I think the difference is here:
- If a compiled C++ program has unsound behavior, the program needs to be fixed
- If a compiled safe Rust program has unsound behavior, the compiler needs to be fixed
That's not completely correct. If the Rust or C++ compiler has a bug in both cases either the compiler has to be fixed or the program can be adjusted to work around said compiler bug If possible. However, It's fair to argue that the Rust compiler makes more assertions and is more complex, which results in a larger error surface.
Which is better, surely? One fix for every program written in Rust vs everyone having to fix it themselves…
> If a compiled safe Rust program has unsound behavior, the compiler needs to be fixed
Can you provide an example of such a program that can’t be fixed at the program level like a C++ example?
> - If a compiled safe Rust program has unsound behavior, the compiler needs to be fixed
Not really: the program still needs to be fixed, but fixing the compiler ensures that this unsound behavior never happens again to anyone. So you have two bugs to solve (vs C++'s one bug) but then you kill an entire class of unsoundness bugs just by recompiling.