OK. First, can you describe PDIS? Say whatever you want about it.

The Public Digital Innovation Space is literally a space. It comprises of physical spaces that’s on the ground floor of the administrative building. We have two offices. On the third floor, we have another space, and here, the social innovation space.

We also have the office here as well as the [non-English speech] , the nobody’s library on the second floor. It’s at a total, five different spaces. Of course, there’s countless -- actually, countable -- digital spaces online, as well. We have a GitHub. We have a slide channel. We have a Sandstorm Rocket.Chat, a shared comment board, and a lot of spaces.

The idea of the space is very easy. Everybody joins here to do something for the public good. People get to decide what project they work on. There is no commanding relationship between the PDIS members.

People who join PDIS, primarily through three means. I’m a political appointee. I get to appoint two other secretaries. The three of us is the core co-founders, if you will.

Then there’s people working other ministries. I asked their minister to be stationed here. We have seven different career public servants stationed here. The implicit promise is that I won’t ask for more than one person from any ministry. Also, we have the technical team from the III.

From the what?

From the Institute of Information Industry, the III. The III, again, supports six regular staff and three contractors, or consultants.

Nine people came from the III. Again, I ask each department of III for one person only, and not two people. Basically, it is a very crosscutting skill set, as well a very crosscutting composition, including NPO — if III counts as an NPO — and career public servants.

The idea is that whenever we lack any skill, we try to recruit people who are givers, who give more than they take, and they excel at something that none of the existing PDIS members are good at. That’s our only two recruiting criteria.

What are the benefits of social digital innovation and, more broadly, technology for the government and the citizens?

The benefit of the digital are twofold. First, that is can overcome space restrictions. We can connect these connected spaces. We can connect remote islands, rural places, and so on, using digital technology so they can be part of the policy-making process by having people speak in their own habitat and having the people in Taipei, or in anywhere, really, in the world to understand their life story more. That’s overcoming the space difference.

We can also overcome the time difference because everybody has two minutes of kindness. If we can let people join in their ideal two minutes, whether to sign a petition, to do online voting on pol.is, or whatever, then we engage people in their best, most altruistic public-minded time.

If we only say, "This town hall only opens for 15 minutes, and you have to travel four hours to Taipei for that," then it only includes a very small kind of people. By extending the time and reducing the time commitments, we get the best part of everybody.

How would you define social digital innovation?

Social innovation is anything that has a social mission, a social purpose. "I want the society to function in whichever way."

That it changes the social organization. It used to be like this, but now it organized like this, so you changed the social organization.

Then it, finally, has some social impact. Usually, we now define impact in terms of sustainable development goals, but there could be other metric, as well.

Again, a mission, an innovation in process and technology, or even organization, so a innovation, and then impact. Mission, innovation, and impact together defines social innovation, as opposed to, say, industrial innovation.

If the innovation is digital in nature, instead of purely face-to-face, then we say it’s digital innovation. It doesn’t mean that it’s wholly digital. It only means that it use digital technology as part of the enabler in the innovation part.

The mission, the purpose, and the impact, the outcome, these two are the core, and whatever models that’s operating in-between. It doesn’t always have to work, but always it has to make something new. If it fails, then it fails in a public way, so that everybody else can learn from it, and then do some other innovation.

The most important part is the experimentation part, right?

Exactly.

How does PDIS work in practice? What kind of culture are you trying to implement inside government institutions?

We are trying to make a culture based on trust. We have the mutual trust between sectors as the core, because in our members we have people of all the different parties in Taiwan. A lot of us, like us, are Independents.

We have people from all the different generations, different discipline, different skills, different ministries, different stakes.

If we are to increase trust at all, we have to first build trust between people here, in PDIS. We do it by, first, being transparent. Everybody work out loud, so everybody knows what everybody else is doing, even though they may not partake in it.

Then accountability. Every week we review as the comment board, or every couple days we do a stand-up meeting so that everybody knows what people are blocked on and what kind of help any member wants. We can freely enlist each other’s skills to do worthwhile projects.

Can we say that you’re trying to implement an open government culture?

Yeah, but first within the team, and then trying to use the same model to influence the peer network, the vTaiwan network, the user advisory council network, and, basically, everything we touch.

