2020-05-11 Interview with Andrew Leonard

Hi! Would you like to repeat your question?

[laughs] Actually, that’s kind of a question for maybe later, but the description of you as the face of Taiwan’s soft power I thought was pretty interesting. I’m wondering if that rings true to you.

In Taiwan, we’re translating this hashtag, #TaiwanCanHelp, into 暖實力, which means “a warm power,” and not a soft power, because soft power is maybe contrasted to the more hard, the coercive power, so the hard power.

At the moment the PRC is developing something that’s not hot power. That’s sharp power, which is different from hot power in that they use it in a very strategic fashion. We also think that our soft power need to upgrade. We upgraded to warm power, or 暖實力, and symbolized by the hashtag #TaiwanCanHelp.

I’m happy if people see me as one of the ambassadors of this new massage. TaiwanCanHelp is not just soft power. It’s not just about credibility and trust, although these are, of course, of utmost importance, but rather credibility and trust on solving the issues you care about. That’s the essence of #TaiwanCanHelp.

That’s a big part of what I want to try and get out in this story. I want to start very early on and figure out how you joined the open source tribe.

I know a little bit about your background. That you had some rough going early on in school, some bullying, some health problems, and then you dropped out and self-taught yourself over the Internet.

At what point did you know that information technology was the path that you were going to take? How did you first get exposed to the core of gift economy, the open source principle?

Computer programming I started very early on, when I was eight years old, but it’s not until when I was 12 years old when I first dialed in to the Internet. I always call myself a digital migrant, not quite a digital native, because I did remember the analog days, going to libraries, tape cassettes, and things like that.

When I was 12 when I joined, it was 1993. There was no graphical browsers, yet. The World Wide Web was just being invented.

Gopher and FTP Rkey and so on were the dominant players. I personally fell in love with the Gopher technology, and also was very frequent in the dial-up BBS that are being transplanted onto the Internet into a kind of massively-multiplayer BBS. They are still around.

The point is that I witnessed for myself as an enthusiast and as an early adopter how World Wide Web just boardified everything, and helped I think mostly by Gopher’s restrictive copyright license at the time. Gopher was refusing to hand out license in a non-exclusive, open fashion.

You have to actually talk to the Gopher makers to operate your own Gopher server, whereas as we all know, Tim Berners-Lee did nothing of that sort and said everybody is welcome to fork my ideas. I paraphrase.

That’s changed the landscape immediately. I see most of the Gopher sites I frequent just become World-Wide-Web-only sites in a couple of months’ time. By the time that Gopher decided to offer the software under GPL, I think that’s already after Google. [laughs] It’s already gone.

That really taught me the importance of open innovation. It’s not just open source. It’s about a culture, about a humility of a great designer, as Tim Berners-Lee is, that says there is many things I have never thought about, and I don’t want to be this benevolent dictator. Instead, I will just let each webmaster master their own changes to my idea. I think this is something that really affected me ever since.

It’s interesting what you just said about the benevolent dictator. I, too, right at that time as a reporter, was falling in love with these same things, and I ended up spending most of the ‘90s tracking down all these probably pioneers, and profiling them, and writing about them, and really buying in to this gift economy, the power of the collective intelligence.

Then I watched over the next 10 years as those promises did not seem to be realized.

Oh, really? Which? [laughs]

I thought access to this global information network that leveraged all of our individual talents to build this would increase our intelligence, we would be able to solve problems better, and it would be a progressive force for democracy.

And we’d make new problems even quicker. [laughs] Cause new problems, that is.

I feel it’s even worse than that. In the United States specifically, I feel like the Internet allowed everybody to seek their own truth in echo bubbles and actually contributed to polarization and weaponized the culture wars.

One of the things, when I first encountered your work, I was like, “Wow, this guy still believes in all these things from that time.” I wonder how you’ve kept your faith in all of this and how Taiwan has managed to avoid the information polarization that is true in Europe. It’s true in the United States. There’s been a big divergence, and the path is different. That’s a big question. Go for it.

Sure. I think it’s mostly fun and games, literally. [laughs] The counter-disinformation strategy, for example, is called here. It’s called Humor Over Rumor. You don’t see that in the US much nowadays.

I think that shows a fundamentally different driving sentiment. Whereas the US politics, speaking as purely an outside observer, many narratives that you just mentioned about escalating culture wars takes on a battle or war, a warfare metaphor, and therefore privileges outrage as your primary sentiment.

Here, we use, for example, when we talk about disinformation, we don’t use the F-word, which alienates journalists such as you and my parents – fake news, I mean – but rather we say that this disinformation does intentional harm to the public, much like a virus.

We used an epidemiological metaphor long before coronavirus was a thing. If you would use that as a metaphor, then basically, you know that mental hygiene, [laughs] hand sanitation – that is to say, Facebook Feed Eradicator or whatever interventions – to decrease the basic transmission value of disinformation, of the intentional harmful untruth.

Also, inoculation. Making sure that people receive a different kind of sentiment that is based on humor, fun, and things like that is going to be what’s at stake. Just as you cannot really sit down and negotiate with the coronavirus, people cannot sit down and negotiate with memes either.

It all depends on which are the driving sentiments. We happened to chose humor and fun as our driving sentiment. I think that is where it really diverges.

That seems to be maybe coming in a little later. I’m also curious about, I talked to a guy who studies disinformation for the Institute of the Future, and he’s spent time in Taiwan, has really been tracking things. He thinks that Taiwan has the most active civil tech society in the world.

Oh, yeah, certainly.

Where does that come from? What is that rooted in? What’s unique about Taiwan’s development that has given you such a vibrant civil tech society?

First of all, I think this is just this DIY culture. We are, of course, still and were – when you’re around in Taiwan – producers of much of the component that led to the personal computer revolution.

If people are not happy with how IBM designs its personal computer, they just try out a new design. It’s fundamentally very democratic. Around the same time, the martial law was lifted. People see how the press freedom and how the freedom of assembly and so on can also shape the society together.

