"This post could have been a tweet", and so on.

It's a lovely metaphor for the soulful business of doing something really well, even if there is no 'market' for that kind of quality

Art is often its own reward.

I've been getting great results from Codex. Can be a bit slow, but gets there. Writes good Rust, powers through integration test generation.

So (again) we are just sharing anecdata

Heh,I read SF as San Francisco; point remains true. Except the Valley wants to force a future, not describe it

It's not inconsistent to say there's a financial bubble and also genuinely think it's a new era for software development.

There aren't enough programmers to justify the valuations and capex

These are a few well-chosen dependencies for a serious project.

Rust projects can really go bananas on dependencies, partly because it's so easy to include them

So the cognitive reality of being a billionaire is now available to everyone?

But this is one thing that Gen AI is genuinely good at, constructing computer programs under close human supervision. It's also the most profitable (but not enough to justify valuations) Also, it may be a big thing here but its pretty niche in the larger scheme of things

The article is about it encroaching in the domain of human communications. Mass adoption is the only way to justify the incredible financial promises.

I use Claude at least weekly to help write documents for me. And I’m a good writer, who spent a lot of time and energy getting that way. I have a friend who is a terrible writer who I do proofreading for. He uses chatgpt and it’s made a world of difference for him in getting things accomplished and communicating what he wants.

I think there are lots of valid arguments against llm usage, but it’s extremely tiring to here how it’s not useful when I get so much use out of it.

So much for the old thermonuclear ramjet idea....

This is how Option works with ? in Rust. But short-circuiting in an expression does make sense

Is this not more a Cargo thing? Cargo is obsessed with correct builds and eventually the file system fills up with old artifacts.

(I know, I have to declare Cargo bankruptcy every few weeks and do a full clean & rebuild)

A little of both. Incremental compilation cache is likely the single largest item in the target directory but it gets cleaned up on each invocation so it doesn't scale in size with time.

I believe the next release will have a cache GC but only for global caches (e.g. `.crate` files). I'd like us to at least cleanup the layout of the target directory so its easier to track stuff before GCing it. Work is underway for this. A cheap GC we could add earlier is for artifacts specific to older cargo versions.

Correct builds != never running garbage cleanup. I would settle for it evicting older variants of a build (I also dislike the random hash that’s impossible to determine what specifically is different between two hashes / which one is newer).

Automatic garbage collection of old build artifacts* is coming in Rust 1.88 (currently on the beta channel, will become the new stable release in two weeks):

https://github.com/rust-lang/cargo/issues/12633

*EDIT: For now this only deletes old cached downloads, not build artifacts. Thanks epage for the correction below.

That is not for build artifacts but global caches like for `.crate` files but its a stepping stone.

Oops, thanks for the correction.

That’s as an additional flag and not the default?

In versions earlier than 1.88, garbage collection required the unstable -Zgc flag (and a nightly toolchain). But in 1.88 and later, automatic garbage collection is enabled by default.