How do you interact, collaborate, exchange with the administration and public servants of all levels and ministries?

We have this participation officer network, which is every ministry, by regulation, has to assign a team of at least one person -- but now they’re all three, or to five people now -- of all the different skill sets.

There’s people who work on public communication. People who work on information management. People who work on design. People who work on law, policy, and things like that.

Their core task is to engage with any stakeholders that show up. Instead of just with the representatives or with the media people, the professional journalists, they interact with e-petitioners, people who propose a referendum, people who somehow try to engage in any participatory agenda setting, is the participation officer’s business.

How would you define stakeholders? Who are they?

Anyone who could be impacted by a public policy and can declare that this is my stake is a stakeholder. Basically, we use the so-called open multi-stakeholder governance model, in the sense that anyone who can demonstrate to other stakeholders that "I really have a stake here," they don’t have to find a representative. They can, themselves, show up and become a stakeholder.

How did you come with this open stakeholders model?

This is actually the first political model that I learned about when I was 15. That was in 1996.

I participated in the early transition of the early web, like Gopher, HTTP, Archie, to the modern web, which is based on the World Wide Web, HTTP, and the World Wide Web Consortium, the World Wide Web governance model.

The W3C model, again, derives from the Internet Engineering Task Force model, which, in turn, it derived from the early Internet culture. All this is based on the idea of rough consensus and running code.

Rough consensus meaning in our meetings we don’t make a voting. That the idea is we reject voting, kings, and presidents. The idea is that we ask what are the solutions that works for everyone and what are the core values that people can agree, despite their initial disagreements. What’s the non-controversial essence?

All these are written in a very Taoist way in a document called "The Tao of the IETF." It described rough consensus and all this idea.

This is the first political system that I ever engaged with. I was 16 at the time. I couldn’t really actually vote. I got the participatory democracy first, and then I learned about modern representative democracy, which is a pale shadow of it.

I agree.

I’m having this indigenous digital native viewpoint and try to always test the applicability of the rough consensus model, multi-stakeholder model to actual day-to-day governance.

What does running code mean?

The what?

Running code.

Running code means that as long as we have a rough consensus, or as soon as we have a rough consensus, people start implementing it in every kind of way. It’s called about early experimentation to test whether this rough consensus makes sense.

It’s running code in a sense that everybody can see how it functions. It applies to, for example, our new sandbox act, is the idea of running code, in a sense that if you want to try fintech experimentation that could be against the law, you get this six months of experimentation period.

Then you have to publish the data, the algorithm, the things that you want to test in this law-breaking sandbox under the new fintech sandbox act. This is the idea of, in Taiwan, at any given moment, there may be dozens of experimentation running. They’re all running code of maybe one central idea.

Maybe only one or two of them would turn out to be socially beneficial, and then the rest of them will actually go away after the sandbox period, but one or two good ideas will influence the regulators and become the new national regulation.

What kind of obstacles did you encounter when collaborating with public servants or PO, and how did you overcome those obstacles?

Not at all. There’s no obstacles.

Really?

Yeah, because we’re a voluntary organization, so people only join voluntarily, and they collectively vote on what to tackle. The only thing to manage is the culture, or the expectation. If anyone want a quick fix of anything, this kind of open model slow process is not for that. It needs time. It needs space. It needs trust.

Once people learn, "OK, so this process is only good for this kind of stuff, and other kind of stuff, maybe, is about matter of national defense, national security, so we don’t bring this here," as soon as we do this kind of expectation management, then we encounter no obstacles.

You didn’t encounter, for example, timeline obstacles?

The what?

Timeline, because institutional timeline and civic hacker’s or civic maker’s timeline are quite different.

Definitely, which is why we make time for each other. The joint platform is known for its 60-day timeline, so every stage, like from the petition to a response, or from a regulatory announcement to the end of commentary period, everything is on this two-months’ period.

It’s not too long to waste the patience or the relevance of participation, but it’s not too short so the trust cannot form. We usually settle on two months, give or take one month, timeline.

In the PDIS map you formulate PDIS objectives, and one of them is to facilitate or ease the way public servants work.

It’s our values. It’s not objectives.

How do you do that in practice?

