-
How many words of transcripts are you at now?
-
You mean as in since I become digital minister?
-
Yeah. You must be at millions of words of transcripts.
-
(laughter)
-
It’s not quite millions. [laughs] It’s this number of speeches, like 200k.
-
237,841 speeches.
-
Each is a paragraph, essentially.
-
Wow.
-
I can get you the word count. Just a second… This is the word count, not quite a million.
-
875,336 words, there you go. Do many people read it?
-
Yes.
-
You get hits on it?
-
Yes.
-
Interesting. There are a couple of things I wanted to talk about. The first was the current election. You’ve been one of the people involved in some of the efforts on disinformation and that sort of thing.
-
That’s right. Countering disinformation, but yes.
-
Exactly. The election’s tomorrow. I guess we can now speak about the election process. Can you give me any sense of what you’ve seen? There’s been lots of warnings that this was going to be an issue. There’s been lots of anecdotal discussion about some of the various memes and other bits of disinformation that have spread.
-
There’s also been efforts by, as you know, Facebook and others, ministries, parties themselves, to counter some of this disinformation at a very fast pace. What have you seen? Do you have any statistics? I know these things are hard to gather.
-
It’s plenty of statistics. Have you seen the LINE dashboard of the fact-checking?
-
Uh-huh.
-
That’s statistics right there. This is the number flagged as disinformation. This is unique message count.
-
Do you mind if I take a picture of that?
-
Not at all. This is a public website.
-
That’s great. It’s just easier to have here.
-
This is the baseline, based on what people reported. Obviously, there’s more disinformation that’s not…Like there’s more spam than email flagged as spam, but you can get a trend from the number of email flagged as spam obviously.
-
By warning, you say there’s early warning of the scope of the issue?
-
Yeah.
-
Which kind of warning are you referring to?
-
There’s been a lot of people talking about it, from the academic side, from the political side. There’s been a lot of people suggesting that they were concerned about this election.
-
Certainly, but this is a perennial issue. This is not something that’s early become public awareness since the previous election. I wouldn’t say that. There’s always disinformation in each and every election.
-
The main thing that’s changed is the prevalence of social media and video-based social media and the unlimited 4G connection, 10 megabits per second, every place in Taiwan at no marginal cost. People can share large videos or start live streaming.
-
That’s new, the prevalence of 4G, broadband as human right. That is new as of the previous election. Otherwise, I wouldn’t say that either the motivations or the incentives have changed much. The prevalence and, indeed, the amplification is new. That’s that.
-
You asked about the patterns, right?
-
Right.
-
This time around in Taiwan, there’s more overt pieces of disinformation. In the previous election, if you ask people, “What do you think about disinformation?” especially on the LINE system, usually they say, “That’s just something people remix.” There’s no obvious attribution line to who create it, where it was created, how it was promulgated, and things like that.
-
Now, there’s more and more disinformation that the original poster isn’t hiding that they manufacture it or they made it. There’s more and more of the fact-checking ecosystem that also does the attribution and makes the attribution visible, which is harder than the attribution itself. That I think is new.
-
Baselines are hard. Are you able to do any comparatives in terms of the volumes or types of disinformation this time, perhaps compared to 2018 or 2016?
-
That’s a DoubleThink Lab question. I don’t have it myself, but people in the DoubleThink Lab have been doing precisely this.
-
That’s Puma Shen, yeah?
-
Yeah, that’s Puma, precisely this analysis. They have the numbers.
-
I know Facebook has had a war room here looking at some of this stuff. Can you tell me about some of the stuff that you’ve seen the social media companies do? Is it as simple as shutting down ads and flagging certain content as potentially disinformation, or is it much more sophisticated than that?
-
It’s sophisticated, a little bit more. I wouldn’t say much more, but somewhat sophisticated. First of all, it’s not about taking down ads.
-
Not about checking?
-
It’s about making sure that the ads are properly attributed. If they take down, that’s because they came from a extra-jurisdictional source. What we’re saying is that the norm around campaign financing here is that we have a separate branch of government called the Control Yuan that publishes the raw data of campaign donation and expense.
-
I call this a norm because people expect to see such raw donation data in everybody who is involved in the election campaigns, regardless of whether they’re a candidate, a candidate supporter, somebody who are volunteering to buy some ads, and so on.
-
Previously, Facebook was not conforming to these ads. The ads were…
-
Facebook was not?
-
Yeah, were non-transparent. There’s a lot of talks about people using hyper-precision targeting to discourage certain parts of the population from voting. Everybody else doesn’t have this ability into this information operation. Indeed, there’s a lot of evidences in the Russia to the US election report precisely using this tactic.
-
This time around, we’re saying to Facebook and other social media companies that you either conform to this Control Yuan standard, which is publish the raw data for a long time, keeping it after the election, and also making sure that people can independently reuse it in a structural way, which is what our Control Yuan did, so a step more than honest ads.
