 to make open source, to make Bangkok suckless. So let me introduce quickly first. So my name is Prum and I'm an indie hacker from Bangkok, Thailand. So I'm originally from the southern of Thailand, which is Songkwa. I'm also co-founder of Creator's Garden. Before this, I was VP of engineering at Belgium. So he wants a lot of distance, as you all can tell, but all of this has the same goal, so you can feel like you're all created and better off in an area. But there are all kinds of students. So there's a hackathon that was done. And it was called tuition. No one needs an experimental idea at the moment. So he wants a lot of distance, where everyone can talk to you basically anything that is stupid. So this time it's going to be about it turns out people really enjoyed it. And we have been here for six years, great. So yeah, so this has been our six year now and the seven one is going to be organized soon. So we have been building that since 2017. And we have gotten over 900 developers over the years. And they have built so many projects. So let me show you a couple from our students. First, how do you build a bug free code? Maybe you use a language model, right? To correct the code, but we have a better solution. You just put like that insignia and now your code is perfect. This is a common solution near like Songkran in Thailand. Yeah, in Thailand we call this germ, which is giving your server a boost. It's better than if firewall. Now, how do you improve friendliness of websites? Thai websites are known for being very hard to use and not flashy. So we build a Chrome extension to make it so much better. So you can randomize from the color palette you want. I warn you, this is not epilepsy fairly. And this is the best one. How can animals code? If you're a seal, you don't have hands. You only have your voice. How do you code? You see, you do a programming language for them. So this is a programming language that our friend built called Unlang. Un is the sound of seals. So it uses just five four characters but it's fully tiering complete. You can build any application with this and it compiles to C. Okay, another one. How do you, can you be more productive when you have a lot of windows? You use a window manager, right? No, we do a trolling window manager. So if you use keyboard, you're fine. If you use a mouse, you are screwed. So it tries to move the window if you move your mouse. Okay, and this one, right now we're moving towards the age of quantum, right? So how do you supersede Moore's law? Which is the law that indicates silicon gets smaller. It's, well, basically harder to jam more stuff on the ships. Well, you enslave the entirety of human and then you make them run assembly. So every assembly instruction is basically just doing arithmetic, right? So technically you can write a program that transcribes your assembly code and send it to a human to answer because we have seven billion people in the world. You can basically multi-trade better than SIMD. Yeah, so this is obviously very, very stupid. There's a couple more. So like obviously a rig row. Oh, this one is cool. So basically if you have a logo but your client say it sucks, the logo doesn't look like it's properly thought out. We use OpenCV to randomly draw circles. So it looks like it's properly designed. Yes. So yeah, the project itself is obviously extremely silly but technology behind us. We use OpenCV, we use AI, we use machine learning. So people got to learn from this obviously stupid things. Yeah, so if you're interested in at 100 projects, just scan this code and you can see a lot more gems for yourself. All right. All right. So Thai people, they are very creative as you can see but there's a weird thing where I saw a few developers engaged in open source. Right now in this conference, there's only three people. There's me, there's Mishari and there's one more. So like what happened? We have like so 900 developers building open source but they're not getting outside of Thailand. They're not building anything useful. Here, look at commits at top. You can see that people, though yeah, the first 10 ones are my friends. I ignore that. But basically other people, they are not engaged with open source as much. So this comes to a fundamental question. What motivates you to build open source? This was me when I was 14 years old, I think. So I got in love with Linux because my Windows file system broke down and my files is gone. So I cry for over a day until I remember I have Linux. So every proprietary system didn't work but basically I downloaded the source code for NTFS and I program it to remove all the error and suddenly all my files are back. So I fell in love with open source because I can modify the Linux driver, the Linux kernel and I can make anything possible. And if you use Linux, you can have this really totally fancy Windows. There's the KDE booth if you're interested. Another motivation is what you like. So my friends, he like a certain kind of Japanese cartoon. I'm not gonna say what kind it is but basically he's really into that so that he built a whole tech stack around it and it only takes less than a dollar to serve a