I arrived at the same conclusion recently from a completely different angle.

I tried learning Rust several times and I found it impractical. I didn't really see the point, to be honest.

So you can imagine my surprise when a few weeks ago, I ended up reinventing Rust from first principles.

I had been forced into a bit of PHP/WordPress development due to my job. And I simply investigated the sources of my pain in this environment. Everything is mutable. Everything is global. Everything is dynamically typed. You get the picture.

I simply made note of each of these pain points and inverted them to approximate the solution space.

What I arrived at was an almost exact description of the Rust programming language.

(I think Haskell might be an even better description, but that's a story for another time.)

One small issue. Rust is not a general purpose programming language. It is explicitly a systems language. For systems programmers and for projects which meaningfully benefit from that approach. This makes it unsuitable by definition for the majority of programmers and the majority of projects today.

(The Low-Levelness Tax and by extension the Borrow Checker Tax are not necessary for the benefits which the Rust programming language brings! These same advantages can be had, in the context of a garbage collected language.

(Although, as far as I can tell, the borrow checker may bring significant advantages when it comes to concurrency, In terms of both correctness and performance, even in a high-level language. So the borrow checker may have serious advantages beyond the domain of low-level programming.))

I consider this very unfortunate because it currently feels like almost all the good ideas in programming are trapped inside of the Rust programming language. Which, again, is excellent, within its niche. But niche it is!

They are slowly finding their way out. But I have to wonder if there's anything we can do to accelerate the process.

So my hypothetical ideal language is basically Rust with a garbage collector, and with the numerous ergonomic improvements that this facilitates.

---

Curiously, I found Ruby surprisingly close to my ideal language. It is very high level and very expressive. It just has this fatal flaw that it has given basically no attention to correctness at the level of the type system.

So, I have to consider it unsuitable for any serious work. But I found it very interesting that Ruby and Rust had arrived at very similar designs, despite approaching the problem from very different angles.