Thread 105838456

New demo for the Jai language:
https://www.youtube.com/watch?v=RIYGaSBKy3w

>>105838456 (OP)
>demo

>>105838456 (OP)
for plugins if it.message it.message(it, message)

>all this shit about metaprogramming
who asked for this?

nice i've been waiting for this talk to be uploaded

>>105838456 (OP)
I am having a hard time watching and especially hearing this video with this mono shit.

>>105838456 (OP)
so he was working on his language for over 10 years, is he a neet or how did he earn money in this 10 years?

>>105838456 (OP)
crazy, this looks really good what he did but I think we don't need another programming language.

>>105838622
"for plugins if it message then it message of it and message"
lol no hate on the syntax, just observing that this example is a clumsy sentence to convert to and say in a natural language (and this sentence too.)

>>105839193
>Get millions from successful game
>Invest it into dividend paying stocks that prioritize dividend growth
>Live rest of life off of dividend yields that increase every year.

So, has he released the compiler to the public yet?

>>105838710
sounds a bit complex. I thought he hated complexity.

>>105839905
if he did that it could be used by the unworthgy

>>105838456 (OP)
JAI SHRI RAM
JAI SHRI RAM
JAI SHRI RAM

JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN

JAI SHRI RAM
JAI SHRI RAM
JAI SHRI RAM

JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN
JAI SHRI RAM SOKOBAN

>>105839847
I'm pretty sure he mortgaged his house by now but yeah he could have done that.

I wonder how many people here on /g/ actually have access to the compiler beytah. I have it. It's not hard to get access to if you're not a retard. Like just ask him lol

>>105838456 (OP)
holy fuck he's old now man wtf

>>105840792
Why bother if Odin and C++ exist?
Jai is stupid as hell man.

>going to a functional conference under the pretense of shilling for a new systems language
>immediately start shilling your game
this guy has balls

I gave up, he being is extremely interactive with the video but it's delayed by 20 seconds. What the fuck. It's unwatchable without constantly fast forwarding to see what he's talking about. Hopefully this was just a throwaway talk and he does another one.

>>105838456 (OP)
joe blow

Stopped watching here. A shitty Sokoban game with a level editor shouldn't take 300k lines of code. Jai sucks

>>105840736
>I'm pretty sure he mortgaged his house by now
Why are so many people so shit at anything involving money.

>>105841516
3D games unironically require like 100x the code than that of a 2D game. Some people just can’t resist making their lives hard while most of their game mechanics would probably work just fine in 2D, certainly for just a Sokoban game.

>>105842181
Jon Blow isn't going to make a 2d game to show off his c/c++ killer, get your head out of your ass

>>105838456 (OP)
Jonathan blowjob
Jai sri ram

>>105840593
This.
Jai is a language for brahmins and kshatriyas.

>>105842181
He mentioned before he'll release most of the source code (under a special license) for Sokoban to serve as a full example of the language.

>Braid and TheWitness
But those suck. Why is this guy such a big deal?

BHARAT MATA KI JAI

JAI MATA DI

>>105842963
>>105843080
why are you people on /g/

>>105838456 (OP)
>first time I hear about it is on /g/
>unemployed tards immediately start arguing over it
So the gist is this language doesn't need to exist and almost nobody will ever use it?

Hi Jon. You're 15 years too late. Personal computing is long dead and it's never coming back. Sorry.

Enjoy your retirement. I'm serious.

>>105838456 (OP)
damn, that intro slaps.

>>105842181
I've written a 3D game engine with a low-level graphics API and it doesn't approach anywhere near 300k, after the initial ~3k lines of boilerplate you would be surprised how easy it is to get skinned meshes drawn to the screen with shadows, post processing, etc

post source code of gtfo

>>105843568
engine is proprietary, sorry im not a freetard

>>105843623
Then it won't succeed, sorry. Competition with Godot was feasible for you, but on Unreal Engine level going solo?
Grow a community, it's ok to get help. There a lot of awesome developers around waiting for the right chance. You're better than this.