-
Indeed, we have seen projects like the g0v, the vote.ly.g0v.tw. For any pair of candidates, you can very easily compare not just where their Facebook page is, but how much they’re spending on advertisement in an hour-by-hour basis and the reach of such advertisements.
-
The idea is that people can very easily see not only what they did in the previous election, as provided by the Control Yuan with a very detailed breakdown, but a hour-by-hour breakdown of who exactly is being targeted by this candidate’s advertisements, where, and the reach.
-
Wow.
-
The subtext is that if anybody try to pull a hyper-precision targeting movement that discourage people from voting by spreading disinformation through hyper-targeted means, then there could be a social sanction against it, literally, the next hour, instead of waiting after the election. That’s somewhat more sophisticated.
-
Facebook is providing that data?
-
Yes, and basically conforming to the same expectation that the Control Yuan is providing.
-
To your knowledge, has Facebook ever done that for an election somewhere?
-
This is a new system for them. They’ve been prototyping this, for example, for the upcoming US election, I’m pretty sure. This interface, this structured data and so on, is because they have to work with the investigative journalists, the civic tech community, and so on.
-
As far as I understand, the criteria, the structured data, the presentation forms, the ad flag redesign, and so on, is being co-evolved with the demands of fact-checkers and investigative journalism community here.
-
I don’t think Facebook has provided something that’s exactly the same as this before, but if this turns out to work well, I’m sure they will take it elsewhere.
-
That’s one. One of the questions I had was in which ways do you see some of the response in Taiwan through this election cycle as developing solutions that could potentially be helpful elsewhere? This is one example. Where also are there examples?
-
The CoFacts system has been ported, as in adapted, to Thai. This is CoFact.
-
To Thai?
-
Yeah. This is CoFacts. The Taiwan project is called CoFacts. You probably heard of it.
-
Yeah, right.
-
If you take away the S and go to CoFact.org, as in singular, [laughs] then you’ll see the Thai version. LINE is as popular in Thailand as here. When CoFacts first started, there was no indication at all that a CSR arm of LINE Taiwan is going to provide the dashboard that you just took a photo from.
-
In fact, they were pretty much saying, “We’re just a end-to-end encrypted messaging platform,” and so on, “mere conduit.” At that time, CoFacts was the only social-sector response to what we call the problem of getting the clarification back to where the disinformation are from, a circulation problem.
-
By CoFacts being a viable prototype, it not only influenced LINE Taiwan to work with CoFact-like fact-checkers, four of them now, providing a CoFact-like dashboard from the LINE dashboard.
-
It also influenced other jurisdiction where LINE is available – Korea, Japan, Thailand, and so on – and the civic-tech community to easily – because it’s open-source, there’s no copyright restrictions – start a prototype there and use that community to collectively negotiate with the chapters of LINE in their jurisdictions.
-
Did you say there’s four fact-checking groups behind that now?
-
With LINE, yes. That’s MyGoPen, CoFacts, Rumor and Truth, and the Taiwan Fact-Checking Center.
-
We’ve talked about two very different things now. We’ve talked about Facebook taking a corporate action in conjunction with the government. We’ve talked about…Is it fair to say more of a crowd-sourced approach when it comes to reporting disinformation in the same way you might report spam?
-
What are your observations about, in general, the things that are most effective? Is it as simple as saying, “You need a whole bunch of different ways of attacking this”? Presumably, there’s no silver bullets here. Are there silver bullets?
-
The people who innovate to sow discord and spread disinformation are very innovative and sharing each other’s playbooks. There might be a silver bullet for any particular tactic ongoing. 20 years ago I worked with the Freenet project. I saw first-hand the co-evolution of the Great Firewall and the VPN bypassing community.
-
There really is no silver bullet, per se, as long as the people who sow disinformation are also evolving. It’s a ongoing thing. This is just like cybersecurity. You may have found a 100 percent good solution to a certain tactic, but that only means that from tomorrow they switch tactic.
-
I’ve got to tell you, from the firewall perspective, it’s not getting any better. It’s getting a lot more difficult in China these days. When you look at this in sum, there’s still been a great amount of disinformation that is circulated. Do you feel like, from an overall perspective, this election has been cleaner from a disinformation perspective? Or do you feel like it’s been a struggle or perhaps even a failure?
-
The surface to sow discord has…
-
The surface?
-
The potential surfaces to sow discord have been reduced by tenfold.
-
By tenfold?
