 Hello everyone, this is Ding Hong, it is, hi everyone. I see there's already somebody getting to the slide-up, and if you have a mobile phone, please stay beside the phone, and with 00412 as the code. And you can ask me any question even during my presentation, and I will try to work my speech, so it includes people's questions. And also, at the end of the 20 minutes talk, I will also reserve another 20 minutes for this content of my speech entirely determined by your slide-up questions, and if you see a question somebody already asked that you want to like, you can just press like, and we'll go to the top, and I will start from the start. So without further ado, I'm very happy to be here at Codegate to share with you some stories in the past couple of years that we have worked toward the open government. This is our president, Dr. Tsai Ing-wen, when she was elected last year, I was very happy. I voted for her, not knowing that I would become a cabinet member, and I voted for her for many reasons, but the good feeling reason was that I'm an animal lover. I lived with seven cats and two dogs, and she is a fellow animal lover who lives with two cats and three dogs, and she's very much into animal welfare, even animal rights, and a lot of rights for the region needs, for diversity, for LGBTQ, and so on. And because I agree with her politics, I voted for her, but also more importantly, she believes in a political system that is post-partisan system, and this is our first family, by the way, literally our first family. And then as I mentioned in Taiwan, we have four months of transition time between the presidential election and the new cabinet actually taking into office, but those four months were very peaceful, and one of the reasons is our previous minister, our Prime Minister, Simon Zhang, who was a Google engineer, belongs to no parties, he is independent, and his successor, our current Prime Minister, Lin Quan, is also independent. So between the two independent ministers, they did a transition based on the principle of transparency and open data. Back when Simon Zhang was the Premier, he mandated that all the ICD systems, which was constructed for less than about $1 million, must be opened data by default, of course protecting national security and privacy, but otherwise it's opened by default. This made Taiwan into the first place into the global open data index. So continuing with this kind of thought, during the transition, he ordered all the ministries to produce a checkpoint document, basically have a checkpoint image of all the ministries' policymaking and upload it to the public internet for the next cabinet to download. So this is not a transition between two parties, but a transition to the general public with the checkpoint documents, and for everybody included to study the documents and understand where the country is going. So in her operations speech, Dr. Tsai Ing-wen, our president, expressed her idea. She said, before democracy in Taiwan was a clash between two opposing values, but from this point onward, we need to become a conversation between a diverse, many different values, and it is by this vision that we further the idea of open government. But Taiwan was not always like this. The idea of people participation, of radical transparency really early started during 2014, where we had to have a million people on the street and occupy the parliament for 22 days. It is a sunflower movement. During the sunflower movement, it is not just a protest because it starts by having the students taking the parliament doing the job that the legislators refused to do. The legislators refused to deliberate a trade service agreement, and the students and occupiers are trying to find a way for half a million people, anybody who shows up on the occupy to deliberate, to see what people's feelings really are to work this agreement, to build consensus. So during that time, the movement that I participated in, the Gulf Zero movement handled the information and communication technology infrastructures for the occupiers. And the Gulf Zero movement started in 2012 was this very simple hack of the government. We fork government websites. If you see a government website in Taiwan, it always ends with GOV.TW. So for example, the parliament will be LI.GOV.TW, and then we build a shadow legislative government and we have the URL LI.G0V.TW. And this solves the discoverability problem because you don't have to Google for it. You just take the official website, change the O to a zero, and you get into the shadow government, which is built using the principle of data and crowdsourcing and participation. And after forking the government websites this way, like I did with a lot of friends for the dictionary website, we abandoned our copyright, we donated to the public domain so that when the government renews this website again, it can take our contributions into the official government website, and many ministries already did that as well as local governments. So the idea is hacking the government but not demolishing or causing chaos but making patches, making a better fork in the hope that government will merge it back in. But why are there so many civic hackers in Taiwan who want to work collaboratively with government this way? I think it was because when I learned programming, when I was eight years old in 1989, it was also the year in Taiwan where we have pressed freedom after 30 years of martial law and censorship of speech. Finally, we have freedom of speech and then we have personal computers in the same year. And in the same pattern in 1996, when we had our first presidential election, was also the year where war, I went becomes affordable. So in Taiwan, when we see free software, we don't think about free as in not paying but free as in protecting the freedom of speech, the freedom of assembly, the freedom of creation, the free, the software freedoms and culture freedoms that we're, we fought so much to get and would not want to lose. So I think this is why when we see policy making no longer responding to the newest technologies and the government system doesn't make the full use of its ICT capabilities instead of just shouting and blaming the government, we say, okay, we fork the government. We tried to work some experiment away so the government works better with us. So I will very quickly describe one case where we collaboratively did a policy making on. Back in 2014, there was this company called Uber which starts its service in Taiwan. It starts operating illegally but then it starts hiring drivers with sub-professional licenses. And the whole idea of Uber was this meme, this virus of the mind, is basically saying algorithms dispatch better their regulations and so people don't have to file their regulations. And this kind of memes is very different from traditional interest groups because the interest group you're facing was one person but with this kind of meme, you're basically continuing trying to find out where this virus come from and it's very little that the state can do to a virus of the mind like this. So our idea, which we learned during the sunflower movement was just to get all the same haters, the passengers, the drivers to sit down and talk and listen to each other and come to an understanding because deliberation in this kind of common understanding really is the vaccine, the inopulation against virus of the mind. When we have considered other people's positions and arrived to something that people can live with, people become immune to future propaganda or rumors or things like that. So we started by collecting all the facts that we have and then determined using machine learning and a gamified interface, people's feelings toward those facts. And then we start with the ideas and the best ideas are the ones that take care of most people's feelings and then we turn those ideas into regulation. So before we introduced this kind of policy making, the old way of making policies was just the government talking with some private sector association heads and some academics or some professors, but they use an expert language and do not speak to the public before they arrive to a conclusion. But people on the street can also talk about this, but they more and more use one word for different meanings so that people live essentially in different realities. And when people don't have the same agreement on the basic facts, they start paying attention to each other's feelings and ideas become ideologies. Ideologies to me are even more dangerous kind of buyers of the mind because they blind us to new facts and blind us to each other's feelings. So we start always by releasing open data and asking the private sector and civil society to share as much factual data as possible as the ground truth for discussion. And then on a fielding stage, we use this Polis system for online facilitation for three weeks. The whole idea is that you can go to this website using your phone and then answer, yes or no, to the question of, do you feel that the insurance is important, for example? And so on. As you answer yes or no, your position changes in this principle component visualization graph which dynamically recalculates your position among your Twitter friends and your Facebook friends. And this has two important effect. First, even people who are against you are your friends. It's just you didn't talk about this over dinner. So it makes people not antagonize each other, consider each other as enemies. And second, it makes it people see that people's position can change because we say we only take the sentiments that is agreed by a super majority of people, 80% or more of people as the binding agenda. So the idea is that the terms that we negotiated with Uber is determined collectively by everybody involved or you have to convince everybody of your feelings. And once you answer a few yes or no questions, you can then propose your feelings and then for other people to vote on is an iterative, recursive process. So after three weeks or so, we actually have a strong consensus of seven items that more than 80% of people agree with, which are consistent across all the different sectors. And it is through to this consensus, we run a live consultation, everything is live screen, everything is on the record, people can just say anything knowing that it was listened to by thousands of people so they won't just randomly say things and go and deny it, but everything is on the record. And so after getting everybody understanding the feasibility of these consensus items, we then translate it into regulation and then finally ratify it. And so this ratification is very important because then Uber just today announced that they will agree to play by these rules starting tomorrow. And so this is knowing that the passengers and of course the drivers and so on can all get behind these consensus items. So after I become the digital minister, I started this public digital innovation space, which is a way for agile open source hacker like mentality in the public sector for people who work in the career public service to join. And I started by having this direct line to the digital minister so anybody can ask me any question and I will try to answer it within 24 hours. But this is not one-on-one because every answer is then public and send to thousands of subscribers. The journalist everybody is equal on this platform so it's a frequently asked questions and I won't have to answer the same questions twice. And the second thing that I did was I practiced radical transparency so that all the meetings I held, all the interviews that I give and every venues that I chaired and all the stakeholder meeting are turned into this real-time transcript which has this XML encoded and so you can do a lot of semantic analysis on it. So even Uber's David Bluth when he visited me gets 360 recorded and turned into a transcript so that the missing stakeholders in the room can nevertheless participate