[00:00.000 --> 00:11.040]  So, good morning everyone. I am Adrian, a PhD candidate from Lidon University in the
[00:11.040 --> 00:16.800]  Netherlands and I'm going to, today I'm going to talk about my project, my software, so
[00:16.800 --> 00:22.840]  called relativization, which is a weird thing, an interstellar social simulation framework
[00:22.840 --> 00:27.440]  and a turn-based strategy game as well. And I hope this can be an interesting start of
[00:27.440 --> 00:35.960]  today's schedule. So, before I go into the details of my thingy, I would like to do a bit of self
[00:35.960 --> 00:43.400]  introduction because it relates to my overall open source development experience. And I was
[00:43.400 --> 00:50.200]  educated in physics, I did my bachelor in physics, I did some master's research on so-called
[00:50.200 --> 00:57.160]  gravitational waves under the LIGO collaboration. And I must say, like before my master's research,
[00:57.160 --> 01:01.280]  I would say, kind of mathematical students, I didn't really like computer, I like pen and paper
[01:01.280 --> 01:06.640]  more. And, but like my research forced me to appreciate computer because I was forced to use
[01:06.640 --> 01:12.560]  doing things like supercomputers, Linux, terminals, and writing programming language, programming
[01:12.560 --> 01:19.120]  like C or Python. So, I started to appreciate like software or open source software in general. So,
[01:19.120 --> 01:28.560]  that's why I get into this world. And after my master, I decided to kind of switch my field to
[01:28.560 --> 01:33.280]  do something social science. So, that's why I came to the University to do something called like
[01:33.280 --> 01:39.760]  quantitative science study. And I'm supposed to work on, supposed to work on social modeling,
[01:39.760 --> 01:45.720]  simulation, or data analysis on some kind of social data, actually most specifically for
[01:45.720 --> 01:50.480]  academic system. And as a hobby, I kept contributing to open source software because I think it's
[01:50.480 --> 01:56.160]  fun to work on, I mean, just use the software and if you find something wrong, then you just submit
[01:56.160 --> 02:02.200]  a PR or a great issue for that. And I learned a lot from that kind of process. And then like,
[02:02.200 --> 02:08.960]  after, because I started my PhD right before the pandemic, as you can imagine like things go
[02:08.960 --> 02:15.400]  when quite long, because I partially blame the pandemic, because for example like the data that I'm
[02:15.400 --> 02:22.440]  supposed to use is still not ready, even now. And the whole collaboration become mass, the
[02:22.440 --> 02:27.560]  infrastructure is mass. So, I partially also blame the engineer in my collaboration because
[02:27.560 --> 02:36.400]  they are so annoying to work with. So, I have to rescue my PhD, right? I can't wait for that,
[02:36.400 --> 02:42.160]  the things to happen. So, I need something to work it out. I can work it out by myself. And
[02:42.160 --> 02:49.000]  because I, because it's pandemic, so I have to work from home. And I was also kind of traumatized
[02:49.000 --> 02:54.000]  by the collaboration that I experienced. And I know something about physics because of a
[02:54.000 --> 02:58.160]  background. I know something about social modeling. And I'm also kind of familiar with
[02:58.160 --> 03:04.600]  software engineering. I'm not trained in that field, but I learned a lot from my hobby. So,
[03:04.600 --> 03:10.880]  it is natural for me to try to combine these things together to create something for a thing,
[03:10.880 --> 03:17.800]  a weird thing, interstellar simulation. And then like, perhaps some of you guys are a fan of
[03:17.800 --> 03:23.320]  science fiction, movies, or games. And you may ask these kind of big questions like,
[03:23.320 --> 03:28.160]  will human civilization become interstellar one day in the future? Or like, does alien exist or
[03:28.160 --> 03:35.960]  not? And if they exist, or if we will become interstellar, we will be curious about like, what
[03:35.960 --> 03:41.000]  is the form of society of those interstellar civilization, right? And of course, it's very
[03:41.000 --> 03:45.920]  hard to study those kind of thing, rigorously, in academic settings. But perhaps one way,
[03:45.920 --> 03:52.240]  at least I claim, I argue that one way to try to explore this is to assume that some social
[03:52.240 --> 03:57.600]  theory we know nowadays can still be a good approximation for that kind of society, like
[03:57.600 --> 04:02.880]  perhaps if we expect that they to also be utility seeking, or there are some collective
[04:02.880 --> 04:07.320]  actions there. And of course, we can study that rigorously because we don't have the data,
[04:07.320 --> 04:12.080]  we don't have observations, but we can try to explore this kind of domain by simulation.
