Seems fun! It can take the ball a long time to reach a stop, so maybe either skip ahead or use a friction coefficient that is higher when it is rolling?
This is great and I’d love to hear more about what worked / what didn’t / how you achieved these results.
A couple of months ago I worked with my 9 year old to have Claude build a little game with threejs. We’ve now got three levels with characters shooting silly string and banana guns. It was really fun to see him imagine something, have it show up on the screen, and iterate it.
Anyway, I tried building an LLM-backed workflow that could guide kids through the game creation process and kids can see their idea come to like (think: “sparkly purple unicorn shoots stars at a dragon”). I couldn’t _quite_ get it to work like I wanted so I shelved it pending future ideas/improved models.
Some background. I'm a 20-year software full-stack developer (started with Ruby on Rails, then moved to Node.js. These days doing a lot of Next.js/React, and some Svelte in my spare time) - as well as Astro.
So, last year after finishing a contract at Volvo Cars, I took some downtime and dabbled with learning Svelte, and managed to recreate a silly little prank I made when I was working at a web agency called New Bamboo many years ago. That little prank then got iterated on, and is now the music editor app at https://lets-make-sweet-music.com.
After that, I created a simple ball-table game using Svelte and Threlte, which is a wrapper around the excellent ThreeJS library. Once I realised that Threlte had support for Rapier physics, I realised that I could make a game, so I made this: https://3d-garden.vercel.app
Usually when you hit a blocking issue, you'd google around and probably end up on Stack Overflow. But last year I increasingly found myself using ChatGPT (I have a $20pm subscription) to ask questions and copy/paste snippets of code for suggestions on how to resolve a blocker.
That process turned out to yield some good results, so I was able to iterate on my ideas and get quite far. My only real limitation is that I have a tendency to lose interest in projects and end up jumping from one project to the next. I also started looking at making a LLM-bot interface to try and create a cyclical loop of bots making other bots do things and implementing the feedback loop.
I also managed to implement an isometric game engine based on an example I'd made using Löve2D (a game engine in Lua), but this time written in TypeScript and using HTML5 canvas. I managed to work on that for quite a bit, but as usual ran out of passion.
In theory, if I can resurrect my passion on it, then I could get it to the stage where it would be possible to play a SimCity-2000/3000 like game entirely in the web browser.
Based on the work across those side projects, I managed to extract out a library for emitting events http://github.com/anephenix/event-emitter - the purpose being to easily decouple Svelte components and make the logic within those components more modular.
I took up the opportunity to work with a former boss at an e-commerce company based in Sweden, so I recently moved here. It was during that time that I learned that the team had been dabbling with Claude Code, which I hadn't used at all.
I then decided one weekend to have a go at trying to build a pixel editor for Babsland, but rather than do it by hand, I decided to try out Claude Code with the Zed Editor (I needed to switch from VS Code because my apartment didn't have electricity for the 1st week, so I relied on charging devices at work and a great hotel - Scandic Centralen in Gothenburg). I realised that VS Code being an Electron app (after all I wrote a book on Electron and Nw.js) was using a lot of battery, so I switched to Zed.
In a couple of hours at the hotel, I managed to build the beginnings of a pixel editor - which is now https://www.babsp...
/skill:make_hn_frontpage_submission create a golf game the domain is the-golf-is-golfing.com deploy using <...> use svelte.
I don't know, it just feels so low effort when it's just "look what AI made", not a writeup of how the golf game was made, problems experienced, back and forth needed.
I know this works because I have a real /skill:create_website command that does just that, except that it only has to create dockerfile + push + kustomize apply and the domain is automatically taken care of
The choice to make a golf game with AI seems a bit ironic. Getting a project onto the green seems tantalizingly close, doesn't it?
Making games people actually want to play is hard with or without AI and developer skill has little to do with it. Even the most pro-AI individual has to see the long tail of tedious work left here. At least on mobile, the controls are floaty and imprecise. The camera is at a strange angle making it hard to see what you're doing and made even worse by the HUD in the way. There's only one course: a straight. The course itself looks glitched out. The hole seems out of proportion to the ball. This is just my initial impression after 5 minutes. There's likely a lot more.
What's worse is that to really fix these issues, a human at some point has to comprehend and undo a lot more than they bargained for.
One would hope we're finally done with the dismissive arguments that "AI will only keep getting better" or that these are just nitpicky refinements. The remaining work is the most time consuming, and even when finished the result will just be mediocre. Mediocrity isn't mere incompetence, but being asleep at the wheel when the ideas that are foundational to a project are being made. Any statistical model, such as generative AI, is wholly concerned with broad brush strokes.
I do see some value in using AI to find (rough) examples of code that wouldn't exist otherwise, but I wonder how much this further limits the creativity of people who are already stifled by media overconsumption and already conditioned to overlook details.
Programming is hard, even for someone who has an interest in it. i know some c# just for selenium work- honestly im quite terrible and have too little concentration or innate skill to get anywhere. i’ve made attempts but they fizzle. Those barriers of esoteric knowledge are folding like card castles now. would it be helpful to understand perfectly the underlying principles, heck yeah, but i want to see my visions for different programs come to life! apps that i believe don’t exist. I sincerely don’t mean it as jab to the author but programmers tend to be rigid thinkers and it’s totally practical to do PoC on golf but yeah, rigid thinking doesn’t always result in novelty. let’s let things play out and see what happens when this sort of capability reaches other creatives/types of people. and as the tools strengthen and expand.