Instead of a bunch of geeks in IT doing digital things and another bunch of people studying public administration and politics doing democracy, in Taiwan, it’s literally the same generation. The same generation could start to do both.

Because of that, our first presidential election in 1996 is already very much by all sides powered by the World Wide Web and related technologies. For us, there was no democracy before the Internet. The democracy comes with the Internet. I think that makes sure that the talent is not split between different subdisciplines, but rather integrate in an interdisciplinary fashion.

Actually, the very first piece I wrote for Wired in 1994 was about Taiwan’s success in becoming the component-maker of the world. Further research learning how you first got into semiconductors, it’s almost an open source story. You got a really sweet deal on IP from ICRA and ran with it.

That’s right.

That, the story of Taiwan and semiconductors has been told over and over and over again in Taiwan. Is that connected to what you just said there, this knowledge that…?

Yes, I would say so, definitely. In Taiwan, I think it was never a place where we think that, during the democratization, that the technologies come from top, above, and that people must adhere to some technology.

That, I include actually the constitution, the fundamental rules of democracy. During that period, Taiwan changed its constitution many times. [laughs] Even the constitution is something that you can amend very quickly and in rapid succession for, I think, at least five times in just a few years, span of a few years.

Actually, seven times, one of which is unconstitutional. Anyway, from [laughs] ‘91 to 2005, the constitution itself went through seven amendments, six of which are legal. All of those is an imprint on our generation that even the constitution itself is a social technology.

That democracy itself is a technology, and much as you can try different semiconductor design, you can try different design. If you look at the revision history of the seven constitutional amendments, they prescribe very different politics.

It’s nothing like the original version anymore. That’s what happens when people like Linus need to adapt a very large system design to run on very large machines to run on very small personal computers. That metaphor, I think, is also very apt to describe our democratization process.

That’s great. That very much explains, I love that metaphor of the generation that grows up with both democracy and computers. What connects that same generation to open source principles?

Open source principles, I think, in Taiwan, we never had a schism of free software/open source. I think I helped choosing the English name of the first nonprofit in Taiwan furthering that ideal. I think it was in ‘99 or 2000, and it’s called the Software Liberty Association Taiwan, or SLAT.

The reason why we – why really I, but [laughs] we – chose the software liberty, as opposed to open source, which was an economic argument, or free software, which is a communist – in all due respect, communist – argument, I think, is to avoid this basic left/right narrative.

It’s two movements, actually, but with the same technological implementation. It’s very interesting. I’m sure you know all about it. In Taiwan, we say software liberty by saying that no, this is part of democratization process.

People who participate in software making, if they run it as a democracy, then we can enjoy collaborative governance with all the, not users anymore, but rather really citizens in your community and so on.

That formed our initial argument, which with time, is also what the floss movement would mainly adapt with the idea of public money, public code, and things like that. I think we did not have that schism and so were able to leapfrog a little bit of the implementation.

Then many companies would think, “Oh, I will just then…” Open core or whatever adaptations that they have to make is seen as both practical but also democratic. The focus was also on governance, or even primarily on governance.

Does that lack of what you said, left/right schism, does that translate more generally into the political sphere? On the one hand, it seems like the success of just your appointment and the intercommunication of g0v and Taiwan, you can explain it by saying civil tech is more aligned with the DPP.

It also seems like, looking a little deeper, that you were originally brought into the consultative process by the previous administration…

That’s right.

…and Jaclyn Tsai. Again, in the US, this is our core problem, is that everything is a political issue. Even now, we’re arguing about how many people have died from COVID-19. Whether you’re on the right or the left, you have a different opinion. Long-winded, but how does the civil tech intersect with the party politics situation?

There’s only one party in Taiwan that I publicly supported, but I’m not a party member. It’s the Can’t Stop This Party, or literally, the Very Happy Party. It’s just exactly what you would imagine such a party would be like.

It’s very much like the Icelandic Pirates or whatever. It’s a mock party, but also serious. They have a civic counselor, and all the YouTubers are in-line with that party manifesto, which is to put a stop to 出征, to “expeditions”.

Basically, outrage, rage posts, to put an end to rage posts, and to facilitate democratic conversation, which is a very laudable goal. Their logo is anything but serious. It’s a YouTube logo with the triangle switched to point downwards. [laughs]

I think that that really symbolizes a lot of things. For us, first of all, it’s not that they are not professional YouTubers. They are professional YouTubers. It’s not they are not political. They are very political.

In fact, right after this interview, I’m going to film a very short clip for a creator in Thailand that want to share the Taiwan anti-COVID thing. Instead of reaching to our Ministry of Foreign Affairs, he reached to Adi, which is the most high-profile knowledge YouTuber who teaches English.

He started a crowdfunding campaign that builds this website, taiwancanhelp.us. Empowered by not only professional designers, but if you scroll down a little bit, the MOOC that they feature is actually from a leading epidemiologist who also happened to be our vice president.

Track two and track one diplomacy is very blurred in this formation. That, I think, shows the fluidity of non-partisan or trans-partisan politics. It helps that Vice President Chen Chien-jen is independent.

He doesn’t belong to any political party. Neither do I. In the cabinet, there’s more independent members than members of any party. In our nine horizontal ministers – minister to coordinate other vertical ministries – I think there’s only two out of nine, even not sure about the two, have a party affiliation.

I’m not the exception. The horizontal minister that’s sitting next to me, Minister John Deng, in charge of trade negotiations – very important now in a Taiwan-US FDA relationship – he was the Minister of Economy from the Ma Ying-jeou administration, from the previous KMT administration.

This is the norm. What I’m trying to say is that it’s not because of the civic tech. It’s about that we have a different constitutional modality that makes the ministers twice-removed from elections.

