LISP is a black hole

Those kikes can suck any ideas you can think of into their money swindling machine.

aoe

Lisp is good for domain-specific solutions, but it doesn't adequately address the issue of multiple inheritance out of the box.

>>1

LISP

1.Garbage Collected. Only a few LISP has a decent GC, but it still lowers performance substantially.
2.Horrible prefix-only syntax with no separation of data and functions.
Requires syntax highlighting to be usable at all.
3.Inefficient data structures(list and list-based abominations such as tree lists)
4.Lack of bindings to latest libraries/APIs. If bindings exists there are often flaws in their implementations, making them harder to use.
5.Overall reliance on recursion and list comprehension in idiomatic, normal code(low performance).
6.Not actually functionally pure or safe.
7.Self-modification of code, doesn't seem a great feature today.
8.LISP macros the selling point of LISP, are actually used to fix LISP deficiencies or overlay some boilerplate in 99% of cases. DSLs created with macros are not actually as useful as actual languages of that domain or specific C/C++ libraries doing the same thing.

JACKSON FIVE GET

>>4

no separation of data and functions

that's the point

>>4
2. You missed the point. Also most langs are awful without syntax highlighting.
3. That's why most major dialects provide array/vector/hashmap types OOTB...
4. FFI is easy. Autogenerating bindings is easy.
5. Hurr Lisp has first-class functions so it's functional/Durr Lisp has object system so it's OO!
6. Never claimed to be.
7. Great feature in the right hands. Disaster in the wrong hands.
8. A language that can *fix* its own deficiencies? You're criticising that?!

Troll?

>>7

8. A language that can *fix* its own deficiencies? You're criticising that?!

She probably thinks "real" languages like Java++ don't have deficiencies.

check em

>>8

she

JACKSON 5 GET

But that's Haskell.

>>12
Haskell is the only language I've never had fun using.

Haskell is the black hole because it's always expanding with myriads of new extensions that it would take lifetime to learn to use, and its type theory is getting ever and ever more complex without leading anywhere, it just sucks up the money that goes to these "type theorists". doesn't lead anywhere. It started with purity and monads, but now they're in batshit-crazy land with stuff like "A Strongly Normalizing Computation Rule for Univalence in Higher-Order Minimal Logic" and fucking "Cubical Higher Type Theory as a Programming Language". Nobody even knows what all this crap is for, but they're always researching more and more of it as long as they get money grants.

>>13
You've reminded me of Republic of Haskell... Industry of safety and happiness... I hereby pronounce you guilty of side effects...

>>15
I'm here to inform you that you've exceeded your daily quota of ellipses.

Behold the glory of Time Cube >>14
https://github.com/mortberg/cubicaltt/blob/master/TypeChecker.hs

>>14
I don't think you understand the purpose of research. There is no inherent requirement for pure research to have immediate applications in the real world.

>>18 pure research can't have side effects

How many lists can you fit in a clown cdr?

>>22
Nice dubz

>>14

I agree a bit with you.

E.g. TypeFamilies provide crippled typelevel lambda's. Why not add the real thing? Now you get BlaType is not injective errors if you do something minor interesting with them.
But they fixed functional dependencies, so it evens out there. And they added some interesting generalizations over type classes.

DataKinds (and PolyKinds as well) are fucking useless without KindSignatures and also they fucked up by only providing one Universe. Where normally you will infinite universes. values : types : kinds : sorts : ..... Haskell only has values:types:kinds:sorts end of story.

Other problems is type equality from GADTs. Haskell doesn't allow you to prove type inequalities, which makes your code much more complex than needed.

On the other hand, haskell is quite practical and one of the reasons is the type extensions. Extensions, which work, get merged into the language. Type extensions which fail get removed after a while.

>>22

I didn't explain why having not a stack of universes is bad. It means that the last universes contains itself, potentially leading to logical inconsistencies. Therefore the type system is inherently unsound.

But things are getting better. Some guy is busy implementing dependent types into haskell: http://cs.brynmawr.edu/~rae/papers/2015/thesis/eisenberg-thesis-proposal.pdf

Which will clear up a lot of problems, we have now. Together with some extensions, some new and interesting properties of the type system will arise. Now they need to be faked, which sucks.

oh my god fuck off you haskell tards, there is an entire subreddit for your garbage. stay there.

>>1
That's true. The difference is, we don't think it's a bad thing.

>>26
That group doesn't think 9/11 is a bad thing either.

>>27
Yes, we Lisp programmers all love 9/11. We're all fucking nigger kike terrorists. How did you see through our clever disguise?

>>27
People did vote for Bush/11
Trump/16 should be fun

inb4 the terrorist predictors say predicting terrorism is terrorism

>>30
Fucking meta-terrorists. They're all nazi faggots, and I'm definately not one of them. Yep I definitely didn't write >>28, and I'm not using this post to mock you now.

Hmm... the sarcasm was good, but the subtlety needs more work.

>>30

le pedophile sage

>>32
Checkem

>>33
Nice pedophile dubs!

>>34
why is "33" pedophile dubs? what's special about 33?

>>35

there are some pedophiles who keep saging every thread for no reason. this is pedophile saging. we must counter those pedophiles and bump the threads.

>>36
Was not talking to you...