To facilitate and empower the civil society or innovation?

The way public servants work.

It’s very simple. It’s by introducing digital tools to simplify the frontline staff’s work. It’s by reducing or absorbing the risk of mid-level executives by having the minister or the president absorb the innovation risks. Should things go wrong, it’s always our fault.

For the senior public servants, usually they actually have a very good idea of where the country should be going, but they may have issues in explaining it. Previous ministers was taking all the credit and giving some, or most, of the blame to the senior public servants. We see that during the Sunflower Movement very clearly.

In this case what we’re trying to do is that we share the credit, so whenever a PO does something really well, we put him in the slides or put her in our communications, so everybody knows that it’s a senior executive that actually get this thing right and do the innovative thing.

Our full transcript of the drafting area of the pre-meetings of all those policies also help, because journalist can go back and see who actually proposed this good idea. Basically, by sharing the credit, absorbing the risk, and reducing workload we try to win the heart of all the levels of public servants.

Those things are not fungible, so we must not do two of them at the sacrifice of the other. That always backfires. All the improvement we do may be small, but they are Pareto improvement, meaning that they don’t sacrifice any level.

What kind of digital tools do you use for example?

A lot, like hundreds.

Some examples?

The main working tool is that after every internal meeting we have this Etherpad, which is a collaborative document. It’s like Google Doc, but it’s a cyber security-hardened, white hat hacker tested.

We get the white hat hacker team that won the second place in the DEF CON to test this system. It’s open source, so it’s actually line-by-line full auditing. It’s not just black-box testing. They say it’s really safe.

We have this very safe collaborative document in which, after each meeting with me, there’s 14 days or 10 working days for all the public servants involved to go back and edit the transcript for their comfort. First, it makes them all look very professional, because the unprofessional parts are edited out.

The second is that it makes them actually listen to the different side’s idea, because during the meeting all you want to do is to talk, but you don’t actually listen. During the editing they absorb some of the good ideas.

This transcript, after the two weeks, is then published to the public Internet for everybody to see. We get visitors from all over the world. Sometimes I get emails, and sometimes there’s public messages in Taiwan responding to parts of this conversation.

Again, I’m acting as a channel to bring only the professional parts and translating it to professional language to the next meeting, so that public servants think, "Oh, the civil society, the netizens, they’re so professional."

It increases trust both ways, you see.

It kind of interprets...

Exactly, it professionalize both sides as a facilitating role. That is the core system.

We have a lot of collaborative chat system, collaborative task management system, collaborative spreadsheets. Basically, take everything that you expect from the office off-web, but make it collaborative.

Which changes have been made, thanks to the PDIS, within the government and the administration?

Hundreds.

Literally. Larger ones.

Some examples?

We passed quite a few central government regulations regarding to innovation.

Related to innovation. For example, the Presidential Social Innovation Hackathon.

What’s that?

It’s a great example. It has an English page, so you can check out the English page.

I’ve read it.

The Presidential Social Innovation Hackathon is a three-month-long hackathon, a real marathon, in which the first few weeks we ask the public servants to work secretly, or not so secretly, with their social media collaborators, their friends, in the civil society to try to rephrase their often neglected social innovation ideas, maybe hidden in their drawer desk for a very long time, into civil society proposals that they happen to provide the data for.

We try to encourage public servants of all parts to collaborate with civil society to bring up proposals in the Presidential promises.

Presidential promises?

The promises. All of Dr. Tsai Ing-wen’s platforms. When she ran for president, she promised improvement in these areas.

To achieve those improvements?

To achieve her agenda.

It’s great.

That’s why we voted her. We might as well participate. The idea is that we outline the parts in the presidential promise that are not yet fully fulfilled. Then we asked...

The project has to focus on that only?

Exactly, yeah, on the parts, but it’s very broad. It’s long term healthcare. It’s youth employment. It’s traffic and tourism. It’s Medicare. There’s a lot. Anyway, anyone can propose such a cross-sectoral idea.

The only requirement is that it has to be innovative, it has to be feasible, and it has to be having a social impact. Again, social innovation. It has to be a social innovation. Basically, what we see from the 108 applicants so far, many of them are not just declaring the before and after of social impact.