-
Literally, because each referenda topic is a division in the society. In the previous election, it’s not just a election. It’s a election plus referenda. Literally, anyone can still campaign on the referenda there, but no candidates can campaign on the election day except it’s a sick day, and some candidates are also proposer of referenda. [laughs]
-
It made it very difficult to devise counter strategies or clarification information that can respond to all the different referenda topic at once. As long as there’s a particular one that escapes the attention of the multistakeholder ecosystem around fact-checking, then all the disinformation package can go through that channel instead of by any particular candidate or other referenda topics.
-
Which makes it extremely difficult to defend, but this time around, we say maybe we do a year of election, then a year of referenda, then a year of election, and then a year of referenda. For everybody going into the voting booth, there’s just three things to think about, not 20 things to think about.
-
You’re talking about restructuring the system of elections in response to disinformation in some ways.
-
The previous referendum was the first meaningful one. That, by itself, is a experiment. We learned from the experiment and – what’s the word? – we pivoted.
-
That’s interesting. I understand what you’re saying. There’s a lot less scope for disinformation.
-
That’s right.
-
There are fewer questions out there.
-
There’s three surfaces, one around the presidential candidate, one around the legislative candidate, and one around parties.
-
That said, this is, in some ways, more important. This is the presidential election. The stakes are higher. Do you feel like all of these efforts have in large measure succeeded, or are you disappointed?
-
We’re still 36 hours…
-
(laughter)
-
From experience, the most concentrated efforts will appear 24 hours before the election, sometimes not intentional, sometimes intentional. Everybody here probably still remember the previous presidential election. The deciding factor occurred after the banning of further campaigns in the form of a viral video. I wouldn’t say anything about success or not…
-
(laughter)
-
…until the tallying have been done. [laughs]
-
Are you able to speak with any authority on the source of disinformation that you’ve seen, as in domestic versus foreign?
-
Of course, the TFCC will tell you which disinformation packages are attributed…Like this one, which went pretty viral, is from Zhongyang Zhengfawei, the Central Political unit, the Chang’an Sword. This is pretty clear-cut because that’s right there on their Weibo, and the timestamp is earlier than any remixes or comments.
-
There really is no other possible explanation but the fact that 中央政法委长安剑 – I’m trying to think of a charitable word – adapted the photo from Reuters and put a very misleading frame around it. That’s very clear-cut. I don’t have to speak to any authority about it. I can check it myself. I haven’t spoken to 长安剑 about it. [laughs]
-
They’re not answering your calls?
-
(laughter)
-
I didn’t call them, of course. It’s right there on their Weibo.
-
That’s one example. In aggregate, is the problem, as you see it, largely foreign-inspired or Chinese? Is it largely domestic do you think?
-
This is like asking where do memes come from. Each popular Internet meme may actually originate from a movie, a particular discussion group in Reddit, or in one of those forums that specialize in making memes. You can find, of course, the original memetic engineer, but the repurpose of that meme for political purposes is somebody else altogether.
-
Add to that the remixers. They remix mostly end-to-end encrypted channels, sort of AB testing, find the ones that go viral, and then start surfacing it on the public social media. By the time you see the payload, it’s kind of too late. [laughs]
-
Every point around this is a different actor. Somebody do it for profit. Somebody do it for fun. Somebody do it for the sense of control. Some do it because they’re being paid for it.
-
Really, to say what is extra-jurisdictional and what is not make very little sense except for the very last-mile where it does make sense because you want to know who is paying, if anyone, for the final delivery of the package. Asking, “Where do the memes come from?” is very difficult to answer in a meaningful fashion.
-
I understand, but it’s an important issue isn’t it not? It does feel like it’s very different for democracy for people inside a country to be having a debate as opposed to a foreign power trying to influence, right?
-
I understand, but you see, I just caught a cold today, and I can’t negotiate dircetly with the virus. We are not in the same category.
-
They speak a bit different language, I mean the virus. It’s the same for the virus of the mind. We, of course, can try to attribute the epidemic, but it only makes sense in a aggregate sense. We try to reduce the spread. We try to increase the distance.
-
For each individual message, for each individual strain of virus, it is important to protect against that as a kind of public health matter and increase personal resilience and drink some coffee so that I’m not as affected.
-
It doesn’t quite make sense to look at it in a piecemeal fashion. By the time we’re done looking at it in a piecemeal fashion, another strain came. It makes more sense to solve it in a systemic manner, to look at inauthentic, coordinated behavior rather than any specific disinformation message.
-
I presume there will be an exercise in review after the election is over.
-
Of course.
-
Do you have any initial ideas on what might come next? What next steps have seemed to be necessary based on what you’ve seen?
-
When in doubt, we throw a hackathon.
-
(laughter)
-
Of course, this time, we have a lot of capable people. The DoubleThink Lab is connected with a lot of folks to keep a historical archive of what has transpired during the election. Post-election, it makes sense to look systemically at all those different data sources.
