4 ms·

Maybe Haskell is bad from practical perspective, but not from theoritical. So it has a strong appeal towards academics and purists.

by smlckz 6y ago

smcl 6y ago
Yeah this article was bizarre for me to read. My understanding is that Haskell doesn't aim to be big and popular (as Rust and Go have as a nice counterexample) its intended use is in academia by CS researchers, and its users seem to be quite satisfied with it. The fact that it gets any sort of traction at all outside this suggests that it's actually exceeded expectations of its success. > It is as if Haskell doesn’t want “normal” people to understand it Well, yeah kind of. I don't think it's as hostile as that though, the language isn't actively against "normal" people (I guess, that means your common or garden OOP programmer) it just doesn't actively court them. The comments about how the community treat newcomers are quite interesting though and if they are true then that's pretty shitty behaviour. It's one thing if the language isn't aimed at a given demographic of programmers, but it's another thing for people to use it as an excuse to beat down and berate others.
- Typhon 6y ago
  The one valid definition of success for a programming language is that it looks pretty and thus the most successful programming language is Piet ( http://www.dangermouse.net/esoteric/piet.html http://www.dangermouse.net/esoteric/piet.html ) I mean, joking aside, it's always fascinating when people can't wrap their minds around the idea that their criteria for success or quality or goodness are subjective and not universal.
  - smcl 6y ago
    Piet is lovely and a really creative idea :-) But yeah you are absolutely right.
- Blikkentrekker 6y ago
  I know many academics that exclusively use Haskell or even Idris and Agda which are even more obscure but more correct. Agda is so “correct” that the language is total and not Turing complete with the exception of a “partiality monad” that allows for partial and Turing complete computations — similarly to how Haskell isolates effects on the world from the main language and encapsulates them into a special type, Agda isolates partial computations. And this language is very much used within specific academic fields.
  - smcl 6y ago
    Wow ok Agda sounds ... challenging :-)

Maybe Haskell is bad from practical perspective, but not from theoritical. So it has a strong appeal towards academics and purists.

by smlckz 6y ago

Maybe Haskell is bad from practical perspective, but not from theoritical. So it has a strong appeal towards academics and purists.

smcl 6y ago

Yeah this article was bizarre for me to read. My understanding is that Haskell doesn't aim to be big and popular (as Rust and Go have as a nice counterexample) its intended use is in academia by CS researchers, and its users seem to be quite satisfied with it. The fact that it gets any sort of traction at all outside this suggests that it's actually exceeded expectations of its success. > It is as if Haskell doesn’t want “normal” people to understand it Well, yeah kind of. I don't think it's as hostile as that though, the language isn't actively against "normal" people (I guess, that means your common or garden OOP programmer) it just doesn't actively court them. The comments about how the community treat newcomers are quite interesting though and if they are true then that's pretty shitty behaviour. It's one thing if the language isn't aimed at a given demographic of programmers, but it's another thing for people to use it as an excuse to beat down and berate others.

Typhon 6y ago
The one valid definition of success for a programming language is that it looks pretty and thus the most successful programming language is Piet ( http://www.dangermouse.net/esoteric/piet.html http://www.dangermouse.net/esoteric/piet.html ) I mean, joking aside, it's always fascinating when people can't wrap their minds around the idea that their criteria for success or quality or goodness are subjective and not universal.
- smcl 6y ago
  Piet is lovely and a really creative idea :-) But yeah you are absolutely right.
Blikkentrekker 6y ago
I know many academics that exclusively use Haskell or even Idris and Agda which are even more obscure but more correct. Agda is so “correct” that the language is total and not Turing complete with the exception of a “partiality monad” that allows for partial and Turing complete computations — similarly to how Haskell isolates effects on the world from the main language and encapsulates them into a special type, Agda isolates partial computations. And this language is very much used within specific academic fields.
- smcl 6y ago
  Wow ok Agda sounds ... challenging :-)