[04:12.080 --> 04:19.200]  And in case if you're not familiar with social modeling, or like,
[04:19.200 --> 04:28.200]  in case if you're not familiar with social modeling, and nowadays when we talk about social simulation,
[04:28.200 --> 04:33.880]  it's majority, like they're talking about agent based modeling. So, social scientists design
[04:33.880 --> 04:39.240]  some behavior rules based on the understanding of how human works, or based on some experiments to
[04:39.240 --> 04:44.360]  design computational rules for agents, and they put the agents into a world just like
[04:44.360 --> 04:49.760]  machine learning people do, and to see how to interact with each other and interpret the
[04:49.760 --> 04:57.200]  overall simulation outcome from that. And there are many existing frameworks that can help us
[04:57.200 --> 05:05.600]  to do this kind of simulation, but I have to create my own framework because for existing
[05:05.600 --> 05:11.240]  social simulation framework, they don't support the physics well, because we're going to work
[05:11.240 --> 05:17.840]  on like interstellar society, and it is a scale that is not on the earth, so it's not easy to
[05:17.840 --> 05:21.960]  deal with, it's possible, but it's not trivial. And there are also some kind of physics simulation
[05:21.960 --> 05:25.480]  framework out there as well, but they are not built for social scientists, so it's also a bit
[05:25.480 --> 05:34.640]  hard to work with. So that's why I have to create my own framework. I think it's interesting and
[05:34.640 --> 05:42.880]  useful if I want to create something meaningful for this thing. And when I talk about the physics
[05:42.880 --> 05:49.240]  for interstellar society, I'm talking about special relativity, which is simple undergraduate
[05:49.240 --> 05:55.400]  level physics, because it is a fiction when we consider a scale like 100 light years or 10
[05:55.400 --> 05:59.520]  light years, I think it's a good starting point of like try to imagine about this kind of thing.
[05:59.520 --> 06:07.200]  And special relativity tells us that there's an intrinsic time delay out there, so everything
[06:07.200 --> 06:13.040]  cannot travel faster than the speed of light, and also special relativity tells us that like
[06:13.040 --> 06:18.840]  there's a time direction effect, so if you move very fast, close to the speed of light,
[06:18.840 --> 06:26.040]  you're close to the store, so experience the store time essentially. And in the setting of
[06:26.040 --> 06:33.080]  social simulation, we have to deal with problems like what can you see or what can an agent see,
[06:33.080 --> 06:39.360]  and how an agent interacts with other agents, because of time delay, and how we change the
[06:39.360 --> 06:49.400]  state of the agent. I will go into this one by one in the following slides. Because of the time
[06:49.400 --> 06:54.800]  delay, information travels at the speed of light, so we are not seeing each other instantaneously.
[06:54.800 --> 07:02.080]  We are seeing the path of each other because of the delay, so here the red agent, the red people
[07:02.080 --> 07:08.360]  here see the, the red person here see the, the, the path of the blue, blue agents of people,
[07:08.360 --> 07:15.000]  because they have, they have time delay, and it depends on the distance. And technically they
[07:15.000 --> 07:24.640]  see the, the path like comb of the, of the, of the universe at that point. And how the second problem
[07:24.640 --> 07:28.840]  is how the, how the agents interact with each other. I mean if you're a programmer, you're
[07:28.840 --> 07:34.960]  probably familiar with the command design pattern, and this is used for like dealing with, if you
[07:34.960 --> 07:38.120]  have something you might execute, but if you have, if you have a delay, then you have to create a
[07:38.120 --> 07:45.120]  command, wait for a delay, and send it to, send it to a target, and this is what I, I, I, I used it in, in, in, in this framework,
[07:45.120 --> 07:49.840]  so that like if you have some, some, some agent need to interact with each other, they can't, they can't
[07:49.840 --> 07:55.800]  just do that directly, they have to, have to send the command and, and the, the command travels at a speed of light in, in, in the
[07:55.800 --> 08:07.760]  universe. And lastly, the, the, the last problem, how, how we change the state of the agent. We define
[08:07.760 --> 08:13.840]  a term here called mechanism, which is essentially the dynamics of the model. It, it, it changed the state of the