-
Something like that has been done when Twitter did its data disclosure around antiELAB around Hong Kong. Revealing not just the tweets but rather the metadata pointing to coordinated, inauthentic behavior from actors that are state-blessed, so not needing VPN to go through the Great Firewall to access from a block of IP in the PRC to Twitter to work with to 200,000 fake accounts.
-
Facebook and Google, of course, corroborated on that, saying, “We’re also seeing similar activities.” Twitter says, “We’re not the ones to do analysis. We’re publishing all these data, data disclosure, and you, the civic-tech analysts, journalists, you do the analysis.”
-
That’s exactly what the journalistic community in Taiwan had done to look at the Twitter data disclosure. Something like that will probably happen with the post-election data.
-
This is maybe more of a personal viewpoint perspective. I don’t know how much you deal in these sort of things. When you look at this, there has still been a lot of misinformation flying around. Have you been disappointed with the tenor of conversation?
-
The tenor of conversation?
-
Yes, the types of conversations, the content of conversation around this election. In an ideal world, we think of elections as places where there’s a contest of ideas.
-
There still is a contest of ideas.
-
Right, but it’s been a bit shy on policy discussion, this election. Have you been…
-
At a MP level, I think it’s pretty OK. We do see very creative platforms and the ways to deliver platforms. I personally work with the CEC, the Central Election Committee. Everybody is using QR code and layout. Not by everybody. A majority of MP candidates are using these sort of things to make sure that they can visually communicate their ideas on a grayscale printed bulletin.
-
The freedom afforded by QR codes, URLs, and so on is you look bad if you only have a few emotional slogans on your bulletin. There’s a real contest of bulletin materials and how best to format it in the bulletin. Because it’s released now also as structured data, the vote.ly team, there’s quite a few, the votetaiwan.tw team. I’m sure I’m missing a dozen.
-
Many have been reusing these candidates’ platform graphics to analyze the trends and things like that. There’s even a crowd-sourced attempt for people with blindness to type out the main ideas there. I think there is very creative, and because of its creativity and it’s just plain fun, people spend a lot of time looking and talking about.
-
There was a particularly good one from 李正皓. I don’t know whether you have seen that one. It’s really good, and I want to give him special mention.
-
[laughs]
-
Just a second.
-
(pause)
-
The platform, this one . That’s him. This is very interesting. That’s his…
-
I’ll grab a picture of this.
-
…platform.
-
(laughter)
-
This is really pretty good. Super-creative.
-
When was this released?
-
Last week or something.
-
Wow. There’s traditional elements. There’s cartoon elements. It’s very cool.
-
Also, look at the the signatures.
-
(laughter)
-
Very cool.
-
I’m fine with this tenor of conversation. This is awesome.
-
(laughter)
-
In general?
-
In general, the MPs are doing this pretty well.
-
How about at the very top levels of conversation?
-
The fact-checked community who do the real-time fact-check around the presidential conversations like the debates and the platform forums have been trying very diligently to put some sense into the policy debate. Have you seen the fact-check project?
-
I’ve heard about it.
-
This one?
-
Exactly.
-
They just type, in real time, the transcripts and find out the differences in disinformation versus misinformation versus mal-information in the various candidates. There’s, I think, three from James’ side and five from the Dr. Tsai side. All of it are just she being too humble.
-
For example, this one was saying that Japan…There’s one. Japan correspond delegate ambassador have been visiting Taiwan for 14 times, but when she said that, actually that was the 19th time of this type. Of course, she makes mistakes, though not on the same direction as this candidate.
-
Mr. Han.
-
Yeah. As you can see, the fact-checking community has been classifying the substantial statements versus the plain emotional statements. There’s a stark contrast between one candidate and the other.
-
Again, from the high-level perspective, yesterday you had a big rally in town. You had people up there talking about President Tsai’s London School of Economics credentials and that sort of thing. It feels like, from that very high-level debate, it has not been the cleanest of electoral debates.
-
That’s the same in every other presidential election. It’s not particularly bad by presidential election standards.
-
Interesting. I wonder. You’ve also done a lot of the work and you’ve received, I think rightfully, a lot of international acclaim for some of the work you’ve done on using digital tools for policy and to use very sophisticated ways of marshaling public contributions and that sort of thing on policy.
-
I wonder. You’ve worked on that for a number of years. Assuming that this government wins reelection, do you have plans for the next four years? Do you have plans for the next four years? What do you see as next? I know that there are other democracies who are looking to Taiwan as an example of how to marshal some of these tools for good as opposed to evil. Where do you think you can go next?
-
The fact that we’ve moved referenda to its own year means that we have a full year for deliberation without the feeling that whatever deliberation we do will be eventually recaptured into partisan politics because the same day as election.
