Hi. I’m working on youth engagement, social innovation, open government, and digital semi-ambassadorship to semi-sovereign global multinational entities. Those are my main portfolio so far, which are connected but they’re not really the same thing. I wonder what are your interest?

I would like to learn more about your theory of change, and how to sustain the theory of change within a government setting.

Very simply put, my office is one delegate from each ministry. Taiwan has 32 vertical ministers, and each one can send a person to my office. While in my office, they still report to their minister. Their salaries and the scoring, ranking, whatsoever, they do it themselves. I don’t score or rank them. It’s a purely horizontal team. That means that I may have at most 32 colleagues each from each ministry.

At the moment, I have maybe 20, meaning that some ministers never sent anyone, like the Ministry of Defense never sent anyone. My scope is limited to the ministries which has sent delegates to my office. Mostly people-facing ones, the Ministry of Education, of Communication, of Interior, of Culture, of Law, of Finance, and recently Foreign Service for the past couple years. That’s the people-facing ones.

The idea very simply put is that I don’t order them to do anything, all I ask is that they work out loud, meaning that we seek opportunistic synergy between those different ministerial programs. For example, the e-sport, which is an emerging phenomenon, and the Minister of Culture, Education, Economy, all have a stake in it, but they also have a stake in not absorbing all the risk associated with it.

My work is just to moderate this internal discussion, making sure it’s made into a transcript as we’re making now, making sure that outside stakeholders learn of this transcript, and they all have their own suggestions, and I bring it back to this cross ministerial collaboration settings. At the end they’ll settle on something that leaves everybody better off, a Pareto improvement.

Of course, it would be sustained, because then they would feel like they have a stake themselves in sustaining the systemic change. I’m more like a crowdfunding site, where we solicit peoples’ commitment and only after people commits actually put it into strategy.

Are all secondments joining your office a top-ranked official so they can command their staff?

On average, people who join my office is around section chief level. That’s a mid-level management. There are of course exceptions such as the communication delegate was Director General of Legal Affairs, so that’s pretty high, like a high-level public service.

There are also people who joined my office just as section worker without section chief position, but after rotating back to their ministry, we always make sure they get promoted to section chief. There’s two people have done this way. On average, they’re around section chief level. They’re not exactly what we would think as top-down, like deputy minister level.

Those people are actually where we face the most resistance when it goes about digital transformation, because for one, they absorb the risk if things go wrong. The ministers can blame this level. On the other hand, if the digital transformation isn’t done well, they must also absorb the extra work associated with it by making it right.

If, for example, you build a healthcare website, [laughs] then it goes wrong, and you have to clean after it, the section chiefs bear most of the burden, they work overtime, and so on. Finally, they are the level that gets the least credit because if they’re the front-line staff, like social workers and so on, at least the constituents know them personally.

If they are very high level, like director-general and higher, then the media loves them, and so they get that kind of credibility. Mid-level is anonymous. For most digital transformation tasks, they absorb the risk, they don’t get the credit, and they may have to do extra work.

By making the team composed mostly of mid-level management, we make sure that whatever innovation we bring out actually saves time, gets more credibility, or reduce risk for this level. It’s never in sacrificing the other two. It’s a very Taoist way of change making.

Would you like to give me one example?

One example. We have this mechanism called Presidential Hackathon where everybody can propose a solution to a structural problem, a global problem using digital means, using collaboration. Every year, there’s more than 100 proposals. There is a wishing pool or wishing well where people can write their wishes.

We choose from the 100, 20 teams every year to incubate them for around three months. Then they do a pilot. I understand that the US government has this 10x program. Like in Germany, they have Prototype Fund and so on that’s more of a social innovation part. What we have in common is that we are solving structural issues. We are not just solving on a case-by-case basis. I don’t know too much about 10X, but in Prototype Fund, even if it fails, it’s very good, because everybody learns that this approach doesn’t work.

There’s a postmortem, there’s a discussion, there’s a cohort that can support those internal innovators. Their main problem is that when they work, there is no binding mechanism to guarantee them budget, personnel, and regulatory change to actually scale out the impact. In our Presidential Hackathon, usually after three months…For example, last year, we give out five trophies every year.

