Maybe it's because of Linus' age or because the players involved are bigger than many past issues, but the way I see it, both decisions are really costly:
- Endorse Rust with little reserve and over half of the C++ devs will feel betrayed and quit supporting the Linux project. They've been working on C++ for Decades and things mostly worked, so they won't pivot for a new language and way of developing for something that exists for less than 30 years.
- Ban rust contributions and the entire Linux foundation goes directly against some big players, like DARPA and other departments of the American government[1], which itself is a trend setter. Some big sponsors might also pull out and that ALSO removes devs from the project.
So, which decision would be so overwhelmingly more advantageous that's worth taking all the negatives of the other on the chin rather than trying to minimize harm and wait to see if either Rust software proves to be not so magically immune to memory leaks and vulnerabilities or if some tool makes the transition less contentious?
> over half of the C++ devs will feel betrayed and quit supporting the Linux project. They've been working on C++ for Decades and things mostly worked, so they won't pivot for a new language and way of developing for something that exists for less than 30 years.
You mean C? C++ has been dead and buried for years and there are 0 kernel devs thinking that C++ will ever be allowed in (that idea was killed in the 00s if I recall correctly).
I don't think the situation is quite as extreme as you're making it out. For example, the employers of Linux kernel devs are getting pressure to use Rust because of the push from within & without the industry. I think push comes to shove, most people have a stronger preference for their paycheck than for the language they work in.
And I still think the culture clash angle should be taken more seriously.
An accomplished senior C dev can find good jobs writing drivers or microcontroller code more easily than starting over into a junior rust dev and being talked down by whomever joined the rust wave before them.
(which would be even more aggravating if the C dev is doing it since before the rust adopter was born).
- Endorse Rust with little reserve and over half of the C++ devs will feel betrayed and quit supporting the Linux project. They've been working on C++ for Decades and things mostly worked, so they won't pivot for a new language and way of developing for something that exists for less than 30 years.
- Ban rust contributions and the entire Linux foundation goes directly against some big players, like DARPA and other departments of the American government[1], which itself is a trend setter. Some big sponsors might also pull out and that ALSO removes devs from the project.
So, which decision would be so overwhelmingly more advantageous that's worth taking all the negatives of the other on the chin rather than trying to minimize harm and wait to see if either Rust software proves to be not so magically immune to memory leaks and vulnerabilities or if some tool makes the transition less contentious?
[1] https://stackoverflow.blog/2024/12/30/in-rust-we-trust-white...