-
It makes it much likely, higher payoff for a Swiss-style, full deliberation before referenda. Whichever side of referenda I talk with, they all want their ideas to be deliberated in more nuance rather than just in very short few weeks with a few people talking and not debating on TV. Then people go into the referenda booth being captured by party rhetorics.
-
No referenda proposer truly want that because each referenda proposer is also a advocate for a new direction for the polity. Just winning the referenda is not enough for them because referenda, they all are binding only for two years anyway.
-
What they really want is to deepen the understanding around that particular issue in the whole society. That’s what only deliberation can do. A more deliberative referenda is definitely on the horizon no matter who win the election.
-
Using some of the tools you’ve built or just purely for that scheduling reason?
-
Pure for the scheduling reason, but also we’ve been seeing the tools that we built, including the Join platform, which is a very popular platform, like 10 million visitors out of 23 citizens in Taiwan, being used as a jump board to referenda.
-
Virtually every referenda topic in the previous referenda session have seen a more mini version of it on our petition platform before. They use it to mobilize, to rally, to find talking points and things like that, to raise awareness. There’s no reason why we can’t use the tools we developed for the petition platform to enhance substantial deliberation leading toward referenda topics.
-
You would have an interesting perspective on this. This is a very big-picture question. As you know, we are at a moment of perhaps global fear about the prospects for democracy. There’s a fear in places like the US that the democratic system is not working as it should.
-
There is a fear that authoritarianism, both in non-democratic systems like China and in democratic systems, like the US, like India, like the Philippines elsewhere, is leading to a perversion of democracy and that democracy is on the ropes in the current context. I wonder. Do you share that fear? Do you share that sense of pessimism?
-
No.
-
Do you feel optimistic about…And I’m talking about the state of democracy, effectively.
-
Well just look at the CIVICUS Monitor.
-
Liberal democracy is predicated on a vibrant civil society. It’s based on the idea of what I call a trans-cultural republic of citizens, trans-cultural meaning instead of silo multiculturalism, people are willing to look beyond the culture that we were brought up with, that we went to school with, and look at our culture from the perspective of a different culture co-existing in the same polity.
-
From a indigenous point of view, for example, the Tongku Saveq, the Jade Mountain is a spirit.
-
From the viewpoint of Лариса Бакурова who is voting also this week, she was originally from Ukraine, with a different perspective on our democracy.
-
There’s a lot of people who are adopting a very Swiss view that the only thing we have in common that binds the republic and citizen together is that democratic process and not any particular religion, ethnicity, or whatever.
-
If you adopt this view, then you feel that democracy is just another social technology that everybody can participate in a civic-tech way. We can all think of better ways to do referenda, to do participatory budgeting, to do sandbox experiments, to do Presidential Hackathon, to do petitioning. I can go on.
-
All this is reinvigorating democracy instead of it being fossilized in a particular ritualistic form and people play game theory like gerrymandering, around it. We can instead do reverse game theory, which is mechanism design, to respond to the trans-cultural demands from the social sector, from the civil society, to democracy as a social technology.
-
Whenever you see on the CIVICUS Monitor somewhere that’s completely open, which is kind of rare nowadays in our part of the world. There’s New Zealand. I’m happy about that. Then you feel this sense of reinvigoration.
-
In the obstructed, narrowed, or even entirely closed jurisdictions, of course you feel the opposite. How much optimism you feel is directly correlated to how much civic space there is in your jurisdiction.
-
It feels like there are broader forces marshaling against the desire or the ability to do that reverse game theory.
-
Mm-hmm.
-
Is that not…
-
Not in Taiwan, though.
-
Not in Taiwan?
-
Not in Taiwan. If you ask a random person on the street do they want democracy to evolve in a more modern form to respond to people’s need here and now, whether they support any of the three presidential candidates, they’re going to say yes.
-
If they ask them whether you want, through democratic innovation, to link Taiwan more to the jurisdictions around the world that have a fully open civil society, New Zealand, Nordic countries, Canada, and so on, they’re going to say yes regardless of which presidential candidate they respond to.
-
I say this because there’s a real investigation into this. People went to the rallies and asked those questions to the core supporters of the presidential candidates. That’s remarkable similarity. If you only look at the answer of those two questions, you can’t tell a Tsai supporter apart from a Han supporter.
-
That’s interesting. To go back to what we were talking about at the outset, to what degree do you see disinformation – here, I mean state-sponsored disinformation – as a threat to some of that?
-
We don’t know about state-sponsored, which is why I’m very careful to say state-blessed.
-
Fine.
-
To say sponsored, you have to do a financial flow attribution. I think there’s two layers. One is at the cognitive-spaced level, this whole rhetoric around “breakaway” territory, this whole rhetoric of the “great rejuvenation” and the “Chinese Dream”… All these are rhetorics. And also note how I put “re”-unification in air quotes.
