>java without docs or types
>everything is a hash map
>clojure maps are somehow 130x slower than java maps
>popular libraries fall back to java for performance
>users shill its concurrency primitives, only use atoms
>atoms just piggyback off java.util.concurrent

>>105911079 (OP)
Scala wins yet again!

There are only two kinds of languages: the ones people complain about and the ones nobody uses.

>>105911086
thanks for reminding
>shills claim Clojure "embraces" the platform
>cant use clojure on other JVM langs unless you want to compile Clojure code by hand
>cant use async/await in ClojureScript because they'd have to redesign the entire compiler from scratch
scala and scala.js dont have these problems btw

>>105911079 (OP)
>>105911086

C# > Java > Scala > ... > Python

>>105911086
>>105911122
>le scala meme
Clojure is much more useful than Scala.
ClojureScript, ClojureCLR, ClojureDart, Babashka, Jank (C++), Clojerl, Wolframite...

>>105911119
pearls before swine

>>105911079 (OP)
You use Lisp to learn how to code and then you code in C++ in C style.

You forgot that it lacks cons cells which means it isn't even a real Lisp dialect. Common Lisp can't stop winning...

>>105911204
No one ever used any of these with the exception of clojurescript

clojure is for golems who are different than the other redditors and for some god unknown reason want to use java inside of lisp
java would be dead if it weren't for minecraft

>>105913593
>java would be dead if it weren't for minecraft

>>105913263
I don't mind that it lacks cons cells. It's not a deal breaker. I think the sequence abstraction is a good idea.
What I do mind is that Clojure feels like a very thin layer over Java. Its REPL feels a lot worse than any other Lisp dialect I know because at some point it just starts spitting out Java errors at you. And I also mind that its macro system is cucked beyond all recognition. I see no reason to intentionally take one of Common Lisp's most powerful features and then decide you just won't let people do that anymore because you hope businesses will like you. All that happens is retards see parentheses and get spooked anyway and now you have a language that is still slowly fading away, except it's strictly weaker than other Lisps.

>>105911079 (OP)
Babashka is nicer than fucking with Python venvs. Even the regex engine alone is an improvement.

>>105913578
This is a good example
https://github.com/agent-kilo/jwno
>>105913593
I don't like Java but this is delusional as fuck

>>105911194
And F# > C#

>>105915509
Too bad. Microsoft doesn't care about it. It'd have been great.

>>105913578
>To NuBank who approached us very early for sponsorship.
>To Roam Research who bet their mobile apps development (now in the App Store and Play Store) on ClojureDart and allowed us to make steady progress since Summer 2021.
https://github.com/Tensegritics/ClojureDart#thanks

>>105915760
I wonder if Google will kill Flutter. Dart failed, but the toolkit is usable.

>>105914726
clojure actually did manage to eke out a niche in industry. walmart labs, nubank etc. it did a pretty good job of mogging every retard who thought lisp was inherently useless for real SWE.
but it ended up not really mattering cause the language stagnated which happens to every JVM language the instant corpos realize they can't foist it into java jeets

It always smelled fishy to me
>>105911204
ClojureCLR is dead

>>105916777
For a time yes but it's still dying. Hickey just can't accept that corporate types will not love Lisp no matter how badly you mutilate it for them.

>>105918024
Nah. As long as Nubank keeps bankrolling it, it will be fine.

>get clojure job
>codebase is a complete mess because it was written by some dude that just freshly learned it 8 years ago and then left the company
not that different from any javajeet job to be honest

>>105918024
>>105918349
You're both right to some degree, it won't probably die but for me, it has plateau'd already. Unless one of the implementations that branched out gains traction by fixing some of the problems or being more compelling on its niche (like Babashka for CI/CD tasks or ClojureDart) it's not going to gain many more users. Maybe Lombok will change that.

>>105911079 (OP)
>java without docs or types
Hold the presses, dynamic language has no types. It does have docs.
>everything is a hash map
That's the whole point. Object encapsulation kills function re-usal.
>clojure maps are somehow 130x slower than java maps
Somewhat slower to enable immutability (hash-array mapped trie), but you didn't mention that property did you?
>popular libraries fall back to java for performance
Java is very fast but has shit start-up and it's a memory hog.
>users shill its concurrency primitives, only use atoms
true
>atoms just piggyback off java.util.concurrent
Not related in any way whatsoever.