>>105841516
For some reason, Blow does his state control like a caveman, so he litters a bunch of if bool checks everywhere.

>>105844620
How the fuck do you do conditionals then? Polymorphism?

>>105844668
state machines to start. I have no idea why he doesn't use them for a game like his. He's wasted so many hours manually debugging state issues, and he ends up "fixing" them with the same ad-hoc if bool "guards"
And since he already has a transaction system, he should be using contract programming too.
But I imagine he considers both of my suggestions bad and has rants for each on why they're actually not useful or productive.

>>105838456 (OP)
In 10 years he'll release something that's just a crude version of rust 0.1.

>>105842934
Wait he literally did, retard. If that codebase is unable to show the essence of the language then maybe it’s overcomplicated and shit.
https://youtu.be/jqtHi9XRhr0

>>105838622
Saars we invant the Jai Hind language frum the omakase principle of Ruby and the Rails okay?
It is like magic. You write the coding in natural flow of the langbage. Just like me speak the english.
Johnanthan sars is the icon of Jai has tens of years of experience in building perfarmance critical applications such as video games which have tens of downloads.
You are a nobody.

>>105838456 (OP)
So he made an entirely new language just to do what visual studio profiling already does?

>>105844775
> You are a nobody.
no need to hate on yourself this much mate

>>105842181
>100x the code than that of a 2D game
it's literally +1 axis and that's it, what are you on about

>>105847046
Holy shit, nocoder. I hope this is bait.

>>105842074
let me guess: you are a daytrader

>>105844709
>But I imagine he considers both of my suggestions bad and has rants for each on why they're actually not useful or productive.
I suspect he's not an idiot and in practice the more sophisticated methods have important drawbacks, yes
I've only dabbled in game development, and state management has been the most difficult part for me to do cleanly
I either do it in the simplest most straightforward way (which I suspect will not scale well???) or I get lost in over engineered garbage that makes me want to kill myself before I even dare to touch it again
I've been reading and watching youtube videos with multiple different approaches, and nothing feels satisfying to me
maybe blow just went down that road and figured out the "dumbest" solution works better for him

>>105838456 (OP)
Is this video's audio right panned or my speaker is broke?

>>105847416
It autofixed itself.

/SOLVED

>>105844709
People shit on pirate furfag and yandere dev all the time but jon’s code also screams unmaintainable to me. Every time I open his streams he’s balls deep in a 20k loc file within like 7 layers of indented if else blocks. Maybe it’s just my OOP modern C++ mental illness, but his code doesn’t look very nice to work with.

>>105842074
Intentional lack of education on the topic. You can't let the people know hot to get ahead, and if you do that's eating into your profits. Why help them when you can exploit them.

some fucking demo, he just showed his game and then skipped through a shitload-of-loc files and talked through it. wow the compilation took 2.3s but its too slow, what the fuck are you doing with your life you fucking nerd.

>>105847546
Multiple people have made multiple languages in the spare time they had waiting for C++ builds to complete.
The irony in talking about wasting life while spending time defending slow compilation that is quite literally an objective waste of everyone's time.

>>105847478
>but his code doesn’t look very nice to work with.
it's classic solo-dev code.
his co-compiler dev has only talked about it being fast, and jon being smart, but never compliments jon's code.
Which is an achievement in itself, despite all the bad, blow atleast actually makes things, which is his generation of programmers i guess.

>>105847610
He used to do contracting for game studios in the 2000s like Ion Storm Austin.

>>105843423
cool, now write math code, add some editor features etc.

the renderer is not the issue when that dude said 3d games usually have more loc

>>105843623
kys shilltard

>>105843423
>I've written a 3D game engine with a low-level graphics API
Let's see it then

>>105847610
>his co-compiler dev has only talked about it being fast, and jon being smart, but never compliments jon's code.
you sound like an insecure girlfriend

>>105843423
Retard

>>105847572
im not defending it, i live that shit daily my nigga. just go outside when its compiling or at least dont complain about 2.5s compile times.

>>105840571
he's entirely for complexity when its complexity he understands

i know types filter out 90% of the workforce (and 100% of this board by extension) but a lot of the shit he complains about is a non-issue with types, and a lot of the bugs he solves on stream could've been solved with types as well.

>>105848155
>a lot of the shit he complains about is a non-issue with types
what

>>105848155
>"with types"
sounds like you don't understand them either

Impressive compile times in jai, cool. :-)

jblow is solving circular mental puzzles of his own making rather than just creating a good game. There is absolutely no reason to create your own programming language for game development. It doesn't solve any real meaningful business problem, and in fact it creates a lot of new ones.

Unfortunately the guy is now knee deep in tech debt and has staked his public reputation on this terrible mistake and now its too late to turn back. He'll look back on all the time wasted with lots of regret.

I'd guess he is in a position in life where he doesn't ever need to hear criticism. He certainly doesn't seem open to hearing any.

>>105838456 (OP)
C-descendant syntax is so fucking ugly. I don't understand how anyone writes a language today and decides to go anywhere near it.

>>105848199
maybe he is developing a new programming language because there is a need for one and not to develop his game - what are you even saying you retard? do you think he wrote that language only to develop his sokoban game? why do retards like you come up with the weirdest strawmans.

>>105848199
>jblow is solving circular mental puzzles of his own making
couldn't have put it better myself

>>105847572
Slow C++ compilation speed is a skill issue and is nothing compared to rust. Have more than 4 cores in 2025, use forward declaration or precompiled headers and your build will be sub 5 seconds.
t. heavy C++20 ranges and boost user

>>105848179
>>105848177
modeling effects as types and having a type hierarchy for your data models makes it easier to do a lot of the shit he complains about, like serde and having specialized behaviors for specific actors, and not having stupid shit like his "wall" entity having a "is charmable" attribute.

if you're still confused then get a job and read a few books you might understand then

>>105848231
what does he complain about? give me a concrete example here

>>105848216
no shit, you are also a dumb retard who didnt work on any projects with actual complexity and performance goals.

>>105848231
>modeling effects as types
Anon I know you're not about to shill effects systems to the retards on /g/ who experience immediate neural shutdown the moment the word "monad" is used

>>105848238
I gave you three concrete examples he talked about, re-read the post. You do know what serde is and what it means in this context, right?

>>105848241
wow did I strike a nerve?
nothing jblow is doing is practical, he's just solving problems that exist only in his own mind

>>105848217
sub 5s for *incremental* build before some sperg doesn’t believe me. full rebuild of like 30 cpp files is around half a minute. Also incremental compilation is a solved computer problem so I don’t see Jons reasoning for not supporting it. It is not 1990 anymore.

>>105848258
it is exists in our codebases - maybe its just you being super smart, right?

>>105848257
Serde is a serialization library for Rust iirc. When he has complained about this? What's wrong with having a wall entity and a charmable attribute? I don't understand the actual problems here

>>105848263
what exists in our codebases?

>>105848263
>problems that only exist in shitty imperative languages
>let's solve this with a new shitty imperative language instead of using one of the decades-old functional languages
big smart

>>105848270
yeah great, lets write the new triple a game in a functional language - this dude solved all our problems!

>>105848270
what problems?

>>105848271
Right, retards on 4chan are writing AAA games. I forgot.
>>105848272
No idea since nobody will say. I'm just baiting retards.

>>105848278
>Right, retards on 4chan are writing AAA games. I forgot.
i have met a lot of people from the industry here actually or vice versa.

What are you writing?

>>105848278
>I AM JUST BAITING HAHA no dude wtf its not like I was wrong, I was merely pretending to be right

>>105838456 (OP)
They reuploaded the video with synchronized audio.
https://www.youtube.com/watch?v=IdpD5QIVOKQ

>>105848254
its just funny to see the unearned arrogance

jon blow would fit in here for sure

>>105848293
I think you lost a few of the messages in your context window bot, try again:
>>105848270
>>105848263
>>105848258
Notice how the term "problem" keeps being used throughout the conversation but nobody ever defined a concrete problem. That means that my first post >>105848270 was likely written tongue-in-cheek since I'm claiming functional languages solve the mysterious problems that are yet to be defined.

