God I wish Apple offered first party support for Linux on Mac computers.

Then realized Rust wasn't that hard.

The benchmark claim is based on a naive, simple "hello world" GET req/response, comparing the Go stdlib vs the equivalent handler using uhttp. I used oha to measure.

https://github.com/alshdavid/uhttp/blob/main/benchmarks/go/s...

vs

https://github.com/alshdavid/uhttp/blob/main/examples/basic/...

Output: https://github.com/alshdavid/uhttp/blob/main/crates/uhttp/RE...

Often a meaningless benchmark, but it does serve to measure the overhead from the http library itself. Happy to hear suggestions on the benchmark approach as I don't want to be misleading

Comments URL: https://news.ycombinator.com/item?id=47635731

Points: 5

# Comments: 2

If a human reads GPL code and outputs a recreation of that code (derivative work) using what they learned - that is illegal.

If an AI reads GPL code and outputs a recreation of that code using what it "learned" - it's not illegal?

If that is the case, then copyright holds no weight any more. I should be allowed to train an LLM on decompiled firmware (say, Playstation, Switch, iPhone) in countries where decompilation is legal - then have the LLM produce equivalent firmware that I later use to build an emulator (or competing open source firmware).

The first thing to highlight is that multithreading is possible but extremely impractical on the web, and even if you succeed, it's slow. Check out Surma's video on it: https://www.youtube.com/watch?v=7Rrv9qFMWNM

Using threads unlocks so much performance, from everything to time-to-first-paint, time-to-interactive all the way to the general feel of the application.

That sort of optimization is trivial in languages designed for it (like Rust or Go).

Developer experience on the web is unbelievably poor. TypeScript is, semantically, fantastic, but then you have to build it - do you use swc or tsc, do you use webpack, rollup, rspack, vercel's stack, etc? Once you build it, what are your transpilation targets - did you even configure them or are you building bundles that are 40% larger for no reason? The runtime overhead of bundles is also non trivial.

Languages like Rust are maybe 20% more complicated than TypeScript - but they have TSX (also Vue templates, Angular templates, whatever) baked into the _core_ language - requiring no external transpilers or LSPs.

That doesn't mean every application is a good candidate for native/wasm - but the applications that would benefit would benefit end users and developers immensely.

The push back really makes no sense to me. Everyone complains about Electron (and OP about Linkedin, and everyone about Jira) being slow or eating up half their ram - yet with the same breath say that wasm would have no benefit.

> The ease in reverse engineering JavaScript

It's not as easy to reverse engineer native binaries - but you can do it pretty easily and produce the equivalent of minified source code from it. This is how emulators and custom firmware is created. It wouldn't prevent security researchers from identifying negligence.

2) It's not just me, all of the apps listed have the same issues. I know that in addition to Atlassian, Meta and Canva have sunk millions of dollars into writing custom bundlers.

2) wasm makes serving static content practical as server side rendering is economical. My previous employer, for instance, spends tens of millions of dollars every year running SSR servers - almost all of that would be eliminated if the backend could just run the client as wasm.

3) Scrapers (think puppeteer) would be faster and more resource efficient because they wouldn't need to start a JavaScript runtime to load a page.

4) You don't use Electron apps?

5) You may not, but everyone else uses rich interactive web applications. Think of the energy usage and cost savings there would be to the world.

Most trivial apps don't need to be optimized, and for them, JavaScript is fine.

But for complex interactive web applications - it really fuggin matters.

Think; - vscode

- facebook

- jira

- linkedin

- reddit

There's no reason these applications should be slow, single threaded, and consume gigabytes of memory - but that's a limitation of the technology.

I know first hand that Atlassian has spent millions of dollars building bundlers in different forms just to save a few milliseconds of load time.

Just let me write the front end in Rust and if the browser detects that no JavaScript is running - don't start a JavaScript engine.

While you're at it, improve the SharedWorker story so I can effectively share data between tabs (enables cross tab sync, great for chat apps and local caching). I recently tried to make an offline-only application with a wasm-based sqlite implementation in a SharedWorker and the API just doesn't work.

Yeah Atlassian. 1/3rd of my team were given the boot sadly. One guy had 12 years at the company - crazy times

But yes, that's very expensive and surprising to me.

Take a litigious company like Nintendo. If one was to train an LLM on their works and the LLM produces an emulator, that would force a lawsuit.

If Nintendo wins, then LLMs are stealing. If Nintendo loses, then we can decompile everything.

Are you saying that you believe that untested but technically; models trained on GPL sources need to distribute the resulting LLMs under GPL?

The thing that leaves a bad taste in my mouth is the fact that my works were likely included in the training data and, if it doesn't violate my licenses (GNU 2/3), it certainly feels against the spirit of what I intended when distributing my works.

I was made redundant recently "due to AI" (questionable) and it feels like my works in some way contributed to my redundancy where my works contributed to the profits made by these AI megacorps while I am left a victim.

I wish I could be provided a dividend or royalty, however small, for my contribution to these LLMs but that will never happen.

I've been looking for a copy-left "source available" license that allows me to distribute code openly but has a clause that says "if you would like to use these sources to train an LLM, please contact me and we'll work something out". I haven't yet found that.

I'm guessing that such a license would not be enforceable because I am not in the US, but at least it would be nice to declare my intent and who knows what the future looks like.

I find it funny that the headline is "JavaSCript is enough" - yet this is a compiler on top of JavaScript that introduces magic and behavioural changes to syntax. How well does this work with testing frameworks? Can this run without the compiler?

A lot of the thinking behind this compiler comes out of the box with Rust. If only wasm worked.

MacOS is slightly better but Apple's anti-competitive practices targeting developers makes it a tough sell. Linux is better still, technically, but due to the fragmented desktop environment landscape and distribution difficulties, it's also a hard sell.

I think it's fair to say that the entire story for native app development is a mess - which is most likely why Electron became so popular. There just isn't a practical sane alternative.