They’re actually outline the shortage of skill sets in their respective agencies. We have people asking for data scientists, for professional experience designers, or civil designers. We have a lot of people asking for AI expert, whatever that means.

In any case, what it outlines is a shortage of skill in the public service for some particular roles. After this initial stage, we select 20 proposals, along with their human resource requirements, and then ask people to participate.

Anyone, any hacker who fits these skill sets can then join, add them into the collaborative, ad hoc team to implement those social innovation ideas. Then we will not give out any cash awards, because it’s impossible to compare a disaster relief network with a youth employment program. It’s just impossible.

What we will do instead is having the presidential office serving as the project manager of those projects. Whatever silo effect, whatever cross-departmental political will issue they had, we will no longer have.

The president herself and her staff will help the other UN, as well as the administrative, to procure the data they need, the political will they need, the communications they need for these social innovation to happen. The president promised that, remember?

I don’t remember.

That’s the idea. Then after three months of intensive collaboration, maybe five of them will actually show some promise into being integrated into the normal, everyday work in the public workforce.

We have now the justification saying, "We should hire more people with these skill sets." The president herself gave out trophy awards. That is a hack, really, to try to circumvent the existing bureaucratic structure of reporting, and then really promote social innovation at the core front line staff level.

The judge panel includes the CIO of local city governments, including digital minister, including the director MIS and National Human Council, including people from Academy Seneca, from the social impact people.

Basically, we have people of all the different sectors, related to all the different stagings. If some issue or some proposal is flagged by anybody in a judge panel as infeasible, not innovative, or whatever, we will actually have a face-to-face discussion right here in the social innovation lab, actually, about the feasibility.

We try to select the 20 cases that will make the president’s time worthwhile. That’s the basic idea.

The objective for the hackathon is to select 20 cases?

Yes, for this year.

Is the presidential hackathon a project from PDIS?

It is a project that PDIS, we wrote the voting rules for judges. We helped building the website, the main English website’s visual design. We did that. I did all the typo fixes. Also, I tried to design the judge rules and the timeline.

That’s our main contribution, is to make it a noncompetitive, truly collaborative hackathon. All the administrative parts are handled by the III and the GACC, which is the president’s NPO. You can look it up.

In any case, and every other thing are jointly participated by the NDC, as well as all the ministries and local parties involved.

Who has suggested the idea to organize that kind of event?

One of the youth councilors suggested, actually.

Maybe it’s a bit redundant, but what kind of PDIS projects have been successfully implemented and audited by public servants or inside government institutions?

A lot. For example, we pioneered the idea of the teleworking intern, which was a very new idea in the public service.

What’s that?

An intern is someone who works in the office of a public service agency, such as your or somebody else’s linking profile. A teleworking intern is someone who can work at home. We have a teleworking rule. That’s part of one vTaiwan’s deliverables back in 2015.

Applying that rule, anyone can work at any place, satisfying some requirements. The internship program of the government agencies, funded by the Youth Development Agency, the YDA, has never connected with the teleworking initiative.

PDIS actually did a reinterpretation of the rules, and made sure that last year, we have 25 interns who are all over Taiwan, whose whole job is to make sure that our iPad, as well as our MacBooks, the Safari browser should look the same as a Chrome on Windows or Internet Explorer on Windows.

To check all 508 government agency websites and make sure that they work across the board. It’s an improvement in the sense that all the young people who reported issues -- they may be a usability issue, or whatever issue -- are then handed to a team of people who are skilled designers and coders, who are also interns, who actually propose changes in HTML, JavaScript, and CSS.

From the agency point of view, they bring gifts. They can just give those things to their vendors and contractors, and instantly get a facelift. I think this is awesome, because it makes the agencies trust the young people more.

It also encourage participation, because otherwise only people living in Taipei City get to participate in this kind of internship programs. We have people all over the island. Then this year, we have expanded to 35 people.

The manager are no longer PDIS people. They are veterans from the previous, from the last year. They now become senior interns, who design this whole program of this year. Our management cost is zero. They are on the same Slack channel.

This year, they are using iOS and Android phones to check the responsive websites of all the government services. Again, to ensure the same experience across different phones. I think this, again, is very worthwhile, because people, that’s just everything.