[08:13.840 --> 08:21.280]  agents and, and send commands out, so you define the whole, how, how, how the model actually works scientifically. And
[08:21.280 --> 08:30.720]  recall that because clock takes over for, for, for, for moving agents, so I divide this into, into two types of
[08:30.720 --> 08:38.120]  mechanism, a regular mechanism executed one, a process once per turn, and dilated mechanism process once per end turn
[08:38.120 --> 08:44.240]  because it depends, and, and, and the end depends on the speed, because if it is moving faster, then some things, it's
[08:44.240 --> 08:52.080]  slower, time is slower there, so executed, the execution is slower there. So it is, for, for, for convenience, if you, if you
[08:52.080 --> 08:57.000]  model your thing space, and if you, if you, if you model your thing space on these two, two types of mechanism, then it's
[08:57.000 --> 09:05.280]  easier to, to, to, to adjust, to, to follow the physics. And as an example, regular mechanism could be something like
[09:05.280 --> 09:11.000]  you observe things, and update your information, and it's not something that you, it actually takes time so it can, it can
[09:11.000 --> 09:16.360]  be executed once, once per turn. And for dilated mechanism, it can be something like manufacturing, it actually takes
[09:16.360 --> 09:24.160]  time to, to, to produce some product, and to, to, to, to make, make it and send it out. So if you have a slower time, then
[09:24.160 --> 09:33.720]  you, you manufacture that, sorry. So to deal with this, this kind of problems, I, as I said, I created framework, and the
[09:33.720 --> 09:41.040]  framework is written in Kotlin, there are many reasons behind that, I'm not going to, to, to give you that, and, and
[09:41.040 --> 09:48.480]  basically the framework and force a specific way to, to, to, to model your, your, your, your social model, so that it, it can
[09:48.480 --> 09:54.560]  be automatically, automatically account for the physics. And it also provides some functionalities to, to, to help you to
[09:54.560 --> 10:01.800]  develop the so-called in this data social model. And it deal with some technical subtleties like parallelization of the, of
[10:01.800 --> 10:10.800]  the, of the simulation, and how do you serialize things into, into, into data, or like how do you make deterministic
[10:10.800 --> 10:22.280]  simulation using, under a parallel process. And this is the, the kind of the core of the, of the framework. And you can
[10:22.280 --> 10:27.880]  check my archive repeat, and actually it's going to be a proceeding for, for some other conference. And, and for the, for the
[10:27.880 --> 10:37.320]  algorithmic details, and for the mathematics details. You want to take a picture? I, I, you can find it like in, in the
[10:37.320 --> 10:48.280]  last slide, but anyway. So, I mean, I, I, I didn't begin, I, I wasn't an experienced programmer when I started this
[10:48.280 --> 10:54.640]  project, so I, I make some decision that perhaps I would change if I, if I do it in another way. If, if I start it, start
[10:54.640 --> 11:02.280]  from now. But like, I, I, I, anyway, like I make some decision, for example, here on, on how, how we should use this
[11:02.280 --> 11:30.520]  framework, for example, I decided that like this should be used, say, I say, I say, I say, okay, here it works. So, there
[11:30.520 --> 11:35.520]  I have some technical reason behind why I made this decision because I think it's always easier to provide a library for people to use.
[11:35.520 --> 11:45.520]  I mean, because the reason behind it is that I try to keep it in a pure Kotlin project and it's quite not possible to do things like this,
[11:45.520 --> 11:52.520]  serializing string into objects because the diffraction library of Kotlin is not very powerful.
[11:52.520 --> 11:58.520]  And anyway, this is the decision I made. I may change in the future, but it's how it's now working.
[11:58.520 --> 12:08.520]  And to use this, you have to create a template using the source code and build a model based on the template.
[12:08.520 --> 12:14.520]  And I have created a few effect samples on how to use this framework.
[12:14.520 --> 12:24.520]  I mean, I will do a brief illustration on the results because it's very interesting to see how things actually work out.
[12:24.520 --> 12:31.520]  So the first example is the flocking model, which is kind of the standard of the 101 HMAS modeling.
[12:31.520 --> 12:38.520]  It's a very simple model. The bird follows a very simple alignment bruise and they align with each other.
