too much risk.

  • Here is someone who can absorb all the risk, do all the pilots. If some of those things happen to work, then it trickles back into the ministries.

  • (applause)