My daughter just asked me, she would be hearing about this, "Could you please explain it?" She is in the car racing somewhere. I had like eight minutes to try and explain vTaiwan to her.

She flipped out, appropriately flipping out. Are we going to see you, Audrey?

Yes, as soon as I’m fully dressed and something...

Excuse me. Good morning.

All right. It was probably good.

I’m, again, literally in bed, and since you’re recording...

I won’t push anything. I went back over my old list of questions. I have things there that I’ve extracted. I’m still curious about a lot of the new stuff came from your New York tape, Shuyang. I just took lots of notes during it.

I now know enough to get down to more details. One of the things I am interested, I get this picture from Audrey that everybody just sort of floats in and organizes things and floats out. Whatever happens, the only thing it could have...and it’s, yeah.

That’s very intriguing, and I love it. I love the imagery, and trying to apply my experiences in open space, particularly five-day open spaces, which are totally different from one or two-day open spaces, and saying, "How would you actually live that?"

There’s all these staff and interns around Audrey. There’s this sense of g0v. It’s floating around and all these things intersect and stuff. I am just curious because there’s a number of times when you, Shuyang, say, "We did this, so the community did that."

I’m going, "OK, well, who is ’we’? Who is ’the community’? Is there any sense of roles that are...? Does the staff hold a container of process and support and options and you can come talk to them? Who is doing what?

"Then we decided when we were finished with the pol.is..." I’m going, "What do you mean? Wait a minute! What do you mean when you [say you] are ’finished’ with pol.is? Who’s deciding ’we are finished with the pol.is’? Who’s inviting these stakeholders? Which stakeholders are being invited to the conversation? What’s going on there?"

Anything you guys could say that will help clarify all that, I would love to know.

A clarifying question would be, would you be more keen to learn about vTaiwan as of this week, or more about telling how it was like in 2015 because...

This week is more of my interest although I am interested in the evolutions and how it became that way. As I said before, I made the mistake of saying - and I’ll make the same mistake again - I’m looking for a place to stand in the quicksand, in the river. You can never step in the same vTaiwan twice, I gather.

I am interested in that fact and as it’s evolving, there’s stuff... there is some kind of "here’s where we are" sensibility and here’s what we want to check out or try because of X and Y, and just to get a better sense of it because I want to communicate about this in a way that’s realistic.

I am not in any way resistant to the constant change and the plethora of feedback loops and all that.

Sure. Let’s take a snapshot.

If you just say if something is open space, everything happens, and that doesn’t tell anybody what they might do to. I got it. It’s you guys and the way you look at your scene and interact together and all that, which is fundamentally at the core.

It’s not the ABCs that you do. You are creating the ABCs as you go. If there’s no ABCs, there is nothing remotely like what you have which could show up anywhere else.

Why don’t Shuyang describe a snapshot of the vTaiwan meet-up this Wednesday, which is literally two days ago, maybe the week before, and I’ll try to illustrate in the meanwhile and see if that helps clarifying things.

All right.

Two days ago on Wednesday, because we always spoke around at a certain timeframe, Wednesday afternoon to evening. Two days ago, we started from three o’clock, around 3:00, 4:00-ish. We booked Open Kitchen inside the social innovation lab.

I remember I wanted to work on something with my friends there to work on a vehicle that’s supposed to carry a camera around so people online also wanting to participate in vTaiwan could participate with a live stream video. They can have a control to a vehicle which is carrying the camera.

I invite some friends I consider as makers and I asked them if they can help. That was two days ago. There’s some other people from different backgrounds, like two other researchers working on observing open governments seen in Taiwan. There are people who are there to discuss about the current issue on vTaiwan about data policy.

People gathered together and shared whatever they want. There’s also a friend who’s making a start-up. She really need help for some brainstorming sessions, so she worked here and grabbed us whenever we have five minutes to put a Post-It to her wall. [laughs] That was last week. That was two days ago. Last week, there’s some...

Also, the POs came, right?

Yeah, the POs came. PO from Ministry of Transportation and Communication, they came.

They just came? There wasn’t an issue they were responding to or something? They just came because they’re part of this coming-and-going community and see what happens, kind of?

That’s exactly right. She’s here to share the birthday cake of our court reporter, our stenographer on that day. People came for her birthday cake, I think. That’s the general idea.

[laughs] There you go again.

It’s open space so she can actually make a cake. It’s Open Kitchen.

That’s what we...

You have lots of fun talking about the fun you have, don’t you, Audrey?

That’s right. The data integration excuse, I mean, the topic, [laughs] is currently at the snowballing stakeholder stage, so data integration. We have a topic.

Data integration.

We’re not totally topicless.

Is that somehow all the people are involved or that was some kind of formal invitation sent out to people saying, "If you’d like to come talk about data integration, we’re gonna be here," and people know, "Oh, that’s the, that’s a big thing. Anything can happen around the edges, but that’s the big thing," or what?

What does it mean that that’s the topic?

The snowballing survey was finalized last week. We have the URL to the snowballing survey. It was coedited last week. Actually, no, Shuyang got everybody drunk and brought them to dinner last week. Lisa and I remained to finish the questionnaire. It’s actually two spaces, last week.

Before they were drunk, they were very contributive, I think. Anyway, so what we have a...

The snowballing survey.

Contributing on some part.

This is the context of the snowballing survey. The only part I’ve seen a reference to that is in the asking stakeholders, "Who else should be involved," kind of, but I don’t know...

That’s exactly right. If you go to vTaiwan now, so the front page says, "Data integration. We have a, we’re having a snowballing survey."

What happened...

Then, you have this slides prepared by some...Actually, a law firm, the Guo-Ju Law Firm explaining the current situation, challenges, and whatever around data integration and open data and data governance in the current administration. The snowballing survey, I think, is this link. It starts by...

You’re already at the stage of getting...

Stakeholders.

...getting stakeholders.

Right. It starts by explaining what data is, what anonymization is, what encryption means, all of these small lexicons to define a key concept. It begins saying, "Have you ever requested data from the administration?" If you have, you do it for the citizens supervising purposes, academic purposes, commercial purposes, other public interests, or whatever.

You can choose as many as you like. It starts saying, "What have you requested? Are you looking to request next?" If you have been refused before, or you haven’t, that’s the lucky option. Whether it’s because of it’s confidential, it’s private data. There’s no rule for providing it or there is rules prohibiting it. [laughs]

If yes, what? We have, currently, a platform to request more data. Do you think the platform is good enough, and the benefits, and whatever? At the end of this survey, which is rather long, we ask people to identify more stakeholders and provide their email, if possible.

If not, well, at least their names, and we’ll try to find their emails. Here is the snowballing part, and very interestingly, actually...

Snowball, it’s actually a survey about your relationship to this issue, and you think that...

...people.

The name of the person filling in is optional because people may want to anonymously filling this. Email is also optional.

The results of the survey is a sense of what’s going on in this field, on this issue at this time. It’s snowballing because you’re getting more and more people who are theoretically interested in this topic, who are having interest in this topic.

The final question is, do you agree to receive more emails if we have more rolling snowball surveys like this one? Of course, we have sent it to everybody who have previously agreed on previous questionnaires like this. We send it out to all the different communities. That’s the people are associated in the Open Data Taiwan group, the g0v group, the whatever group.

That’s people’s contribution before they get really drunk last week.

[laughs] I’ll have to digest that for I can ask more questions, but I have an overview. I understand the strategic placement of the getting drunk.

There’re some real issues going, real topics we’re discussing in the vTaiwan Hackathon. Before last week, we were discussing about how to formulate the questionnaire, and before that, we probably were discussing about end goals to look at this issue.

This issue surface from somebody saying at some point, "We really need a thing on this topic," and other people going, "Yeah, that’s a good idea," and a bunch of people get together and they start working on it.

That’s exactly right. There was a proposal initiated by five, six people. Only one of them are actually a regular in vTaiwan community. The other ones are from the wider g0v community, from the open data alliance. There’s also one running for councilor now, actually; he used to work in the Ministry of Economy Affairs. It’s a diverse bunch of people, all wants more data integration who raised the subject.