They could be usability. They could be hard of understanding. They could be friendliness, up to some handicapped people or whatever. We don’t restrict what they report, but what they report, again, an information expert intern team takes care of fixing it.

Again, for the agency, it address the problem of them not able to communicate these requirement very quickly to their vendors, and then bring them with real gifts. Again, this is a way to engender trust. This idea, once we brief them in the cross-ministerial meeting, actually every ministry’s MIS has heard of this story, because they receive those reports.

We see the member of the cabinet, including National Palace Museum, which is a member of the cabinet. It’s a very special thing. The National Palace Museum actually now started to work with this kind of engagement through teleworking.

We see many promising developments in other places also. The participation idea, this program, increased a lot after we did this participation.

How PDIS achieve to change the policy or the law?

The law.

Right, we don’t directly work on the law. We work on the lawmaking process. For example, the fintech sandbox. That is something that we oversee from the beginning to the end, but we didn’t write anything in it.

We used the vTaiwan process to get the stakeholders. We get the stakeholders to meet each other. We try to absorb the common grounds. We translate the common grounds to legalese. Again, we did these things, but we here is a very broadly speaking multi-stakeholder team.

The core PDIS people just ensure this process happens. What we make sure is that everybody has the patience to go through the process, and to wait until the right person has the FSE, and then this passes into law.

What we do is basically let people not lose in each other during the process, and also increase trust because of the process.

Which projects have been aborted and why?

Hundreds of them. We tried many, many silly ideas, and most of them doesn’t work. There’s no risk, because there’s no cost, either. There’s many. For example, Shuyang brought this business origami thing that we use to plan our weekly road map.

It just turns out people here prefer iPad much more than real, folded origami paper models. As soon as Fang-Jui joined, everybody switched to RealtimeBoard. Nobody really use the business origami anymore, but we thank the Hitachi people for providing that initial inspiration.

There’s many things that we try, but ultimately doesn’t work, or gets replaced by better digital tools.

Last question. How do you imagine the future of the PDIS?

The future is perfect. The idea, very simply saying, in this environment, where experimentation and failure is the norm, and the occasional innovation gets hyped beyond belief, it’s very difficult to do anything wrong in this environment.

What we’re trying to do, very simply put, is to package the value-based culture, the virtual ethic that this thing is we are working, and the ontological policies that we build out of as the manifestation of those core values and try to spread these parts around the world.

As soon as any of our processes, the little green cards, there is the outer circle. As soon as anything, like a dandelion seed, lands anywhere in the world, it is a victory. I think that the future of PDIS really is not in Taiwan. Taiwan is just where this governing model first gets hatched.

Again, we learn a lot from our oversea compatriots. As soon as we try something that we feel really works, we document the hell out of it, and we do a lot of English fellowship with mostly English-speaking, but now, also Japanese-speaking places.

We try to expand that more. The idea is that anything that increase the mutual trust is a gift to humankind in this era of Anthropocene’s recession of democracy. Basically, we were witnessing this phenomenon of, the more democratic a country is, the less trust it has between the stakeholder groups and the government.

So true.

It’s certainly possible that democracy could fall out of favor. However, today we still feel that democracy is a worthwhile experiment, so we are committed to carry on the experiment.

It may or may not work out in the end; maybe authoritarianism turns out to be the way to go...

Possibly, there is no democratic state at all.

Possibly, yeah, exactly. So, what can we lose by trying, right?

OK.

So we might as well play. This is like a sandbox, where we try with radical freedom, absolute freedom of speech, of assembly, and things like that, and see what works. What works, we can export or share with other democracies around the world to maybe convince them to stay with democracy, as low-trust as the idea is.

With the new National Referendum Act, with the new Digital Communication Act, which legalized the vTaiwan process, with the new participatory acts of land use planning and things like that, Taiwan is the right ground for this kind of mass-scale participation.

They have to go through digital tools, because otherwise, the public service is just overwhelmed with all this participation. We are in a very unique time point where there is not just external pressure, but also internal pressure for us to do those additional innovations.

Just like Taiwan itself, we may suffer earthquakes because of the plate tectonics, but we grow five centimeters every year because of this.

Thank you.

We’re good?

Yes.

Thank you.