I don't actually use Clojure in production but it sounds to me you don't have a single clue about the language. It's worth learning just to understand the different perspective, inherited from Lisp.

Clojure is for when you are completely sick of the bullshit and just care about the problem domain that your software is targeting.
- functional/immutable defaults for complex data transformation
- trivial calling of Java or Java libs when needed (or when you want to get imperative, typically in some small hot path)
- battle tested to hell JVM

>>105919180
>Object encapsulation kills function re-usal.
only in cases where reuse would cause higher coupling and lower coherence. it's possible to have functionality that's both encapsulated as well as highly reusable

also, Primitive Obsession is a code smell of its own, and having everything as a hashmap is a big symptom of it

>>105919180
>Hold the presses, dynamic language has no types.
rich hickey cultist has never heard of strongtalk or how it became the basis for the HotSpot JIT

>Somewhat slower to enable immutability
the 130x figure compares property access in immutable hash maps...

>Java is very fast but has shit start-up and it's a memory hog.
clojure makes the startup and performance even worse thanks to loading 6 million classes through clojure.core alone

an empty keyword consumes over 200 bytes of memory.

>Not related in any way whatsoever.
all of the "good concurrency" that Clojure supposedly "enables" is not inherent to anything in the language itself. everything it can do is trivially done in Java and even other languages (like Rust).

>I don't actually use Clojure in production
I can tell. Unlike you, I have and it's a miserable experience.

I don't like but Babashka is literally the only thing besides Perl that just works for scripting and isn't retarded like POSIX sh.

>>105911079 (OP)
I wish the CLR had its own Lisp dialect.

There are 3 kinds of programmers
1. Those who like computers and optimizing things
2. Those who like building real products
3. Those who like neither and just jerk off to non-von neumann architectures or whatever

you guys just dont get it

>>105922543
there's a clojure port but nobody cares because .NET fags only use C#

>>105921092
>everything it can do is trivially done in Java
Javajeets trying to understand why other JVM languages exist will never not be funny.

>>105923262
The Clojure port is barely maintained and in the middle of a complete rewrite to F#, you can't expect anyone to use something like that for serious projects.

>>105923368
yeah and it's unmaintained because nobody cares

>>105922586
#1 is just a dumber, mathlet version of #3. You "like computers" in the general sense but don't realize that a computer is an abstract mathematical concept that we can only imperfectly implement into a physical system.

>>105923362
>if you dont like clojure you must be a java jeet
the coping has begun.

I'll preemptively post more truthnukes about this language and its dead ecosystem

>core.async
btfo by kotlin coroutines. then btfo by java virtual threads.

>clojure.spec
somehow takes the worst of prismatic schema and malli. Retards at cognitect decided this must be a hard dependency of clojure.core itself since the 1.8.0 release. the compiler literally hard codes spec checks in macroexpansion

Rich's own "Maybe Not" talk is essentially a whole discussion of problems that malli literally solves (but I disgress, malli is useless for any static typing or compile-time guarantees. have fun with edge-cases that arent caught in production because spec checking is so dogshit slow that everyone turns it off)

>muh types bad
https://ask.clojure.org/index.php/2800/clojure-set-should-check-or-throw-on-non-set-inputs

>muh tools.deps, tools.build
Ant with s-expressions. But since its blessed by Relevance^W Cognitect, rich hickey cultists treat it as some breakthrough in build systems

inb4 you pretend Leiningen is any better

>muh test.check
half-baked implementation of shit Haskell has had for decades

>muh datomic
the only database where swapping two :where clauses turns a 10ms query into a 10s query

>muh babashka
an interpreter that barely implements half the core language and resorts to literal binary blobs for extension (see: pods). installed on maybe 2 computers in the world, 1 of which is borkdude's macbook

>>105911079 (OP)

There may or may not be some problems with maintainability of Clojure code, but your comparison is surface level.
You also didn't point out any actual strengths of the language or the s-exp notation (fast prototyping, referential transparency, seq abstraction, lazy computation). You're basically just shitposting.