-
All these are essentially cognitive-space moves that wherever and whenever people, including foreign correspondents, use these words, [laughs] they implicitly propagate this state-blessed ideology that decimates the civic space of Taiwan.
-
However, we’re also reasonably sure that our use of…I think it’s easier if we use Mandarin words here.
-
When I speak to people and I say that we need to work on technology for 透明, transparency, people correctly understand, whichever political side you’re on, that I’m saying making the state transparent to the citizen. Whereas, in PRC, that would mean the opposite, making the citizen transparent to the state.
-
Wherever I say that the social innovations build trust between the social groups so that everybody gets credit, people understand talking about like credit union or co-ops. Whereas, in the PRC, the words 社会信用 now means a state-sponsored scoreboard that prevents people from…
-
Social credit, right?
-
…from buying train tickets. But the same words sound the same. Whenever we talk about the importance of people’s mutual respect and peace, 和, that’s also a Japanese concept. They also use that word all the time.
-
The norms here are literally to respect the words you say, to have a sense of justice and solidarity, and to ensure peace together. The same words will probably be twisted to say 和谐 in the PRC, which means something else altogether. It’s censorship, actually automatic censorship.
-
What I’m trying to say is that I’m reasonably confident in the resilience of the civil society here precisely because, first, these words are not being twisted quite as much as the PRC have done twisting these words.
-
Second, the trans-cultural setting, just like biodiversity, Taiwan has the world’s, I think, 10 percent of marine biodiversity in our seas. That means that there’s no single strain of virus can infiltrate or massively change a majority of the population because people are so different in their configurations of ideas. This inherent diversity protects against this kind of propaganda.
-
You made an interesting analogy to your cold.
-
That’s right. Viral outbreak. That’s the only thing I could think about now.
-
Have you studied public health? When you look at fighting disinformation, do you use public health models in some ways?
-
We do use public health analogies. Memes, being virus of the mind, naturally lead itself to analogies like making sure there’s media competence, education, starting from the first grade all the way to life-long education. It makes sense to say that we’re developing civic tech so that people who are empowered by a journalistic education can contribute to crowdsource fact-checking.
-
It makes sense to say that, if you do hyper-precision targeting and we can attribute it to a foreign source of income, it makes sense to put on a quarantine and say that, during election sessions, only domestic memes are allowed to post advertisements.
-
These are natural analogies. It’s not a stretch. It’s very easy to draw a parallel between them. But, I have not studied public health, no.
-
This is kind of a personal question I had. It seems to me one of the interesting elements of China’s interactions with the world – and we talked about, for example, Chinese expansionism and the Chinese model – is that, in many ways, the problems that confront China’s leadership are the same as the problems that confront the leaders of democratic countries. Fake news is one of them.
-
China has developed solutions to many of these things.
-
They spend on more on “harmony and stabilization”…
-
Than on defense.
-
…than defense.
-
Exactly. In China, the solution on the spread of disinformation, what their state classifies as disinformation, is things like real-name registration. It’s things like…
-
Facial recognition on the Internet cafés.
-
…facial recognition, censorship, automated tools to delete the spread of content, as well as human content tools to delete the spread of content.
-
And just downright blocking Wikipedia now.
-
Exactly, but they have a solution. That solution is actually an effective solution. Do you see it as a given that that Chinese solution is the best solution when it comes to some of these issues?
-
Taken to the extreme, you can say what Russia have currently demonstrated, that they can just decouple from the Internet and become a intranet, that is also a “solution.” If your computer, you don’t want virus, you can unplug the Ethernet cables. You won’t get virus if you seal the USB ports and unplug all the Internet connectivity.
-
If you don’t have any freedom of speech, you won’t have hate speech. That’s true, yes, but calling it a solution is weird. This is like saying to solve a privacy issue, we just make everybody accept state surveillance by default. Without privacy, there is no privacy concerns…
-
This is a solution only in a very macabre sense of “solution.” I wouldn’t quite say that it’s a solution if you want to make sure that it works elsewhere. When we call it a solution, I mean in a mathematics sense. It means that whenever you see a problem like that elsewhere, you can solve that in a similar way if they’re isomorphic.
-
We’re not saying that refusing to solve is a solution, although it is a solution. If your math teacher give you a hard problem that you can’t solve, you can say, “OK, I quit.” That is a solution, but that solution doesn’t transfer to other classmates.
-
Unless you’re talking about a chemical solution, which is another thing altogether. If you just dissolve the paper on which the problem is stated. Yes, it is a solution, but it doesn’t make sense to comment on it.
-
Your remit is obviously Taiwan. Your job description is not to fix other people’s problems, but I know you travel. How many times have you been to Canada?
-
Quite a few, Ottawa, Vancouver, at least two times to Toronto, three, four times.
-
I know you travel, and I know people want to hear what you have to say. I wonder. What is your observation in terms of what are some of the elements that are preventing other countries from going down some of these paths?