thousand people. And here's even worse or even better because he was building this cartoon website and he think it's not fast enough. So he wrote his own web framework in bun and now it's the fastest. So there's a lot of motivation that can make people do open source or when you see a lot of bad syntax and you want to make it better, you make it open source. Now, unfortunately in Thailand, there are a lot of hostility towards open source and that was part of our culture in the beginning because we were afraid of people taking our IP, afraid of family foreigners taking our land. So that kind of fear was kind of ingrained but like as the newer generations grow, they want to change that. They want to try to make things more open and more shared. So what motivates you? I would say it's really two things. First, you fix the shortcomings of your software. You're using a library but you see, hey, the software is not really doing what I like. So I fork it, I write a plugin. Another thing is you get to learn so quickly because if you do open source, you get to learn from really good people. You get people like trying to teach you by comments. Now, what do Thais want to fix? This is like the big question. If I can't answer this, I can't get as many Thai people to join me. So let's talk the elephant in the room, shall we? Okay, what do you think often you think of Thai culture? You might think of kao sa, na na, all the clubs. You might think of beaches. You might think of temples and pad thai. But if you take the red pill, well, it looks a little different for us in Thailand. This is the freshest day in Bangkok. Yeah, when you wake up, this is like what you see. So yesterday it was 152 AQI for us in Bangkok. So basically if you forgot to wear your PM mask, well, you're gonna have to go to the clinic. Many people are feeling their eyes burn. So, oh, dryest feet in Bangkok. So we got so many floods. We have like really bad water management problems. And here's the light test traffic in Bangkok. I'm not joking. So there are so many cars and it's very hard to travel. And this is the cleanest street post. Some people say it's kind of like an installation art in Thailand, but well, trust me, it makes it really hard to travel. So everyone knows that Bangkok has a problem and all of us want to fix it. So how can we do open source tooling with data science to make it better? So this is where we assemble people like the Avengers. And I want to tell you one thing right now. Even the topic of the talk is Bangkok open source hackathon, but hackathon itself is a lie. It's not really a hackathon. So it's all started when the type of grammar association, they asked me, hey, can you build a hackathon that would take all the ideas from hackbtk and make it happen? So I thought, what the heck is hackbtk? So it turns out it's an idea of fun, you know, that kind of hackathon where you don't actually get to code and people are actually like brainstorming ideas. Do you think it went off with hackathon? Well, it went exactly as you mentioned, the ideas remain ideas, most of them, and the startups that say they want to take that idea and make it happen, some of them can, but it's a proprietary solution. So I actually met this guy, this handsome guy, he's the newest governor in Thailand. He's like, Thai people call him Mr. Strong, so he's Cha Cha. So we talked to him about open source in Thailand, and this is what he told me, and it kind of struck. So he told me he wants to see the software being maintained and adopted, keep developers, they should spend more time with the communities, they're working to solve problems. Also, he wouldn't like to see projects be abandoned. So this is really the heart of open source. If it's an idea in a hackathon and it dies, no one can use it, but if it's open source, everyone can fork it, they can use it. So here's another red pill. As much as I like organizing hackathons, I have done like over 20 of them, but the problem is it's not really the answer to this problem, neither is proprietary startups because for hackathons, they're really perfect for like ideation, getting people to learn, right? That's its own goal, but it's not really for being something that sustains. Another thing is called pick startups. In Thailand, there are hits, but no one's making it open source. They're not sharing their code, but we are the one that's paying the taxes. So there's the same public money, public code. Now, what's the solution? Let me tell you one project we have been doing in Thailand. So we had the election several years ago, so we have this really fancy dashboard, but the thing is this dashboard was built in less than eight days. So how do we do it? It has so many features, it has so many things. Well, we make it open source. So in this project, in the span of eight days, we have over 30 contractors, I believe, and people in over 90 PRs. Now, because Thai people are so, they're really passionate about elections because of our current situation. So you can see why there's an influx of Thai people opening PRs. Oh, and either way, if you're interested, we have a goal mind of how we use several open source techniques, like