>>105848311
we are talking about the problems he mentioned all the time. i thought you were aware and actually know what you are talking about. it is in the fucking video lmao.

i know this is not bait, stop acting like it, you are just low iq.

>>105848311
Your post sounds exactly like something a geniune FP retard would write

>>105848317
What problems does he mention all the time?

>>105848322
skip through the video.

>>105848317
>we're talking about the problems he mentioned
>what do you mean name ONE problem, it's in the video!?!?!
lmao
>>105848318
nice contribution retard

>>105848327
I did
The only thing he really complains about is build system languages, which is not a problem so much as something he doesn't personally like

>>105848329
you just sound mentally ill at this point, if you were trying to be funny or troll then it didnt work

>>105848335
>still unable to name one problem Jai aims to solve
lmao

>>105848348
I'm also asking that question

>>105848322
>>105848327
actually you know what, i will be nice to a retard for once:

"modern" languages introduce garbage collection, abstractions, oop etc. to handle complexity and issues like memory management etc.
while those approaches have benefits they also come with a cost.

jonathan blow claims that the cost benefit analysis aged badly and we should rethink what actual approaches one should attempt.

jai is a experimental programming languages that ignores lots of "decisions" made by modern languages and attempts to come up with their own solutions to common issues.

e.g. memory management: instead of the runtime handling the memory management, what if we give the developer more tools and ways to a) control memory more easily and b) have a better overview of what is actually happening statically.


happy?

>>105848358
So Jon invented manual memory management, something we already had in C and C++? Bravo

>>105848283
everyone send dm pls if you work in the game industry

>>105848358
>linear types or affine types
you don't need a new language for this

>>105848364
nope never said that, have a good day though!

>>105848373
Jon's language does nothing to do with memory management that you can't do in C++

>>105848366
>yep you know what? the answer is types!
>what do you mean we cannot use my favorite functional language on this soc? did you try types?! what about linear types? maybe quadratic type!??!

>>105848373
>attempts to come up with their own solutions to common issues.
nta, but what solutions lmao

>>105848382
seems like you arent aware what jai is actually.
wont bother responding to you at this point anymore. id rather argue with types dude

>>105848392
watch the video

>>105848401
I'm aware what Jai is, I also watched the entire video
It doesn't do anything with memory that you can't do with C++

>>105848387
Correct. There are a dozen functional languages and even a few imperative ones that already implement linear or affine types. They solve the issues of bad ergonomics of manual memory management without relying on GC. This is a solved problem and Jai is bringing nothing new or interesting to the table.

>>105848405
ill take the bait once again.

a few things c++ cannot do but jai can:
- actual comptime stuff that helps with static analysis in many forms.
- that whole context stuff
- actual sane "generics"

>>105848401
>watch the video
Bro literally showed visual studio profiling tools of ““code style”” and shilled his game which memory leaked like hell, just like you would in C/C++.

>>105848431
You're correct, but it's got nothing to do with memory management
The question here is what problems is Jai solving
Still haven't recieved an answer

>>105848411
you still didnt show me a single functional language that could be used in those fields. those fields being ANY field near to hardware where and high perf or latency requirements.

>>105848444
how is checking for memory leaks, memory allocations etc. statically not memory management? thats the whole idea here, don't manage the memory during runtime but manually WITH significantly better tooling and analysis.

>>105848461
>how is checking for memory leaks, memory allocations etc. statically not memory management?
It is, but it's something you can do in C++

>>105848461
>>105848468
Sorry I missed the "statically" part
The static visualizations Jon does are cool (can be done easily with a linter though) but they don't actually solve problems, they just help you look at your code

>>105848468
the tooling in C++ is very slow - most games I have worked on could not even stay sub 100ms with sanitizers etc. on. You know that - so why even compare it? The point is that in Jai its easier to use and causes less perf pressure ergo your game still stays "interactive" instead of being a slide show.

look dude, ive been working in C/C++ for a while, there is a lot of tooling and you can do a lot with C++, its just not great and no actual C++ developer (especially in game dev) is happy with what they got.