I am trying to understand the pattern here. The petition is posted on the vTaiwan site somewhere? There’s a space for putting up petitions, and then people sign, sign, sign, and the petition.

We have 20 people signed this now, let’s get together and talk about what we’re going to do about it, or what’s the line for petitions showing up?

There’s infinite lines of input. The easiest one is just to show up in Wednesday, during my office hour and talk to me, so the office hour. It just takes one person. You don’t need 20 person, it’s not online. You just come and visit me. That’s the fast track.

OK, stop there. The fact is the fast stream go talk to Audrey, so you have talked to Audrey, and Audrey says, "Oh, good idea. Why don’t you do X?" or Audrey says, "I’ll take it and post it where..." I mean what happens, they’re petitioning you, the king. [laughs]

"Oh, king, here’s my petition." Audrey does what, why would they come to you?

In all fairness, they can also petition to the Premier, who is also touring around Taiwan the same way I do. It’s not just me who are doing this office hour thing. The Premier is doing this thing too.

Anyway, people are petitioning because they understand that we’ll have a complete transparent made of whatever material...

Whatever happened while they talked to you.

Right, what they talked to me. This will be useful because that’s what the vTaiwan community will use to elicit people’s interest in talking about it, or to determine whether it’s not interesting.

We have of course...

They’re taking their conversation with you, which has a link to it and they send it to their friends and whatever, there’s a networking thing that happens and people notice it, and then start to get together and act on it.

Right and in this particular case, the data integration petition, actually having a national open data advisory council, which is a council of experts on open data and so on, its advisory. I am not a chair for that. The chair is another minister with a portfolio, Wu Tsung-Tsong, who also insists that all the open-data related consultation meetings are published in full transcript.

That also results in a transcript. Alternatively, people can go to join.gov.tw, which is a place where people who has more than 5,000 petitions will always get a response but actually what they petition is of a high political interest, like the Mac tax filing system, is exclusively difficult to use, then we actually make a response when it’s 50 people, not 5,000.

The point is that we always capture...

The petition goes and post is posted there, and various people are checking out the petitions and signing this one or that one. Depending on what the dynamics around the topic is a lot of numbers or a few numbers will get a response.

Here is where POs come in, because POs always interview the petitioner here. In the official National Open Data Advisory Group meeting, it’s a face-to-face meeting with very high bandwidths and very firm understanding of what’s going on. And we’re making a transcript of it.

In the written petition in join.gov, there is no such high bandwidths. We just have 500 watts to work with. The POs always arrange for someone to interview the petitioner, so that they result in the transcript.

That’s with the 5,000? Is that the 5,000 level or what...?

Yeah, they always interview, as a rule, when it reaches 5,000. For cases that are really interesting to the PO themselves, they may arrange the interview well before.

The PO or the other people in their ministry?

Their responsibility is in the PO. PO may delegate to other people in the ministry.

The head of the ministry doesn’t necessarily decide if this petition is worthy until the PO go interview them? It’s more the PO is that’s their job is to...

POs are extremely autonomous. They look at it and say, "Hey, we got to handle this one."

Thank you. I have a few steps here. [laughs]

Awesome. Anything you would like to add, Shuyang?

Yeah, I think it’s clear that the POs can choose any subjects actually, not only from joint website also. They can bring up any issues they think that it will be nice to discuss with their broader network of people, maybe from people they couldn’t even imagine to talk with from the beginning.

If there’s any case in their ministries and they feel like, "OK," but they want to interface with smart people, oftentimes they actually can bring up in the monthly PO meetings. We’ll also do a transcript from that. It’s possible to be brought up to vTaiwan, as well.

The Committee for Transitional Justice is a new member in the cabinet that we’re going to have in a couple of months. They will take on things like on our coins and bills, there are the figures of Chiang Kai-Shek.

Do we want the figures of Chiang Kai-Shek to remain on our bills and coins? Very soon, we will have a new PO from the Transitional Justice Committee. I’m sure that they won’t wait for 5,000 people to petition to bring this sort of discussion.

The central election committee also has a PO already, but they have not finished building the National Referendum Act electronic system. Once it’s there, of course, I expect that things will surface from that platform, too.

The PO function is one of the accelerators or attention enhancers, points of initiative in the system? I’m trying to see if when you look at how something comes in to be, at any given point, in the system, any decision is made, if everybody is always totally equal in that, or if there are points where initiative is expected or is institutionalized. This is an example of that kind of thing.

Even though I understand there’s nobody stopping anybody from interviewing anybody and posting it [laughs] on the web, there are certain things which create emergent data order out of all of this.

Yeah, it could be PO who made the initiatives. For example, two days ago, PO also came to vTaiwan Hackathon, but for vTaiwan’s context, anyone can take the initiative to bring up an idea from the civil society. The only requirement is to have a contact also from the government side.

Every issue has to have both commitment from the society’s side and also the government’s side to be able to launch on the vTaiwan’s platform.

That’s interesting.

That’s probably the only requirement, because we need to be able to communicate between the two different...

The POs are the point people for the government side.

They could be.

They could be...

For some cases, it’s some other people, like the National Development Council. The NDC people are also...

Yeah, I forgot to mention, NDC has also its own platform. The law.ndc.gov.tw platform, where people can have petitions, but it’s more like they pinpoint the parts of regulation that are out-of-date as regarding to technical or technological progress. Law.NDC is also in what we call the platform economy regulation.

The platform economy adjustment regulation is a regulation which you start to nominate here is a process in which that anyone can propose a platform economy case and say that the current law is currently in gray area in this. The Law.NDC people will make sure that it’s deliberated both within the government and also with the civil society.

The platform economy regulation is one of the empowering regulations and in which case the POs are not necessarily involved. The NDC can just become the government contact and then ask for civil society. By the way, the platform economy regulation is one of the product of the vTaiwan process, so it’s...

[laughs] Right, again, the...what do you call it? The public?

Platform economy regulation.

The coming around? When I’m losing my words already, not regressive...

Recursive.

Recursive. Thank you. Recursive public, recursive dynamics, that’s a new dimension I hadn’t seen before. Basically from civil society, anybody can play that role. From the government, there are particular positions that play that role. Is it particular parts of the government or people like the POs?

Mm-hmm.

Thank you. Then, there comes the process itself. As you get in, I gather the creation of the educational materials or what we might call "briefing materials," here is a...

I’m assuming that that’s sort of an open space, open source, anybody can put their two cents in. You have some structure that can make sense of that? A lot of people dump stuff, information in and then somebody sorts it out? Is it like a Wikipedia? How does that unfold, and when do you know the process is done and ready to go onto the next step?

That’s two questions, actually. Shuyang may want to handle the first one, how is the informative materials done.

We always say vTaiwan community actually means whoever is having accounts in vTaiwan’s platform and also who have interest to participate in the vTaiwan Hackathons and so on. Essentially, a very broad community, and probably you will have around a thousand people right now.

They have the status and identity. They are specific identities that have associated themselves with vTaiwan.

People...

Yes, yes, they are on the select channel, for example.

People bring in their skills. There’ll be people who are good at translating more difficult law terminologies to easier-to-understand description. Those people will take the lead to make an open dictionary. We call it "mini-lexicon".

They are volunteering for specific roles, or they have said, "I’m interested in doing this." Nobody’s contacting them?

That’s exactly right, yes.

Also, probably people who are in charge of the case in the ministries, especially public servants who have been working on this case for a long time and they understand the regulation very well.

They’ll help out to put out the slides. They’ll be on the slides trying to describe what the issue is about. People who are good at, for example, design can try to make the slides more with a better presentation. Maybe sometimes we also make a logo for a topic.

Sometimes we made a tech line just for the viewers of vTaiwan’s website to understand [laughs] not only a specific case, but also a tech line to understand what this issue’s potentially about and what other controversial points people can probably start to look at from beginning.

It actually does that. There’s a sense of parallel to Wikipedia. It’s an emergent organization, but it is an organization. People have roles, and there’s certain kinds of lines of communication that are...