>>105921092
>>105923362
>>105923588
he's right though, the entire issue with Clojure is that on one hand, it wanted to be "the future of Lisp," but it is true that Clojure thought the way to do that was to take nearly all of Lisp's most powerful features and throw them away.
Macros are neutered, CLOS was scrapped, despite both macros and CLOS being two things Common Lisp does in an objectively more powerful way than any other language.
At some point you have to ask why not just stick with Java at that point? And even then, Clojure failed to even get popularity because he just never understood why Lisp was not popular with businesses. It never had anything to do with the changes he made.
It has already peaked, and it's just going down from this point on. The niche support it's got will gradually fade. Except its peak was lower than the original Lisps, and once a corporation drops it, it's a strictly (far) worse version of Common Lisp, so what reason is there to keep using it?

>TL;DR: The best way to advertise your favorite programming language is by writing programs. The more useful the program is to a wider audience, the better advertisement it will be.
https://www.stylewarning.com/posts/write-programs/

clojure was a mistake

>>105923443
Maybe people would care if it was maintained. F# has a small but dedicated community even outside of what support Microsoft provides.

>>105925539
>>105925523

>>105915659
it's okay, we are getting F# but better anyway

https://www.roc-lang.org/

>>105920070
if you want that use F#

>>105925990
>F#
>microꜱoyft™
No, thanks.

>>105925904
https://en.wikipedia.org/wiki/GNU_Emacs

>>105925523
Asking CLfags to code instead of evangelizing is too much, bro.

>>105926070
Not CL
Also RMS hates CL too.
>Common Lisp is extremely complicated and ugly. When I wrote GNU Emacs I had just finished implementing Common Lisp, and I did not like it much.
https://lists.gnu.org/archive/html//emacs-devel/2010-04/msg00612.html

>>105925921
>Written in Rust
>No .NET
What does this have to do with F#? Might as well just use OCalm.

https://www.youtube.com/watch?v=GedrGWu16_I

>>105924939
>>105925111
Clojurescript is nicer than plain Javascript and i hate Python, those are my use cases mostly.
>binary blobs
You're half-right (it also supports running Clojure libraries that don't use those unimplemented features) but i don't really mind desu, using libraries is overkill for the tasks i use it.

>>105911119
>>105911204
Trvke
OP is a faggot

>>105926151
i did not mention common lisp retard.

>>105925523
>stylewarning
the pedo tranny physics crank from the lisp IRC is not someone to trust, at all.

>>105926166
What's even the use case for this? Another Lua look-alike embeddable language?

>>105926802
Meant for>>105923588

Why should i give a shit
about this lang? looks bloated.

>JVM in 2025
Just use Go

>>105933162
>JVM but worse, slower, less features, and primative GC
Thanks for the bump though

>>105933162
People go to Lisp specifically to escape from Dijkstra's procedural hell.
Go is like old-ass Java but without bloat.

>he doesn't use the metadata feature in order to tag certain values as sensitive in order to redact them in logs

https://www.youtube.com/watch?v=X__jCh_Yj9w

>>105933597
>attach metadata to string
>java.lang.ClassCastException
>attach metadata to vector
>none the seq or transduce functions preserve it
>all functions must use with-meta and copy the value's metadata
amazing "feature"

test

>>105932127
https://www.youtube.com/watch?v=SxdOUGdseq4

>>105935297
Watching it again years later I understand more than I did before, especially when it comes to things involving system design (architectural boundaries). Interestingly, I don't think people writing Clojure use protocols much, they defaulted to this old "inline everything" mindset from CL/Scheme.

>>105933518
Much less memory consumption and startup
>inb4 Graal
Not even Java ecosystem supports this reliably yet
>>105933526
Go as design is niggerlicious, i can't contradict that
>Capctha: PAWG4

>>105936772
>less memory consumption and startup
doesn't even remotely come close to making up for
>slower, less features, and primative GC

>>105934046
that's a man isn't it?
just prefer to know so i can goon more accurately

>>105936803
It does, tuning Jeetva's GC to use less memory cuts down the perfomance to its knees

>>105938203
At least Java can make that tradeoff. Go has no tuning ability, RAM is cheap, and performance is more important.