>>105848451
Because you're not being sincere. Real Time Haskell has been a thing since the '90s. Since the introduction of linear types https://ghc.gitlab.haskell.org/ghc/doc/users_guide/exts/linear_types.html you don't even have to use hacks; just disable GC and live entirely in Linear types. There are forks of the runtime that you can use that cover whatever stupid "but it won't run on a smart toaster from 2004" cases. Now you'll move the goalposts to "but people aren't doing it right now" and that's almost entirely due to the power of retards like you pretending it can't be done or that functional programming is too ivory tower for normal people. It can be done and it's only too hard for true retards like you, not for normal people.

>>105848491
We're talking about his linter for Jai that dsiplays a graph of various shit in your program right? That's not a runtime tool

>>105848485
>don't actually solve problems, they just help you look at your code

thats the idea - give more power to the developer to solve the issues instead of having an authoritarian runtime that costs more than it does - at least in low latency stuff.

>>105848485
>they just help you look at your code
This is what happens when you treat your codebase as art. It goes against all engineering principles and is a characteristic of a junior dev.

>>105848504
You can write linters for C++ too

>>105848519
nope, not linters like that. read properly, if you dont understand some words, you can kindly ask a fellow retard here what the words mean!

have fun learning programming dude!

>>105848500
are you seriously suggesting haskell for fintech and game dev? lmao, why did i know that you were some haskell retard.

>>105848527
>nope, not linters like that.
Yes you can
All the stuff Jon is doing with his metaprogram can easily be done with a C++ linter
Not that it's particularly easy to parse C++ code but it's not hard either

>>105848485
>hey just help you look at your code
what else are you writing as a developer?

> HEY this tool doesnt solve my problem!! ARGHH i can only see the TRASH i have written more clearly

>>105848527
there is a shop I know of that uses haskell and jane street uses ocaml

you must not be in fintech (or any job for that matter)

>>105848536
Its a neat tool, but it's not solving any big problems, which is what he talks about all the fucking time so it just makes him look foolish when he talks so big and shows off something that essentially a hobby project

>>105848540
>i know this one shop in bangladesh, they are very hip!

>>105848534
oh really? great, i did not know! finally, dude. where can i get that linter? does it work on my custom allocators? wait, it misses some allocations SOMETIMES? what do you mean i have to run it a few times and get different results?

just stop replying to the FP retard, they're unreachable

>>105848527
https://www.janestreet.com/
https://serokell.io/projects
You should've stuck to just saying "games" retard. ML languages are massive in Fintech.

>>105848550
I'm saying you can make a linter yourself, I'm not saying you can get one, maybe you can I've never used C++ linters

>>105848552
you are right, im outtie.

peace

>>105848552
>>105848560
dude had to samefag himself a reddit-tier exit because he claimed functional langauges aren't used in Fintech

>>105848557
>massive
>less than 0.1% of all codebases in fintech
ok, kid.

>>105848558
do you even know what writing a linter that could do all that means for C++? what are you even saying? of course you could, you could write anything in any turing complete language. thats not the point.

>>105848582
>do you even know what writing a linter that could do all that means for C++?
Yes
The stuff that Jon is doesn't require complicated program analysis

>>105848568
why do you get this mad for being wrong? major issues this dude. peace out, brother.

>>105848582
source?
By the way, Jane Street is hiring for several roles that all require OCaml RIGHT NOW:
Base salary is $200,000 - $300,000
https://www.janestreet.com/join-jane-street/position/4274288002/
Base salary is $175,000 - $300,000
https://www.janestreet.com/join-jane-street/position/5108180002/

>>105848591
>Yep, its REALLY easy - but nobody has done it yet though idk but its easy and everyone would love to have it.

>>105848604
since when is ocaml an alias for haskell?

>>105848612
>but nobody has done it yet
Who the fuck said that? I'm sure there's countless C++ linters

>>105848612
try harder next time

>>105848622
i asked you several times, show me a linter that shows me where my code is leaking. i would love to chug into my emacs config.