Direct election of the president, who appoints the prime minister, who appoints the cabinet. Because of that, we are able to enjoy a remarkably party-free deliberation. We propose draft bills which are then sent to legislature for amendments, and party politics enters then. The core of the proposed bills come from the administration, which is decidedly, have a non-partisan feel.

That is really interesting. I wasn’t very clear on that before. That’s an actual structural design.

Constitutional design, yes.

When you say how Taiwan can help, when my editor was like, “How can we make this relevant to an American audience?” That’s one aspect of it. You’re asked all the time, I’m sure, how do these lessons translate to other polities? How do you approach that answer?

That’s a very, very good question, and I’m going to fetch a cup of coffee before answering that.

Please.

Be right back.

No problem.

I can’t hear you now. There’s no sound.

OK, right. I’m back. Shared experiences help, so this is Starbucks. [laughs] When I work with Silicon Valley companies, mostly telecommuting, social text, there was a colleague who sent me bottles of wine. Not very expensive ones, just bottles of wine from Napa Valley so that we can video conference and then drink the same wine.

[laughs] That’s nice.

A social object. [laughs]

A lot of that going on in the United States lately. We’re all doing weird Zoom…

Hair-cutting at home and all that. I think I wouldn’t call it a translation, just as I wouldn’t call the Sunflower movement occupy of the parliament. What we are doing now in the administration for open government, I wouldn’t say that we translated the occupiers’ work to the administration.

First of all, that’s called cooptation. [laughs] I really don’t do that. I would simply say that they offered an existential proof. They made the fear, uncertainty, and doubt go away. Previously, when people say that…

The Sunflower movement?

Yes. Previously, when people say, “Oh, there’s half a million people on the street, many more online,” people always associate that with chaos. They’re even call themselves leaderless. How is that even possible?

Well, actually, there is more than 20 NGOs, so it’s polycentered. It’s not really decentralized. In any case, it’s more than one representation that people are so accustomed of. Then after 22 days of occupy in the parliament, there is nobody dead, no one missing.

It’s all very civil. People converged on the shared vision four demand, not one less. Anyone who participated in that changes from within so that they are much more willing to trust that a bunch of strangers in a well-facilitated place can produce something like a rough consensus out of differing positions.

We take that and then designed the administration to fit the new political mandate, the new societal norm, the new societal expectations. In a sense, we’re just channels of that post-occupy energy.

I wouldn’t say that we’re translating anything that occupiers have done. I’m rather saying that we’re taking that energy and designing an interactive platform that can carry that energy. In that sense, the US also saw its social media campaigns of hope and fear that take the post-Occupy Wall Street psychic energy and channel it into novel organization capabilities.

It’s just, of course, the culture is different, so the channels are different. I’ll stop there. [laughs] In any case, what I’m trying to say is that it’s never too late. Simple social innovations, if you amplify it across jurisdictions.

For example, recently, I’ve been sharing this incentive design that you can use right away. It’s how we spread the use medical mask. We build medical mask as something that protects the person who wear it, because A, it reminds people to not touch their face.

B, it reminds themselves of proper hand sanitation rules. This is all true. This is not about the property of the mask material itself. This is more a psychological reminder of self.

Because of that, it enabled if you have a 50-people gathering and only 5 people wear mask, they will be able to ask the other 45 people, “Hey, are you taking care of yourself? Why don’t you put on a mask?” in a very individualistic and not at all collectivist way of design.

We’re just a friendly reminder that you need to protect yourself. Actually, we all know the primary benefit is that these five people gets protected, because a mask primarily protect other people. We don’t touch on that point. We don’t even say that. It’s just implied.

Because of that, it was able to spread as a community norm much more quickly than many other jurisdictions, because it’s capitalizing on the self-interest, instead of the community interest, of people. It doesn’t rely on altruism.

Many mechanism designs like that, that comes from the civil society, what we at the administration do is just to amplify it and communicate that on the daily press conference. There’s many nuggets like that, I think, transmits very quickly and easily without any cultural translation needed.

If that kind of participatory mechanism design eventually becomes the norm, then you will see a sea-change, and then people will start to think how to collaborate with other different people, rather than to cast them as others.

I’m very optimist on that, because even five, six years ago, in early 2014, if you ask a random person on the street, “Do you think there will be ever a stop to the bipartisan fight?” they will say, “No.”

If you say, if people who voluntarily participate in online platforms can actually take away, for example, the plastic straws…Which is not here anymore, and just ban plastic straws just by a 16-years-old starting an e-petition, they were like, “You’re crazy.” In just five very short years, we’re here.

Just to drill down on the masks, because that’s…I’ve been following you on Twitter since I started first writing about COVID-19. I actually have my own little newsletter.

Oh, you do? [laughs]

It’s about Szechuan food, globalization, and Taoism.

Szechuan food? [laughs] OK. That’s great. My grandpa came from Szechuan.

Oh, really?

Yeah, my father’s father. No kidding, no kidding. I am still enjoying spicy food.

Well, my first experience of it was in Taiwan. I started writing about Taiwan, because you were doing so well. Then I saw the video where you talked about the mask hack. I know that there was an engineer in Tainan who came up with the idea.

Then his server got overloaded, because it was so popular. Apparently, you connected him with some Google engineers. Can you walk me through the granular details of that? That story is fascinating.

In the very beginning, we sent all the medical mask in store that was available to non-medical workers to the convenience stores with a rule that each person can only purchase three at a time. There’s no way for the convenience stores to do cross-chain check.

The same person can purchase three mask and then go to, like from 7-Eleven to Family Mart and then get another three mask. Add to that people know that the supply come in on 2:00 AM or something.

People can ride their scooters and then just go back and forth and collect all the available stock before anybody else has any time to do that. By the people wake up at 6:00 or 7:00, all of the stock is gone.

It’s not very friendly, to say the least. However, Wu Jian-wei in Tainan found that his messenger – Line, actually – gets swamped with a lot of feedback from his friends and families on saying, “Hey, this convenience store still have some mask.”

