Rust has some safety guarantees and it is strictly better in that regard than C++. But it’s not a panacea, and the footprint of `unsafe` buried in so many crates is worrisome, especially when novice systems programmers see “guaranteed safety” and make assumptions that aren’t necessarily valid.
Aside from that I very carefully and clearly did not lump all Rust evangelists in a single bucket.
You know that your points aren't very interesting. We value languages like C# and Python for their safety, even though we know that libraries sometimes contain C code which may have bugs and that the implementation is in C or C++ and might have bugs and it runs on an operating system which contains C and assembly code and might have bugs. There's a huge value in knowing that the code you write won't have memory problems, and you recognize that with all other languages. Your vendetta against Rust prevents you from recognizing the exact same thing when applied to a language you don't like.
> But it’s not a panacea, and the footprint of `unsafe` buried in so many crates is worrisome, especially when novice systems programmers see “guaranteed safety” and make assumptions that aren’t necessarily valid.
On top of the tens of thousands of crates that (ab)use unsafe everywhere and programmers importing lots of packages and installing hundreds of transitive unsafe packages, using an NPM-style package manager and marketing this as "guaranteed safety" does sounds very much like the Rust hype-squad are over hyping their snake oil to push the cargo cult.
My guess is that you're lumping all Rust evangelists together in one bucket and thinking of them all as "toxic evangelists", when a lot of them are just people who want the computing world to be better.