-
Is it just the bureaucratic, institutional resistance? What do you feel are some of the obstacles standing in the way of others using some of the tools that Taiwan has built?
-
Canada is a really good model. We held a workshop there, and they are adopting the same tools as we use. They contributed even for the Listening at Scale pol.is conversation. They contributed the bilingual capability, which we then use to work with the AIT on our digital dialog to allow people to comment in either English or Mandarin and automatically get a translation out.
-
That’s a contribution from the Canadian government because everything has to be bilingual there in the federal level.
-
Really?
-
Yes. We do work together, quite closely actually. I would say that Canada is one of our main allies. That’s no coincidence because Canada is, according to CIVICUS Monitor, another jurisdiction that need to solve these issues confronting democracy in a way that doesn’t encroach the freedom of the press and the people.
-
They can’t take a shortcut, saying the minister can “harmonize” any journalist. That probably wouldn’t fly in Canada.
-
Not yet.
-
(laughter)
-
No, I don’t think.
-
That was a very, by the way, Chinese use of the word “harmonize.”
-
I always put air quotes whenever I say it. [laughs]
-
Canada contributed funding or contributed expertise. Sorry, can you help me understand that?
-
They contributed research too. We publish our digital democracy in, for example, social archive, which is open-access, preprint server. If you search for Taiwan in social archive, you can easily see our contributions, for example, around vTaiwan, so I still publish, even being a digital minister.
-
If you look at the follow-up and citations of that paper, you will see then input from the Canadian. What used to be called the Energy Board. I think they changed their name to something else.
-
That’s right. We used to be the National Energy Board.
-
Yeah. That’s right, the NEB, and who really want to use these ideas to do a meaningful consultation with indigenous nations when it comes to energy, like oil pipes. The courts keep telling them that they are not doing meaningful consultations with the indigenous.
-
It turns out that we’ve started some meaningful consultation with the indigenous. Although Canada led us, like the transitional justice equality reconciliation process. It’s a few years before we did the counterpart.
-
We think that in certain indigenous nations, like the Orchid Island, which is a good example where they have sovereign, blockchain, cryptocurrency thing, this data collaboratives, and negotiation around the truth and reconciliation around the nuclear waste issue and so on.
-
These are good examples of using modern digital tools to empower indigenous nations to reflect properly their worldview and put a idea of transculturalism on meaningful consultation. Instead of asking everybody to learn the one, official language, we changed the law last year to make the Dawu language also a national language. Now we have 20 national languages. [laughs]
-
All this is something that is very relevant to the continued existence of the National Energy Board. [laughs] They did quite a huge research, jointly, and published papers on their findings of how the Taiwan idea can apply to their case.
-
Interesting. Before I move to China, I was based in Calgary, which is where the headquarters of the National Energy Board. In effect, I did a lot of writing about pipelines. I wrote the entire route of two proposed pipelines. That issue of consultation is a central one, so that’s interesting. They’re the ones who stand to be the early adopters in Canada of some of this stuff.
-
Very much so, and that’s because of necessity and also because similarity in configuration.
-
That’s super interesting. I might reach out to them on that, because that’s super interesting.
-
Yeah, the main contact is, I think, Annette. Let me make sure. Yeah, Annette Hester.
-
I’ll look her up.
-
They may actually make a visit sometime this year.
-
Oh, so this is an ongoing thing?
-
Yeah.
-
Do you see yourself as building models that could be…I mean, do you see yourself as working? Obviously, yourself is working for Taiwan. But when you’re building some of this stuff, do you…?
-
I’m working with Taiwan, yeah.
-
Do you have a bigger goal in mind for you, personally? Do you feel like perhaps you can help to build things that are useful elsewhere?
-
I’m also a board member of three international NGOs, while being a digital minister in Taiwan. That’s RadicalxChange in New York, Digital Future Society in Barcelona and soon, the CONSUL Foundation in Amsterdam.
-
The CONSUL Foundation?
-
Yeah, the CONSUL Foundation. CONSUL of course, being the major participation platform for Madrid, just like our Join Platform, and very popular in Spanish-speaking jurisdictions.
-
We’re all in this together, right? We’re the resistance – if you take a “Star Wars” analogy – to counter authoritarianism. Whatever we try out in Taiwan, for example, quadratic voting which is a core radical exchange idea to use a mechanism designed to incentivize people to input more in the process of, for example, choosing which team out of more than 100, to receive a coach for three months to build the presidential hackathon product.
-
This is to solve say, marine debris, or whatever. Anyone receiving the trophy – we hand out five of them every year from the president – gets a presidential prominence. Whatever the present prime minister wants will become national public policy and rolled out within the next 12 months.