Maybe they are off a big road, so harder to discover, and “Hey, this runs out of stock,” and so on. It’s not very helpful, because not everybody have in their mind the location of each store. There’s a lot of noise and not enough signal.

He thinks, “Oh, there’s something called the Google Place API. I could just take all those incoming messages and visualize that in a map.” Once he got the visualization done, he found out, “Oh, you can just report on the map,” which is not a new idea. It’s Ushahidi. It’s exactly the same idea as Ushahidi. Then he just code it up in a couple hours.

What he coded up was people reporting individually what was in stock. It wasn’t automatic?

Right, so exactly like Ushahidi. Then he intended that to use for his friends and families, but one of that went viral. Basically, all the major media started reporting it.

Then he went to lunch, he went back, he owes Google 20K US dollars because of the API usage. He had to shut it down, [laughs] and then redesign it with a bunch of back end engineers how to reduce the use of Google Play’s API.

I was part of the people who use that map. I was part of the people who contributed to his bill of API usage. [laughs] Even though he shut it down, I still have screenshots of when it was working.

The very next day, it’s a Monday, I went to the premier’s cabinet meeting on mask rationing, and I showed that to our premier, saying that, “Here is a really good idea. Why don’t we just show the availability of mask?”

He looks at it, and he understood it very quickly, saying, “Oh, this is just like a GPS navigation map, right? It shows red, means that this road it shorter, but you shouldn’t take it, because it’s out of stock. It shows green, because this one is further away, but actually it has stock. You should move toward it.”

It’s exactly like a navigation aid. It’s a very good metaphor. I am like, “Yes.” My work should be that we amplify that idea and make sure that it become a public infrastructure. He is like, “Sure, what does it take for that to happen?”

I’m like, “Why don’t we ration through pharmacies, but publish the pharmacies’ stock in real time?” Now, the National Health Insurance Agency, actually, all the credits should go to them, because they implement that literally in a day for you to swipe your NHI card and get your mask rations and then report directly back to the NHI database.

It is to the credit, also to the pharmacists, for agreeing to publish. In many other jurisdictions, the FOIA law only mandates you publish by the end of the day or the end of the week. That’s considered sufficient disclosure.

What we’re asking is very different. We’re asking publishing every 30 seconds. That means that no person would reveal those numbers before it goes out, which is very contrary to how public administration usually operates.

They, meaning the National Health Insurance Agency, didn’t agree on that in the first day. He said, “OK, we will have our own querying platform, and then we publish maybe every hour, or every couple of hours, every hour for the outside third parties to implement.”

I am like, “OK, it’s not good enough, but I’m not going to argue. I’m just going to start working with these guys.” I take the schema and then started publishing it to the g0v collective. Then the g0v community, it’s not just Howard Wu.

It’s also Kiang Jian Ming-soon, and also a very popular Line bot called [non-English speech] from the Center of Disease Control is an interactive Line bot. At the first stage, there’s at least three very mature implementations, two maps and one chat bot, that shows the availability of masks every hour.

Very quickly, the NHI discovered that OK, their website offered real-time information. The third-party tools offer per-hour information. Their website got denial of service really quickly. They really couldn’t sustain the tens of millions of people who want to know their real-time stock, because it’s not useful to know the numbers one hour ago.

Maybe the stock is gone. By now, it’s only useful if it’s real time. Because it literally DDoSed their entire network, they decided to take that down and redirect all the querying information to this makeshift website that I personally coded up, which is mask.pdis.

It shows a list of, at that time, only a handful of applications. Now, if you go to it, you see more than 140 applications, including many maps, and also these things designed for people with blindness and things like that, voice assistance, and so on.

I think that’s a lesson to the NHI, because they very quickly see that the only way for them to absorb this kind of workflow is to redirect people to a place that will never get DDoSed. The mask.pdis.nat.gov.tw website is served from GitHub Pages. They have plenty of experience of that incoming traffic.

Then the link is essentially designed so that the first few are interchangeable. If any one of that goes down, you can go into a comparable service and guaranteed one of the top five or so will be up at any given time.

That’s how we survived the first day. By the end of the first day, it’s changed to every 30 seconds release of every open data. Nowadays, it’s not queueing much any more, so there is no need to publish every 30 seconds.

It’s now published every three minutes, still pretty fast. What people are interested in nowadays is just analysis. You can see the civil society contribution that shows the availability of stock, how the supply line is growing, the time series, the over and undersupply, and so on.

That is very valuable. I also showed that to our premier during our meetings. I think there’s two lessons. One is that the government trusts the citizens with open data for not abusing it, and the citizens trust back by making much more inclusive access that could satisfy all the different needs of people of all the different kinds.

That, I think, is one of the largest single-issue hackathon ever, even though it’s purely virtual, and the Tainan people never had to travel to Taipei to enable this kind of co-creation.

That is great. That is so interesting. This intersects with some of my personal interests, but I was watching the very informative interview you did in Amsterdam a few months ago. The interview said, “How do you go about achieving this?”

You said, “I take a very Taoist approach.” That says to me Wuwei, which has always been a tricky concept for people from the West to understand, because at first, it’s like you’re doing nothing. Then, you talk of yourself as a channel, which seems to me a very Taoist metaphor.

Then at the same time, you actually coded something. Code is ones and zeros. It’s like if you start with the first line of the “Tao Te Ching,” it says, “The thing that can be said is not the true thing.”

Code is the true thing. How do you mix those two things together? You’re both a channel, but you coded up that website. You made something out of nothing. How do you resolve that contradiction?

Since we’re in Tao Te Ching territory, let’s quote the scripture. Chapter 11, “The Uses of Not.”

That sums up the idea of making a portal. I just literally took it from the g0v HackMD .

That’s like a Hackpad. It’s a collaborative list that people are seeing how many maps there are. I just literally take that hack.md, put it into a Jekyll template and give it a domain name, mask.pdis.nat.gov.tw , and that’s it.

