Hello.

Hello.

Hi, there.

Just a minute. [laughs]

You look very red.

Is it better?

Is it really blurry.

Much better. Hi.

Hi.

Hi. Thank you for taking a short time with us. We read really carefully the link you sent to us, so most of our question are actually answered in those links. I mean, lots of our questions, but we still have some, so we are happy to Skype with you.

We are at Superpublic, the place where you came last time in Paris. We have a great memory about those times.

Me too.

We’re about to move to a new place, a new building, a Civic Hall that the Municipality of Paris is giving to us. It’s quite exciting. We’ll see how it goes.

I was asked to be on the advisory committee actually for the new Civic Hall, but I cannot really travel to Paris, I had to say no. I know the plan. When are you moving?

We are moving actually now. [laughs]

Wow. That’s great.

We have to launch the new place in January probably, so it means we have to work from now to December. The place is only for a year, so it’s really short. We have to be quite quick to move. If not, it doesn’t worsen it. You know how it is. Sometimes things are temporary and then it goes...

It lasts a bit longer.

It lasts a bit longer. If it’s really for a year, the municipality should give us another space just after, a place which is actually better. Let’s see. Let’s see how it goes. [laughs]

We’ll see.

Sure.

This conversation will be on the record, and at your preference, we can publish either the text transcript as you have read, or the video, or both. I think we can decide after the conversation if it’s OK with you.

Can I ask you a question?

Yes.

What software do you use for the transcript? It’s brilliant.

It’s actually not software. It’s a Web service. It’s called CastingWords.

CastingWords?

CastingWords? OK, because...

Yeah, it’s only half-automated. It’s castingwords.com. Yeah?

OK.

You can see my whiteboard, right? It’s good?

Yes.

We prepared some questions that we didn’t read in your interviews. Maybe we missed something, so sorry if you answered this question.

No, it’s fine.

Our first question is about the software tools that you are using while you work -- civil servants. Do you use special tools? Did you find that difficult because some of the civil servants must not be used to only all kinds of tools, this kind of question?

On PDIS website, we have a page just for the tools. We refer to tools, both software tools and also process tools, and it’s quite different, the two of it. The software tools, we use mostly the open source suite called Sandstorm.io, and Sandstorm is this secure, self-hosted, completely open source tool that I also mentioned in "Nuit des idées."

It’s a combination of a Slack clone, a Cielo clone, a Google Dot clone, a Google Spreadsheet clone, like everything, [laughs] that one can use during team collaboration. It’s a productivity suite that everybody can just add to it.

Because it’s open source, the PDIS team -- P-D-I-S, that’s my office -- also wrote our own apps on top of this Sandstorm platform, such as the Hackfoldr, which is a collaborative bookmarking tool.

We also wrote, I think it’s called Biandang or something. It’s a way for people to order lunch together. The main advantage of that over the QuickSurvey form or whatever, is that it provides tallying, and also a place to upload the photo of the menu of the restaurant that’s being ordered, and things like that.

You see, it’s very practical tools that we are introducing as the software collaboration. It’s meant to be what we call in the flow of work, meaning people don’t use these tools to fulfill some other process. It’s designed so they fit into the existing process as a smarter alternative or collaborative alternative to existing tools.

Because of that, it requires very little training to use the collaborative lunchbox ordering tool. You just have to use Typeform or Google Forms before, because the UI is exactly the same. The same goes for collaborative writing, spreadsheets, chatrooms, and whatever.

Because of that, it doesn’t really require special training. We also don’t meet much resistance, because we designed so that they look like their desktop counterparts, or their online SaaS counterparts. That’s one part of it.

For the process, we also use process-specific tools. Here, we don’t necessarily use only open source. For example, we very heavily use GoodNotes, which is this app that I’m using now to communicate with you.

It is not open source, by any means, but we don’t have very good open source alternatives, and because we are not modifying and distributing it, anyway, we see smaller motivation to innovate on the software part of this.

We introduced GoodNotes as a way to promote facilitation, collaborative note taking, more focused discussion during meetings, and things like that. That’s one. During meetings and large town halls, we also use Slido -- which is, again, not open source -- a way for people to enter Q&A, enter questions during the talk.

Usually, when there is thousands of people, when people hoards the microphone forever, it really diminishes people’s ability to talk. Now, using Slido, we were able to see, for example, one question or one statement that has hundreds of likes.

If I answer one of them, because it’s sorted by order, I answer, essentially, a hundred people’s questions. It’s a much better use of people’s time when you do things this way.