[12:38.520 --> 12:43.520]  And it creates some interesting microscopic pattern as shown in the figure.
[12:43.520 --> 12:47.520]  And here we are interested in interstellar flocks.
[12:47.520 --> 12:54.520]  I mean, it's a very hypothetical scenario, but you can interpret it as kind of a group of spaceships
[12:54.520 --> 13:00.520]  and they try to align with each other in the universe where some kind of mysterious creature like living in space.
[13:00.520 --> 13:07.520]  And I assume for the sake of physical realism, I assume they are propelled by a proton rocket
[13:07.520 --> 13:10.520]  and we can measure how well the flocks align.
[13:10.520 --> 13:17.520]  And this is the simulation result. I'm also not going into the details of this thing.
[13:17.520 --> 13:23.520]  But you can see some interesting things like if you can adjust the speed of light because this is a simulation
[13:23.520 --> 13:28.520]  and if the speed of light is stored and they take more time, it takes longer to communicate
[13:28.520 --> 13:35.520]  and they are less well-ordered and you can also tune the perturbation on the flock
[13:35.520 --> 13:41.520]  and so it takes more, it's harder to align with each other and it takes more energy to do so.
[13:41.520 --> 13:49.520]  And you can see in the figure on the left hand side or in the right hand side in your alignment
[13:49.520 --> 13:57.520]  then it takes half the mass of the population to align with each other for one year, for one turn.
[13:57.520 --> 14:04.520]  So it's not a very environmentally friendly way of traveling, I would say.
[14:04.520 --> 14:14.520]  And the second example is some kind of knowledge dynamics which is closer to what I'm supposed to do in my PhD.
[14:14.520 --> 14:22.520]  So we study how I try to model how research or knowledge is generated in the interstellar setting
[14:22.520 --> 14:28.520]  and there are some assumptions that are made like cooperation takes time,
[14:28.520 --> 14:35.520]  we need cooperation to stimulate innovation but cooperation takes time to process.
[14:35.520 --> 14:39.520]  You have to invite people to cooperate with you and you have to wait for the response
[14:39.520 --> 14:43.520]  and if it is very far away then it takes a long time to work with people.
[14:43.520 --> 14:47.520]  So this thing is affected by the time delay.
[14:47.520 --> 14:51.520]  And there are some other factors like if there is a small time delay
[14:51.520 --> 14:59.520]  and so you have a more rapid information exchange then it essentially creates a more competitive environment
[14:59.520 --> 15:04.520]  because when I do something and you want to compete with me and you update information very frequently
[15:04.520 --> 15:11.520]  then it creates a kind of stress to work your research out.
[15:11.520 --> 15:15.520]  And also like research of the next time, it is a dilated mechanism.
[15:15.520 --> 15:18.520]  It is affected by the time dilation if you want to move.
[15:18.520 --> 15:23.520]  There is a cause if you want to move closer to each other to do research
[15:23.520 --> 15:30.520]  like there is a cause if you want to move from your home country to this place to work with each other.
[15:30.520 --> 15:34.520]  So this is also some kind of simulation results.
[15:34.520 --> 15:37.520]  This I would say is kind of interesting.
[15:37.520 --> 15:44.520]  So we can also tune the speed of light and as you see in the figure on the left-hand side
[15:44.520 --> 15:52.520]  the blue curve was lower at the beginning and it becomes higher eventually.
[15:52.520 --> 15:57.520]  This is a curve when you have a very slow information travel.
[15:57.520 --> 16:03.520]  Because the innovation or the cooperation started very slowly at the very beginning
[16:03.520 --> 16:10.520]  so it rises very slowly but at the finals ultimately because of its less competitive
[16:10.520 --> 16:15.520]  overall research outcome becomes better eventually.
[16:15.520 --> 16:18.520]  I think it is an interesting result.
[16:18.520 --> 16:22.520]  The figure on the right-hand side shows that there are two kinds of strategies.
[16:22.520 --> 16:26.520]  You can stay in your home country and do all the things remotely
[16:26.520 --> 16:29.520]  or you can move the conference physically to work with each other.
[16:29.520 --> 16:33.520]  The orange curve represents the moving strategy
[16:33.520 --> 16:37.520]  and the blue curve represents the remote strategy.