I literally did all of that during my office hours, where people can talk to me for 40 minutes at a time in the public park in the Social Innovation Lab. I only had 15 minutes of time slices between two appointments, and I take three of those slices.

Taking, I think, only 40 minutes at most of time, simply just giving the g0v community’s collaborative hackpad an official domain name, getting the SSL certificate for it, and putting it on fastly.

That’s all I have done. This is just a space. I just opened up that, and it was very ugly then. All the colors, all the buttons and everything is done by designers from the g0v community. The picture, the favicon, everything, I didn’t do any of this.

I basically just amplified a small community into an official domain, opening up the official domain on GitHub Pages, and say, “If you don’t want our official domain to look so ugly, send pull requests,” and people did.

Then I talked then to Google, saying that this can absorb all the traffic that you can point our way, so feel free just to redirect on it. I think after a couple days, if you asked Siri, “Where are the maps?” the Siri team also redirected to that website without me doing anything.

All I did was to hollow out the clay to make a pot, to make the initial carving of the uncut wood, and then I didn’t do anything afterwards.

That’s fabulous. As you know, there seems like in Chinese history not that many examples of actual Taoist philosophy being employed by the rulers. There’s this great concept in the era of the warring states, but then it’s just legalism, Confucianism. Is one of the things that’s powering Taiwan’s digital democracy tapping into this deeper stream of philosophy and culture?

Definitely. I personally practiced in the 黃龍丹院, which is the 全真 school of Taoism from very early on, when I was four years old or something. They have this idea of…

Actually, it has a Western counterpart. It’s “The Secrets of the Golden Flower.” It was adapted by Carl Jung in his psychology, so exactly that sort of thing. The 小周天, the inner cosmos, et cetera. I’ve been practicing that for, ever since I had memory.

I really do think that it is part of this stream of tradition from Laozi and Zhuangzi that really powers the work that I do. I actually, when the Perl community interviewed about why I started Pugs, the Perl 6 implementation that actually worked, using Haskell, all of my answers came from the scripture, from the Tao Te Ching.

Really?

Yeah, it was very intentional.

Is that available online somewhere?

I think so. If you google for it, it will probably be in some web archive.

One of the stories I wrote in that crazy era of the ‘90s was a profile of Larry Wall and Perl on its, the way it is a medium for connection. That was very powerful. It’s amazing. Now, no one ever talks about Perl. It seems to have disappeared into history, but it was very powerful.

I’m going to steer a bit away from Taoism. I think I won’t have much space to talk about that in Wired.

In 2017, you appeared via telepresence robot at an Internet governance forum. After your remarks, a representative from China was upset, and, “so-called country, so-called Taiwan independence.” This is a big problem.

He did not leave the room, which means that it was diplomatically accepted.

OK. I have a couple of questions about that. The first is how did it happen in the first place, since technically speaking, Taiwan isn’t in the UN? How was that organized? Was it in any way intentionally provocative?

No, not at all. I have done that many, many, many times. It’s just that IGF is live-streamed, unlike many UN meetings. The IGF is live-streamed, because it’s a hybrid. It has the UN and ITU multilateral on one side and the IGF community multistakeholder on the other side.

It abides by both rules. Within the UN venue, it’s operating by IGF rules, but the UN venue itself, which incurs checking with the passport at the entrance, of course, is by UN rules. It’s very different from other UN meetings that I also participated which are not live-streamed.

In any case, I prefer telepresence over flying. Due to carbon emissions, that’s an actual reason, but also that I just jet lag very slowly. I prefer not to travel, if I can help it. These two reasons are really good.

Of course, nowadays, no contacts, coronavirus, much better reason now. In any case, back in the day, those two reasons – to reduce air travel – was my standard answer to anyone who asked me to contribute to a panel or something.

I always ask, “Can you do a telepresence robot, or gauze holographic projection, or VR, or whatever?” Even though these technologies may sound expensive, they are less expensive than an actual airfare plus accommodations. They are reusable, circular economy, and all that.

Because of that, it’s not meant to be provocative. It is actually very practical. I think they also did not realize that it would be live-streamed until I actually spoke, because there was no interference when they were setting up the robot and things like that.

Then it started dawning on the PRC delegate that this is actually getting live-streamed. This is not something that they can say that, “Oh, we’re just watching a movie in a closed room, and the movie was recorded one second ago from Taiwan.”

It’s not easy for them to say that, because it was live-streamed and will be open to journalistic interpretation. We did not issue any press release. It discovered independently by investigate journalists. We had to then say, “Yeah, it was just a time-saving equipment,” to the journalists afterwards.

I noticed that after the PRC representative spoke, a representative of the Solomon Islands spoke and defended Taiwan’s presence. Then two years later, the Solomon Islands basically gets derecognized as Taiwan.

I feel like that’s just a micro example of constant pressure in every forum that you get, Taiwan, from Mainland China. Some people have suggested to me that that’s one of the reasons that has focused people on making your democracy strong, on using every technology available to really be healthy, because you have this existential threat across the straits. What do you think about that?

It’s a generative adversarial network. [laughs]

Generative adversarial…

Yes, it functions exactly as how a GAN would function. It’s true from the other way around, too. I work with the translation of the Freenet from its very early versions. Back in the time the Great Firewall wasn’t great at all.

It was very technically extremely not useful. It was more a token than anything. Perhaps because people like us working on Freenet, Wuxia, and many other technologies, the Great Firewall had to escalate in its technical prowess, simply because otherwise, it cannot justify the budget they are spending on it.

After literally decades now, a couple decade, of GAN, both the VPN, Tor networks, and so on become very sophisticated, but so have the Great Firewall, and they can repurpose it as a Great Cannon, and so on. That, I think, is just how the nature of things are. I think I do see it as a generative adversarial network situation.