You made a talk in Paris, where you presented a case about social housing in Taiwan?

Right. That’s what was used during the talk, yes. We also use RealtimeBoard, which is, again, not open source. This system is really key, because it puts the note taking, the Post-It notes, into somewhere that many people can collaborate online. It’s like a smart whiteboard.

We use it specifically to do service design, in the sense that we get people’s ideas in through, for example, e-petition. Then we have designed a form to make those e-petition sentiments into de facto, the filling and suggestion part, but then we draw a tree that makes the RealtimeBoard apparent, in the sense that it distills people’s arguments into this service design, diagrams that people can add on without losing focus. Then we print it so people have a conversation around it.

I can show you some examples, just a second. It’s all in Chinese, but I think it’s still possible to see the general form of it. OK, let’s see if this screen sharing works. Can you see my browser window?

No, we see the GoodNotes screen.

OK, like this?

Oh yeah.

Great. This is a typical meeting where we have a 5,000 people petition and then the participation offices or ministries agree to work on it. This particular one is called Lemon Cars, which is cars that was manufactured poorly but it was not discovered until a few weeks into driving and people would like to return it.

We don’t have that kind of law here. Currently it’s mostly individual contracts that covers only 30 days of warranty. It’s worded poorly and so on. First, we gather the stakeholders, including the Ministries of Economy and Transport, Consumer Protection, as well as the private sector, as well as inviting the proposer of the petition as well as the five counter signature people who responded first.

Then we use this RealtimeBoard to separate into the problem statements, the feelings, the suggestions, the factual information, responses so that people can very easily see the policy map, everything that’s related to this.

This is the final form. At the beginning the orange parts are not that many. Basically, we get people into this five-hour meeting every Friday to collaboratively fill in this policy map so we can settle on what’s the best question to ask at this moment and then work on ideation development together with all the stakeholders.

Every Monday we take it to the Prime Minister and other ministries with our portfolio. Usually, we mix them together so that even though there’s like two people from the same organization, they end up becoming different groups. Every group work on one aspect of the problem.

You can see that the RealtimeBoard is actually printed out in the handouts so that people can look at the paper and use real Post-It notes to add to the parts that they want to add. We, in Realtime, transcribe it back to the online counterpart.

There’s a lot of online/offline interaction this way. By the presentation time people just have this idea, development form, which is then, again, Realtime transcribing to online parts for the people who are watching the live stream or the Slido to have input.

This is a pretty sophisticated offline/online collaboration that we’re practicing every week. That’s how we use RealtimeBoard. We also use virtual reality a little bit, but it’s more experimental. The everyday tools, the process tools are just the three. I think that’s the answer to your question.

Yeah, for sure, it does. It might look silly because I know in Taiwan you are generally more numeric-friendly than in France, but we have often this problem in our program. We have a service that is in charge of security, safety for the network and Internet in the administration.

This service doesn’t allow us to use some tools. For example, for most of French Administration, using Google Map is not possible from their workplace. Do you have this kind of problem here?

Yeah, if it’s pertaining to national security, of course that’s another matter, but because by definition I don’t even touch national security it’s less of a problem, also because we host the core tools, the Sandstorm tools ourselves.

You can see that one of the first things that I did was to ask the same people as we mentioned, The Department of Cyber Security, to commission a security revue, a penetration testing of the Sandstorm platform.

Because the Sandstorm platform is also a security product, it’s not just a productivity product, it also makes it much harder to attack those individual programs. Once the Department of Cyber Security did the audit of the Sandstorm platform itself, it is less worried about a specific application that we run on the platform because the platform itself is hardened. It protects against the intrusion.

I think it’s essential that we have this security-hardened platform underneath the individual applications, which all may have security problems. They are mitigated by the underlying platform.

Also, another question that might or might not be relevant for your context. In France usually if, for example, an administration wants to create a new public service, probably the administration will make a really expensive contract, a public procurement with some developer.

Usually, the person writing the public procurement, they are the person in charge of the policy. They are not really technical. Usually, they are asking something like a little bit, I don’t know. They are asking for, I don’t know, a map, for example.

They know what they want, but they still have to explain everything in the public procurement. Usually, it’s a big contractor, like a big company, that will do the app. Usually, the app in the end will not really be exactly what it was supposed to be at the beginning.

Do you have these kind of big contractors the administration is working with them, or are you developing internally?