In Wikipedia, there’s ways of tagging things so that they show up and somebody particular can attend to that piece of the puzzle, whatever.

The current status is we know to launch a topic on vTaiwan’s website. We probably need the dictionary. We need the slides. We need a name and tech line and some description. To fulfill all of that, we make few kengs.

The last video already described about keng. We make this all requirements to launch the vTaiwan case as different kengs and ask people if they want to join and help out.

The symbols you’re showing me are different roles that people have in the system that they choose which labels they want?

That’s right. These are the standard g0v stickers from actually two years ago. There’s many more stickers now. That’s the usual stickers that we use.

Somebody can have six stickers, and "I’m interested in participating in these six ways," kind of...

That’s exactly right. This is to dig a keng for other people to fall in, I guess. This is to volunteer to fill a gap that somebody has identified. I think these two are the most important. Oh, wait. No. This nobody one is the most important.

That’s where you are, right?

Exactly. As Shuyang said, for a vTaiwan case to be published, we need to have a title, preferably in English and in Chinese. We need to have pretty pictures. We need to have a summary, which is a few lines. We need to have slides. We need to have a small lexicon.

We need to have some timeline. We can have a lot of other supporting materials, but these are the essentials.

It’s funny, I think about the pattern language that Martin and I put together. A lot of these, I would love to have a community of practice on a pattern language where this kind of organization happened and the patterns would emerge with all those things attached to them. Anyway, so all this is happening, bustling along, weeks, months go by.

It’s getting fuller and fuller. There’s more and more people adding more and more stuff. Where does it stop? Does it not stop? Somebody just decides, "Let’s, we have enough to start with. It can continue on developing, but we’re gonna do, start the next step."

To start a snowball rolling, usually we have people on Slack, on the mailing lists, in face-to-face meetings, and quite a few ways to signal their comfortableness or uncomfortableness of going on.

Once we have a general sense of rough consensus usually in a weekly meeting at the end of the meeting when people feel it’s generally OK for both people online and offline, then we will publish it.

We won’t publish it if anyone identifies any blocking issue with it. We’ll just say, "OK, we’ll spend another week on it." It’s a very, very slowly convergent process.

The weekly meeting has people who are involved in significant roles in this together, and they’re looking at where are we, where are we, and then it sort of emerges one time, people are going, "Huh? OK, It feels like now is a good time to move on." The weekly meeting is mostly the people who are...

Yes. You’re right.

...involved in doing this particular process.

Yes, and who are excellent pushers who will enlist people who are not qualified, and they are if they require some other expertise. Shuyang is an expert in this.

It’s not just that they were participating, but they’re going to be participating in the next stage also, and their ability to network out, and reach people is as important as how ready the informational material is.

Yes, and find facilitators for it which is necessary for the next stage.

Find facilitators for what?

Who will facilitate a face-to-face meeting making the digestion of the snowballing survey.

OK. I had imagined a pol.is thing in here after the snowballing, but you’re saying...

That may happen if we reach let’s say a few hundred people that we think OK pol.is is warranted.

Now I’m understanding a thing about it being efficient. There’s going to be some kind of reflection on what has unfolded, and what they could unfold through the rolling survey. I had the rolling survey as something that is used to identify stakeholders, and that is...

It is.

...one of the processes that you’re also gathering information in the survey. The information from the survey is going to be given to a number of the stakeholders for a face-to-face, or if there’s lots of people there’s going to be a pol.is, and the results of the pol.is will be given to a bunch of stakeholders.

That’s exactly right. Yes.

There’s like two separate...

Identifies the stakes and the holders.

Identifies the stakes and the holders, interesting way to put it. When is it decided to do the face-to-face and who decides that?

Anyone can host one. However, to get community support, you will need a neutral facilitator.

At any time you’re doing the survey, and the results are being compiled as you go, there’s not a stopping point board?

Because we have to check in every two days to snowball it.

This is where we’ve gotten to, this is where we’ve gotten to, and this is where we’ve gotten to, and any time along that where somebody goes, "I feel we have enough for me, and the stakeholders," I have in mind to go ahead, and do a facilitated conversation.

Yes. A facilitated conversation.

I talk to my friend, Joe, and he’s a facilitator, and he’s going to facilitate that. I’m going to pick seven people out of this, and pull them in, and we’re going to have a conversation. That could happen 17 times and it’s all grist for the mill, and I think the more times, the better.

For the social enterprise case, that literally happens 17 times, so yes.

[laughs] I picked that up out of the field. It’s not like there’s going to be the stakeholder conversation which is somehow decisive. All these things are overlapping, and with the...Again, anybody can do it. OK, I could see how it’s organized to allow that.

Anybody can do it, and anybody can view it. This is the way for people in the public, or the government to take action in the ways that...You provide a whole pile of ways to take action to reflect on this issue, and try and make decisions, and all the rest, and you just sort of create...

Seriously, the only thing we ask in return is for the facilitator to be somehow known to the community. Ideally come into the weekly hackathon, but it can also happen in a remote fashion. Like in the social enterprise case, there are things that are held by at the MP actually. MP Karen Yu, and she’s holding it in the capacity as a social entrepreneur herself.

She was one of the pioneers at Fair Trade coffee in Taiwan. She holds the stakeholder meeting, and again, all the vTaiwan community ask in return is that for them to relinquish copyright to a degree so that we can include this in the vTaiwan website and circulate it, and use it as a material for next stage kind of deliberation.

It’s not like there are standards that are required in order to either facilitate, or convene a conversation, but part of the transparency is this is the nature of the person who facilitated it. This is the nature of the person who convened it. These are the stakeholders that were there, and people can pass their own judgments on what’s most dependable.

You are just providing space for people to do these, and presumably people who aren’t being paid attention to because they’re incompetent will also ultimately drop out. They’re not getting anything out of it.

That’s right and because every week, we review what has been happening in stakeholder communities. If they need a facilitator, or if they need facilitating technologies, the technologists including communication technologists also have their own workshops or collaborations.

For example, Shuyang knows about this person who came a lot in the previous weekly meetings, Hsin-Yi Chen, is now working on a technological tool that lets people look at official documents like PDF files, and identify the points, pro and con to it, and automatically surface them into a mind map.

That works pretty well in other technical tools that we’re working on. That kind of tech, that pol.is idea clusters that I drew down to the relationship between the issues identified as with potential solutions as well as the potential stakeholder who can provide the solutions as core issue based mapping.

There’s going to be workshops in the Sense team, and the IBM team, are going to co-create the tools using concrete cases of course, but working more on the tool level.

That’s an example, the expansion of toolset that g0v is always poking around in, and trying out.

Yes. That’s the design part in it, the UX part.

The tools just appear, and get talked about, and after a while, they start getting used more, and more, and that’s the evolution of the system.

If it’s proprietary, then we try to railroad the creator to open source it, but it was...

...Seattle.

Exactly. It’s already open source. We’re now trying to connect the railroad to a central dispatch. Otherwise, if it’s open to begin with like Sense or like IBM, then it makes it easier for people to integrate into it, and be part of the process.

I think it’s interesting when project merge like when the Sense team came to IBM team. Because I think both teams are in lack of some resources the other team actually has.

They’re complimentary.

Yes. It’s complimentary too. It also makes sense to merge, and to talk about what possibilities. Of course we can create something together, and fork more different versions afterwards, but it’s really a bit tough. It’s kind of collaboration between different skillsets.

I can understand the theoretical level, but I have never been in a geek group doing this kind of thing. I can feel what it’s like, and go, "Wow." It’s definitely a growing evolving ecosystem, and I know ecosystems well enough to go, "OK, this is what’s happening here."

You can think of it like the Chinese Wikipedia article on acupuncture being pulled into the English article on acupuncture. They have different viewpoints on the same thing, and they have to merge its conflicts, but it ends up being a better article on acupuncture which is by the way my first contribution to Wikipedia, so obviously...

Something is just dancing. I wish I had more sleep to be able to be more zippy in my responses, but I’m getting the...