>>105848629
I have never used a C++ linter

>>105848206
>C-descendant syntax is so fucking ugly.
as apposed to ?

>>105848634
>apposed
https://www.erlang.org/
https://ocaml.org/
https://www.idris-lang.org/pages/example.html
https://www.haskell.org/
https://lisp-lang.org/
https://racket-lang.org/

>>105848647
Watch him call them all ugly because they don't look like C

>>105848231
>type hierarchy
oh no.
we already walked down that path in the 90s.
but you go right ahead, and don't let us stop you.

>>105848670
In unsound type systems; apples-to-oranges.

>>105848647
are you the "curly brackets are ugly" dude?

>>105848647
i love erlang, but prolog langs are pretty noisy and ugly at times

>>105848647
thanks but not thanks

>>105848686
>prolog langs
not a single one of those languages are descended from Prolog

>>105848719
>erlang isn't descended from prolog
anon...

>>105848723
It's like saying Lisp was descended from assembly because that's what it was implemented in

>>105848734
anon, erlang was literally first written in prolog.

>>105848741
Anon, Lisp was literally first written in Assembly. Did you read the post you're responding to?

>>105848743
ok man, good luck in life

>>105848673
>In unsound type systems;
i'm a system programmer.
on the side i also uses Go a lot and some c# (and thankfully i'm not using java anymore)

for system programming, "types" that are implemented by the compiler itself i.e. but by a library, and relay on facilities that are inaccessible for the programmer is SWEEPINGLY a no go.

end of conversion.

>>105848914
>"types" that are implemented by the compiler itself i.e. but by a library, and relay on facilities that are inaccessible for the programmer is SWEEPINGLY a no go.
This is incoherent, please try again.

>>105848934
remove
>but by a library

>>105848937
An example would help. Or are you saying that built in types like int, char, float, bool, etc., are bad in C?

>>105848950
i'm responding to
>>105848231
>>105848673
an not-so-good anecdote would be mutable references
not-so-good anecdote would be rust's
the last time I tried crating my own "container class" in rust, i ended up throwing so many RefCell<T> I just drooped rust all together.

>>105843083
Because the name Jai makes it sound like a streetshitter language.

>>105842181
you are literally going to get B T F O eternally when the game sokobans you from interacting with 4D space and revolutionises gaming.
QWOP ain't got shit on the sokoban king
it's gonna sokobang your brains out souljaboy you gon' get told

>>105848217
>>105847572
or just avoid templates and STL entirely and your C++ will compile as fast as C

>>105849071
Best (You) in a while, thank you anon.

>>105849095
But what if I want to use stl and templates smartass? I’m not writing hashmaps from scratch in current year.

>>105848451
2025 and i am forgotten
https://antelang.org/

>>105849912
You can't be forgotten when you were never known.

>>105838456 (OP)
we live in the age of llm vibe coding. no new language will ever take off

>>105840571
complexity in the right place allows you to remove a lot of complexity overall -- me

>>105849912
>written in Rust
https://github.com/jfecher/ante

>>105850197
quite the opposite, new languages (when a good one will be made) are what is going to bring down vibe coding

>>105851272
RAII and smart pointers make me not have to think about memory and I think they are great. But noooooo, THAT kind of complexity is bad because Jon blowjob told so. What even is “complexity” anymore? It is like saying your code is “clean”, same energy.

>>105852498
>What even is “complexity” anymore?
Stuff Jon Blow is filtered by.

>>105852498
I had more in mind using metaprogramming to make generic data structures or custom allocators with a lots of option in such a way that you produce the exact function definitions you want without having to rewrite yet another X data structure or allocator. This could also apply to parsers and decoders in general.

About RAII and smart pointers, this is overhead that can be avoided using other methods. A combination of custom allocators and powerful type systems.

>>105852585
Doesn't matter.
Most games are GPU bottlenecked.