Yeah, we do have these kind of procurements, but one of the work we’re doing is that we’re separating the backend, which is the function, the self-storage cybersecurity, and the frontend, which is actually the way that it reaches people, like through a website, through a bot, through a app, or whatever as well as the experience itself, the design.

I think it is important that we can use tools. For example, we just adopted the OpenAPI 3 standard, what we call the OAS, as a national standard.

This means that during the procurement, all the units, all the organizations, even though they may not know the technical details, they can demand that for all the parts that interfaces with other systems, the back-end developers must offer not just human-readable functions, but for all those functions, it must also offer machine-readable APIs.

Their APIs need to be produced in a way that conforms to the OpenAPI standard, which means it’s not proprietary. It doesn’t use proprietary extensions. Also, we changed the procurement regulations so that any organizations can demand this at little or zero cost, meaning that the bidder cannot use this as a reason to charge a lot of money to add the OpenAPI support.

This means that it is what we call decoupled architecture. For example, one very concrete example is the tax-filing software.

Taiwan’s tax-filing software was this kind of monolithic procurement as you said. It was required to be cross-platform so they used Java applet for Mac and Linux, but Java applet is actually deprecated by Oracle.

It makes a very bad user experience because you have to skip a lot of security warnings and also downgrade Java and whatever just to get the tax filing system to run while the Windows people can just file it in 10 minutes. It is obviously not digital equality.

Once one of the participation offices brought it to our attention, what we did is that we restructured the procurement so that the original contract only covered the back-end parts. For the HTML5 actual cross-platform part, that is the front-end. It also runs on a cloud like automated Elastic Cloud.

This part is contractually separate from the back end so that they can be done by different vendors or even many different vendors because maybe one wants to make one for specific needs. What this means that it opens the gate for many different front ends without worrying about separate security issues.

Then we, as the petitioners on the e-petition system, who are themselves user-experienced designers and ordinary people who just want a better tax filing experience to run workshops to redo the flow of the experience. Again, this team is completely unrelated to the other two teams.

They have produced what we call a service design, a user journey, blueprints and other artifacts. Then we take those artifacts, and as the one of the front-end bidders saying whether you can just incorporate it into next year’s design. This part doesn’t really touched the back end at all.

What I’m saying is that maybe it’s not possible to completely make the organization aware of all the different disciplines involved, but using OpenAPI and other service-defined methods, we were able to decouple them into smaller contracts.

Each one has a different development methodology that can involve the user from the very beginning while the back end one doesn’t really need to change or to rewriting so much.

That’s really interesting because in France, we have this kind of specialty of, for example, for national health insurance. They wanted to make a new platform, mostly internally, but it take ages and millions of euros. In the end, it doesn’t even manage to make it because of back-end problems. It’s been interesting what you just said.

For example, what was exactly your role when you suggest to other administration to use this process of, what’d you say, decoupled?

Right, decoupled.

Decoupled. Is it like a note that you will give to other administration? Is it a training? How do you communicate with them?

That’s a great question. Mostly, we provide suggestions. We don’t issue orders. Suggestions take the form of templates. The idea is that the public doesn’t really need to know anything, but if they know that for example, they want to involve early stage user participation, then we provide them with examples that correspond to their situation.

For example, this is the Agile RfP template. Then we have provided links that corresponds it to the part that actually have used Agile RfP, so they can compare whether this suits their need or not.

We also make very public the cases that we actually did as PDIS for the general public so that once the general public knows about it, it’s possible for the public servant as part of the general public, to go back to their units saying, "Can we do something like that?"

For example, we integrated the two national disaster response agencies. One has a map view, and one has a lot of data into this shared map-based view. This is entirely transparent. The order meeting logs and everything is online so people can know exactly how we made decisions and when we made it.

Another case is, again with a decoupled architecture, a data-driven architecture, is the national dashboard on the vegetable and fruit prices because people will speculate on those prices after a typhoon or something. Now we try to make it entirely transparent whether to import, export. Reserve levels and things like that are all published.

Again, this is done in the open, meaning that the data providers from the agricultural agency, from the local markets, from the weather bureau, and so on, they become aware of the possibility of doing this kind of things. They can write their own proposals saying, "OK, we want do to things this way because this lowers everybody’s risk overall and share the highlights."

That’s the basic idea. It’s not so much as telling all the ministry that you have to do this but to make two or three really prominent cases visible and make very transparent what are the decisions involved, and the provide templates to every step. That’s the basic idea.