No. You shared some feeling also when I joined the g0v community before. You come out with some more questions.

I have a lot of questions, but there’s some questions that are popping in my mind, and then fade into the distance. I have ones written, too, to ask, but go ahead. You can stay on a little bit longer, right Shuyang?

Yes.

Yes. I can stay on until noon before my lunch meeting, but yes.

We’re going to...

She has an hour more than I have.

Part of me wants to make sure I’m getting whatever I can from Audrey before he goes.

Of course.

I think if you can give me a sec to look. I got what that is. It’s funny how well you’ve covered all this. I would like to poke all that. It was a breakthrough for me to see the parallels to Wikipedia which again, I haven’t engaged in detail, but I have dived into the background community a number of times, and seen the complexity, and sophistication of how it self organizes. I see that as a parallel.

Then there is the talking to Taiwan which I gather is just one of the many players that can convene a conversation.

Yes. They are the film people here.

They decide who they want to interview, and when and all that. It’s not as if they are the key players. They’re a group. They’re just in the group. They’re not an official channel organized like a business or something. They’re just a group of people who have certain competencies and material, and they just do it.

At the end of the day, they’re just a loosely couple of team with a bunch of Slack channels, and a crowdfunding website that has some amount of money, and that’s it.

They have managed to garner a particular attention, but that’s just because they garnered a particular attention. They don’t have any consensus or official role.

I’d say that’s because they have excellent designers. Their website looks really, really pretty. Their content is released under a creative common license which enables the connection to pol.is and to the vTaiwan community.

You say the outcome of the conversation can then be used to make a draft bill. Is there a group within vTaiwan that does a let’s translate all this material into a draft bill?

The pol.is.

Is it in our group or is that something that’s done by legislature officers?

My answer is the same as the last time. It depends on whether it’s a regulation or it’s a law, and the process differs slightly. If it is a regulation, usually the ministry will have the final say of how it looks like.

The government’s point of contact here will be actually the primary people translating the consensus points into regulation, and going back vTaiwan in the draft stage for another round of direct feedback including the name of the regulation.

Their platform economy regulation was used to be called the sharing economy regulation, but during one of the consultation meetings, everybody hated that name, so it’s now platform economy regulation.

On vTaiwan, the URL is still sharing economy because we can’t change that very easily. That dynamic is cross feeding dynamic. Happens well into the draft stage for regulations is what I’m saying, if it’s a law, ultimately, we depend on the parliament.

We more consciously include MPs in the process in the hope that the MPs will know the context of the drafting stage. We will not always write the legal text as suggestions for the MPs. We may translate it into a little bit more clarity, a little bit more coherency, but at the end of the day, the actual text is between the administration, and the parliamentarians.

You’re basically playing a role of briefing the parliamentarians on here as what a lot of people think, and here’s a lot of the issues that are involved, here’s what they think should be done over to you. It doesn’t have the same feedback loop that POs have.

Yeah. Because once it’s in the parliament, of course they may hold the public hearings, and so on, but the community, or the committed ministry officials can just be summoned there. We’re not agenda setters in the parliament.

That’s one of the interesting places because one of my...There’s ways in which one of the images, narratives, metaphors I use is of you guys creating a demonstration parallel government. Not one that necessarily is implementing things, but the decision making dimensions of government, you say, "We can do this."

That’s why it’s a part of the story of the Sunflower Movement was that you guys demonstrated you could do it better than the...

It’s a demo.

That that’s an ongoing thing, and there’s a way in which in some contexts where the government was not particularly receptive, but there was a lot of grassroots energy to try something like this, it could be sold to grassroots supporters as we’re going to show that we can do this even better than the government joining us to pull it off.

The government may or might not listen to it at all, but there may be some ambitious politicians who recognize a wave to ride when they see it, and say, "Hey, we’re going to pledge to do what the people say," and that gradually infects the government that way.

This happens in 2016. The people’s judiciary reform movement, and we really want Billy here, but I can’t do a pale imitation of him. We have Billy Lin who wears many hats. He is one of the PDIS colleagues, but also a vTaiwan regular. Also a leader of the people’s judiciary reform movement 2016. Also adviser to the official presidential judicial reform forum a year afterwards. Also a founder of the Social Democracy Party, one of the cofounders... he wears many hats.

In any case, the PJR started by the Judicial Reform Association, and is a grassroots totally bottom up.

No government binding power whatsoever. A way to apply the vTaiwan process to the IBS staff judicial reform to increase trust between the people, and the judicial system by deliberating things like citizen’s jury, and -- we don’t have a jury system by the way -- many other possible innovations.

It is applying this, but broadly speaking, it’s not listed on the vTaiwan platform. They use the vTaiwan discussion forum for one aspect of it, but it has its own website, and its own branding, and everything.

It really resulted in a demonstration to the presidential office that this process is feasible, and so then the same process, adjusted a little bit has, then became justice.president.gov.tw which is the president’s take the same process which has completed and the results are being tracked.

This is another example of piloting, prototyping something, because it worked successfully and there’s a mixture of public support, plus pressure, plus the willingness of the politicians to go along with it, suddenly a version of it has taken up and adopted by the government.

That’s exactly what I meant.

In itself is an example of vTaiwan was modeling something. They’ve picked up that to adapt to their scene, and then their scene was picked up and adapted by the government. There’s a whole evolutionary pollination kind of activity.

Yes, crosspollination and all that. The folks who did the electronic platform for the presidential version of this is called Watch Out. The Watch Out group is again very fluid organization. The three Watch Out engineers and designers was part of the original vTaiwan team in its original version who did the initial version of website and everything.

Now, Watch Out has...for example, you’d see how as a designer, who again from time to time come to vTaiwan meetings and get drunk. Officially, they are also the vendor to take on cases like the presidential judicial reform process, and the CKS memorial hall reform process and other processes.

They’re a vendor, a social enterprise-ish organization. That itself is institution. We have very close but not formal relationship with lots of Watch Out folks.

I had this funny impression that within the last couple of days, somebody was communicating to me from Watch Out, and I didn’t recognize it and threw in the trash. I’ll have to go and dig around my trash, [laughs] see if I can find out what it was to recognize that.

Shuyang talked about challenges at the end of your presentation, Shuyang. I don’t know if that’s something you want to address with Audrey. Curious what’s the content of those four items that you put out.

I’ll bring those on my slide. I think we talk about challenge some current state as in how we could possibly improve vTaiwan from now on. Those are very flexible recurring system already. It actually depends on what are different types of people and what are different skills we have in the community.

There are some bullet points probably to be pulled out that can describe the challenge of vTaiwan right now. We like to attract people who can help us on that. We’ll be up to attract more people in a sense of people with different dynamics in their skill sets to participate in vTaiwan’s process.

In terms of there are different cases, different issues, launched on vTaiwan website, but sometimes we need to have a different mindset to help us out on reformatting the slides or have a different look on open dictionary. It would be nice to have people from different backgrounds to participate as well, how to increase the participation on the process and also on the comments of every issue on vTaiwan’s case.

There’s different branches of participation I’m sensing, that some of it are the people who are part of the geek-skill community, or the people who are actually creating the platform for people to engage in these different ways on. You would like to have diverse skill sets in that.

Then there’s a question of increasing the people who use those platforms as stakeholders or ordinary public citizens who come. You want to have more numbers and more diversity of those.

All of them. We want to attract all of them. When we talk about stakeholders, we actually are thinking not really so much about the quantity of the stakeholders, but actually the dynamic of the stakeholders. If there’s a issue, we can interface more stakeholders from, let’s say, different groups, different backgrounds.

Then it’s very good enough instead of having 4,000 people from the same background and talk about the same thing.

The thing is that a snowball works better if it starts from the different sides in the opinion groups, so that the snowball worked well like this. If we started from these corners, then we don’t actually reach that much people from the other sides.

Right, but there’s many different ways to cut the pie as they say. There’s functional roles that people have in the dynamics of the system. When the people say get the whole system in the room, they’re usually looking at the different functional parts of how things are unfolding.