>>105923468
You will NEVER be a woman.

>>105923362
>Java is a jeet lang because reddit told me so
How about you "go" back to India, gojeet shill.

>>105933526
Go is so niggerlicious only shitjeets use it.

>>105923468
#1 is engineers focused on practical solutions.
#3 is academics detached from reality.

>>105934046
>>105936878
Kill yourself
>>105938595
Industry is so brainfucked by dealing with Node.js and Python they really settled for, quite literally the first crap with a better tooling that had corporate backing. Why didn't Google develop a native compiler for Kotlin is beyond me.

>>105938637
At least it's trending in the right direction. 15 years ago the industry was full of bootcampers using Ruby and PHP.

>>105911079 (OP)
>lisp is dogshit
who could have guessed?

>>105923588
Your post has made me more interested in Clojure than any amount of Lisp weenie cope. I've seen far more damning (and true) complaints about all-around decent languages.

>>105926151
>Scheme is elegant, and it is a better direction to move in.
Absolutely GNU/Based & SICPilled RMS!

>>105938637
>Kill yourself
I accept your concession

nobody in clojure uses metadata and even users of libraries that bother returning metadata (cognitect-aws) aren't even aware of it

>>105939134
As someone who writes Common Lisp for personal projects and Clojure professionally, everything that the retard said is either not true, a subjective opinion, or an exaggeration. Clojure is an effective and a very tasteful wrapper over Java. The only complaint I have is startup time, but its an amazing lisp overall.

>>105941884
why Common Lisp for personal projects?
my biggest love for Clojure/Elixir/Rust is iteration. so, so often we iterate over complex data structures (hash maps, sorted maps, hash maps with sorted maps with sorted sets with integers), and the aforementioned languages make this trivial by exposing generic functional mechanisms for doing such. Common Lisp does not (as far as I am aware). you are stuck in a world where you must explicitly use the correct function for the correct data structure, or must manually wrap these explicit functions with your own interface, no?

>>105939134
>im a contrarian retard
you'll fit right in with the rest of the Clojure community, then
just make sure to never go against Rich

>>105942690
This is bordering schizophrenia
>>105939311
Is the Emacs engine adding support for Guile? Elisp it's falling down the cracks

>>105942468
How does CL not give you this, you have everything you could ever want

>>105944123
please point out what I am missing
and no, Common Lisp does not come with everything I need (such as hash/tree sets). you have to pull in a lib for this, which I am fine doing

>>105920612
Defining an object you're coupling functions with data, which means those functions are no longer usable in larger contexts and this ends up as libraries of (partially) duplicated functions created specifically to work with instances of a specific class.
Having a strong generic abstraction (like seq in Clojure or an array in APL) with a library of declarative functions on top of it is a very powerful thing.

>>105925539
Lisp Curse is lispers shitting on other Lisps

>>105944947
clojure is not a lisp. never will be.

>>105945646
why

>>105911079 (OP)
Are LLMs good at Clojure?

>>105945688
it would be too long to explain.

>>105913263
>>105925539
>>105945646
Sneed, faggot.

>>105945836
Fuck no. LLMs are mostly trained on Python and maybe some JS/TS if they're feeling adventurous. They hallucinate badly on other major languages, let alone something obscure like Lisp or Clojure.

>>105945836
It's alright. They do often hallucinate non-existing functions on the first try, but after pointing that out to them once or twice they come up with working code.

>>105946088
>common lisp
lel

>>105946088
that's because people don't even bother saying "common lisp," they just say "lisp" lol

>>105946593
nice cope

>>105946088
clojure is dying, emacs lisp is now above lel.

>>105946727
>pull request
braindead ꜱoydev award

>>105946643
cope is pretending Clojure hasn't already plateaued and is quietly dying

>>105925523
>The first page of Zig’s results alone have far more programs than the whole of Common Lisp’s corpus under consideration.
Kek

mmmmmmmmmmmmmmmmmmmm delicious seethe