OK. On your everyday work, your team, with who are you working in each administration? Is it with the top direction? Is it with some people working on specific projects? Is it a mix of all of that? Why are you...?

Right. We are...

...that you had some in each administration, you have someone dedicated too?

Yes. That’s right. We have two different networks. The PDIS team itself is actually already very dynamic. I’m an anarchist, as everybody knows, I don’t give commands. Everybody who work with me are actually volunteers. Aside from me and the two assistants that came with me, everybody else are actually coming from some other agency and volunteer to work in PDIS.

For example, our core legal person is actually from the National Communication Commission. We also have staff from the Ministry of Culture, the Council of Agriculture, and the Ministry of Finance. These are the initial team.

Then we are joined by a lot of additional people from the National Development Council, one of the more respected, previously a director of Information Department, and many other people as well from the III -- the Institute for Information Industry.

It’s very interesting because around this time, I asked all the different ministries to form a participation office in their work. The PO Network, now we are going to have a regulation about it, is that in each of the 32 ministries, like a lot. I can’t draw that many.

They select, one, two, three or more people who directly report to their CIOs. Our CIOs are either secretary general or deputy ministers of that ministry.

The POs, they join a chat room. They joined a Sandstorm collaboration platform just like the other PDIS people. We add to it five people, six now, one from each of the institutes in the III, in the Information Industry Institute. They are mostly people who can actually write software, but on the other hand, also participate in the national policy making information industries.

Throw in some filmmakers and random occupiers, [laughs] and artists, and designers. That’s how the PDIS looks like.

It is about 20 people here, about 6 people here, about 50 people now here. Basically, we just run weekly lunches together.

The people in the PO Network, anyone can raise any issue as agenda for us to talk about. We meet with POs deputy ministers every quarter to review how the open government thing is doing. III people here serve as a kind of hit team to solve any information issues.

Usually, when there’s some issues that requires timely resolution, you don’t really have a contract. This is like an in-house coding team that can very quickly, within one day or two days, code up a system from scratch just to solve some issue that really needs to be solved. It’s the rapid response team for the participation officers.

The PDIS team mostly just work on whatever they want to work on. It’s entirely anarchistic within PDIS.

Usually, we also play the supporting role of if one of the POs say, "You know, I want to run a participation tax filing software, but we don’t have designers here," one of the designers from PDIS will just randomly join the Ministry of Finance on that team for a couple weeks just to get things bootstrapped. There’s a really good two-way relationship between those teams as well.

It’s, as you can see, very organic and loosely coupled. So far, it’s been working pretty well.

Cool. I don’t know if you have other questions, Lilas, but maybe what we thought would be nice if you have still a bit of time...

Sure. Of course.

...be to explain to you a little bit more why we are doing this interview today in which project and maybe have just some insights and maybe some advice for the participants of our project that we will just describe. Do you feel like describing the project or do you want me to do it?

You can do it. Maybe you can start.

What we are doing tomorrow...It’s not tomorrow. It’s the day after tomorrow, actually, and the day after the day after tomorrow. We will gather 15 persons coming from very different backgrounds, so there are civil servants from city level, county level, national level and also people from private organization and NGOs and a bunch of people working on numeric...

Also, designer.

Designers, sorry. Lila is a designer. [laughs] Designers, developers.

Our project is about trying to imagine collectively the future of administration so that’s a foresight project. It’s one day and half workshop altogether. The aim is to imagine some scenarios for the future of the administration, and then in the second part, we will work on the scenarios.

Then we will prototype some of the ideas, at least one of the ideas, that was developed during the workshop and try it in different administrations. We have both legs. We really blur foresight scenarios for administration in 5 to 10 years and also concrete prototype that we want to try in a month.

It’s like something we have done three times for now on different topics. It’s doing really well because we think that foresight on these a little bit...

Too blurry.

Too blurry but we need it to imagine thing that are really new or not happening for the moment or not to imagine new steps. The topic of the workshop we will do tomorrow is...How do you translate that actually? I don’t know.

Digital product frugal.

Do you say frugal? No, I’m checking for the transcription, I mean translation of the word. Actually [French] . Yeah, that’s frugal. It’s a...How do you say?

Digital.

Oh yeah, digital.

Digital administrative project with a frugal point of view.

Do you understand frugal?

Yeah, like shoestring budget?

Yeah, that’s kind of a play on the word that we...Because for us, it’s more when we say frugal, we think more environmental problems.

I see that too.

Some of the administration can obviously see also what you just mentioned, like less budget. It’s a way to play on those words because we don’t want to make an administration with less budget. That’s not what we think, but it’s often what we hear. I don’t know.

For us, it looks a little bit like Taiwanese already the administration of the future in a sense. I guess we could begin the workshop by just documentary about how is it working in Taiwan. It looks like crazy for French administration.

We’re not too far in the future. We’re just a few hours in the future. [laughs]

Anyway, I think frugal is a really good angle, because in Taiwan of course, we also say things like circular economy or sustainable design. I think it’s a very good angle to focus people’s energy on the digital, because too many times when people think about digital, they think about something additional to paper. They think about something that’s very show business-like.

I think in the end, digitalization of administration should not just be five different agencies and each replacing the paper or face to face workflow with online counterparts, because that means if the user want to access the service, she or he also have to visit five different websites. It’s just websites instead of offices.

I think digitalization has a promise to really simplify the working environment of public servant. I think that’s the core message of why I call myself a public servant of public servant, because instead of telling public servants what they need to do, I always ask, "How can I help to simplify your work?"

I think "simplifying work" is a really powerful strategy to introduce digital, because it means something that’s worth learning. Because if it actually increases their work, it’s not worth learning from their perspective. Even if it increases citizen satisfaction, there is nothing in it for the civil servants themselves.

To introduce digital design, they really have the potential to not just simplify the user experience but also eliminate redundant or unneeded work from the civil servants themselves or even automate them away.

Of course, by this year, people are asking, "Is the minister trying to use AI to make us jobless, to take civil servants out of the loop?" I always say, "No, what I’m trying to do is to improve your working condition." Basically, anything that can be replaced by a machine gets done by a machine.

That means that human beings are now listening and are now making value judgments. These things cannot really be replaced. I think simplifying work is a really good driving value if you’re having the frugal angle because nobody will argue for complicating civil service.

Yeah, that’s true. I don’t know if that rings something for you, but we thought maybe you could just...Since the conversation is recorded and will be published in a way or the other, we thought maybe if you have some advice for civil servants lost in French today administration.

What kind of mindset and posture are...Advice can you give to them so they don’t give up and try to imagine the administration of the future in frugal digital administration?

Sure. I think the work all of us are doing now is essentially rebuilding trust among people, among all the different sectors. Trust as we know is a relational concept, meaning that it doesn’t really exist only on one end or the other. It really takes a reciprocal relationship.

I think there is every reason for the citizen to mistrust the government. I think it’s mostly not because the government has done anything wrong, but because of the social media and a new generation of digital technologies that makes the civil society much closer to each other.

For example, the NGOs or the designer developers, they can get gatherings like the GitHub, the other Meetups, everything, that the new generation of technology really brings people really, really close together.

Once people are this close -- sometimes I think unhealthily so because the emotions and subconscious also overlap -- but in any case, they feel really close to each other.

Once this kind of feeling permeate, it suddenly feels like the distance between the civil servants and the citizens are very large, a large distance. We see that all over the world, but it actually has not changed at all. It’s just the people have a different way to organize with each other.

I have two message to say here.

The first is that the civil servants can also use the same kind of tools to build solidarity across ministries, in the administration itself, and basically cultivate a digitized solidarity, to demand the national government to offer working conditions that doesn’t treat them as machines or paper-pushers, but actually improve the workforce. That’s the first thing.

The second thing is that because trust is bidirectional, if we can let the NGOs and the individual contributors see that we are willing to trust people first, then eventually they will trust back, because that’s just how human nature works.

It doesn’t work the other way around. We cannot say, "We should do nothing and have the citizen trust us." That would be fascism. I think what we really need is to conditionlessly trust the public.

Whether it’s through radical transparency, or whether it’s through open data, or through open policymaking, or through this kind of imagined scenarios of foresight, all this is basically saying, "We trust you to not be mobs, to not disrupt the public administration, but actually bring something that we have not thought of to the table."

I think just this posture of trusting is already making the public service much closer to the people, because the NGO’s designer and developers know this kind of condition-less trust. We’ve been working that on the open internet culture for decades now.

Once civil servants adopt this posture, they will be seen as authentic allies in this project to make everybody trust each other more, not as faceless people serving in the civil service.

So first, to organize using the same tools internally; second, to trust people externally. That’s I think the two messages.

Nice. I don’t know if you have other question.

No.

Maybe I have just one last that it’s not really directly related to...

That’s fine.

I would like to know a more general context about how are things going, how things in Taiwan? How is your government going?

