your volvo is a car. just not a sports vehicle. i write kernel modules and other "system" libraries. Is the problem that my definition of a "systems language" is too narrow for you? Yea you could say we resort back to the C FFI whenever i need to think about dynamic linkage with modules that may or MAY NOT have been compiled with the same version of rust but jesus is it a tedious process for me.
> Is the problem that my definition of a "systems language" is too narrow for you?
I'm suggesting that the exclusion of Rust from that category is wrong. Even with it as narrow as you just defined it, Rust still fits into that category.
Is it the best systems language available at the moment? It depends on your context. With TockOS, if you can use that, I would say it is. With Redox and other full OS impls, it's clearly capable as a systems language. With examples of kernel modules for Linux written in Rust, it's again clear that it can work in that environment. By excluding it from that category, it continues a line of FUD against Rust and in favor of C.
> Yea you could say we resort back to the C FFI whenever i need to think about dynamic linkage
Yeah, this is a bummer. And for the foreseeable This is probably going to be the case (C++ has this same issue, but to less of an extent because it is generally easier to flow between C++ and C).
> jesus is it a tedious process for me.
Yeah, again, where a strong ABI is needed it's a problem, but there are people working on some solutions to make this easier, for example:
