A Linux maintainer admitting to attempting to sabotage Rust for Linux project
social.treehouse.systems
social.treehouse.systems
(I am a big proponent of language interop as an alternative to big rewrites. But opinions differ, and my opinion is worth nothing because I'm not a maintainer of the relevant code.)
passive aggressive threats (and that is exactly what they are) are not a technical discussion. you can be polite or blunt. you can be nasty or you can be good. but don't confuse polite for for good or blunt for nasty - some of the nastiest meanest behavior is packaged in a nice and polite delivery.
it seems like to me that the rust team fundamentally doesn't understand their role and position. they're showing up to a house that somebody else built and instead of saying "how can i help", they are saying "i have this cool thing, can we add it to your house so we can play too". at first, being bohemian open-source people, the kernel guys say "sure". but then it becomes problematic because it isn't free or easy and there are impacts. and they never asked for it in the first place.
100% the best thing for them to do is go build their own house. why do they even want rust in a 40 million SLOC "C" project? go create a kernel in rust. they could even leverage existing Linux drivers or other components. then it's their house to do what they want, the way they want. figure out how to box up "unsafe" "legacy" filesystems and drivers in their own rust ecosystem.
Definitely not what is happening here. btw what the point of CoC if it's not enforced? I'm not saying this person needs to be removed, but someone needs to talk to a person that says "You might not like my answer, but I will do everything I can do to stop this." in regard to R4L just based on personal preferences.
Code of Conduct is not about demanding absolute perfection and then selectively using it as a cudgel to beat people you disagree with. Doubly so since Hector's behaviour over the years has frequently been less than stellar, including in that very thread where he calls Hellwig's comments "distractions orchestrated by a subset of saboteur maintainers who are trying to demoralize you until you give up".[1] Yikes!
Using "cancer" to describe "it will spread everywhere and it will become unmaintainable" is not great, but at the core still a technical disagreement. Outright dismissing people's technical opinions and ascribing malicious motivations as part of a cabal is a mean-spirited and nasty personal attack, and essentially just an insult.
And it's really not "sabotage" to disagree or to be against something and being upfront about it. If that's "sabotage" then anyone saying "I don't think we should go ahead with this" is guilty of "sabotage".
[1]: https://lore.kernel.org/rust-for-linux/2b9b75d1-eb8e-494a-b0...
For it to be technical disagreement, there should be _anything_ to back it up. All I've heard is "another language would spread and become unmaintainable" is one, that's just emotions. There is nothing technical about his reasoning on why he doesn't want those patches to land.
> If that's "sabotage" then anyone saying "I don't think we should go ahead with this" is guilty of "sabotage".
Sabotage part is saying "I will do anything to stop rust from landing in Linux code base" (paraphrasing). Calling in cancer just unprofessional and rude, but that's another story...which probably also violates CoC.
Paraphrasing to the point where it says something completely different than was actually said.
An actual quote is "keep the wrappers in your code instead of making life painful for others". So it's about where and how.
And of course this is a technical disagreement; you just don't agree (which is fine). But please, don't pretend the disagreement doesn't exist.
> An actual quote is "keep the wrappers in your code instead of making life painful for others". So it's about where and how.
Literally the same? or you do not include "wrappers" in "rust landing in linux code base" ? Absolutely no one suffers from them landing.
I'm done arguing over this.
"Suffers"? No. But "makes the code harder to work on from my perspective"? Well, some people think that it will. And the Rust people are agreeing with that in the very email thread: their solution is "you don't need to write Rust, we can work together on this". Whether that's a good solution is the point of disagreement.
And of course "I don't want Rust landing in Linux code base" is not the same as "I don't want Rust added to this subsystem, I think it should live somewhere else".
Seems pretty clear cut to me.
Why do rust developers demand everything be re-written in their language? Especially one of the longest running, largest and most successful C projects of all time? It was never going to work out.
There are a few brand new operating systems being developed in rust, why not contribute to them instead?
I'm pretty sure Torvalds is the one who decided to add Rust to the Linux kernel.
> I was expecting updates to be faster, but part of the problem is that old-time kernel developers are used to C and don't know Rust. They're not exactly excited about having to learn a new language that is, in some respects, very different. So there's been some pushback on Rust. - Linus Torvalds https://www.zdnet.com/article/linus-torvalds-talks-ai-rust-a...
No it really wasn't. He just said let's see how it goes when the rust devs proposed it.
> I actually was hoping that we'd get some of the first rust infrastructure, and the multi-gen LRU VM, but neither of them happened this time around.
If we're appealing to authority here's his response a few weeks after the mail you posted:
The notion of "Linus accepted Rust in kernel, therefore it can be used everywhere" is a major point of conflict.
Also note the interview you linked continues with "Another reason has been the Rust infrastructure itself has not been super stable".