4 ms·

A programming language with Python-like syntax and a Hindley-Milner type system plus typeclasses. Basically, we bring Haskell’s type system an imperative progra

A programming language with Python-like syntax and a Hindley-Milner type system plus typeclasses. Basically, we bring Haskell’s type system an imperative programming language. An easy-to-learn scripting language with a strong, sound type system with reliable type inference will reduce bugs in codebases of every size across the industry.

zengid 7y ago
I'm not an expert, but that kind of sounds like F# (maybe not the typeclasses though?).
imiric 7y ago
Have you given Raku (fka Perl 6) a try? It seems its roles[1] resemble Haskell's typeclasses, and it borrows heavily lots of good features found in other languages. Disclaimer: I haven't used either language, just learned about Raku recently from this[2] excellent FOSDEM talk. [1]: https://docs.raku.org/syntax/role https://docs.raku.org/syntax/role [2]: https://fosdem.org/2020/schedule/event/rakulang/ https://fosdem.org/2020/schedule/event/rakulang/
salt-licker 7y ago
I think languages which bolt some trait / generic / role system onto an object-oriented system are generally a bit too bloated, non-opinionated, and confusing. Typeclasses not only encourage an elegant separation between data and functions, they allow you to design guarantees about how your functions interact with each other into your program — basically, not only data inheritance but also functional inheritance. Object oriented syntactic sugar is an easy add-on to a typeclassed language, but not the other way around. Beyond immediately reducing bugs, I think widespread use of this sort of language designed with a good type system from first principles would allow many more programmers to write programs with some provable security guarantees.
bordercases 7y ago
Disagree about scripting. Compile times are fast if you are willing to tradeoff performance. But you can decide whether you whether or not you want a long compile fast program, a short compile slow program, or an interactive compiler. The example I can give is StandardML. You have SML/NJ and MLTon as two possible compilation options among other. SML/NJ has an interactive mode. MLTon compiles for longer but results in faster programs. (And it's also Hindley-Milner) Just going to the "it should be interpretable" route without there being any option for whole-program optimization at compile time results in unnecessary bloats in performance which by all means we can avoid now as development hardware has become faster.
- salt-licker 7y ago
  Yes, I agree the language should be compiled for performance, types enable compiler optimizations and it would be silly to abandon that benefit. By “scripting language” I mean the syntax will be similar to existing interpreted languages and it will be easy to write small scripts without specifying types since they are inferred.

zengid 7y ago

I'm not an expert, but that kind of sounds like F# (maybe not the typeclasses though?).

imiric 7y ago

Have you given Raku (fka Perl 6) a try? It seems its roles[1] resemble Haskell's typeclasses, and it borrows heavily lots of good features found in other languages. Disclaimer: I haven't used either language, just learned about Raku recently from this[2] excellent FOSDEM talk. [1]: https://docs.raku.org/syntax/role https://docs.raku.org/syntax/role [2]: https://fosdem.org/2020/schedule/event/rakulang/ https://fosdem.org/2020/schedule/event/rakulang/

salt-licker 7y ago

I think languages which bolt some trait / generic / role system onto an object-oriented system are generally a bit too bloated, non-opinionated, and confusing. Typeclasses not only encourage an elegant separation between data and functions, they allow you to design guarantees about how your functions interact with each other into your program — basically, not only data inheritance but also functional inheritance. Object oriented syntactic sugar is an easy add-on to a typeclassed language, but not the other way around. Beyond immediately reducing bugs, I think widespread use of this sort of language designed with a good type system from first principles would allow many more programmers to write programs with some provable security guarantees.

bordercases 7y ago

Disagree about scripting. Compile times are fast if you are willing to tradeoff performance. But you can decide whether you whether or not you want a long compile fast program, a short compile slow program, or an interactive compiler. The example I can give is StandardML. You have SML/NJ and MLTon as two possible compilation options among other. SML/NJ has an interactive mode. MLTon compiles for longer but results in faster programs. (And it's also Hindley-Milner) Just going to the "it should be interpretable" route without there being any option for whole-program optimization at compile time results in unnecessary bloats in performance which by all means we can avoid now as development hardware has become faster.

salt-licker 7y ago
Yes, I agree the language should be compiled for performance, types enable compiler optimizations and it would be silly to abandon that benefit. By “scripting language” I mean the syntax will be similar to existing interpreted languages and it will be easy to write small scripts without specifying types since they are inferred.