This isn’t Linux, but Linux-like. Its a microkernel built from the rust programming language. Its still experimental, but I think it has great potential. It has a GUI desktop, but the compiler isn’t quite fully working yet.
Has anyone used this before? What was your experience with it?
Note: If this is inappropriate since this isn’t technically Linux, mods please take down.
It’s a system programming language that isn’t C or C++.
Edit to add: How did Go get on that page? That’s a stretch.
But what is wrong with C and C++ apart from the ISO fuck-up (ahem, slow updates)? There’s a lot of technical debt, so wouldn’t it be better to create an alt-language compiler that adds improvement over C, so that migration is possible in multiple stages?
Something like:
Then we arrive at Rust as a natural outcome.
And it’s of course possible to migrate to Rust from C or C++ progressively, fish has almost got it done.
Did fish migrate progressively tho? I thought they swap out everything at once as soon as the rewrite is ready
Well, they are not going to release in between, but their rewrite still “works” at each commit being a hybrid of Rust and C++.
Ah, alright
Rust isn’t just a new improved version of C or C++. It’s completely new and it feels completely different to use Rust. In a positive way
Guaranteed memory saftey
(notice: I am not a Rust or C/C++ expert)
Doing all that is creating a completely separate programming language from C. Rust is that programming language.
Rust does that with modules and crates.
You mean having consistent/universal style guidelines? Rust pretty much has that with rustfmt.
Safe Rust is memory safe (using things like the borrow checker), and Unsafe Rust is (usually?) separated using the
unsafe
keyword.Although Unsafe Rust seems to be quite a mess, idk haven’t tried it
Rust has macros, iterators, lambdas, etc. C doesn’t have those. C++ probably has those but in a really weird C++ way.
I should have framed my words better, I guess. Rust is a radically different language, and honestly, none of the feature it offers fixes the main issue, that is technical debt - I mean yes, there’s incline C or FFI, but that’s still going to be a radical migration.
What I’m trying to propose is an alternative project, independent from the ISO. Maybe it could be a C-to-Rust, or a C-to-Vale migration project. It could be any of the modern language, I don’t really care. But that particular compiler/transpiler/migrationpiler/<something>-piler should have the ability to do step-by-step migration.
I’d say no. Programming safely requires non-trivial transformation in code and a radical change in style, which afaik cannot be easily done automated.
Do you think that there’s any chance to convert from this to this? It requires understanding of the algorithm and a thorough rewrite. Automated tools can only generate the former one because it must not change C’s crooked semantics.
I was planning to learn C23 for quite some time. It’s a pity that I’ve been planning to learn RISC-V with it, sigh. I guess I’m gonna move over to Rust or Zig, whichever makes sense. But I’d probably switch, when Vale becomes a legit language.
I think there’s no need to stick with one particular language. It benefits to learn more languages and bring the “good parts” of their design into your code whatever you are writing it in.
Btw It happens that I’ve learned a bit of RISC-V, with Rust.
Idk what the iso fuck up and I don’t code enough to appreciate whatever technical debt exists in either so I am probably sound like an idiot but…
Since I do infosec, the glaring issue for me is not being memory safe.
C and C++ can’t be fixed retroactively because old code must remain compatible.
If you’re going to implement your own C dialect, you may as well just write a new language.
For C++ that’s Rust, for C that’s probably Zig (Zig will just let you import existing C files, which helps with porting). Carbon and experimental languages like Jakt may also work, it all depends on what your priorities are.
Conservatives always loose, especially in tech