We’re doing well. The approval rating of our administration is at a historic high. We have a new prime minister, that’s the main reason, but he was selected and the hand-off was coordinated by the previous prime minister, so again, a very peaceful handoff.

I think one of the focus at the moment is to make this kind of deliberative or participative process even less expensive than before. During 2015, when we did the Uber case, it was a one-shot case. It’s really expensive, we put lots of hours into it. What we’re doing now is to regularize it so we can run it weekly, and there’s a standard operation process and everything.

This is what we call mainstreaming of participation, because previously they see it as something they have to do extra work for. Now, it’s just something that happens every Friday. It just takes them an afternoon’s time to prepare, because people have so much experience.

There’s more than 20 different collaboration strategies we’ve run now, every single one involved more than 5,000 people. The whole society is aware of it. Taiwan is 23 million people, I think 4 million people are now on the Join.gov.tw platform, the participant platform, which is a really good penetration rate.

Generally, people can trust the government to talk about things seriously if they bring people to the platform. That happens very predictably.

What mainstreaming does, I think, is it lowers the fear, uncertainty, and doubt. Because the uncertainty about the process was the main reason why people don’t adopt this kind of process in the first place, or doesn’t adopt digital tools in the first process, because they were not certain that the risk will be worth it, worth the trouble.

Because of the mainstreaming, now the people can say, "The Digital Ministry just does it every Friday." There’s much less uncertainty, and the fear over risk-taking is also much lowered because instead of blame-avoiding and credit-seeking, I adopted the reverse strategy. I’m blame-seeking and credit-avoiding.

Meaning that if a public servant decide to run this process, it’s an experimental process. They can always justify saying, "It’s the Digital Minister doing experiment." PDIS just absorbs the risk. If it goes wrong, it’s all Audrey’s fault basically.

It goes well, then the attribution, the credit is actually to not just the ministry, but actually to the career public civil servant themself. Because through radically transparent records, everybody can see who is the one that actually proposed this wonderful idea.

Using this strategy, we see a lot of public servants now proposing creative ideas that may be have only 20 percent chance of succeeding. Then PDIS collaboratively make it a higher chance of succeeding, but the motivation is there because they know that if it works, it’s their credit. If it doesn’t work, it’s Audrey Tang’s fault. I think it is a really important atmosphere to make this happen.

We resolve many regional cases this way also. We’re actually going to the most rural areas and even offshore islands and try to get people into this kind of deliberative meetings without learning any digital tools, because it’s the meeting itself being digitized using what we call ambient computing rather than having the people going to a website or whatever. That’s our current situation.

We’re now being handed more and more challenging cases. At the end of this month, we will finally meet 工鬥, the Worker’s Solidarity Struggle Union, who argue for seven more national holidays.

That was the classic case. I remember when I had an interview with Amaëlle Guiton, she said that that’s one of the classic cases that there’s just no room for compromises, it’s an either-or thing.

We believe there is still room for ideation and for creative solutions. We’ll find out whether it actually works or not end of this month.

Next month, we will also go to the only public national park for marine species in Taiwan and where people, again, argues for putting a ban on fishing in an area in exchange for more diversity and better diving and other tourism inputs.

Again, this is a classical environmentalist versus local livelihood debate that many people think is an either-or, but we still think there is room for creative solutions.

So we’re being handed more challenging cases in the past few months. That also means the career public servants trust this method more because previously they were more inclined to just suggest cases that involved only digital issues or things that they know there’s less to lose.

The fact that they are now willing to vote this kind of cases means also that they feel more confident in this kind of methods, so that’s how it’s working here.

Super nice.

Super nice. Thank you. I don’t if you have some question for us. I think we have a lot to think.

It’s really nice to have this conversation. I miss Paris a lot. That’s the only thing I’d like [laughs] to say, and I look forward to visit either in flesh or as a robot to the new place.

For sure you should come to Paris. Please, come to visit either in our old place, either in our new place. We’ll be there.

All right. That’s all I have. Wow. It’s an hour.

Time really flies. Are you comfortable with me just publishing the video?

Yeah, for me it’s OK. I think our video is really blur anyway so...

It’s true. OK. I’ll publish the video and also make a transcript if that’s OK with you. We can just edit it over the next 10 days or so. All right. Best luck on your gathering.

Can you send us the link of the...?

The video and the transcript, once it’s done, of course.

Thank you.

Thank you so much.

Thank you so much.

Have a good day.

Have a good day!

You too.

Bye-bye.

Bye-bye.