Hacker News Reader

Comment by HumanDrivenDev | Hacker News Reader

It sounds and seems like a language worthy of more investigation on my part, but I have to agree that the militant, presumptive attacks on any language besides their Chosen One has left a bad taste in my mouth.

This 100 times over. It's a language that seems novel and strong enough to stand on its own merits - but there's a section of the community that makes me want to stay away.

I wish they'd spend less effort re-writing already working tools in C and more effort writing new, better tools. A drop in replacement for "ls" isn't that exciting for me.

View on HN

Pretty sure that's already happening with tools like ripgrep: https://github.com/BurntSushi/ripgrep

On the other hand part of the complaint in the main article is the lack of basic core utils at a standard that OpenBSD would consider. Or things that could be a base world dependency for them.

But that ties into what I'm saying - trying to 'oxidise' a system written in C that already works.

Or to put it another way, find something new and valuable as your Trojan horse?

OpenBSD developers themselves did that, rewriting a lot of things that already worked. But they wanted to improve them security wise. So your argument doesn't fit with what they did already. Language is tangential here. Their main focus was security. So it's pretty reasonable to ask, why things can't be rewritten in Rust to improve security further.

Note, that the answer doesn't say that it can't benefit security, which would be a principal argument for OpenBSD. It brings side reasons as to why it's not easy to do (like integrated tools, developers' availability and so on).

For those who want the OpenBSD devs to abandon their many, many years of experience with C and instead adopt Rust, I think it's a fair question to ask the Rust folks why they haven't just started from scratch and built their own unix-like operating system. This seems to be a more reasonable approach that trying to force an existing operating system to adopt their language.

Oh wait, that's right, none of them has even gotten around to writing a replacement for ls, grep, .....

> I think it's a fair question to ask the Rust folks why they haven't just started from scratch and built their own unix-like operating system.

Some did: https://www.redox-os.org

> Oh wait, that's right, none of them has even gotten around to writing a replacement for ls, grep,

They did that for some utils: https://github.com/redox-os/coreutils

> This seems to be a more reasonable approach that trying to force an existing operating system to adopt their language.

I don't think anyone is forcing to adopt anything here. The question was about integrating languages like Rust in OpenBSD.

> For those who want the OpenBSD devs to abandon their many, many years of experience with C and instead adopt Rust

Who exactly? Can you actually point to anyone making this desire known in earnest?

> A drop in replacement for "ls" isn't that exciting for me.

And a good getting your feet wet project for someone else :)

> I wish they'd spend less effort re-writing already working tools in C and more effort writing new, better tools.

Is this really a thing? Folks keep pointing out this bugbear of folks going out and asking projects to rewrite in Rust and ... there aren't that many examples of this.

If anything most of the more famous Rust projects (ripgrep, redox, etc) are clean-slate.

https://github.com/uutils/coreutils for instance

Personally, I don't find this a bad idea, as long as it is not the primary thrust. The reason it is needed is that you aren't going to convince everyone to change the default tools they've used for however many years. If you want to make things safe, you have to swap out implementations and keep same/similar interfaces.

Yeah, that's my point, there are scattered examples but it's nobody's primary thrust.

http://transitiontech.ca/random/RIIR this phenomenon has been documented

Reply on news.ycombinator.com