One team looks at the Orchid Island. There was a helicopter that carries the patient to the mainland hospital, but it was raining, it was dark, and the helicopter crashed. They do the root cause analysis and found out that it’s because the local people do not trust the local nurses, because they lack the special medicinal training, and so they would insist that they get carried to the main island.

There is currently at that time no legal framework for a nurse using teleconference to have a special doctor directing her to do medical treatment. The Electronic Signature Act also has not been embodied into regulation that allows for this kind of healthcare record sharing between the helicopter centers, the large hospital, and the local clinic.

They did a prototype within three months and found out it really increased trust of the local people to the local clinic if they can do this kind of three-way video conferencing. After three months and it’s a good idea, they won the Presidential Hackathon Award. There’s no money. The trophy is a micro projector. If you turn it on, it shows the president giving the trophy to you.

It’s very useful, because it’s a presidential promise that whatever they have done, whether it’s fixing the water leak detection using machine learning that shortens the two months detection period to two days in Xilong region, or as I said, a video conferencing, we will do whatever it takes to make it a national program within the next 12 months. That’s the presidential promise.

We changed for the telemedicine case one law, maybe five regulations, lots of budget. If your director-general say, “We don’t have a budget,” you just turn on the projector, and there was, “Oh, we have budget.” If the minister is saying it’s too hard to talk across the ministerial silos, you just project the president, and you have the talk.

In many cases, we found out even if the academy or the social sector proposing, actually maybe a section chief wrote the proposal, but they would say “I’m just in collaboration with this citizen proposal.” After they won the Presidential Hackathon, they will come forward and say, “I’m actually a public servant. I actually initiated this idea,” because they got the political will.

This mechanism, as you can see, it reduce the risk if it doesn’t work, for the ministry just attending a hackathon a few weekends. If it does work, it gives them maximum credibility. That is the kind of mechanism design that we do to support the mid-level career public service.

Is there prize money?

There is no prize money, only a presidential commitment to winning teams.

There is a support team, of course. It’s a incubation process. We have people from III and so on working on those teams.

What I have heard from the 10X is that it’s like an internal rapid prototyping service, but at the end of it, the operation or the maintenance is not for the 18F to do forever. There must be a hand off process to the external vendors. I think that’s public information. It says right on their website. It’s good to know that it’s the same being carried smoothly.

It’s fortunate that in Taiwan, for information services, there is the Institute of Information Industry, which is sort of like 18F, in that they have very flexible staff. It’s a not-for-profit and jointly created by the large IT vendors as well as the ministries to maintain a not-for-profit organization that can do a lot of flexible salary for experienced scientists, researchers and people doing development, design and cyber security while not being restricted by the total amount of public service employees or their paying grade.

It was designed with a very similar motivation as 18F, except it’s maybe 30 years ago. It’s a very long lift, 18F. We worry less about sustainability. The two pillars, the III, for information, industry, and the ITRI for technology research, actual industrial technology, these two are the two pillars that the Executive Yuan, the cabinet, works very reliably with.

They do a lot of spin offs. If one team does it exceedingly well for a ministerial project, they can get spun off to do this service for many ministries and even the business community. That solves a issue that has been going on for over 30 years, is that is it unfair competition with outside vendors when the III staff are literally co-works with the ministerial staff? Are they in a privileged bidding position?

Past couple of years, there is a new doctrine that says the III can do scoping work, research work, and so on, but for maintenance work, it must never join that kind of bid and spin this off. It’s OK for that team to be spun off into a maybe social enterprise, and that carries the maintenance, but then they must bid fairly with everybody else for the ongoing work.

It’s OK for the research part or the scoping part to be done as procurement to III. That’s an equilibrium we’re at right now.

I would like to know more about art of dynamic facilitation, of how to stay calm among rising tensions that’s being argued by different stakeholders.

For me, my work is mostly on taking all the sides. If there is an argument involving five ministries, each with different position, as they should, otherwise they should merge, right?

Five ministries with different positions. My facilitation, as well as declaration meetings, people wear the T-shirt with this symbol. This is the Sustainable Development Goals.

What is this?

The global goals, the Sustainable Development Goals as agreed by the UN back in 2015, that these are the 17 things that we should do as humanity by the year 2030.

The good thing about it is that it was done in a crowdsourced way, meaning they actually listened to one million people around the globe, asking a single question, “What is your ideal future for the year 2030?” A million wishes.