There’s also, I think, a very potent metaphorical battle happening here. I remember sitting with some Wired editors in 1996 or something, and I had a China background. They’re asking me, “Well, the Internet’s going to be the enemy of authoritarianism. The Internet’s going to bring down totalitarians. Won’t the Internet…”

It’s going to be the great amplifier of authoritarianism.

It’s going to beat the Chinese Communist Party. I was like, I still had pretty fresh memories of Tiananmen. I was like, “I don’t know if the Internet can beat the Chinese Communist Party.”

In the current world, where we see so many democracies stumbling, and so many authoritarians all over, this idea that the Internet by itself would lead to liberty seems more in question than ever, really.

Then you have Taiwan saying, “Well, we’re free, and we’re using the Internet.” In a way, it’s a huge rebuke to China. It’s this example of how you can use these technologies for liberty. Do you see? Do you feel it? Does Taiwan feel that at a gut level?

Oh, yeah, very much so. Many of our political debates is hinged on being not PRC. For example, whenever we want to talk about counter-disinformation, anything around censorship is a nonstarter.

People saying, “Oh, we should put on some filters.” The term filter triggers a gut reaction, and the other sides would be like, “Oh, you want us to be like the PRC?” That becomes a nonstarter.

Because of that, we had to innovate this whole idea of Humor Over Rumor, over the social sector-led, much like Spamhaus, the International Fact-Checking Network collaboration, the generational solidarity plans, was born out of necessity.

We cannot entertain – literally, cannot entertain – the ideas that the PRC done in their harmonization efforts. It’s very interesting how our solution is basically being carved out by the also-advancing totalitarian and authoritarian solutions that our PRC are developing.

The more they develop, the more drawbacks that we see from our lens of human rights and democracy. We’re like, “OK, we should totally not go there.” I think it’s not just a metaphor of a generative adversarial network. It’s really shaped like a generative adversarial network.

Haven’t there been some laws passed for fines for misinformation?

No, not at all. There’s no single fine for misinformation. There is a penalty for disinformation, but it’s important to note that there’s no new law for that. We just upgraded our existing laws – “intentional truths to cause public harm” – to also include digital spaces.

These are already penalties if you shout in the public space about epidemiological untruth intended to cause behavior that would kill people, or also during elections. These are laws that are always there. We just upgraded it so that it also applies to additional spaces. It’s not about setting up new censorship laws.

How do you get consensus on what is intentional untruth to cause public harm?

Court decisions, obviously, the judicial branch takes care of that.

I’m fascinated by the collaboration between the platforms, civil tech groups, and the government when combating disinformation, especially in the case of Facebook, which was huge at that time. Correct me if I’m wrong, but it seems like in Taiwan Facebook acted like a good citizen.

They conform to the social norm. Maybe they know that Taiwan people is very good at social section.

If they do not conform to the social norm, such as revealing the role they have, each political advertisement in real time, so that people can call out people who use dark operations, if they don’t do that, then we will say, “Oh, our control yuan does that for regular campaign expenses and campaign donations.”

This is obviously a norm in Taiwan. If you go against the Taiwanese norm, we don’t even need a law or regulation to tell these people we’ll section that behavior. That is how it actually works.

That implies that there’s a majority or a consensus in Taiwan on these norms. That’s where, to me, the Taiwan Can Help model breaks down with respect to the United States. The idea of consensus right now seems so unreachable on a federal level, and Facebook in the United States, I feel, is not a good citizen.

Maybe it’s time that we change the term consensus, though, because you’ve been in Taiwan. 共識 in Taiwan doesn’t quite translate as consensus. What emerged in 1992 is not what any Westerner would call consensus. It’s a respectful “we can live with the common understanding that we disagree.” [laughs] That’s not a consensus.

The closest counterpart in English language may be “common understanding.” Literally, 共同的認識. Or, if you are of the IETF bent, you call it rough consensus. People hum, and that’s it. It’s not like they can put their name on it. It’s not that strong. It’s just they can live with it, go back and write some running code, and stop debating.

That kind of rough consensus is the key in Taiwanese norm shaping, because that enable people to not squander their time on getting the fine consensus out, but rather to agree on something that we can all live with.

Otherwise, in online as well as in real time politic, the person with the most free time wins. [laughs] You start arguing the finer points. Some people would quit arguing and the person who are most pedantic wins by default. Not that it’s a bad thing all the time.

What I’m trying to say is that rough consensus and running code is a really good alternate model. The translation is, again, not a valid translation, but maybe an inspiration. That there is something that polities can learn from the Internet governance in that if we can all live with it maybe that’s good enough. Maybe don’t need everybody to be literally on the same side. This is the idea of overlapping consensus.

Can you explain to me how g0v and vTaiwan implement that? I was very taken in your “TED Talk” by your comment that the way you implemented pol.is it did not have a reply button, and that killed the trolls. Is it really that simple, an interface tweak?

It really is. It really is because it stops the trusted messenger thing. The trusted messenger effect is how you associate a face to a book. Because you know the face, you trust the book. Most of the trolling mechanism is on discrediting the face, and so that you discredit the book.

Any platform with reply button is not immune to that dynamic, which is why all the platform we design, be it the Join platform, of participation, of e-petition, we had two columns, one pro, one contra, but you cannot reply across the two aisles. There’s no voting, per se, on this whole debate, but rather upvote and downvote on the pro and contra.

The one with the most absolute number of votes goes to the top. If you download a lot on the contra, it actually makes it move on the top place.

There’s a lot of very intentional design that makes sure that people can only add, but not subtract or detract from the conversation. The same applies to the real-time interaction, such as Slido. When we use Slido, we almost never use the reply feature, which is off by default. Also, of course, pol.is.

A string of interaction design runs through this. It’s basically saying first we want people who contribute out of self-interest can only do so in a pro-social way. Second, that the people who come later do not suffer a disadvantage vis-à-vis the people who contributed first.

If you can get those two design principles right, any antisocial space can become a pro-social space.

That strikes me as a very profound alternative to building rough consensus than you have across the Strait, where anything that deviates is by definition not accepted.