in future discussions without getting a partial information. So we do our agenda setting using all kinds of very startup-like methods and we also pay attention to cybersecurity. The first thing I did literally when becoming a digital minister is to recompile the Linux kernel because the Linux kernel of our internal system are to owed and they will not run secure computing and Linux containers. And so after we compile in the tunnel and they install the cybersecurity product called sensor.io which then takes a lot of collaborative open source systems into our everyday views and so it got popularized into the national administration and we're now spreading it to the local administrations and it has Kanban board, it has either Calc, either Pat and it has a lot of collaboration tools. So to finish my talk, I would like to play a PDIS movie to introduce the PDIS members who cannot be here physically. There's only three of us here but many others would want to make an appearance so would you please play the video? Hello everyone, let's take a look at what ingredients we've prepared for you today. Ginger, garlic, green onion, the information provided by the public, the PDIS, and a piece of broken heart to prepare this open source system hot pot. Let me tell you, although the open source system is very important but it's also very important to ensure the human privacy. The key to hot pot is the soup. The soup is made with onion, carrot, celery, a small pot for 7-4 hours and then it's made with hot pot, celery, and this is the soup. Wait, wait, no, no, no. Let me tell you, my mother taught us to cook for 800 years. In this soup, you must have green onion, ginger, garlic, chili, pepper, stir-fry it first, then add some Chinese herbs, star anise, bay leaves, cinnamon, and then... What are you talking about? Don't make a fuss. Let the online audience participate in the public government's open source network to tell everyone to participate in the open source system and make a big pot of hot pot. Wait. Did he say wait? The ingredients are fresh. Who is responsible for this? It's me. Who is responsible for this hot pot? It's me. Me. Me. Me. The most important thing is the sauce. So I said the sauce? In general, can we eat the sauce? Can we eat the hot pot? Can we eat the soup? Can we eat the sauce? Don't worry. The process of making hot pot is not only the first step but also the major part of the process. And in the process, we will also take care of the sound of different films. Only in this way can we make all the ingredients make the best out of the happiness. Hot pot, the ingredients and cooking method are all shared inside. Participants and newcomers can do the research and edit the recipe. Let the audience understand the cooking process. Don't worry. I'm not afraid to eat it. Now, we will put the TWS public storage space into the network. So that everyone can update the open-source information at any time. TWS, please subscribe. Happy New Year. Switch to Slido. So that I will start answering your questions. There's now many questions. I may or may not be able to guess through it all. I'll try. The top question at the moment is what do I think of IoT hacking? Because the Internet of Things is evolving day by day and can affect actual life. How do we defend the Internet of Things space? This is a great question. Security for me has two meanings. One is a property of a system. For example, you can do proof carrying code. You can do a lot of verification. You can do a lot of mathematics to ensure the property of a system. But this is a mathematical definition. There's also the social psychological, which is people feeling secure. Now, when these two match, of course, it is the best thing that you have a provable secure system with people who feel secure about it. But not everybody is a mathematician. So you get disconnects. You get things that are pretty secure, but people are not willing to believe that it's secure. Or the worst case, you get security theater, which is the mathematics says that it's all wrong. It doesn't have the property, but people are allowed to believe that it should be secure. And the last case is very valid because when people then get shocked that it doesn't really have the property, people tend to panic instead of finding interesting or useful way around the question. So I really think literacy is the most important thing, which is why in Taiwan, instead of starting programming at a primary school, we start at junior high school. But at primary school, what we focus on is computational thinking, it's design thinking, it's media literacy. It's basically understand how algorithm works instead of just learning to code as one class. And we say you have to integrate it with all the classes, all the disciplines, so people can know that code is behind all those disciplines and how to think in terms of code and collaborate not just with humans, but also with machines. And so when people are learned in this way, I believe people will be much more likely to understand the mathematical explanations and ask of the technology instead of having the technology demand of us. And so when it be IoT or autonomous car or so on, we will get teenagers asking to run in the simulation to test its mathematical properties to really make an effort to understand its system and how you interact with human beings instead of blindly accepting any advertisement or something. So I think this is of utmost importance. And to the person operating the computer, can you click the left button corner? There's a link that says presentation mode. It will make it much easier to read. Thank you. Right, and then I'm very sorry that I didn't read Korean. So I should probably feed it to machine translation at this point, which I will attempt to do unnecessarily. Unnecessarily working, let's see. So the question is, if machine translating is too big-beliefed. Wow, the AI really doesn't do a good job on this one. I don't really understand the question. I'm so