Haskell is not category theory
pema.dev
Haskell is not category theory
1–10 of 54 posts
Re: Haskell is not category theory
#2> However, if you are proficient with any functional language, especially if it is a dialect of ML,
The one, true ML.
Re: Haskell is not category theory
#3I'm confused by his argument that Haskell functors are only endofunctors because everything is Hask.
I thought that even if everything is part of one category, if the input is a different category that is a subset of the one category and transformed to another category, even if a subset of the same category, it could still be a regular functor instead of an endofunctor.
For example, lets say you had a function f(x) that takes unsigned ints and returns the signed int inverse (so f(1) = -1). -1 is not part of the unsigned category since it is below the Initial value of an unsigned number (0). I'm a total newbie to category theory so I'm not sure if he's wrong or I'm missing a key detail here.
Re: Haskell is not category theory
#4> If you are looking to learn the language, I recommend Learn You A Haskell.
I definitely don't recommend this. While cute, Learn You A Haskell is lacking a lot of pragmatic advice about how to write software in haskell. It's good for getting you up to the FAM trio of type classes but leaves out a lot of practical advice around things like navigating base (what to use and ditch from it), the available community libraries, topics like monad transformers, and pragmatic advice around things like error handling, testing, and isolating effects from tests that come up very quickly in real world Haskell development.
Haskell Programming From First Principles is a much better book IMO: https://haskellbook.com/
I haven't written Haskell in a while, but there may be even better books today as well.
Re: Haskell is not category theory
#5> If you are looking to learn the language, I recommend Learn You A Haskell. I definitely don't recommend this. While cute, Learn You A Haskell is lacking a lot of pragmatic advice about how to write software in haskell. It's good for getting you up to the FAM trio of type classes but leaves out a lot of practical advice around things like navigating base (what to use and ditch from it), the available community libra…
I mostly agree with your recommendation against Learn You A Haskell.
However I would suggest people have a look at https://en.wikibooks.org/wiki/Haskell
'Haskell Programming From First Principles' is a bit too fan-boyish.
Re: Haskell is not category theory
#6> If you are looking to learn the language, I recommend Learn You A Haskell. I definitely don't recommend this. While cute, Learn You A Haskell is lacking a lot of pragmatic advice about how to write software in haskell. It's good for getting you up to the FAM trio of type classes but leaves out a lot of practical advice around things like navigating base (what to use and ditch from it), the available community libra…
I mostly agree with your recommendation against Learn You A Haskell. However I would suggest people have a look at https://en.wikibooks.org/wiki/Haskell 'Haskell Programming From First Principles' is a bit too fan-boyish.
I thought fanboyism was one of the first principles.
Re: Haskell is not category theory
#7Earlier quoted context omitted.
I mostly agree with your recommendation against Learn You A Haskell. However I would suggest people have a look at https://en.wikibooks.org/wiki/Haskell 'Haskell Programming From First Principles' is a bit too fan-boyish.
I thought fanboyism was one of the first principles.
Well, you'd want enlightened fanboyism.
Re: Haskell is not category theory
#8I'm confused by his argument that Haskell functors are only endofunctors because everything is Hask. I thought that even if everything is part of one category, if the input is a different category that is a subset of the one category and transformed to another category, even if a subset of the same category, it could still be a regular functor instead of an endofunctor. For example, lets say you had a function f(x) t…
You could probably try to talk about functors between subcategories by using classes, though I think there's no way to define a functor from one class to another in Haskell.
Slightly more annoying is the fact that Haskell has no real product types and instead does more currying than an Indian takeaway. This makes various constructions in Category theory a bit more annoying than they need to be. If you ever wondered what on earth Applicable is and why you need it to fmap a function with multiple arguments, this is why.
There's also the fact that Haskell doesn't really enforce the functor laws, nor does it give any way to really state them. Basically it uses Category theory as a close enough approximation for what it does, but various finer details get lost in translation.
Re: Haskell is not category theory
#9> If you are looking to learn the language, I recommend Learn You A Haskell. I definitely don't recommend this. While cute, Learn You A Haskell is lacking a lot of pragmatic advice about how to write software in haskell. It's good for getting you up to the FAM trio of type classes but leaves out a lot of practical advice around things like navigating base (what to use and ditch from it), the available community libra…
I have tried several times to learn Haskell. The only resource that actually worked for me was Philipp Hagenlocher's "Haskell for Imperative Programmers".
It is incredible. He does an amazing job of starting from the basics and then diving deep. After going through the series I had a solid enough grasp on the language to use it for practical things.
https://www.youtube.com/watch?v=Vgu82wiiZ90&list=PLe7Ei6viL6...
Re: Haskell is not category theory
#10I'm confused by his argument that Haskell functors are only endofunctors because everything is Hask. I thought that even if everything is part of one category, if the input is a different category that is a subset of the one category and transformed to another category, even if a subset of the same category, it could still be a regular functor instead of an endofunctor. For example, lets say you had a function f(x) t…
You could probably try to talk about functors between subcategories by using classes, though I think there's no way to define a functor from one class to another in Haskell. Slightly more annoying is the fact that Haskell has no real product types and instead does more currying than an Indian takeaway. This makes various constructions in Category theory a bit more annoying than they need to be. If you ever wondered w…
Okay this gives me something to go on, thanks!