Might it see a popularity burst though, due to Rust's inherent advantage in AI augmented development? Being more deterministic makes it easier for LLM's to produce working code in it.
There is not such an advancement, Rust is made for dealing with human's mistakes. Chatbots can stop having such a mistakes and generate more terse formal specs instead.
The question by @wookmaster isn't just valid, if we're to follow your argument, chatbots may as well skip C and write directly in assembly or machine code for maximum efficiency! You seem to put too much faith in chatbots that you forget that correcting AI mistakes in code is a major job now. Even the best coders use AI to only fill in obvious code, not to think on their behalf.
This submission doesn't mention the publish time either, and it uses the link that hides a bunch of replies that debunk the article. So I'm just going to repost my comment from the last submission.
> If you're going to post a click and rage bait article, at least mention when it was published. We're close to the article's first anniversary. A lot has changed in between and the premise of this story isn't even valid or relevant anymore for it to be on HN again. And that's a charitable interpretation, given that the 3 cases that the article presents to argue that Rust was 'rejected by major projects' need special scrutiny.
> But first, let's look at where Rust stands now. The Fish shell has been ported to Rust entirely. Git is experimenting with Rust integration. Linux has declared the Rust4Linux project out of experimental stage - which means that Rust code will more or less be a permanent part of Linux from now on. FreeBSD is experimenting with Rust in their kernel. Microsoft has declared their intention to replace all critical code with Rust in the future. A lot of JavaScript development tools are now in Rust. There's just too many other cases for me to continue like this. The 'downfall' part is a bit out of touch with reality here.
> Now let's look at the 3 cases that the author uses to dismiss Rust as a failure:
> 1. Daniel Stenberg's attempt to integrate the Rust Hyper backend into curl. He abandoned it because he found it too hard to integrate it into C. So the case chosen here is a small team attempting to combine code in two programming languages. Such attempts are always challenging, experimental and never guaranteed to succeed. Choosing such an uncertain task to evaluate Rust's capabilities is not a fair deal at all. Yet, they did manage to integrate two other Rust backends - rustls (for TLS) and uiche (for QUIC and HTTP/3). So perhaps they didn't fail at all? And Linux succeeded in the same task at a much bigger scale.
> 2. Prisma migrating its core logic from Rust to TypeScript. How is this an apples to apples comparison? The article itself says that it was because of the preference of a primarily typescript crowd. They wanted access to the core, but their lack of Rust knowledge made it difficult for them. So Rust was out of place there to begin with. Another inappropriate case for evaluating Rust.
> 3. Author Austin Stark's experience with Rust for his trading platform. This is probably the most egregious example. Austin wrote the article complaining about Rust in June 2024 - 7 months before this article. But he reversed his stance and praised Rust in December 2024 - a month before this article. The author just ignored the latter completely. I can't take this seriously.
> I'm an enthusiastic Rust programmer. But I really don't mind articles that are critical of Rust, as long as they shed some valuable insight on Rust or systems programming in general. But this one reads either like a lazy attempt at a hit on Rust or as a rage bait for clicks. I always avoid touching the personal side of such submissions, but I can't help notice this about @RustSupremacist who submitted this one. You've submitted 8 stories so far, including 5 dissing on Rust (including 3 that got flagged) and one dissing on Zig. Why not write an article on your beef with Rust and post it here, instead of reposting misleading baits?
Might it see a popularity burst though, due to Rust's inherent advantage in AI augmented development? Being more deterministic makes it easier for LLM's to produce working code in it.
There is not such an advancement, Rust is made for dealing with human's mistakes. Chatbots can stop having such a mistakes and generate more terse formal specs instead.
Aren’t they trained on human code?
What am I supposed to say, they were trained on Martians' code?
Do I need to agreed with your statement that chatbot will never be able to write a correct C code without stupid runtime boundaries?
The question by @wookmaster isn't just valid, if we're to follow your argument, chatbots may as well skip C and write directly in assembly or machine code for maximum efficiency! You seem to put too much faith in chatbots that you forget that correcting AI mistakes in code is a major job now. Even the best coders use AI to only fill in obvious code, not to think on their behalf.
Why does this misleading article from more than a year ago get posted repeatedly?
https://news.ycombinator.com/item?id=46510073
This submission doesn't mention the publish time either, and it uses the link that hides a bunch of replies that debunk the article. So I'm just going to repost my comment from the last submission.
> If you're going to post a click and rage bait article, at least mention when it was published. We're close to the article's first anniversary. A lot has changed in between and the premise of this story isn't even valid or relevant anymore for it to be on HN again. And that's a charitable interpretation, given that the 3 cases that the article presents to argue that Rust was 'rejected by major projects' need special scrutiny.
> But first, let's look at where Rust stands now. The Fish shell has been ported to Rust entirely. Git is experimenting with Rust integration. Linux has declared the Rust4Linux project out of experimental stage - which means that Rust code will more or less be a permanent part of Linux from now on. FreeBSD is experimenting with Rust in their kernel. Microsoft has declared their intention to replace all critical code with Rust in the future. A lot of JavaScript development tools are now in Rust. There's just too many other cases for me to continue like this. The 'downfall' part is a bit out of touch with reality here.
> Now let's look at the 3 cases that the author uses to dismiss Rust as a failure:
> 1. Daniel Stenberg's attempt to integrate the Rust Hyper backend into curl. He abandoned it because he found it too hard to integrate it into C. So the case chosen here is a small team attempting to combine code in two programming languages. Such attempts are always challenging, experimental and never guaranteed to succeed. Choosing such an uncertain task to evaluate Rust's capabilities is not a fair deal at all. Yet, they did manage to integrate two other Rust backends - rustls (for TLS) and uiche (for QUIC and HTTP/3). So perhaps they didn't fail at all? And Linux succeeded in the same task at a much bigger scale.
> 2. Prisma migrating its core logic from Rust to TypeScript. How is this an apples to apples comparison? The article itself says that it was because of the preference of a primarily typescript crowd. They wanted access to the core, but their lack of Rust knowledge made it difficult for them. So Rust was out of place there to begin with. Another inappropriate case for evaluating Rust.
> 3. Author Austin Stark's experience with Rust for his trading platform. This is probably the most egregious example. Austin wrote the article complaining about Rust in June 2024 - 7 months before this article. But he reversed his stance and praised Rust in December 2024 - a month before this article. The author just ignored the latter completely. I can't take this seriously.
> I'm an enthusiastic Rust programmer. But I really don't mind articles that are critical of Rust, as long as they shed some valuable insight on Rust or systems programming in general. But this one reads either like a lazy attempt at a hit on Rust or as a rage bait for clicks. I always avoid touching the personal side of such submissions, but I can't help notice this about @RustSupremacist who submitted this one. You've submitted 8 stories so far, including 5 dissing on Rust (including 3 that got flagged) and one dissing on Zig. Why not write an article on your beef with Rust and post it here, instead of reposting misleading baits?