Then, there’s perspectives, opinions. People are for or against this or that. That’s a different kind of diversity. Then, there’s demographic diversity. There’s all these different kinds of diversity you can deal with. In my general theory, as I think you already know, Audrey, is that the sense of diversity...

Whatever kind of diversity you use, if you can have some way of having it interact creatively, you’ll end up with something better than you started...

Exactly.

I’m curious, also, because my roots are more in the "One person, one vote" kind of democracy, the sense that we’re all equal citizens, which is a very...

It doesn’t use diversity in a sophisticated way. It’s just saying, "Do you have the right to participate? Are you a member of the community?" Then, you’re stuck into a process which uses you as one data point in the whole thing.

I’m curious to what extent, I know you have a sense of stakeholders, and I’m fascinated by your division of the stake and the whole stake as the "Here are the issues involved, the interests involved." Then, there’s the holders of the stake, the people who are identified with one part of that or not, which I think is a really interesting way I’ve never cut it apart before.

Stakeholders are usually, in the circles that I go in, people who have an explicit interest. They will be impacted in a way that they’re very aware of in the outcome of whatever the issue is, and that either they or representatives of them, like representatives of the union are there as that piece of the system.

The fact is everybody is a stakeholder in everything, but they’re not self-aware, self-defined, organized in that way.

They’re not networked enough, yes.

There’s something, when working with the multi-sector, multi-stakeholder emerging governance group that I’m part of, I learned that this ordinary citizen perspective is place-based. You are a citizen of a place. That’s the way the society is organized, with the mayor or the president or whoever in the legislature at a particular level of place.

I’m sensing all the time that stakeholders is who you’re looking to engage, both in your information generation and in your evaluation of that information. The public in general doesn’t show up in a visible, explicit way. I don’t know if that’s part of your whole approach or if you’re trying to expand that.

Your idea would be everybody in Taiwan is involved in pol.ises on every issues that’s raised, whatever. I don’t know if that would be an idea at a theory level or that’s not even an idea, you just don’t operate that way. The public as an abstraction, you’re not interested in. You’re interested in the diversity of stakeholder at a...

The thing is that if people don’t know, they’re not well-informed, as of their relation, their stake in this topic. Of course, they may be interested in learning about it. We do engage. All our materials can be thought as a introductory material on civics around public issues.

At the end of the day, when we ask all the ministry to publish all its yearly plans, monthly plans, their KPIs, their procurements, their whatever, on a common platform, what we are looking at is not people who do a meaningful vote or referendum on it.

We’re asking people who are interested in it to ask clarifying questions if they have one, to have real-time dialog with public servants, to engage in organization themselves in order to discover their stake and so on. What I’m getting at is that there is no abstract public if that’s what you’re asking about.

There are individual citizens who may not be aware that they are related to this issue in some way. There is many people in the vTaiwan community, and especially the Watch Out group, Talk to Taiwan, and the other more media-oriented people who are trying to connect people’s attention to their stake.

At the end of the day, we’re not the referendum platform, but the community is, to my knowledge, very willing to help to pol.is the concerns before any referendum so people can have an informed choice before the referendum happens. We have no ambition to become the referendum platform of...

There are other things that I’m pretty sure you’re already familiar with, which are based on the individual citizen body considered collectively. Which is what the word "public" refers to, and I’ll get to that in a sec, but I realize in this conversation, any given agency or whatever has a group of stakeholders around whatever kinds of decisions it’s making.

There is also the fact that the resources that most governments work with are taxes that are theoretically paid equally by all the citizens. The money that goes into government is governed by tax laws, and the money that comes out of government is then filtered out through different stake channels. [laughs]

It’s not just budget, it’s that there is some theoretical value in the will of the public as to the general direction of where we’re going as a city, as a society, whatever. The idea of the sortition thing applied at that level, it’s like you have your citizen jury, not in the judicial sense but in the Jefferson Center kind of sense, about some public issue, or the Wisdom Council (Jim Rough’s random selection).

Then we’re going to talk together as randomly selected members of the community. We aren’t in any of these categories. We may be, but we’re not being convened for that reason. We’re being convened to be a generic member of the community, and to look at what the community needs and come to some kind of agreement about that.

A lot of my work has been an effort to generate a legitimate wise voice of the collective public, what’s involved in doing that. I keep looking for connections. I have tremendous respect for all that you’re doing in terms of its potential wisdom-generating.

Covering the ground of what needs to be covered is just very, very powerful, much more powerful than a citizen jury could ever do. You know Frances Moore Lappé?

No, sorry. Frances...

Frances Moore Lappé, L-A-P-P-E, most famous for her first book, which is called "Diet for a Small Planet."

Oh yeah, yes.

She’s since went on to democracy as a major focus. First, it was development policy, and then it was democracy. She created a whole theory of democracy, and in it she talks about the relationship between citizens and experts. Citizens are experts on the values of the community, the whole community.

It’s like, what do we value? Where do we want to go? What is important to us? From a cognitive science perspective, you can’t make a decision without wanting something.

You can use all the rationality and science you want, you can clarify everything about all these issues and options, but you can’t decide unless you actually want something, which is not a rational thing. It’s a simple, that’s what you want.

Frankie’s version of democracy it’s like, the citizens are basically there to articulate and act on the values of the community and the everyday experience of ordinary people. That’s what they’re expert in. Then, the experts are supposed to ideally...The experts are on tap, not on top.

The experts are supposed to help the citizens figure out how to actually achieve where their values are pointing, how to go in that direction given the complexity of the real world. If you’re ignorant you can push in a particular direction you think is going to work for you and make all kinds of messes, and you actually get the opposite that you want.

The experts are there help the citizens understand what’s really out there? How do we go about actually getting where we want to go? That’s the job division. You’re mostly centered at the people who know, there is a heavy gravity towards that. There is people who know in diverse ways, and you’re trying to engage and all that, but the generic citizen is not in the pool.

I think I wrote that in the first paper, a part of me could imagine a scene where there was an annual Wisdom Council for Taiwan which said, this is where our attention is, these are the things we’re longing for, and then that somehow plays into what happens in your existing situation. That’s an influence that shapes how people are thinking about all this.

Yeah. I mean people who come to the office hours, either the Taipei ones or the regional ones that happens every other Tuesday, they’re ordinary citizens.

[laughs] The sense of having a...

A motivation to show up certainly.

No. I’m thinking of having a...There is a different identity. From what I’ve heard so far, the people who show up have a specific interest in something.

Not really.

They’re more stakeholders holding a function of some kind, rather than the man off the street.

That’s why we have this town hall style dialogues in the collaborative meeting process. You see, the experts are learning from each other’s type of fields in just a small room.

On the larger room or across the Internet, on the live stream, there’s thousands of people watching. It cost them nothing, really, to pull out their phone and start watching, or it cost them just a little bit to walk to a local town hall and watch the live streaming with me as the anchor, explaining the moves by the experts.

There is a second tier of involvement that depends on the aesthetics. That’s where the filming, their live streaming crew came in, as long as we can format it. It’s interesting. Lots of people will watch it like any other TV show. This is a interactive TV show. The opinions they send, the messages they send on Pol.is or in chat rooms are then filtered into the live-streamed expert meetings.

It’s different in kind from something to take as an...It’s not an ideal necessarily but as a perfect example of, "We, the people, voice is randomly selecting people," and then having them come to a consensus of some kind is a different approach and then having people watch...

I should mention the other CKS Memorial Hall deliberation is a more sortition kind of citizens’ council. They use random selection and ensure a balance in gender, in age group, in ethnicity, in whatever that is proportional.

What is that?

The CKS scenario workshops are about this building in Central Taipei called the Chiang Kai-shek Memorial Hall. It is a place with very high controversy. There are many people who think of Chiang Kai-shek in a very positive way and many people who think of Chiang Kai-shek in a very negative way and, sometimes, very mixed.

What to do with the memorial hall becomes a controversy. The ministry of culture are used to this kind of citizen council, but they don’t have the power to randomly pull anyone in Taiwan.

