Homie probably went "Claude, rewrite this in rust" and left. Bit off too much. Dont listen to the AI companies, these things cannot be on their own.
post
Jarred was already writing slop well before he had access to LLMs.Oof, that explains everything
Meanwhile, we have an entire ecosystem of JS tools being written in Rust that are absolutely amazing, including oxc, rolldown, biome, and so on. They're not awesome because they're written in Rust though, but because actual thought and effort was put into making them great. They could have been written in any language (see: eslint, prettier, vite, tsgo, and so on).
Bun was never behind for being written in Zig, but was actually pushed forward by the language to accomplish what would have been far more difficult to do in C.
What killed it was when Bun stopped committing human effort in favor of mass-produced slop. Where I used to recommend Bun, I now just recommend pnpm/Node.js.
As an aside, I've been using Boa as an embedded runtime recently, and I've been loving it. It doesn't have the performance of other options, but it's good enough™ and still has room to improve.
Where I used to recommend Bun, I now just recommend pnpm/Node.js.
Do you have an assessment or opinion on Deno?
Not really, no. From what I've seen, I believe it's based on V8? Other than that, nothing really to say except that the imports look interesting.
I generally target the web, and I haven't had a need for Deno. I could technically use it in my current personal project, but I wanted to give Boa a try rather than trying to figure out how to use Deno's bindings while navigating the limited documentation for them.
Looks like Bun 1.4 is out now: https://bun.com/blog/bun-v1.4
Good news: Looks like Bun's competitive with Deno again on node support.
Bad news: Bun still has poor security compared to Deno.
Wait and see news: Is bun's switch to rust going to be worth it over the long haul, or should they have worked on making their code better, or used a mix of languages where it'd make best sense to do so?
I am sure the project will be dead in a year
In the past month, 15.8k commits came from "robobun," 1.6k commits from "autofix-ci[bot]," and 790 commits from Jarred.
No chance anybody is reviewing those. Worse, nobody else can reasonably inspect or review changes. I suspect even from release to release, you're looking basically at a new codebase, the next release invalidating all your previous assessment, and release to release assessment to be impossible.
16/30 = 533 per day. I don't know when 1.3 was, but even just the three month delay means 48k commits. Who knows how many changes per commit. Crazy numbers.
It’s worth noting that in the ~2 weeks since this was published, Bun 1.4 was actually released. I don’t know how the release has gone.
Well considering the Zed Zig lang devs have quietly distanced themselves from the Bun maintainer due to, as the lang devs describe it, “cowboy bullshit and vibe coding”, I’m not surprised that vibe coding in another language isn’t going well either. Note that this project used to be a poster child for Zed Zig lang, but is now (through no fault of Zed Zig lang or the lang devs) effectively divorcing from it.
Edit: derp, lol
I’m an AI coding minimalist…like it for some tasks like creating unit tests and other tedium.
The bun saga of having AI do everything has been a fascinating horror show. Excited to see where it goes.
What was so special about Bun (at least the earlier version)? If it's so special, would someone fork it?
It was a replacement for node.js and npm that was supposed to be faster and had other things built in like minifying and bundling iirc.
Did they deliver? Was it worth the hype?
I've heard about them from time to time. Been rather curious myself, but never really had the chance to use it.
One nice thing about bun was/is that you can feed it TypeScript files directly, to run them locally. No need to transpile first. Much like deno.
I tried it out for a few hobby side projects as someone that doesn’t do much JavaScript development.
It delivered on the dev experience. I didn’t have to learn how to plug a bunch of random tools together because bun had a lot of the basic stuff just bundled in.
Task runner, test runner, typescript, etc.
I think if I spent more time doing JavaScript development work it wouldn’t have really mattered, that project bootstrapping is a one time cost and if you know how to do it it’s cheap enough. As someone that spends most of their time writing other languages though it was nicer than figuring out how to get a bunch of tools to play nice with each other.
I heard of people using it on hobby projects.
Don’t know of any firm using it in production. Node.js never really had scale issues that needed addressing.
I’ve moved more towards deno recently. No real desire for bun on my end

no comments (yet)