Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

raymyers.org

Post reply on HN
Print view View original

331–340 of 862 posts

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#331

Post by piokoch »

RetroTechie wrote: post #97

There's so much good stuff in this post. Can't help to think of a recent HN post about most AI-generated projects being abandoned within months. Why? Because value of a project is not in the code produced. It's in the amount of battle-testing that code has seen. Battle-tested, mature code > fresh rewrite. Existing Zig codebase has seen X amount of battle-testing. Rust rewrite: 0 (except -I'm assuming- passing test su…

"How is that better than the Zig codebase you started with?" - It will be worst, as this will not be idiomatic Rust. That's kind of interesting, BTW, as in next iteration LLM will be trained on tones of crappy code, created as some random rewrites, AI slops, etc., I am curious if someone will be able to curate this or it will be the same process of crapification experienced by Google Search that finally lost the battle with SEO spammers.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#332

Post by cactusplant7374 »

RetroTechie wrote: post #97

There's so much good stuff in this post. Can't help to think of a recent HN post about most AI-generated projects being abandoned within months. Why? Because value of a project is not in the code produced. It's in the amount of battle-testing that code has seen. Battle-tested, mature code > fresh rewrite. Existing Zig codebase has seen X amount of battle-testing. Rust rewrite: 0 (except -I'm assuming- passing test su…

> Because value of a project is not in the code produced. It's in the amount of battle-testing that code has seen.

It is in the marketing. Most software engineers are not skilled at marketing, do not have a large audience, and are not willing to spend money on advertising. It has nothing to do with code quality.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#333

Post by krupan »

small_model wrote: post #317
krupan wrote: post #313

Earlier quoted context omitted.

I love how many people are tech prophets now

Do think the majority of SWEs will be hand cranking out code like it's 1999 in a few years?

No, because nobody has been coding the same way since 1999. But that has nothing to do with LLMs

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#334

Post by Gigachad »

whizzter wrote: post #126
vlaaad wrote: post #3

Did we read the same Anthropic and Andrew Kelly's posts? Anthropic is not in the programming language market; their post about rewriting Bun in Rust is full of technical details that led to improving the end product for their users. Zig's response is a sour opinion piece full of personal attacks. For context, I'm using Codex and have no interest in either Zig or Rust, so just observing this drama from the sidelines.

It's the old craftsmen vs industry issue, Andrew comes from the craftsman tradition that prefers all other people developing to also be proud craftsmen. What Java, JS, Python and C# all did to conquer the industry from a C++ dominance was to provide safety harnesses for less "perfectionist" workers to fumble around without causing a mess, to write C and C++ in an increasingly hostile world we realized you needed a lo…

All I’ve seen is there is literally no programmer smart and careful enough to never create a use after free or out of bounds read in a sufficiently complex codebase.

The state of computer security has moved on from the old model of just patching bugs when you find them. To now where we need to systematically prevent them from happening to begin with.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#335

Post by nivethan »

ramses0 wrote: post #303
TeriyakiBomb wrote: post #244

Earlier quoted context omitted.

A drum I've been banging increasingly often recently is that having friction and time to work ideas over in your mind adds huge amounts of value. Vibe coded projects have this very specific, well, vibe to them where you can clearly see that the lack of time to digest has allowed the person to not challenge their own worst impulses. You can see it in the feature bloat, the lack of depth and polish in core features and…

Fuben-Eki! https://news.ycombinator.com/item?id=37169679

thanks for that link, I've thought that friction was helpful but this gives me a word and a concept to look at.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#336

Post by pseudony »

petcat wrote: post #133
RetroTechie wrote: post #97

There's so much good stuff in this post. Can't help to think of a recent HN post about most AI-generated projects being abandoned within months. Why? Because value of a project is not in the code produced. It's in the amount of battle-testing that code has seen. Battle-tested, mature code > fresh rewrite. Existing Zig codebase has seen X amount of battle-testing. Rust rewrite: 0 (except -I'm assuming- passing test su…

> Now if that's further migrated to safe Rust, put into production & gathered feedback from lots of users, yes then you have something. Obviously they have to start somewhere if they want to get to safe rust with a considerable degree of battle testing. So they decided to start with just a transliteration and go from there. I think the Zig people are really just concerned that maybe Zig itself is a DOA language becau…

I program C for my day-job.

I see Rust encroaching in proposed transitions. It may even happen. That said, it is a poor match for it compared to something like Zig (or Odin). It's hard to make the new Rust code use existing allocator abstractions (so now you have two systems doling out memory, how do you reliably free composite objects with memory from both? How do they share?) and you increasingly have to either abandon any actual benefits of the borrow-checker, or invest increasingly heavily into sufficiently fat bindings to wrap your existing C/C++ in a way where the borrow-checker can assist you. That's before we consider the complexity of the language - I'd doubt a seasoned C programmer has much trouble deciphering Zig or Odin FFI bindings, but in the case of Rust? Yes, there is real friction.

Also if you really value predictable- and higher performance, being in more in control of memory allocations and cleanup is preferable. This is the direction both Zig and Odin cater to.

If you asked me what solves the most issues without adding too many new liabilities, I'd say Zig (or Odin). It would simply be much, much easier to transition a C codebase to either, and either would bring a much improved stdlib with pluggable allocators capable of leak-detection etc.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#337

Post by 15155 »

Keyframe wrote: post #327

Sarcasm hat on - You know what, if LLMs were so infallible at coding and engineers useless for hands-on, then why even bother with anything except asm or even skip the mnemonics completely and go straight for the binary? Why even bother with abstract code when abstraction is now your LLM.

Because underneath the code you have more than one architecture to support.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#338

Post by hatefulheart »

joshuamorton wrote: post #309
hatefulheart wrote: post #231

Earlier quoted context omitted.

Python was invented two years before Java and didn’t move the needle until the mid 2010s. You are the typical mark for hype cycles. Get a clue will you.

Python was being used as a teaching language at flagship universities in the mid 2000s.

Amazing! So was Scheme.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#339

Post by 15155 »

Animats wrote: post #129

Oh, they're converting Bun to unsafe rust. That's easy, but useless. That's the sort of thing c2rust does - transpile to an low level language which is unsafe Rust with a set of functions that unsafely emulate C pointer semantics. You don't need an LLM for that. There's no point. It's not doing the job DARPA's TRACTOR program is trying to do - translate C to safe Rust. I've used c2rust. It works, but what you get out…

> That's easy, but useless

The Rust toolchain is a delight in comparison to Zig's, and the unsafe code only need be temporary until humans (or, gasp, AI) address each class of underlying issue.

This is a no-worse-off state to improve from.

Re: Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

#340

Post by woodruffw »

skilning wrote: post #324
woodruffw wrote: post #299

Earlier quoted context omitted.

I think we can quibble about the Bun post’s factual claims. But I don’t think the post was deceptive or dishonest in its claims. Like most technical writing, it represents a vantage point. As for ghosting in meetings: sure, that seems bad. I would also be upset if someone did that to me. But you can state that factually without making it into a personal attack. It would even be more convincing.

> Like most technical writing, I think you've missed the core thesis of the article. That post _wasn't_ technical writing. It was marketing disguised as technical writing.

It’s marketing that is also technical writing. Most companies do this, and IMO we should prefer this over the alternative (which is corporate blogspam).

Compare Cloudflare, the Google Chrome Security blog, etc.

Post reply on HN