-
You see a real way to promote and uplift civic tech into g0v tech, like immediately. For that, we need social legitimacy. We use quadratic voting where everybody gets 99 points and they can vote. 1 vote is 1 point, 2 is 4 points, 3 is 9, 4 is 16.
-
These 99 points you can only vote 9 votes, but you still have 18 points left, so you have to look at another and vote for and then maybe another one. Then you found out maybe you do a 7 and 7. Basically, making sure that the marginal cost and marginal return of each vote is the same. That elicits a truthful sharing of what people really know about those Sustainable Goals.
-
Each is a Sustainable Goal target and builds social legitimacy. After 200,000 votes, everybody feel that they have won when the top 20 gets announced. As opposed to, if you only allow one vote per person, half people, actually more than half, will feel they have lost after each election.
-
When I say democracy as social technology, I mean it in a very tangible way. That’s a regulatory exchange idea. When we share our findings, that makes the case easier for, say, the g0v.it people in Italy to adopt the same for their budget.
-
g0v.it?
-
Yeah, and so on. When we share our idea of, as I mentioned, data collaboratives where the people measuring air quality negotiated collectively with the environmental ministry here for them to put Airboxes, the measurement devices, into industrial areas because we own the land from the government side. They have the social legitimacy, and so we negotiate, one data producer to one another.
-
This helped solve the issue of each individual user of, say, social media companies have zero bargaining power. Anyone who upload a photo tagging other people have decimated the negotiation power for everybody else involved in the same room.
-
This data collaborative idea then become very useful in international settings where people are figuring out how to make a real union-like structure for the people who produce data to enjoy the return and control of how their labor is being used by the surveillance capitalism apparatus.
-
I can go on, but what I’m trying to say is that Taiwan is one of the labs to try out…
-
Taiwan?
-
Yeah, is one of the labs to try out those ideas. When we publish, we always publish with an eye on international applications.
-
One last question because you’ve got to go. Obviously, one of the challenges for democracy is populism. I suppose I have to ask from a very broad level. How, in your mind, do you address that? You’re building tools for public participation. How do you prevent that from being used for populism or majoritarianism?
-
The problem of populism is not that it’s popular. It’s that they exclude people from population. That’s its real danger. If a strand of populism, in the strict definition of anti-elite, establishment, can include more people – people under 18, people who are not born yet, people are people but are traditionally seen as mountains and rivers, and so on – natural personhood, that’s a thing.
-
If populism can include these people as people, then I’m all for that inclusive populism. Populism is getting a bad rap because it’s exactly the opposite in many jurisdictions. Populism is popular with certain people, so popular that they don’t consider other people as people. That, of course, is the real danger of modern-day exclusionary populism.
-
My point is that if we can build the mechanism of democracy toward transculturalism, then from a transcultural ground, any attempt at populism will necessarily have to speak with all those different transcultural ideas because they cannot capitalize on the silo multiculturalism where each silo consider other silos non-people.
-
Every person will feel that they have different lineages within their own personal stories. Any attempt at exclusionary rhetoric will be an affront to that person’s personal life story, being transcultural. That’s what I mean.
-
I wrote this short haiku-ish thing on the name of the country. [laughs]
-
The one on your twitter?
-
No, this is a new one.
-
This is a new one? When was this written? [laughs]
-
Last month. A transcultural republic of citizens. That speaks to really a very old idea of various flowers or 諸華 you’d say, a very old idea from the warring states, that there’s all the various civilizations.
-
When one is mobile, one can travel between the civilizations and build a transcultural norm from various civilizations. That’s the oldest idea of the rites, or the leader’s job, is specifically designed to say that.
-
But of course, if you look again at exactly the same words from the PRC perspective, they are doing what we call a “夷夏之防”, right? The barrier between the civilized people and the barbarians. That’s the literal translation.
-
When they say they agree to the "rejuvenation of the Chinese civilization," they certainly don’t mean various civilizations including all the Western and indigenous ones. Whatever kind of 华 they mean, is a kind of hegemony of the 华 civilization at the expense of the others.
-
The idea of 華, though, can be interpreted as "between flowers," which is the 天下文明 vision, even literally 萬華, as in ten-thousand flowers. These two are actually two opposing world views, which is why I systemically refuse comparison like what do I think about how fast one is catching up on the other.
-
It’s opposite directions. It doesn’t make sense to compete if you’re running very quickly to a different interpretations of the core idea.
-
Cool, thank you. If I have any questions on anything, I’ll send you an email…
-
Sure, of course.
-
Is that the best way to follow-up?
-
Of course, yeah.
-
OK. Thank you, so much.
-
No problem.
-
Very nice to meet you. I appreciate you making a few moments this morning. Good luck with the cold.
-
Yes, of course. Yeah, it will heal. I’m pretty resilient… Just like Taiwan’s democracy.