Why do Pajeets hate functional programming?
if you aren't .NET you're .NEET
>>54396535
F# is .NET and a pretty decent functional language
because Pajeets are pretty practical, no one actually works on projects in functional programming languages.
>>54396535
underrated
>>54396565
How can it be underrated? It was posted a fucking minute ago.
>>54396557
>>54396565
I was using Neet as a bad word apologies, .NET dev myself here
>>54396587
overrated
>>54396524
your pic related is shit, the empty list case is not defined
>>54396524
>Haskell
>>54396655
>he fell for the JS meme
>>54396524
Functional languages are cute toys to fuck around with but nobody uses them for actual prod work.
>>54396862
https://wiki.haskell.org/Haskell_in_industry
>>54396524
It's hard to hire for functional programming languages. You'll get a lot of tinkerers who can't cut it 40hrs a week. There's a lack of libraries available, so often you have to write it yourself.
>>54397042
>no reason to learn functional programming language because no one is hiring them
>no reason to use functional programming languages because no one to hire.
Why do Paneets love functional programming?
Haskell vs Lisp
What and why?
>>54398172
lisp, because i know it, because it's prettier
haskell's probably faster though, because compiled, static typed etc etc
>>54398245
WRONG
>>54398172
Lisp most of the time because
1. Haskell enforces 'pure' functional paradigm, effectively narrowing the ways a programmer would have to do something. Lisp allows for just about any paradigm you can think of.
2. Lisp can grow beyond it's initial scope. It's metaprogramming facilities allow you to turn it just into the language you need for a particular application.
That last feature is actually what allows it to expand into any paradigm, be it Object Oriented, Aspect oriented, and you can implement new ideas.
The cons of using lisp instead of haskell is less readability, because parentheses constitutes about ~90% of it's syntax. Haskell is way more readable and it actually looks like mathematical notation.
The other downside is that you'll have to learn it, I mean really learn it, not just superficially, if you want to really leverage it's power. You can write simple programs with a basic knowledge, but until you really learn it to depth you won't be doing any of those spooky things that you can only do with lisp.
And the lisp runtime gets bundled into your binaries, so unless you explicitly leave it out, it'll only produce reasonably sized binaries for middle size and up applications
>>54398804
>The other downside is that you'll have to learn it, I mean really learn it, not just superficially, if you want to really leverage it's power. You can write simple programs with a basic knowledge, but until you really learn it to depth you won't be doing any of those spooky things that you can only do with lisp.
This looks like a hell of a downside. It's something that i can cover with a book or it's about experience?
Land of lisp have 508 pages. It looks affordable since I've already some experience with functional programming.
>>54398992
Also the fact "one language dozen of dialects" is scary as fuck, probably the worst downside for a newcomer.
For what I get Scheme is the prettiest, Common Lisp is the feature full and Clojure is the modern.
I genuinely don't know what I'm looking for. I mean, if I'm looking for lisp is 'cause I wanna know what a good language looks like, so Scheme should be the obvious choice. But Scheme doesn't looks any practical, so what?
>>54398992
I'm not sure if I would recommend land of lisp. It has some interesting programs like a genetic algorithm and a wumpus based fun game. But the code style is absolutely horrid.
Practical Common Lisp and PAIP are two good alternatives.
But the real knowledge comes with either of On Lisp by Paul Graham (also read his article "beating the averages") or Let Over Lambda by Doug Hoyte.
Don't expect an easy ride though, and you won't become a lisp wiz overnight. It's mostly experience with a good deal of reading the advanced books for that spooky shit, but it's also real fun and it provides really interesting insights.
>>54399339
>Scheme doesn't look any practical
WELL, Scheme itself isn't practical in it's own right. The standard defines but the core language. Most of what would be deemed useful is implementation dependant.
Also, lisp is a family of languages, just like C is part of a family of language descendants of Algol. C, C++, Java, Python, they are all descendants of Algol.
I advice you go with Racket then: featureful, pretty, and has some good literature on programming language design.
>>54396986
You have to jump through some pretty intense hoops to get good Haskell code.
And by then your code is a barely legible mess. Or you got lucky and the compiler was nice to you.
It's not a shitty as Clojure and much better than Scala, I'll give it that; but stay away from Haskell.
Scheme, Common Lisp, and OCaml are the only functional languages worth using.
>>54399419
Thanks.
Do you have any book to recommend? I know some people would say SICP, but it's not something you can read sporadically in your free time. It's really what I need?
Consider I already know a little of OCaml and Erlang, It's not like I start from zero.
Speaking of Scheme, have any of you seen how fast Chez is
https://www.nexoid.at/tmp/scheme-benchmark-r7rs.html
It was finally open sourced after 32 years.
>>54399853
The Little Schemer!
Racketeers often suggest HTDP but it's too introductory and aimed for a high-school education.
You can read The Seasoned Schemer then, and then choose between SICP (it's not overly hard, does require thinking though), PLAI, the other Little Schemer books, and my personal favorite in the Scheme literature: Lisp in Small Pieces (this one is about implementing lisp in lisp)
>>54396587
downvoted
>>54396557
>and a pretty decent
Where “pretty decent” means “at least it's functional, right?”
>>54397042
>There's a lack of libraries available, so often you have to write it yourself.
https://hackage.haskell.org/packages/
>>54398804
>The other downside is that you'll have to learn it, I mean really learn it, not just superficially, if you want to really leverage it's power.
This is true for Haskell as much if not more so than for Lisp
>>54396564
Third post best post.
>>54401885
It's also true for C++, and Rust. Hell, most "powerful language"s take an effort to actually be able to take advantage of. I'd love to see a language where that isn't the case.
>>54402894
Well it's also amplified by the fact that Haskell etc. are so different from the languages 99% of people know
>>54396524
dear god what the fuck am I looking at? This is complete gibberish. How the fuck can anyone code in a language like this
>>54403302
>hurr I can't read perfectly readable code from a language I've never learned
>it's gibberish
I meah how the fuck do germans even understand their language, right? sounds nothing like english