[16:37.520 --> 16:43.520]  There is a cause imposed by the city because there is time dilation
[16:43.520 --> 16:47.520]  so it starts out lower and eventually pays off in the future.
[16:47.520 --> 16:53.520]  I think it is an interesting result to show up here.
[16:53.520 --> 17:01.520]  This is kind of interesting but you can also always criticize about this kind of model
[17:01.520 --> 17:07.520]  because it is too hypothetical. We are studying something that is not happening in the real world
[17:07.520 --> 17:09.520]  and perhaps we are getting nothing out of that.
[17:09.520 --> 17:14.520]  I kind of agree with this criticism but I am just trying to get something out of the framework.
[17:14.520 --> 17:19.520]  If people are not interested in this kind of model
[17:19.520 --> 17:22.520]  or I also feel like this is not interesting enough
[17:22.520 --> 17:31.520]  and then I made the decision that perhaps I can make a game out of this to make it more interesting.
[17:31.520 --> 17:37.520]  The reason why I decided to make a game is also because I like simulation games
[17:37.520 --> 17:43.520]  and simulation games are kind of similar to modeling
[17:43.520 --> 17:46.520]  or HMAS modeling in a sense.
[17:46.520 --> 17:53.520]  Also, I have many things that I want to test out. The model that I have presented are kind of simple models
[17:53.520 --> 17:56.520]  but they are more complicated mechanisms that you can implement
[17:56.520 --> 17:58.520]  but you can't do this rigorously because they are so complex
[17:58.520 --> 18:03.520]  you can't get anything adequate or interpretable from those ideas.
[18:03.520 --> 18:08.520]  A game provides an environment for me to test things out
[18:08.520 --> 18:16.520]  and to just explore what I can do in this institutional model.
[18:16.520 --> 18:23.520]  Also, perhaps this is one of the sensible ways to get something actually useful from the framework.
[18:23.520 --> 18:28.520]  If it is not very academically interesting, at least I have a game for that.
[18:28.520 --> 18:39.520]  For a game, it can be very complex and I try to address many different things in the game
[18:39.520 --> 18:42.520]  and I try to ask a very good question.
[18:42.520 --> 18:47.520]  For example, we can try to ask how should the economy look like in the international society?
[18:47.520 --> 18:50.520]  Does credit-based economy work?
[18:50.520 --> 18:53.520]  Nowadays, we are relying on currency, credit card
[18:53.520 --> 19:00.520]  but those kinds of currency relies on some fundamental building blocks of the society
[19:00.520 --> 19:04.520]  which might not work in an interstellar setting because the time delay, how do you exchange money
[19:04.520 --> 19:07.520]  and how do you define the exchange rate?
[19:07.520 --> 19:14.520]  Perhaps a simple part of the economy that we use historically is a better choice for those kinds of society here
[19:14.520 --> 19:21.520]  and perhaps we can use some kind of fuel or energy as the money instead of credit card.
[19:21.520 --> 19:24.520]  Also, we can ask how the political system works.
[19:24.520 --> 19:26.520]  Does democracy work there?
[19:26.520 --> 19:32.520]  If you have to wait for 100 years to receive the full-time results from the other planet.
[19:32.520 --> 19:37.520]  We can also ask how we can optimize scientific research in this kind of system.
[19:37.520 --> 19:41.520]  I have my own ideas on how these things work.
[19:41.520 --> 19:46.520]  I put all the things into the game including population dynamics, politics, diplomacy, economics
[19:46.520 --> 19:52.520]  like warfare, science and technology. Some of them are typical elements in strategy games
[19:52.520 --> 20:00.520]  and some of them are not that typical but I still get the inspiration from some other games.
[20:00.520 --> 20:07.520]  I try to get this to work on PC and Android
[20:07.520 --> 20:13.520]  and that's one of the reasons why I choose Collin to implement this kind of thing
[20:13.520 --> 20:19.520]  and I hope this is going to be a multi-player game in the future.
[20:19.520 --> 20:23.520]  I implemented some kind of server-kind architecture.
[20:23.520 --> 20:30.520]  For a game, you have to limit how a client can act
[20:30.520 --> 20:34.520]  because you can't allow them to do everything that they are going to cheat in the game.
[20:34.520 --> 20:38.520]  What they can do is essentially receive a few of the universe
[20:38.520 --> 20:42.520]  and they can send a command to do action.