sorry. So I will read what Google think of it. Good morning. The additional minister is somewhat unfamiliar. There seem to be no exact magic department in our country. The question is, is there a limitation of political power in establishing and practicing digital policy in Taiwan? Welcome to Korea. Well, thank you for the welcome. This is my first idea. So I don't know whether there is a equivalent of PEDIS, but we did learn a lot from white papers published by the government 3.0 effort here. I think that's the closest to ask the open government that we've been doing. And I look forward tomorrow to exchange ideas, not just around IoT as more city, but through the government and government modernization and digital administration with the Gulf 3.0 people. So at the moment, I don't have answers, but I have a lot of questions which I should ask tomorrow. So another question pertains to my future plan. The question was, until when will I be a civic hacker? Do I have ultimate goals in terms of democratic value? This is a great question. So I was a civic hacker since at least the blue ribbon EFF days, that's 1996. So it's a very long time now. I plan to be a civic hacker all my life. So I think the idea of civic hacking is really a refusal to feel helpless. So if you've ever felt helpless by the bureaucratic system as a one, the internet is great because you can always find like-minded people everywhere in the world and somehow link with solidarity and share your contributions, your hopes, your dreams, your fears. And the idea here is that every progress that we make is meaningful in itself. There is no ultimate value that everybody should go to. In a democratic society, everyone is their own dimension just as we saw in a principle component analysis on policy. So everybody determined their life's direction. And if we happen to overlap in our vectors, then we can share and collaborate. But I don't believe in one ultimate value that everybody should go to. If I am, I wouldn't be an anarchist. The other question pertains to what do I think the bio-desecurity, I think I already answered that. The next thing is what social network do I use mostly? This is a good question. If you go to my Facebook page, I have two plugins. One is the link to the Newsfeed eradicator, which when you install, you remove the Newsfeed altogether. And then the next one is called Silent FB. If you install it, all the pictures in Facebook becomes greyscale. So this is how I guard my mind against mental pollution. The idea is that when I use Facebook, I expect what is that I'm seeing. I have to consciously make a decision to go to someone or some group's page and look at this content instead of having the algorithm feed me information. And even if I look at a picture on Facebook, it is always greyscale until I move my mouse. I'm a picture where it turns colorful so that my sentiment will not be manipulated by the algorithm. So I still use Facebook, but with cybersecurity, well, mental, psychological security, defense system. And I also use Twitter. I used to use Google+, and that's not much anymore. But mostly I just join IRC as everybody does nowadays through gateways to telegram, through gateways to Slack. And most use system at the moment is Rocket Chat, which I have set up internally in the administration for all the PDC people to collaborate with all the ministries in Taiwan. Basically, we have each ministry, 32 ministry, pick one to four people as the C participation officers who learn how to use Rocket Chat, how to use Etherpad, how to use those collaborative tools, and they then can teach their ministries to how to use those collaborative tools. So we're trying to set up social networks but internally and in a secure fashion. Do many people in Taiwan contribute to open source projects? Yes, our local annual conference, the Coast Cup conference is thousands of people every year with more people growing. And then practically every language has its own meetups and its conferences. We used to run this open source developer conference, which is also hundreds, not thousands of people, but because each language has grew out of the venue. Now every language has its own large conference as well as new technologies like Dr. Renita's TensorFlow, whatever. So yes, we did contribute to open source projects. And it's also because Taiwan, software is at many times a subsidiary to the hardware manufacturers. So software is a way to save money. And so people would of course want to donate their codes because then the community maintains it instead of having a hardware company maintain their own software system. I think that also contributes to the economic situation. What is the best way for people to encourage other people to join and contribute open source projects? Well, it depends. For me, the easiest way is to work with journalists and people who want to interview me and say I will not agree to an interview unless you accept the Creative Commons license. And people will ask what is Creative Commons and then I give them a lecture on Creative Commons. So the basic idea is that we have a public domain at Commons for everybody to contribute to. And then people can just think it's just for hackers, just for codeers. But I would tell people that no, if you write articles, if you remix music, if you make odds, you can also contribute to the Commons. It's not just for codeers. And I find people much more receptive this way because many creators want their creations to be seen by a lot of people. And if you show them that Creative Commons is an effective way for their work to be seen widely then they will join with a lot more motivation. I don't know about digital minister. Can you explain about digital minister? Of course. What you're seeing now is the analog version of digital minister. I usually present using virtual reality and robotics and whatever. And the idea of digital ministering in Taiwan is to work with all the ministries. I don't have my own ministry. Work with all