>>105847701
>>105847789
>>105847863
It really isn't that hard. Unironically a weekend project

>>105838456 (OP)
Jai Shri Krishna

>>105852602
this. The Witness runs like shit on my 2020 laptop with integrated graphics card. What is the point even of all this mental masturbation as a game developer besides having 3% CPU usage in task manager instead of 10%. CPU fags need to btfo and upgrade their pentiums. Almost all C++ code is still lightning fast compared to any webshit or python ever written.

fucking subhuman jannies

It's embarrassing how much stupid and useless this visualization is. Even the layout is braindead, it should be some kind of fibonnacci thing.

>>105849139
>But what if I want to use stl and templates smartass? I’m not writing hashmaps from scratch in current year.
deepseek can write a templateless hashmap for you within seconds

>>105854053
nigger

To be honest, Jonathan kinda seems up his own ass a little. This language isn't really solving any problems, and it just feels like a waste of time. He has that "105 IQ kid that grew up in a 90 IQ town" kind of vibe.

>>105838456 (OP)
>blowjob

Did he got hit in the head or what? I like his talks and rants but here he didn't say anything interesting at all, he spent all his time awkwardly presenting his dogshit memory visualizer/introspector.

>>105854484
>I like his talks and rants
did you get hit in the head?

>>105848601
Why do people post these (YOU) screenshots like are they supposed to prove anything? Every retard and his dog knows about inspect element brohsef...

by the time i have access to Jai i'll be an ATS expert wizard cyborg with no use for it

>>105842934
>c/c++ killer
I hope he doesn't really believe this

>>105855884
based. Did you managed to make some proofs?

>>105848208
>he is developing a new programming language because there is a need for one
no jonathan there isn't

He should just write a C# compiler. Because it already solved programming. You can't make a language better than C#.

>>105838456 (OP)
>DUUUUDE
>WHAT IF WE CREATE ANOTHER ENGLISH PROGRAMMING LANGUAGE
*Gähn*

>Dude just write compiler for your compiler
Jai "metaprogramming" is retarded

>>105842181
>lol lol why is he making game in a niche market of 3d puzzle, is he stupid
wait until the release and the content starving streamers slurping it up

He sounds like complete schizo with all that toy programs he is showing. He is trying to sell basic profiling tools as some killer feature, lol. Completely deranged.

I watched ~1hr and he didn't show any jai code.

>>105856038
>german programming langs
i really don't want to learn and type 50 character keywords

>>105847265
Just outed yourself as a nocoder there, mate.

>>105856157
this pisses me off, he was supposed to talk about why metaprogramming is important

even this perl profiling library is better
https://www.youtube.com/watch?v=T7EK6RZAnEA

>>105856011
D is pretty much what he actually wants (especially with standard library v3)
But he's too autistic to overlook flaws or hurdles like most autists, so he has to NIH

>>105856227
he was intending to show the power intercepting messages from the compiler provides and the kinds of things you can do with it. I think he just dwelled on the results too much

>>105856420
he's 20 years behind c# compiler as service
https://learn.microsoft.com/en-us/dotnet/csharp/roslyn-sdk/compiler-api-model

>>105856064
He better fix his marketing strategy as no one besides blowfans knew about Braid anniversary releasing.
>>105856221
>3D is just an extra dimension
fucking retard. where are your PBR shaders, normal maps, lighting, vertex skinning, model loading, matrix calculations, …, in your 2D pixelshit game?

>>105842181
the puzzles themselves are 3 dimensional

>>105848647
>lisps
You almost had me until I saw the punchline

>>105856205
kek, checkem aus dieses kode:
#Programmhinzugabe<stdio.h>;
Integerzahlenvariable FrankfurtAmMain {
DrucksachebuchstabeFwieFrida('Hallo, Welt!');
Programmausgabenrückgabewert 0;
}
>UND DAS HEISST EEERIKA BUMMBUMMBUMM

>>105856571
Nice goalpost shift. The discussion was about LOC of the entire project, not just the puzzles/gameplay part.

>>105856644
this makes me wonder what actual german java looks like

>>105839847
I don't think Jon believes in stocks, he just dumps everything into the next project win or lose. That said it should be difficult to piss away $20 million.

