C developer here, so I think to help you all out a little bit on understanding Rust scenario. The problem is that a lot of us prefer to have single language development so that we’re not forced to learn about Rust language or other languages.
Second problem is BIDIRECTIONAL FFI, because when we share code in C, almost every language can use and extend C code, but it’s not true for Rust code however, it’s unidirectional. Other language can’t easily extend Rust code nor can it use Rust’s libraries, but Rust could use C code with ease. I know people would say, “But Rust could expose all of those things.” Well, LLVM which is written in C++ could’ve done the same, but they didn’t, we don’t have full access of LLVM code in C and we can’t expect that in Rust code either.
Rust does have a reputation of being a prosecution complex in some spaces where I did get nagged on by rust folks to rewrite my project in Rust. I told them no and that I could defend my choices for memory safety and reliability by adhering to MISRA-C standard and using Frama-C for verification.
I don’t hate on Rust, but I just want people to understand that there are wider scopes to consider in a larger project like Linux Kernel and that it can hurt us more if we start having more than 1 language in a already complicated project like Linux Kernel. Rust already have it’s own operating system, why not work on that and extend it to compete with Linux Kernel? It’s called Redox OS: https://redox-os.org/
I can see that Rust can make a lot of positive change to the Linux community, just that we need to treat it as a tool just like C, C++, and every other language.
Yeah, this is something I am worrying about in overall open source projects. There are going to be increasing number of protest-ware and that is going to suck.