pony-tutorial VS futhark

I don't think that there is a book written about Pony, but the tutorial and the list of patterns (WIP) are all you need to learn the language.

[Pony](https://tutorial.ponylang.io/): actors, reference capabilities, object capabilities.

Pony is a relatively young but interesting language with capabilities-security.

Well, depends on how you define ownership. Pony 's type system has reference capabilities which let you define who's allowed to do what to a reference and part of it sort of deals with ownership (along the lines of "this actor is allowed to do Y to the reference, other actors are allowed to do Z"). You can eg. have methods that return an isolated value that guarantees that there are no other references to that value, meaning it's automatically thread safe. You can also define things as vals which says that they are globally immutable, refs which give the current actor read/write capabilities but can't be shared with other actors

The alternatives are:

- Division must be impure (because it can throw an exception)

- Division must be partial - i.e. return Option[Int].

Both seem worse compared to defining division by zero as zero. Coq, Lean, and Pony do the same. https://github.com/ponylang/pony-tutorial/blob/master/conten...

Why Futhark? (futhark-lang.org)

There is no on-going work to support Metal apart from the work done by Miles. There's an old issue about it: https://github.com/diku-dk/futhark/issues/853#issuecomment-5...

Functional programming can be a great way to handle parallel programming in a sane way. See the Futhark language [1], for example, that accepts high-level constructs like map and convert them to the appropriate machine code, either on the CPU or the GPU.

[1] https://futhark-lang.org/

Futhark - use a functional language to program the gpu

You might want to look into Futhark, although it's mainly designed for writing GPU code.

Futhark https://futhark-lang.org/