It ends up being an open-application process through face-to-face encounters, or over email, or over Google Forum, whatever. They do get the sufficient population so they can have a fair poll and representation, statistically speaking.

They run many, many scenario-planning workshops with plenty of informed time and plenty of small-group time and so on, all according to the textbook about the CKS thing.

The fact that most of the facilitators overlap with the vTaiwan community at some point is not a coincidence. Because the process, as you said, is completely different philosophically, that case is not listed on vTaiwan, nor we will say it’s a vTaiwan project.

Right. It’s one of the things I’m glad that got clear on. I have an ongoing inquiry into the relationship between the stakeholder way of cutting the pie and the citizen public way of cutting the pie. I’m getting, between the engagement with you and the engagement with the emerging network government from this, rapidly learning how to think about this, but I’m still intrigued.

It feels like there’s a potential synergy. There’s a funny way in which because the parliament is elected, there’s a way they can...like you say, you don’t have the influence on parliament that you have on ministries. It feels like the connection to the we-the-people public would give more leverage over parliament too because that’s the elected parliament.

Right, I do agree. I am going to the Parliament to talk about one possible merger between the two models. It’s part of the Digital Communication Act. I have to go. Shuyang will carry on the conversation.

[laughs] OK.

Send me the recording, OK?

That’s very fascinating. Hopefully, Shuyang tell me what he’s going to say, anyway, bye-bye. [laughs]

Bye.

OK. Yeah, I think it’s all right.

Wow.

I’ll be happy to stay for another 30 to 40 minutes...

OK.

Yeah, for me, it’s a lot of learning also from Audrey.

Yeah. What a position you got.

I think we’re privileged. [laughs] It’s a good opportunity to work with many amazing people.

Audrey is going to be a witness to a parliamentary hearing of some kind?

Yeah, she’s going to the parliament meeting. In a parliament period, she attend the meeting every week. Many will be just sitting there and using her computer, and checking what’s going on, and trying to participate in a very limited way. Actually, she couldn’t be answering questions from the MPs.

She’ll be there to understand the situation over there. I think we’re still discussing about...I don’t know if you have more questions you want to...

I do. Since we’re at this point, I’m curious what...She was talking about integrating some kind of integration between what you’re doing and parliament. I’m wondering what is that piece of the puzzle. Can you describe a bit of that?

I’m not sure if I can describe the whole piece. [laughs] I think a very limited observation.

A little taste....

I think the storyline will be from we are doing this experiment actually called vTaiwan, right? We are trying to experiment an open consultation model from this public/private space. It’s because vTaiwan is public funded but private operate by the community.

This also connects you to the last slides and when we talk about if we should institutionalize vTaiwan in a way if the participation from the government side or from the citizen side should be regulated. Actually, we’re talking about if there should be a regulation to talk about the government should provide the platform that is open for the citizens to be involved in policy-making process.

If that is the case, we are kind of regulating the government’s involvement to chat conversation with citizens on the policy-making process. You can see that. If you are familiar with the fork and merge idea from g0v’s community to the government organization right now.

In the g0v community, we did lots of forking projects to the community. Then when they create some kind of better-experienced websites or projects, the government sometimes merge it back. I think vTaiwan experiments also from g0v to try to prototype a open consultation process.

If the government, in the end, is interested in merging these projects back and writing a regulation around it, then the government is trying to institutionalize vTaiwan. There’s actually a regulation called National Communication Regulation.

The digital communication regulation is writing that government should provide a platform inspired by vTaiwan that provides platform for citizen participation. I think Audrey was talking about that.

That’s still in the parliament. We were not sure if it’s going to pass or not. We’re really looking forward to have that, because that means we are having this vTaiwan experiment in a public/private side but also having a merging-back version in the government as well.

That’s a funny...I can feel trade-offs. You know the word trade-offs?

Yeah.

The sense of if the government is going to make a policy regulation law about something, it has a certain solidity to it. One of the most powerful things about vTaiwan is its liquidity. It’s like it flows and changes all the time.

Yeah, exactly.

You can’t...

You can have guidelines. [laughs]

You can have guidelines, but you can’t have, "Here’s A, B, C." The government is going to want to do an A, B, C, and you’re not an A, B, C. You’re the river....

If this law, if this act is going to take few months to pass, then we can expect the government will run vTaiwan as fluid as vTaiwan’s community. I do think it’s important to have both forks, both the government regulation and the vTaiwan experiment living at the same time.

If you can somehow have, say, an annual review of whatever policy the government makes about this, have it include review of the policy every year by a public participation process or by vTaiwan process so that it has a chance to evolve.

Even though it’s going to be evolving in a more jerky step-by-step way rather than in the flowy way that you guys have, having something in the law that says, "Aw, that law is going to be reviewed and changed," would be valuable.

Yeah, every year, every month, every week.

Yes, I know.

vTaiwan is evolving every week. [laughs]

That’s a whole other thing if you, guys, just review it anyway.

I just had this conversation with Eric Gordon from Emerson College about this vision of having a platform provided by the government, no matter what form it will be in the future, and to have this vTaiwan open consultation process.

He was arguing if we should let this recursive public running this recursive process. Should it be a month, a day, or a week, or a year? [laughs]

Part of what I got...Yeah, if you make it official, then it has to have some periodicity, but you, guys, don’t have anything like that. There’s certain rhythms, but they’re overlapping rhythms. They have different...Some of the rhythms of your work go like this. Some of the rhythms of your work go in with more slower kinds of things. It’s all in the math.

Of course, yeah. The different policies should go in different rhythms. It also makes sense to have this community.

You mentioned abstract public, and there’s also another term called mini-public. I’m very curious on what’s the difference you think around these two terms.

Mini-public and...?

Abstract public.

The abstract public is a pure abstraction. It’s like "the public." It’s like public opinion polls, "This is what the public thinks."

A mini-public is as specific form of public engagement, which is on usually randomly selected, sometimes scientifically, demographically balanced group of people who deliberate in some form, on a specific public issue. That’s a mini-public. Within the world of deliberative democracy, that phrase, mini-public, means that kind of forum.

Mix of diverse people who could represent the public?

Yeah. It’s called a cross-section, a microcosm of the macrocosm, a small version of the larger public. Like you guys know, you can’t facilitate millions of people, but you could facilitate a small group of dozens or hundreds of people and have them come up with something.

You would assume that that was something like what you’d get if you did the same process with all those people, with everybody in the society.

I think pol.is did help us to facilitate more people than one person could. How would you describe vTaiwan’s community in this public term?

I don’t see the organizing principle or...I don’t know what the right word for it is. The way vTaiwan cuts up the pie of the population in its effort to have consultation is by the standard of stakeholder. Stakeholder has to do with one’s relationship to the issue.

Yeah, it’s always different stakeholders.

In the public, it’s more your identity with the community of place. The general public is everybody in a country, and then usually you can have the citizens of the town or the electorate. I guess you could say electorate, because the elections are organized by levels of place-based organization. Here in the United States, you have a city or town, and you have the county.

We actually didn’t limit people from other countries to participate in vTaiwan. Actually, there were so many international people we count as the community, as well.

There’s a way in which that makes sense if your focus is on diversity, because having the more perspectives you can integrate, the more you cover the ground that needs to be covered.

If you were to have all the people who were not Taiwanese be the people who were defining what happened in Taiwan, that wouldn’t make sense, but including people who aren’t Taiwanese into the discussions of Taiwan would make sense, because they bring new perspectives.

That’s true. We always have Taiwanese people.

[laughs] I know.

...more biased. [laughs]

I tend to be biased towards diversity, but up until the last year, I haven’t looked carefully at what all the stakeholder collaborations and stuff that are going around have to do with the "we the people" voice. In the US, we have this "we the people thing."

...tradition.

It’s in the constitution, "We the people of the United States." That’s a meme that is part of our culture. I have been thinking "we the people" is behaving stupidly, at best, if not insanely right now. How could you help "we the people" collectively be wise? That’s been my inquiry.