That’s right.

I feel like that is a way Taiwan can help. Back on the disinformation point, there was a report last year that said that Taiwan experienced more disinformation than any other country in the world.

It’s a survey, as agreed by worldwide experts.

You have the Taiwan FactCheck Center. You have the Cofacts. Do things like that scale up?

They do.

They do? How is that? I find that when I looked at the disinformation that surges through my info center it boggles the mind that 20 editors and a Line chat bot can scale up to the power of the entire population.

First of all, the private sector gets into the game, also. When the social sector prototypes something and it could become profitable, obviously the private sector gets in because of the public sector acting as a hub to the spokes.

This is from our leading antivirus company, Trend Micro. This Dr.Message bot is basically a supercharged Cofacts. Not only it can do Cofacts’ kind of real-time clarification – it taps into the same databases – but it can also work through videos and images.

It’s also mostly a anti-scam and anti-phishing bot. It also counters disinformation, which makes a much easier sell to people, because when people see a pure counter-disinformation they may be thinking it’s a tool that reinforces certain ideologies, but if a tool primarily counters scam and smut, then it’s something that people have more appetite to.

First of all, the private sector, because the social sector takes the open source approach, is able to build value-added products and services on top of that. That’s the first thing.

The second thing is, as you mentioned, Facebook being very open to the social norms, so when the ecosystem begin to from and when, for example, Michael Penn also joined the Internet Fact Check Network, they have multiple fact-checking partners so that they can dial down the virality of the posts that get fact-checked as false by those fact-checking partners.

The reason why this works is, again, by a very active civil society. Spamhaus wouldn’t work if we have not flagging things as fun prosocial voluntary donation of the spam samples that we do. If the voluntary flagging of disinformation is not as active as people in Taiwan are doing, then no amount of technology of IFC or pointer network can work, but they’re very active.

Again, Cofact or so on only served as the initial inspirations. The actual ecosystem is now much more active and larger now than the original Cofact team.

Is there a way to quantify how many people in Taiwan participate in civil tech who are engaging with vTaiwan or…?

Sure. You can just look at a number of people in the g0v Slack channel. It’s a good proxy. At the moment, the general channel is 7,400 people. The active people in the COVID-19 channel is 547.

For each idea is different. There’s the random channel, where people are just typing random things for the Mozilla Common Voice to use as speech synthesis and recognition samples. They have 300 people. vTaiwan is 271 people, and so on and so forth. The general channel is 7,400.

What does that mean for just larger penetration into Taiwanese society? How many people are engaged in…?

The mask maps themselves list more than 10 million active users out of 23 million in Taiwan. The Join platform, which is modeled after the vTaiwan and started after vTaiwan, but is government-supported, also has long past the 10 million unique visitor points.

I use the 10 million as a threshold. Below 10 million people using it, it’s civic tech. Above 10 million, we can call it civil engineering, because like bridges and roads, the bridges and roads that we build has maybe 10 million active users using that infrastructure. If you have the user base that is above half of all the adult citizenship, then you can call yourself a civil engineer.

How does all of this openness, the transparency, how has it helped Taiwan face COVID-19? Three months ago, you’re giving interviews about COVID-19 is not coming up and all of these things, and then it happens.

I have a six-minute video called, “Digital Social Innovation – Taiwan Can Help.” It’s on top of my Facebook and also can be found on my Twitter. To sum it up very quickly, it’s fast, it’s fair, it’s fun.

Fast because of the collective intelligence. Ensure that anyone with a new social innovation, like people should wear pink medical mask, otherwise the young boys gets bullied can translate immediately into government action. Everybody on the press conference start wearing pink medical mask.

The PTT, which is the dial-up BBs that become a massive online Reddit-ish thing, reposted the doctor Li Wenliang’s whistleblowing about a SARS occurring. This collective intelligence works doubly well in COVID because people have a natural focal point to concentrate on.

The fairness, we all already talk about mask rationing and the mask map. It’s equally important that we also, through the analysis like the dashboard I posted, you discover, for example, people who work very long hours, longer hours than pharmacists. They can never get the pharmacy’s mask ration because after they get off work, all the pharmacy have closed.

That’s why we start partnering with the private sector, such as the convenience stores that are open 24 hours a day. In fact, in 20 minutes, I’m going to collect my nine mask for two weeks, which is to be filmed and sent to Thailand for a YouTuber.

In any case, that ensure, again, fairness of all kinds. People who are not national residents, national citizens but rather just residents, all need ways to get access, too. There’s the Digital Diplomacy Lab that just filmed this cute video that shows foreigners in Taiwan how to get a mask and so on so that people can all co-create with the shared purpose of fairness.

The third thing is fun. Not only do we have this counter disinformation humor over rumor campaign, but each CECC press conference gets translated into a dog meme by the spokesdog of the Ministry of Health and Welfare.

The participatory mechanism is called a PO network, participation officer network, which is a vTaiwan-like structure from within the administration. The Ministry of Health and Welfare’s PO operates this social media campaign. They live with a dog that looks very much like the dog there.

Whenever there is a CECC press conference, they just go home, take a photo of the dog, putting hands to the nose and with a “No” to it, or to measure the social distance as the length of like if you’re indoor is three dogs’ length. If you’re outdoor is two dogs or whatever. A lot of it goes viral, so it’s preempting disinformation.

We don’t even wait for disinformation, but rather just spin our own memes to make sure that the CECC scientific knowledge gets viral. That’s also very important because people can remix it in Creative Commons fashion, and make their own memes that reaches to all the corners of the society. The list goes on, but it’s really a party of co-creation.

Do you have a favorite meme of this period?

Sure. My favorite meme is that of our prime minister. When the tissue papers are going out of stock, if people are panic buying, he filmed himself, wiggling his bottoms a little bit, and saying to people that we only have one pair of buttocks each in very large print.