24 comments
[ 4.1 ms ] story [ 49.4 ms ] threadIt's at https://adamtaylor13.github.io/botnet/ if anyone wants to try it.
It's still pretty rough compared to the original but, it's been fun enough for my kids to play!
Sorry but can't wait until this whole AI situation finds a proper balance.
Fun bit, if you hit the ball against the billboard on desktop, it will open the link for the billboard advert and then bounce off.
A couple of months ago I worked with my 9 year old to have Claude build a little game with threejs. We’ve now got three levels with characters shooting silly string and banana guns. It was really fun to see him imagine something, have it show up on the screen, and iterate it.
Anyway, I tried building an LLM-backed workflow that could guide kids through the game creation process and kids can see their idea come to like (think: “sparkly purple unicorn shoots stars at a dragon”). I couldn’t _quite_ get it to work like I wanted so I shelved it pending future ideas/improved models.
Some background. I'm a 20-year software full-stack developer (started with Ruby on Rails, then moved to Node.js. These days doing a lot of Next.js/React, and some Svelte in my spare time) - as well as Astro.
I'm the author of Manning's "Cross Platform Desktop Applications" - http://manning.com/jensen.
And this is my GitHub profile: https://github.com/paulbjensen
I run a 1-man web product studio: https://anephenix.com
So, last year after finishing a contract at Volvo Cars, I took some downtime and dabbled with learning Svelte, and managed to recreate a silly little prank I made when I was working at a web agency called New Bamboo many years ago. That little prank then got iterated on, and is now the music editor app at https://lets-make-sweet-music.com.
After that, I created a simple ball-table game using Svelte and Threlte, which is a wrapper around the excellent ThreeJS library. Once I realised that Threlte had support for Rapier physics, I realised that I could make a game, so I made this: https://3d-garden.vercel.app
Usually when you hit a blocking issue, you'd google around and probably end up on Stack Overflow. But last year I increasingly found myself using ChatGPT (I have a $20pm subscription) to ask questions and copy/paste snippets of code for suggestions on how to resolve a blocker.
That process turned out to yield some good results, so I was able to iterate on my ideas and get quite far. My only real limitation is that I have a tendency to lose interest in projects and end up jumping from one project to the next. I also started looking at making a LLM-bot interface to try and create a cyclical loop of bots making other bots do things and implementing the feedback loop.
I also managed to implement an isometric game engine based on an example I'd made using Löve2D (a game engine in Lua), but this time written in TypeScript and using HTML5 canvas. I managed to work on that for quite a bit, but as usual ran out of passion.
In theory, if I can resurrect my passion on it, then I could get it to the stage where it would be possible to play a SimCity-2000/3000 like game entirely in the web browser.
You can see it here http://babsland.com
Based on the work across those side projects, I managed to extract out a library for emitting events http://github.com/anephenix/event-emitter - the purpose being to easily decouple Svelte components and make the logic within those components more modular.
I took up the opportunity to work with a former boss at an e-commerce company based in Sweden, so I recently moved here. It was during that time that I learned that the team had been dabbling with Claude Code, which I hadn't used at all.
I then decided one weekend to have a go at trying to build a pixel editor for Babsland, but rather than do it by hand, I decided to try out Claude Code with the Zed Editor (I needed to switch from VS Code because my apartment didn't have electricity for the 1st week, so I relied on charging devices at work and a great hotel - Scandic Centralen in Gothenburg). I realised that VS Code being an Electron app (after all I wrote a book on Electron and Nw.js) was using a lot of battery, so I switched to Zed.
In a couple of hours at the hotel, I managed to build the beginnings of a pixel editor - which is now https://www.babsp...
I'm going to play with Svelte for some side projects. I've seen it mentioned a few times and I've been meaning to give it a test drive.
I don't know, it just feels so low effort when it's just "look what AI made", not a writeup of how the golf game was made, problems experienced, back and forth needed.
I know this works because I have a real /skill:create_website command that does just that, except that it only has to create dockerfile + push + kustomize apply and the domain is automatically taken care of
Making games people actually want to play is hard with or without AI and developer skill has little to do with it. Even the most pro-AI individual has to see the long tail of tedious work left here. At least on mobile, the controls are floaty and imprecise. The camera is at a strange angle making it hard to see what you're doing and made even worse by the HUD in the way. There's only one course: a straight. The course itself looks glitched out. The hole seems out of proportion to the ball. This is just my initial impression after 5 minutes. There's likely a lot more.
What's worse is that to really fix these issues, a human at some point has to comprehend and undo a lot more than they bargained for.
One would hope we're finally done with the dismissive arguments that "AI will only keep getting better" or that these are just nitpicky refinements. The remaining work is the most time consuming, and even when finished the result will just be mediocre. Mediocrity isn't mere incompetence, but being asleep at the wheel when the ideas that are foundational to a project are being made. Any statistical model, such as generative AI, is wholly concerned with broad brush strokes.
I do see some value in using AI to find (rough) examples of code that wouldn't exist otherwise, but I wonder how much this further limits the creativity of people who are already stifled by media overconsumption and already conditioned to overlook details.
Perhaps this isn't the place to post demos of what is possible with tech nowadays.