[20:42.520 --> 20:48.520]  I also created a utility-based or DO utility-based AI for the game.
[20:48.520 --> 20:53.520]  This is the Baker architecture that is built on top of the framework.
[20:53.520 --> 20:56.520]  This is the screenshot of the game.
[20:56.520 --> 20:59.520]  I know this is not graphically very attractive but this is like,
[20:59.520 --> 21:17.520]  oh, I can do and... where is it?
[21:17.520 --> 21:35.520]  I'm losing my side.
[21:35.520 --> 21:40.520]  This is how the game actually looks like.
[21:40.520 --> 21:48.520]  You can do some things here to do something and add some commands.
[21:48.520 --> 21:53.520]  This is a complex interface and so I'm not going to actually show everything here
[21:53.520 --> 21:56.520]  but as something is working here.
[21:56.520 --> 21:59.520]  What's the scale?
[21:59.520 --> 22:01.520]  The spatial scale.
[22:01.520 --> 22:10.520]  For this, it's like three times three light years keep.
[22:10.520 --> 22:12.520]  This is a very small universe.
[22:12.520 --> 22:14.520]  It can be larger.
[22:14.520 --> 22:20.520]  At least it can handle a thousand light-year keep game without pressing my computer.
[22:20.520 --> 22:23.520]  Thanks for the open-source ecosystem.
[22:23.520 --> 22:27.520]  I can rely on all kinds of open-source library to build my game.
[22:27.520 --> 22:32.520]  I use DeepGDX for the working interface, I use Coroutine for the parallelization,
[22:32.520 --> 22:37.520]  and also thanks for the open-art community so that because I'm not an artist,
[22:37.520 --> 22:39.520]  I can't create all the art by myself.
[22:39.520 --> 22:43.520]  I can rely on creating component-licensed assets and I can use open-source software
[22:43.520 --> 22:48.520]  to escape if I want to draw something.
[22:48.520 --> 22:56.520]  It's going to be the end of my presentation but I'd like to have a bit of reflection on my project.
[22:56.520 --> 23:00.520]  I think I've learned a lot by contributing to open-source software
[23:00.520 --> 23:05.520]  and the ecosystem gave me plenty of building blocks from my own project.
[23:05.520 --> 23:08.520]  Of course, because of this, it's natural for me to also open-source my project
[23:08.520 --> 23:11.520]  and at least I've created something interesting.
[23:11.520 --> 23:15.520]  There are also some problems that I face during the development of this project.
[23:15.520 --> 23:18.520]  For example, there's a lack of open-source culture in my workplace.
[23:18.520 --> 23:23.520]  Some people there do use open-source software but most of them do not care that much.
[23:23.520 --> 23:26.520]  Most of the time, I'm by myself.
[23:26.520 --> 23:31.520]  As a one-man project, it's a bit too ambitious, I would say.
[23:31.520 --> 23:35.520]  I don't have time to publish all the things like graphic interface documentation
[23:35.520 --> 23:39.520]  and I can't ensure that things are fun to play and I don't have time to work on the translation.
[23:39.520 --> 23:43.520]  I did try but I don't have time to work on all kinds of things
[23:43.520 --> 23:47.520]  because I'm also doing my PhD, I have to write papers and do presentations.
[23:47.520 --> 23:53.520]  As I decided to go fully open-source, I chose those libraries to depend on
[23:53.520 --> 23:58.520]  but perhaps there are some kind of industry standard solutions like
[23:58.520 --> 24:01.520]  if you're creating a game, people just use Unity for that
[24:01.520 --> 24:05.520]  and perhaps it would be better for the project so I have more time for other things.
[24:05.520 --> 24:13.520]  It's just something that perhaps there's a price to pay for if you decide to go to go open-source.
[24:13.520 --> 24:19.520]  Actually, I did try a bit to commission my process because it's possible to sell an open-source game, right?
[24:19.520 --> 24:21.520]  I know it's possible but it's also very hard to do so.
[24:21.520 --> 24:24.520]  It takes time to advertise it so I gave up eventually.
[24:24.520 --> 24:27.520]  I didn't have time for an hour to push it forward.
[24:27.520 --> 24:30.520]  This is the summary of my presentation.
[24:30.520 --> 24:35.520]  I have created a framework for industrial and social models
[24:35.520 --> 24:37.520]  and I have created some models based on the framework.