Now, I’m extending that. One of the definitions of democracy is that the people who are affected by a decision participate in making it. That’s much more of a stakeholder perspective, so I’m now really interested in what’s happening in the world of stakeholder stuff.

What you’re doing is definitely that, but I still have my roots in this other frame of reference. Because diversity is a resource for better decisions, I’m interested in how those two frames of reference, which are very different, fit together.

It is very different. I’m also very new to this process of implementing or making democracy in Taiwan. For me, it’s very similar to what you’re doing, but I’m trying to understand our definition to what democracy’s actually about.

What kind of democracy? Is it direct, is it digital, or is it fluid democracy we’re trying to achieve in the end? Different powers from different kinds of technology we could use.

Most of the kinds that you’ve named have, at their center, voting. If you’re a liquid democracy, you get to delegate your vote to somebody else. If you’re a direct democracy, you get to vote on every policy that’s being made. For me, compared to deliberation, voting has a very low...

Participation.

Yeah, it’s a low form of participation. It’s a low form of collective intelligence. Part of what’s so amazing about pol.is is it is generating wisdom out of voting.

There’s no interaction, people talking together, in pol.is itself. You have people talking together before and after pol.is, but pol.is itself, you don’t have any back and forth nuance, "Let me take into consideration what you’re thinking." You just have, "Do I agree or disagree with this?" Out of that, you generate a consensus. I look at that, I go, "What the fuck’s going on there." [laughs]

It’s interesting. The voting was just to connect different comments. It’s actually possible to divide not only yes and no or "I don’t know," three options, to more options. There could be a scale of 7 or 10 options in the end.

One of my questions was what would happen if you had a -5 to +5 scale that you voted on? What are the things that would be good about that and what are the things that would be problematic about that? I don’t know how to think at that level.

I’ve got that you’re making clusters of groups of shared belief, and then you’re looking for, among these diverse communities of shared belief, what do they all agree on. That’s a brilliant shift. I love that. I don’t know it’d maybe become too complex to have shades of agreement. How does the algorithm deal with that? I don’t know.

I think the fidelity of this consensus will be different if we have more scales on the vote. When I look at pol.is’s algorithm, it’s exactly the same logic behind Netflix, when people are watching videos, movies on the Internet. If I watch the same set of videos as you do, then we probably will be categorized in the same group on pol.is.

If you can think in that analogy, for me, the measurement of I watch the video, it could be just on and off, I watch or a don’t watch, but it could also be I watch for 20 minutes.

...video watching and so on. The algorithm was more delicate on shaping this consensus on what video to present to you or what kind of consensus you are going to be attached to, in a sense. I do think it makes sense, but also it depends on if it’s really necessary for us to have that fidelity of consensus on every issue. I will think on that.

One of the things I’ve concluded, without having enough evidence to really conclude it, is that there’s two kinds of consensus that pol.is facilitates. If I knew Chinese, I could evaluate this better, because I could look at some of the actual results.

One is a very shallow consensus that avoids the kinds of things that cause people to be conflicted, "Everybody should have a right to breathe." Of course everybody’s going to say that. Who’s going to say you don’t have a right to breathe? That’s a shallow consensus.

There’s the person who has a brilliant idea that nobody else thought of, that happens to cover more ground than anything anybody on any of the sides thought of and is simple enough to recognize that fact. That could surface in a pol.is exercise, and that’s a more wise, deeper kind of consensus.

What I’m used to, in terms of consensus, is largely based on concerns. "Here’s my proposal," and people go, "Oh, yeah. Let’s do that." Then the facilitator goes, "Does anybody have any concerns about that?" It’s related to the rough consensus idea, which has a whole categorizing of what kind of concerns are legitimate on the block consensus and which aren’t etc.

The whole idea of concerns, when you have a concern, it means there’s something you’re not taking into account, or some of the group’s not taking something into an account if a member of the group has a concern. Do you take the time to try to take that into account?

Part of my theory of shared wisdom is the better you do that job of trying to take into account differences and disturbances, the more ground you’re going to cover and the wiser your decision’s going to be. You can’t necessarily, in all circumstances, take that kind of time. Ideas like rough consensus help us navigate that challenge.

When I look at pol.is, my instant thing with pol.is is here’s the agreements, here’s the disagreements. You end up with something that is 90 percent of the people on both sides of the two major disagreement camps like this item, but there’s still some people who have concerns about it and don’t like it.

If you could talk to those people and say, "What are your concerns?" and have teams of people who work specifically trying to figure out how to address those concerns, you could increase the percentage of agreement around that item and make it more wise. That’s part of how I would love to use pol.is.

That sounds really cool. I’m mostly in contact with the development team of pol.is right now, and I think they’re making a new visualization, which I think has that feature. You lay out all the comments on the pol.is conversation along a scale that is the consensus level.

On one end, it’s comment that has super consensus, really good consensus. On the other end, will be the very controversial comments. You can, of course, go through all the comments that is part of the consensus. If you have time, you can also go through more comments that has controversial points.

Did you see that 10-page thing that I sent to Audrey before the first conversation?

Yeah.

There’s a "Using Concerns" section on the eighth page that I talk about the different things that could be done with pol.is and about an online game. Talk about designing for fun. It’s a mixture of competition and cooperation. You take an issue and you get to pick the issue you want to work on, and you are part of a team.

The algorithm of the game helps you decide how diverse a team do you want to be a member of. Do you want to be a member of a team that basically believes like you do? Or, would you like to be a member of a team that’s really different? The more diverse the team is that you’re on, the more points you get for succeeding in coming to agreement.

There’s a way I designed it. I have a feeling there’s a pol.is version of this that would be even simpler than what I’m thinking of. In my design, you have 10 people on a team, and they’re all different. They each write a little statement of how they would solve whatever the problem is, whatever the issue is.

They share their concerns about each other’s statements, and then their job is to collectively try and handle those concerns and come up with an agreement that covers all their concerns. Of course, the more different they are the more difficult that is.

The more different you are and the greater level of agreement you come to, both of those give you points. You’ll get 10,000 points if your team is the most diverse that the algorithm can possibly design and you come up with 100 percent agreement.

Then your solution is posted in the larger field of participants in the gaming community. Other people can give you points for how good your solution is and all that.

There’s a billionaire in the US who’s created a thing called Brigade, which is a way for people to talk about issues and organize to advocate them. I was thinking of this as a game that is attached to the Brigade citizenship software. Ideas which come out as winning ideas in the game would move over into Brigade to organize for actually advocating them and getting them used.

How do you spell Brigade?

B-R-I-G-A-D-E. It’s a military term for a level of military organization. In the political world, it’s like you’re organizing your brigade to fight in the political world for your favorite idea.

I was trying to say let’s try and privilege ideas that are actually wiser than other ideas. We can make a game. It’s a serious game, but it’s still a game. You are in competition. It’s more like Olympics. I haven’t thought of framing it as a deliberative Olympics, but it could be.

We were coming up with this new project. We’re still looking for people and place to work on that. It’s called Holo-pol.is. It’s using pol.is’s library. We’re now trying to run a pol.is server locally and pull some data we have.

For example, we have the data of Uber’s unmanned vehicle conversation, and so on. We can use those data and APIs from pol.is to create a new experience. Now we’re thinking to make it in a VR environment, but it could be just a fun game in any platform.

I can’t even envision what that would be. One of the other questions was what would it be like to have a comment section attached to every item that shows up in pol.is, like Wikipedia? Each Wikipedia page has a discussion section attached to it, automatically, even if nobody’s done anything with it.

What would it be like to have a discussion page, a discussion dimension of each item? When you put an item in, not only do people vote yes or no, but there’s also a whole discussion that can unfold on it. People might read the discussion before they vote their yes or no.

Right, that’s important. That’s why the project Audrey mentioned, sense.tw, it’s more like a study group online, altogether, for people to read different materials and highlight pros and cons online. We can share our perspectives when we read similar documents. That would give a rough understanding on different issues.