the ministries to make sure that when they turn their paper-based paperwork workflow system into digital tools, they don't blindly replicate the workflow of paper-based tools of the same bureaucratic system. The work of digital minister is to look at the existing workflows and then determine whether digital tools can change the workflow so that they can work across units, across departments, across ministries. And this is something unique with digital tools because with paper-based tools or with radio and television, it's very easy for one person to speak to thousands of people. But only with digital tools like Slido can we listen to thousands of people and if thousands of people listen to each other. So the digital minister's work is just to introduce this kind of scalable listening to the administration. The other question says, what is hacking to me? For me, hacking is to understand the detail of a system and you understand it so, so much, you immerse into it so that you feel that the system and you becomes a symbiotic connection you can feel with the system. And then after getting this kind of immersiveness, of course you'll find the shortcomings, the shortcomings of the system and then you'll find that it's not a loophole and so on in the system. And therefore, of course, black hats exploit it and white hats patch it. But for me, I'm a hacker with no hats now. When I immerse myself with the bureaucratic system, I try to find a new system that doesn't suffer from the same loopholes but will not exploit the existing system nor will I want to fix it. I would just be building a demonstration of a new kind of system that doesn't suffer from the same loophole. Is there any difficulty running PEDIS? The open government and open data movement seems really nice but also difficult to running. I think the lesson that I learned running PEDIS is that the public service system is comprised with career public servants and the career public servants are already pretty overworked so if instead of paper data they now have to also prepare open data, it doubles their work and they need to go home later in the day, maybe 10 p.m. and they will rebel against you and this will be a lot of difficulty. But if instead you introduce machines, you introduce workload that makes their workload lighter and you introduce automation to make their lives quality better so that all the automation is done by machines and human do value judgmental work that requires human value judgment, then they derive much more meaning in their work and also they can go home earlier. So if I push all the regulations and everything without thinking about the life quality of public service, of course it will be difficult but because so far all the work that I do is crowdsourced from the participation offices of every ministry who further their own agenda and lighten their workload and they can go home earlier in that way I encounter very little resistance. A person said, hack me. That is exactly what I'm doing. Did you already feel the spirit? There's the other question that's written in Korean. Let me try to look with this LSTM network. Hello, I am jealous that I will meet the minister of the new concept of the department but I do not see the information. I wonder if the minister is involved in making it legal. Well, we don't have a digital department that says for sure but I think what we have is better. We have each ministry participating in this global workspace inside the government in the form of a collaboration platform. The new theories of human consciousness and animal consciousness now says that maybe we are conscious not because that we have more neurons compared to earlier animals it's because we have a few neurons that encompass the whole of the brain that provides a global workspace for all the parts of the brain to work together and this is what I'm trying to do. It is not to set up one more cortex one more compartment in the government but rather have all the departments in the government gradually become self-aware and aware of each other's work. I think this is the main idea. The other question says share your transcript of meeting with Korean government people tomorrow. Yeah, of course. Anything that I do on the record I will first go through this machine learning based transcription system and then edit it. And usually I give 10 working days for me and the guests and for everybody involved to make edits so that it makes sure that we don't do typos and if they can supplement information they can do and so on and afterwards I will publish it on pdstudtw. If you go to pdstudtw you can click track and then you will see all the meetings and everything that I had with government people before. I'm afraid that this is the last question that I will answer because I only have one minute now. It's another Korean question. It says do you think it is necessary to open public data for the public's right to know? Yes, of course. And Taiwan has a relatively new SOIA freedom of information law passed in 2005 and the open data movement in Taiwan is basically saying the public information is good but its public meaning is read early but open means read right. It means that you can take it and make a remix, make a visualization, make an interactive game out of it so we use the CC by permissive license and it's not just public information but also data meaning that it needs to be published in a structured way so that the computer can process it as easy as human beings and this is important because when more we rely on data to determine the policy so if we cannot present evidence as part of the narrative we will never build trust with people and it's only through making the model and the decision evidence possible not just the human readable slides can we win the trust of people so before we ask people to trust us we need to trust people to not misinterpret those data and this is the main work that I've been doing so now we're at time thank you so much for listening, thank you