Fabio Manganiello on Nostr: The opposition to #Rust in the #Linux kernel is turning into non-technical ...
The opposition to #Rust in the #Linux kernel is turning into non-technical opinionated nonsense and gatekeeping.
Even if you ignore the obvious security gains of writing system code in a memory-safe language, there's a compelling argument about the developers community that can't be ignored.
C developers are becoming harder to find. C developers that can write safe system code are even harder to find. Most of those that still fit into this category are now in their 50s/60s, and they aren't getting any younger.
Saying that no languages other than C are allowed in the main kernel tree (not for technical reasons, but because "maintaining multi-language projects is a pain I have no interest in dealing with", which sounds like someone who has never worked on enterprise code for at least two decades) means shutting the door in the face of a whole new generation of programmers.
It sounds like an enterprise code veteran who opposes migrating their legacy COBOL project to SQL or Java because they don't feel like seeing any new technology.
If this kind of friction goes on and Linus doesn't take a clear stance, then the idea of a Linux kernel fork that embraces Rust bindings may not be so far-fetched. And that's a scenario where nobody wins.
https://www.theregister.com/2025/02/05/mixing_rust_and_c_linux/
Even if you ignore the obvious security gains of writing system code in a memory-safe language, there's a compelling argument about the developers community that can't be ignored.
C developers are becoming harder to find. C developers that can write safe system code are even harder to find. Most of those that still fit into this category are now in their 50s/60s, and they aren't getting any younger.
Saying that no languages other than C are allowed in the main kernel tree (not for technical reasons, but because "maintaining multi-language projects is a pain I have no interest in dealing with", which sounds like someone who has never worked on enterprise code for at least two decades) means shutting the door in the face of a whole new generation of programmers.
It sounds like an enterprise code veteran who opposes migrating their legacy COBOL project to SQL or Java because they don't feel like seeing any new technology.
If this kind of friction goes on and Linus doesn't take a clear stance, then the idea of a Linux kernel fork that embraces Rust bindings may not be so far-fetched. And that's a scenario where nobody wins.
https://www.theregister.com/2025/02/05/mixing_rust_and_c_linux/