Imagine we launch an issue. We’ll ask this database with people’s readings and notes already over there. Say we key the issue, say unmanned vehicle, and anything around this topic, any kind of reading materials, notes, comments, thoughts, comments on social media, and so on will be ready there as a big package of background materials to start with.

"You want to work on this issue? Here’s a lot of stuff to work with." Is that where you...

The way we think of pros and cons comes out of the adversarial debate system. I want to highlight the fact that concerns is an example of something that is not part of debate logic. A concern is an invitation to figure out how to deal with a concern, whereas pro and con is, "Do you have more pros and fewer cons? Let’s do it if there’s more pros than cons."

It’s a trade-off thing rather than let’s figure it out and make it better. In the system where you’re planning to put pros and cons, that’s fine but I would say, "Have a place for concerns and things to address those concerns." [laughs] That’s my suggestion.

Of course, yeah.

That’s my suggestion to that.

Thank you, yeah. I will tell the sense.tw team. They do have pros and cons but what I think is very valuable is about collaboratively coming up with the shared notes around this issue.

You think that that’s part of what’s they’re planning on anyways, somewhere in there is having people working together figure out a way to...

Yeah, I think it’s important to figure out how people work together, especially collaboratives, and especially in online space, because we don’t always live in the same place. [laughs]

Yeah, one of the definitions I’ve had for collective intelligence, or standards I’ve had for collective intelligence, which doesn’t fit the pol.is experience as it’s currently organized is that what comes out of it is something that no one in the room came in with, and that is better than anything anybody came in the room with.

In a sense.

In any kind of conversation, interaction, discussion, deliberation, is what you come up with at the end as a group better than anything anybody came into the group with?

What happens in pol.is is the best things that somebody comes in the room with get picked for their consensus value. There’s nothing to redesign any of those things so that they’re even better than they were before.

Yeah. You’re very right. [laughs]

That can happen in a subsequent discussion. That’s part of what the saving grace of pol.is in the context of vTaiwan is that there’s all this discussion before and after the pol.is exercise happens. The pol.is just helps focus things.

We observed some users’ behavior. When they use pol.is they tend to rewrite comments in their own way. I think that would be one example to think of collective intelligence. Of course, it’s not really designed. They don’t have to rewrite other people’s comment.

Right, right, but they’re saying that, when somebody sees a comment, they can disagree with it and they tend to. There’s some people who will disagree with it and then submit a comment that handles their disagreement.

That exactly happens in Uber’s case. It’s very obviously...

[laughs]

You can see many comments like that. You can see two very controversial groups from taxi drivers or pro-Uber or against Uber. You’ll see some people who come up and try to merge this two-sides comment into one comment, which is more neutral in a way and try to get more votes, more likes. I don’t think they have to do that.

It’s very much in a deliberative conversation people are individually saying their piece in an effort to address what other people are saying, that is, in fact, a deliberative conversation.

It’s probably because in the interface they don’t have much to read about. They probably will go through their social media and so on. They could actually go through all the different comments and try to think of their comments. Maybe because all the materials they are looking at is from these two very controversial groups.

When they tend to want to write, they tend to write a comment that is more neutral and try to get agreement with both sides. I think that’s a very beautiful user scenario. When they were in the beginning, four different groups on the pol.is visualization and then merged to two and then merged to two overlapped with one major consensus.

That raises two other issues I had about pol.is, questions I had. What decides what the person who’s going to say agree or disagree sees? Did they have the list of comments and they pick one and they say, "I want to agree or disagree with this" or do they get randomly thrown up a comment?

The system says, "OK. Here’s a comment for you to vote on. Here’s a comment for you to vote on." You don’t see all the comments. You just see the one in front of you the system gives you.

Is the system random, or is it pushing up comments that haven’t been voted on so much? What decides which comment I’m going to see in order to vote on it?

That’s important. I think they changed it throughout the past few months. I can for sure check with the developers there. The statistics shows most people vote for 20 comments before stopping. Only people very interested will vote for more than 20. It’s very important to know what are their first 20 comments was shown to the users.

There’s some priorities that comes into turns. The pol.is team don’t push all the new comments to the first 20 to users. There are some priorities.

The comments which is, for example, created by people who voted more, the comments which is a new comment, for example, and maybe the comment that has a more imbalanced voting is either very controversial, maybe many people agree with that and many people disagree with that will be also shown up more.

There’s always open mind. I’m not sure. I think they thought about it. There’s some behind vision or behind ideal conversation they want to achieve. In order to do that, I do think intensively about what comments comes first.

They’ve been thinking about these different dimensions and trying to weave them into the algorithm?

Right.

I was thinking that the people who gave comments later would be disadvantaged. They’re trying to counter that with part of their algorithm.

That’s more inclusiveness. They’ll be trying to figure out the controversial comments.

I was wondering if a consensus comment shows up, if you track the consensus comment over time, does it drift in and out of consensus? I have no idea, but it just occurred to me that would be possible.

As the consensus gets stronger and stronger or more solid, it’s like 80 percent of both sides like it and stays there, 100 people vote, 1,000 people vote, it’s still the same. Does it go 80, 70, 60, 100, bla, bla, bla, it dances back and forth? I don’t know.

We can plan more on that in our work, I think.

What I know is they only look at the comments. They also try to figure out the persona behind those comments. Now they’re sliding in some questions when people vote. Not only their 20 comments they can vote, but also, there’ll be some questions asking what gender you are, what age you are around, what kind of background you are working with and so on.

Those questions they gather I think will be useful to input to their algorithm in the end.

There’s...You have to go soon.

A lot to play with. [laughs]

You have to go soon, I know, too, but I remember there’s another methodology called...

What is it? Synanym. It’s spelled S-Y-N-A-N-Y-M. It was years ago, I talked to the guy who started it before he put it into private use. Part of its logic was it had groups of 10 people who were doing proposals and then reading each other’s proposals, and then revising their proposals and reading each other’s proposals.

It was noting in that cycle who in the group tended to be chosen. That’s right. After everybody’d write a proposal, everybody would have to pick one of the proposals to revise. The software is watching whose proposal are most people choosing to revise. That’s one thing it’s watching.

They’re watching whose proposal does each person choose to revise. Out of those two dimensions, they are seeing whose mind seems to reflect the group’s mind.

The ideal person would be the person who always chooses the proposal everybody else chooses, and their proposal is the one everybody else chooses. That person has the most points in the group mind.

If they have 1,000 people working in on whatever the question is, then all in groups of 10s, they will pick those people who are the most reflective of the group mind for a next level of groups. One person from each group, and that’s the person who’s picked from each group.

I feel like there’s something like that the pol.is could do...

...who is voting most like other people and studying that person’s responses. Anyway, I think your time is practically up. Our time is practically up.

Yes. Thank you so much for your time. It seems really a good conversation.

Actually, it’s your...

I’ll try to read more. Maybe, soon I would try to arrange another call with you to ask you more questions.

OK, we’re on...

We can answer in email or we can even arrange another call.

Are you going to be in Seattle by any chance?

No. I am not coming this year. I am coming to New York in a couple months.

I am hoping to be in Seattle to see Audrey while I’m there. Next week, I’m hoping to have another call like this. Audrey says the week or two after that, we won’t be able to do it, but your Friday, my Thursday on next week, I’ll be here at the same time.

It’s fun having you involved in it. I would love to have you also if you can.

Yeah, I’ll try my best to be part of this interesting long conversation. [laughs]

I will watch this video again, get myself sorted out and have a new set of questions.

I remember you mentioned maybe some suggestions you want to send to pol.is team, and if that’s the case, I would probably do that, as well.

If you look at that page 7 or 8, whatever it was in the thing that I sent, 10-page thing, it has my thoughts there. As usual I would love to be in conversation with him, which I may when I go up there, at the end of April to talk with Audrey.

He’s going to be visiting with him, I may have a chance to talk with him, also, in that time. I’m more interested in seeing what comes out of the discussion that wasn’t there at the start.

We’re going to do some experiment on that.

...updated also.

Thank you.

Thank you so much.

Thank you.

I’ll dismiss myself.

Bye-bye.

Bye.