They did amazing bureaucratic work that finally condensed them down to 169 concrete targets, making sure that each one reinforces, or at least doesn’t harm, the other 168. Then they sorted them into 17 colors. This is awesome because that means that it shows we all add parts to the puzzle, even though our positions differ, and we have the same value, which is sustainability.

Nobody argued that the earth should go away by the year 2030. None of the people would argue that. People generally agree that we want to leave earth as good when we leave it as when we were born into it. That seems to be a very common value, whatever other people’s positions may be.

My work is mostly towards sustainability. Someone contributes the economical part, but someone must contribute also the environmental part. They may seem at odds, but they’re actually contributing both to sustainability. Someone may contribute to an industrial innovation, but somebody may argue for social justice, which also seem opposite at times, but actually, they’re all adding to sustainability.

The great thing to set the horizon at 2030 or 2050 is that most of the capitalistic investment arguments, if you’re looking for short return, like next quarter, you will do so by creating negative externalities to the environment or the society.

If you’re looking at 10-year, 20-year, 50-year returns, then the capitalistic interest actually aligns with most of the other public welfare interests, because they must take care to not damage their constituents. Anything that takes place 10 years or more in the future is a really good way to facilitate by inviting people to consider for a longer term.

Once we keep doing that, either face-to-face, using dynamic facilitation, or asynchronously by having one side coming to meet and argue for something, and make a transcript published to let the other stakeholders review it, they, of course, have different positions. They also come to meet, and then asynchronously, they converge on something that they can live with.

The next thing, other than taking other sides is that, without seeking consensus, which is too strong – consensus is like you can sign your name on it, the MOU-like thing, a treaty – we just seek a common understanding. If people can live with it, that’s fine. In Internet governance, it’s called rough consensus, roughly live with it. Here, we call it common understanding.

Once we reach common understanding, that’s good enough. Rough consensus is a running code, people just do whatever, and then we emerge again. I usually use an AI system called pol.is to help this facilitation. Every time, you can see that people’s feelings, people have different camps. People have things that they agree or disagree with.

Once you click agree or disagree, your avatar moves toward people that feel like you. You look at the next statement and say agree or disagree, and either use K-means clustering to find the people’s groups, and use principal component analysis to find the most diving points. It dynamically reflects the crowd back to itself. There is no reply button, so you cannot do personal attacks.

After a while, we always end with this shape, which shows people that whatever you have seen on institutional media or social media is but five percent of what we are having in this idea solution landscape. Most people agree with most of each other on most of the things most of the time.

We should just use this as the agenda and just get this rolled out, without any judgment on those large ideological things. Once you’re just seeking common understanding, which is like this part, instead of consensus, which is like this part, it actually makes it really easy to take other sides.

This is the real consultation they run in Kentucky, whether they identify as Democrats or Republicans. Everybody feels the arts are important. Not just STEM, but STEAM should be rolled into K-12. Nobody is against that. It doesn’t cost much. If the mayor adopts it, it will increase the reelection chance. Why not do it? It’s low-hanging fruit.

By keep doing this, we actually make sure that people see, despite their different position, there are these common values and they just come closer with each facilitation.

How do you stay calm?

One idea is troll hugging.

Trolls are people who make people upset as their hobby on the Internet. They will reply to some social media post with messages deliberately trying to provoke people into have a negative interaction with them. If they’re not professional ones, they are sometimes just craving attention that they cannot get from a social setting.

They just provoke people into having a conversation with them. That feels pretty empty, because it’s not relational, it’s transactional. Next day, they just wake up and troll some other people. My hobby is to hug the trolls by making sure that people who address me directly and try to provoke me, I interpret their comments and only reply to the constructive parts. If there are any part of the message that makes me upset, I make sure that I just put on some new music, some new tea, and associate with the negative verbal stimuli with a pleasant, nonverbal stimuli.

If I sleep well that night, it will form a long-term association so that by the next time I see this word or I hear this word, it will be associated with something really pleasant.

It’s like cognitive behavior therapy, except it’s initiated by trolls.

Just encourage the parts that are authentic. Encourage the parts that are consistent, and just simply ignore the inauthentic and inconsistent parts. If that affects me personally, it’s like a massage. I feel hurt, but because I have a kind of complex here, it pays to just relax a little bit, hear some music, drink some tea, and then re-associate that into something that’s pleasant.