Probably. But at least it's useful and runs locally. Better than all the predatory apps that you have to download 100mbs to do something this website can do instantly without downloading anything or sending anything to a server.
I'm amazed by how in-depth the documentation is. Usually projects struggle to get decent docs. (Unless the docs are LLM generated). I'm not sure what's better, the design or the docs.
Like you said in your backstory, it's an awful shame to work so hard on something and have it get absolutely no attention.
I haven't tried out any of the code yet, but I'm curious enough to maybe try it sometime, especially since it has a robust FFI already. I had just uninstalled rust because I'm more of an old school C guy and don't like writing "safe" code.
Thank you. I want to be honest about the documentation: I did use AI to help write and organize parts of it.
I'm Thai, and I'm not fluent enough in English to explain a project this large naturally on my own, especially because I wanted to publish it for an English-speaking audience on Hacker News. I mainly used AI to help translate what I meant, improve the wording, and arrange the information into a structure that would feel normal and readable in English.
The language design, implementation, technical decisions, measurements, and the ideas behind the documentation are mine. I reviewed the text and tried to make sure it represented what X actually does, but I honestly don't always know what wording or document structure feels most natural to native English speakers.
I also tried hard to keep it from sounding strange or overly generated, so I'm glad that the documentation was useful to you.
So you used rust to write a better rust?
I'm just kidding. Looks like a real interesting language with some good features without being verbose or ugly.
The backstory is interesting, though I wish it covered more the rationale behind the language and the thought-process that went into its design.
I especially like that track covers everything from switch statements to if-else statements in a nice concise way.
Curious if the 1 based arrays get confusing in @low mode.
As far as the benchmarks go, I didn't see, was there any other languages compared?
I really love creating technology. Whether it is a game, a program, or something completely new, creating things is something I genuinely enjoy.
At some point, I began thinking that I wanted to create one programming language that could do everything. It was something I had wanted to make for a long time. I wanted it to be safe, flexible, simple, and fast.
I did not want to keep changing languages whenever I moved to a different part of a system. I did not want to constantly switch my way of thinking between different languages, syntax, and models. I wanted to be able to build everything within one language.
For the benchmarks, I intentionally chose not to compare X with other languages. I want X to be X, and I want the results to represent only what X actually produced. I do not want the benchmarks to be used for boasting that X is better than another language, or for speaking negatively about other languages. I want X to exist as its own language without competing with anyone. Readers can decide for themselves whether they want to compare the results.
For the names and syntax, my main goal was to make them as easy as possible. Even if a beginner knows almost nothing about programming, I want them to feel that the language is understandable, and to be able to guess what something means quickly.
I think that creating too many different syntax forms for closely related ideas forces people to remember too much. Reducing a group of related operations into one consistent form can reduce how much the user has to memorize.
More generally, I try not to create a new syntax form for every small variation. I think humans remember patterns more naturally than they remember many separate words and rules about exactly when each one must be used. A familiar pattern can continue to feel natural and simple even when the behavior behind it is still detailed and powerful.
The simplest summary of my design process is that I kept asking what people actually need, and thinking about the problems I experienced when I first started learning programming.
Very interesting and polished. I see that it's got a regex engine written in Pilang as well. I'm curious if there's any benchmarks or comparisons. I'd like to see how big the runtime is, etc.
You can also get really small file sizes by ignoring frameworks and porting unsupported codecs to wasm.
I ported jbig2 (17kb uncompressed) and codec2 (60kb uncompressed) to wasm which enables me to use really small image and audio files in my web app. I also made a custom read only database and search engine with built-in zstd decompressor (39kb uncompressed). It probably wouldn't run on a psp though.
I like optimizing and making things small. I want to use neural audio codecs for even better compression but the model sizes and compute complexity are major hurdles and muddy the vision.
I hope the image and audio bandwidth savings are more than the cost of having to send down the codec JS payload - might be worth it for a web app with a lot of image/audio loading
I was so disappointed that Ghostty doesn't properly render Khmer text. Abugidas are important and you have to be able to render the symbols non-linearly. Cosmic term is the only terminal I've seen that actually works. But it's a bit slow on my 14 year old laptop.
Kitty doesn't work, alacrity doesn't work, foot doesn't work, gnome terminal doesn't work, xfce terminal doesn't work, urxvt doesn't work, xterm doesn't work, the list goes on.
Oop is just Unix philosophy misapplied. Unix philosophy is that each program does one thing and you chain them together. In OOP you have each class do one thing, but at that scale it can become very messy.
reply