password-driven developments and other methods to make open source really easy. So we were actually inspired by HackFall Public Good. So if there's anyone from GovTech here, thank you. I just copy your idea. So that's kidding. So we took the inspiration from HackFall Public Good from Singapore, and we came to a conclusion. Don't stop at prototypes, but really build open source. So how do you do that? This is where we try to come up with the idea of Bangkok open source, where it's not really hackathon that was just a lie. It's actually a two month program where we build two workshops plus three hack days. So it takes a lot more effort than your normal hackathon, but the results were really worth it. So we break it up into four sessions, two workshops, one open hack day, where everyone in Bangkok can join, and a final hackathon. So this was the first workshop. So we first started with getting people from different open source circles to meet because no one knows each order. And here's a couple of cool talks from that event. When we think of open source, we think of software, right? But this guy, he tried to be an open source community where everyone from multiple programs can adopt that idea. And as you can see, most meetups are whole, like in a conference hall like this, where it's really nice and fairly, but we actually held it in the middle of a barn. And we have the whole village making their own food and producing drinks. So it turns out it's a really nice way to bring communities together. Okay, let's do one better. How about an open source house? So most of the house plan in our world are closed source, right? But if you're trying to build better communities, you've got to make it open source. So this is why the team from Muse Foundation, they try to build a whole village where every house is open source. So everything has a CAD model where you can do printing, you can do the very precise operations. Another talk is about a system in Thailand called Trafi Fung Diu. So I'm not sure if Singaporeans have something similar, but if you see something wrong, like there's a flood or there's the lights doesn't work, you can report it and it floats into a database. The nice thing is that the data is completely open source. There's an API, so we took some data analysis so we can know what problem we can focus on. Another talk is about CWIC tech. So in Thailand, we do a lot of data storytelling and data visualization because people have difficult time understanding data. So that's why we use data a lot in CWIC tech. So you can look at the YouTube channel if you want to see these kind of videos. So now let's look at workshop two. At workshop two, we try to get people to brainstorm ideas, not just me, she's shorter, and actually try to come up with prototypes. So we have speakers from the domain site and I think this is very important. Many hackathons, they only have hackers, developers like us, but they don't have people from government, from the citizens, from community that actually knows that. But of course we also have people who are good at tech to give advice into how to structure the rate post. So a big part of our success was the Bangkok Open Source Discord because even if the hackathon is really five days, but it's actually two months. So most of the teams, they were actually on Discord night and day. This team called WeSpace. The funny thing is the hackathon has been long over, but there are no signs of stopping. So this team, it's just like done another satellite programming. So you can see like the community is getting up very fast and there's already at 125 people in the Discord. Now, when it's a hackathon, many times there's clear competition, right? You have to get first place, second place, but we scrap off that because if in Open Source, there can't be an internal competition, you have to cooperate. So the goal is you can join any team, you can leave anytime and there's no first to third place, but we come up with something a little different. We also did an open hack day. So this is not only limited to people who are in the hackathon, but let's say if you're someone who's, for example, we got a guy who was an abarist. So an abarist is someone who work with trees. So he comes into the project and he tells the guy like how like tree management works. And it turns out to be really good because people can come together and work. And fast forward to the hackathon day. So in most hackathons, like the hackathon vibe is very stressful, right? You don't really get to sleep and everything is tense, but because they have been working on it for already two months, it's not really stressful because like it's mostly finishing up the project and trying to make sure everything's polished. So we have three judges from different sites. So we have someone from CWIC tech and someone from growth tech and from the industrial side. Now this is the fun part, the awar because usually we have first to third place, but I think that is very anti-open source. So we actually make three kinds of awards. Contributability, which is how easy it is to open a PR or contribute something. Concept, which is how like how good