[24:37.520 --> 24:40.520]  I have also created a game on top of that
[24:40.520 --> 24:43.520]  and perhaps it's lacking some immediate practical value
[24:43.520 --> 24:48.520]  and that's not what I thought was that like it didn't exist tomorrow, right?
[24:48.520 --> 24:51.520]  But I think it's not going to happen in a short while.
[24:51.520 --> 24:56.520]  So perhaps there's no immediate practical value but I still believe this kind of meaningful
[24:56.520 --> 24:58.520]  and educational exploration.
[24:58.520 --> 25:03.520]  If the game is fun to play eventually then you can ask your kids to play for that
[25:03.520 --> 25:06.520]  and perhaps they can learn some physics from the game, right?
[25:06.520 --> 25:10.520]  I don't know. And at the very least, at the most important,
[25:10.520 --> 25:13.520]  unless I feel like I've achieved something in my PhD,
[25:13.520 --> 25:16.520]  I'm not wasting my time in my pandemic.
[25:16.520 --> 25:23.520]  So that's all my presentation and this is the link of the project
[25:23.520 --> 25:27.520]  and if you'd like to make it a star, just check it out.
[25:27.520 --> 25:31.520]  And the thing that I show is, I think you can find those things
[25:31.520 --> 25:33.520]  in the readme file of this page.
[25:33.520 --> 25:36.520]  So thanks a lot for your attention.
[25:47.520 --> 25:51.520]  We'll take a few questions while the next speaker will come and...
[25:51.520 --> 25:52.520]  Thank you.
[25:54.520 --> 25:56.520]  Okay, Hafer.
[25:56.520 --> 26:01.520]  In your simulations when you said that the simulations then cluster,
[26:01.520 --> 26:05.520]  how many years was simulated in the simulation
[26:05.520 --> 26:07.520]  before you got to sort of the result?
[26:10.520 --> 26:12.520]  Yeah, I'm talking like...
[26:12.520 --> 26:15.520]  Can you repeat the question for the recording?
[26:17.520 --> 26:24.520]  So the question was like how many years did I simulate for the model that I presented here?
[26:24.520 --> 26:28.520]  I typically simulate it for 1,000 or 2,000 years
[26:28.520 --> 26:32.520]  and because of the spatial scale it's like 10 times 10 times 10 light year
[26:32.520 --> 26:37.520]  so it has affection time for the spaceship
[26:37.520 --> 26:42.520]  to move towards each other and to do the cluster.
[26:42.520 --> 26:47.520]  I assume that they move at a speed of light velocity.
[26:47.520 --> 26:49.520]  And you have a question?
[26:49.520 --> 26:52.520]  Yes, I guess it's more technical.
[26:52.520 --> 26:55.520]  So your framework is written in copper.
[26:55.520 --> 27:01.520]  Does it support any form of extension like in another JVM image?
[27:01.520 --> 27:06.520]  I mean, in principle you can use Kotlin with other JVM images.
[27:06.520 --> 27:11.520]  In your, for instance, or I guess customized agents,
[27:11.520 --> 27:14.520]  maybe even maybe the physics model that say
[27:14.520 --> 27:20.520]  I want to implement faster than light and travel or something like that.
[27:20.520 --> 27:28.520]  So the question was like can you use other JVM language to extend it?
[27:28.520 --> 27:32.520]  I think in principle it is possible because you can always,
[27:32.520 --> 27:36.520]  the way to use this is to create something from the source code
[27:36.520 --> 27:40.520]  and you can always work with JVM language in the source code.
[27:40.520 --> 27:45.520]  But I'm not entirely sure because I use some Kotlin feature like inline function
[27:45.520 --> 27:48.520]  to simplify the code so I'm not sure for that kind of thing
[27:48.520 --> 27:51.520]  can you use it nicely with other JVM language?
[27:51.520 --> 27:54.520]  I know you didn't go into that, but why Kotlin?
[27:54.520 --> 27:56.520]  Why not like Azure or Java?
[27:56.520 --> 28:00.520]  I mean, because I want to work on, I want to actually work on Android
[28:00.520 --> 28:04.520]  and you know, Android Java support is not great.
[28:04.520 --> 28:07.520]  Okay, sure.
[28:07.520 --> 28:19.520]  Thank you.