Of course, there’s never a meme without a payload. The payload is a table that says, “The tissue papers are made from materials from South America, and the medical masks are made out of materials locally. You do not need to panic by, but rather you can just make sure that you buy whatever you want because it’s no use stocking this anyway.”

This is how the meme looks like, literally. I’m sharing my screen. You can see even the format itself is shaped like a tissue paper box, with all the right decorations. This is our premier with his wiggling bottom that says, “We only have one pair of buttocks each,” and then a factual table that shows how the producer differ and the material is different.

It is super-effective. 48 hours after this gets posted, there’s no panic buying anymore. We discovered the person who spread the disinformation in the first place, and they’re a tissue paper reseller. Go figure.

It’s hard for me to imagine some previous premiers doing that, like Hau Pei-tsun. I’ve been intrigued at the stats on gender parity in using Internet services and on presentation in parliament.

Over 40 percent now, women.

When I lived in Taiwan in the ‘80s, it was much more traditional and male-dominated. China’s going the opposite direction, Mainland China. It’s getting worse. What explains Taiwan’s success in gender?

There’s many things, but I would focus on gender mainstreaming and the Gender Impact Assessment Network as one of the prime examples. This is a story that’s not told a lot. The international community hears, of course, the story of Sunflower movement in Taiwan. It’s an easy story to tell and, of course, the national parliament.

The real story is, again, one of a continuous co-creation with civil society. We have the Gender Equality Council for 12 years now. They measure through the gender dashboard all the important things about gender equality.

Wherever there is any measurement of inequality that is being added in as one of the things to track by any government project or any draft bills, which is a lot every year. They get written to this gender dashboard and are continuously monitored even after the project goes out of their life cycle.

This leads to a theory of change. People can witness each individual project and how it’s negatively or positively influenced gender parity in all walks of life, not just political representation. Their theory of change gets reviewed by the Gender Equality Council, which, is by design, one more seat of civil society leader than minister. It also includes 17 ministers, which is most of the ministers.

This council with this unique assembly makes sure that all parts of the government are aware of gender mainstreaming. For each draft bill, like the Employment Service Act amendment, usually you wouldn’t think that it has a gender impact, but they have to go and make a gender impact assessment anyway.

They have to check a lot of things like analysis of the current status of problem, how it affects people of different genders, how to protect foreign workers, reemployed women, necessity of amendment, the policy goals, consultation procedures. All of this assessment are they rated by independent civil society leaders.

If they think that the consultation is not serious enough or it’s not meaningful enough, they can send a draft bill back to the drawing board for them to redo a consultation. The benefits to the gender equality will be then listed into the general dashboard. They’re measured starting actually right away even before the bill enters the effect.

Also, the statistics analysis and policy analysis, and how it connects with the CEDAW and things like that. After 12 years of evaluation, all civil servants, even the ones working in labor, in finance, and things like that, linked their main work with gender and proactively discovered gender related issues.

We have this huge amount of evidence-based policymaking that we can then use to make theories of change about gender aspects of all project and legislation. That’s why when the two referenda and the one constitutional ruling about marriage equality gets passed, the solution space is extremely narrow.

In the professional public servants, because they have 12 years of gender mainstreaming, they immediately discovered the true solution, which is to legalize their bylaws, but not the in-laws. That is to say, the couples who wed enjoy the same rights and responsibilities by law, but it doesn’t marry their families in law.

That is the solution of gender mainstreaming, this contribution to the marriage equality. That’s something that other East Asian jurisdictions is a true Taiwan Can Help situation because they had great difficulty convincing the families to wed together because for them, marriage is a social ceremony, but then the rights and duties are individual.

We have the word to describe that. I call it 結婚不結姻. Marrying the bylaws but not in-laws.

That 12 years, that would date to the Chen Shui-bian administration?

That’s right, and the CSOs who very intentionally, just as we’re now, rolling out the open government partnership’s idea of a national action plan. They wrote out the CEDAW ideas of gender mainstreaming in a very systemic way from the civil society and embedding themselves into the administration for the Convention of the Elimination of All Forms of Discrimination Against Women.

Has Taiwan figured out how to be a civilized democracy while the rest of the world flails?

To be honest, our e-petition platform, in the two column, you can reply across aisles. That’s a carbon copy of Better Reykjavik from Iceland. We learn from international counterparts. It’s not like we figured it out by our own. Polis is from Seattle. It’s just like we take the component’s bits and pieces and improve them.

We’ve covered a lot. One last thing. You talked many times about how the government should trust the citizenry. Citizens should be wary of the government.

That’s right.

Looking across the Pacific at the US, we have all these video now of armed protesters and state legislators protesting social distancing, protesting measures that are for the public benefit. Is that a situation you could see where we could…I’m not articulating this very well.

That relationship of trust in my country is broken. I understand the direction that you want it to go. How do we fix that? What do we push on to reestablish that faith in the people?

There’s a crack in everything, and that’s how the light gets in.

We could go by without that.

It’s really the people. Even if it’s in a small community in the municipality, they suddenly decide to start working based on rough consensus. That can already serve as a great inspiration for everybody around them. It doesn’t need to start at a federal level.

Instead, I would argue that it almost never work on a federal level where each citizen may not have the same kind of empathy as any other citizen in the federation. If you started with a small community, you can almost always get everybody to take all the sides of the multi-stakeholder conversation.

That’s how the Internet governance really shines, because once people can take all the sides, you do get rough consensus. If the lived experience is too different, then it’s actually very difficult to even start to talk about the same experience because same word would not work on the same thing.

My main suggestion is to start small and do not prescribe anything. Don’t make long speeches. Instead, just start designing for spaces for people to participate. To quote the scripture again and to conclude this conversation, from the 23rd chapter:

Thank you. I feel honored to have been able to take part in this conversation.

Great questions. I’ll send you the transcript.

That would be great. That would save me…How fast do you turn that…

Just 24 hours or something.

That is great. Thank you.

Cheers. Bye-bye.