the concept is and implementation, which is how good the software works. It can't be a prototype. It has to work. So there is eight group as you can see here and I'll quickly go over them. So here's a question, not a stupid one. Like how do you know which party to work for when you have so many political parties and all policies look similar, you use AI and you use natural language. So there are like, I think the students are 13 to 15 year olds so they're assuming they're like middle school, high school but the really cool thing is they wrote their own like text clustering algorithm to try to cluster the policy. How do you know where to buy land? There's so many things you have to think of in buying land, but they actually wrote a software that would incorporate multiple data source. For example, is it near the airport? How much are the income of the citizen near there? Is it near the sea? So they can make better decisions for business and for buying the land. And again, these students, there are 15 to 18 years old but they're already building stuff for people of Bangkok which is really cool. And as you can see here, it's not mock data, it's actually live. Another, this is the fun one. How do you make money while making lives better? Well, you report people who doesn't follow the law and get money. So there are people who like park their motorbikes in like the walkways and in Bangkok you see this every day and you're really angry at them. Like why do you park there? You cannot walk. So they have a channel where you can actually submit this photo and get money but we're like, wait, we have AI. We can use the YOLO model to basically do like segmentation and we know like, whoa, this is the bike. We can report them to the police. And so we did that. So they take the data from all the CCTV and auto report to the police and you can probably make like a hundred thousand bucks from just doing that. So yeah, this is how you get rich from open source. And how do you know changes are actually being made? Because like so many things are happening in Bangkok. So my friend Tai, he wrote a tweet on Twitter board that aggregate the data from traffic which is the system for reporting problem. And he posts a summary every day on Twitter the before and the after. And this is really nice because people can actually criticize the government with data so they can reply on the tweet, hey, why is my light pole not being fixed? That's not good. All right, what about building tools? So this woman, she's a bonus. So she's trying to build a tool to break all the silos in Thailand because in Thailand you need 10 signatures to do anything. So she wrote a tool for everyone to connect. Peer sharing is a nice one. So this is a software where people used to do peer sharing in SHAP groups and they're basically losing a lot of money because of people are cheating. So they write a system so that it's impossible to cheat anymore. Okay, we have three more minutes. So let's talk about the award winners. So this one is the concept award winner which is called Rasa Kicha. Rasa Kicha is government cassette. So basically it's the messages from the government but right now in Thailand it's all in PDF and it's not properly digitized. So they try to digitize this together and the next one is respace. So this is the best one. They try to digitize all the trees in Bangkok because right now they are actually paying the government a lot. I lost that slide. But basically right now they are paying 180,000 bucks in order to digitize 40,000 trees. So you can see we're wasting 100,000 maybe million bucks of money paying for proprietary software. And it would take them 300 years to digitize all the trees in Bangkok and I'll be dead long before that. So the nice thing is they use data from Street View and Aerobu. They use a lot of AI like deforestation and tree detection, also segmentation and they submit all that data to OpenStreetMap. So you can see this is how people are actually building with open source and there are people who are not only developers but others who are working on that. And you can see we get so many pull requests over 110 comments. And we also get so many contributors. So a quick special thanks to Konmishari here from OpenTech because he has been helping us so much with like providing ideas for the hackathon and also thank you for everyone who sponsored. So quickly what we learned. First, in order to get people to like come collaborate to open source you must get them motivated. So this might be building something silly. Next is keeping the momentum. How do you keep people engaged? And finally to facilitate the adoption because an open source software that no one uses isn't really open source. So you need someone, you need to get them adopted by people. And finally, this is the second governor of Bangkok. So he was taking the photo for us. So it's not only students building but also the governor of Bangkok. So the next step is we're keeping up that momentum. So we're building a next hackathon for the upcoming election where we can catch if the hackathon, sorry, the election is being done transparently. And that's it for me. Thank you.