>>105856664
>>105856571
I thought someone else replied. Ignore this shizo rambling.

>>105856242
D is shit, stop suggesting that crap
most of its features are half assed implemented and bugged. its a sandbox for the creators to play around, thats it. they always have to go for the latest trend, implement as prototype and leave it as is. its pathetic really. even the GC leaks, thats how bad the backlog of bugs is.

>>105856693
the greater point is contributing to improving something, instead of fragmenting things further. Made worse by the fact of Jai not even bringing anything new.

>>105856693
>half assed implemented and bugged
>sandbox for the creators to play around
and Jai is somehow different?
>the GC leaks
Jons game leaks

>>105856708
>>105856709
both D and jai are shit. better now?

>>105856786
no because i'm sick of programming langs staying shit

>>105838456 (OP)
It's kind of wild to me just how many people seem so bothered by him for one reason or another. Like, whether his language is good or not, we'll see when it comes out. What's the big deal? If it's trash, we'll find out. And if it's good, then great.
Why get your panties all wet over something that's not even out yet? For the time being, I think it looks promising and I'm impressed by what it seems to be able to do. I personally couldn't have made that.

>>105838456 (OP)
In 20 words or less, what is the point of this language, what does it bring to the table that we didn't already have.

>>105838710
everyone asked for this.
real metaprogramming tools RAPE C++.

>>105856011
There's already Beef, a language written by a game developer, for game development.
https://www.beeflang.org/

>>105857708
>let rand = scope Random()
>defer delete e2;
it would be nice if it was possible to compile C# with beef compiler.

>>105857591
fast compile times and sane error messages even when using templating
contexts that make custom allocators way more convenient
being able to statically run literally any part of your code as part of the compilation process for metaprogramming
hygienic macros
a build process that doesn't use any external tools, everything is defined in the language itself

>>105857708
>class Program
first line of real code and I already don't want to use it

>>105857761
You don't need it.

This works too.
using System;

static
{
static void Main()
{
Console.WriteLine("Hello");
}
}

>>105857853
why do you need the first enclosing static? and why does Main need to be static?

>Console.WriteLine
I'm sorry but everything about this just sucks.

>>105857761
>waaaaaaaa there is too much syntax and "oop" for my brain to handle
nocoder argument
>>105857743
>a build process that doesn't use any external tools, everything is defined in the language itself
zig exists
>>105857354
>Why get your panties all wet over something that's not even out yet?
Because calling your software "not done" indefinitely doesn't shield you from all criticism.
>I personally couldn't have made that.
I wouldn't either as Odin, Beef C++ and C# aot already exist.

>>105857354
>It's kind of wild to me just how many people seem so bothered by him for one reason or another.
Because he constantly talks shit about everything

>>105857969
Enclosing main in a fucking class is peak niggercattle retardation and a prime example of what OOP brainrot does to someone. It tells me everything I need to know about the language. I have no interest in using it.

>>105857993
>Because he constantly talks shit about everything
when it comes to the absolute state of modern programming, he's usually right

>>105858045
Yeah you should listen to him talk about something that isn't programming

>>105858055
there is a reason I don't, he's only good at 1 thing

>>105858029
>using a class with only static methods is OOP
the state of cniles

>>105858089
calling functions "static methods" and putting them in a class is OOP

>>105858107
The compiler output isn't any different, so no.

>>105858140
The compiler output doesn't determine what is OOP or not

>>105838456 (OP)
I don't see how this programming language is good for gaming when it's between C and C++. Maybe it's good for simple games but with the complexity of nowadays games you need guaranteed and thread safety. Micro-optimized performance is not a requirement anymore.

>>105858150
It does. There would be an allocation if you put your main logic in the constructor of the main class or in a seperate run method (like Java fags do). Using the "class" keyword with only **functions** marked static would just behave like a namespace in the case of C++.
The code in >>105857708, >>105857853 is not OOP at all.

>>105858208
You don't know what OOP means, it's a human design paradigm, it has nothing to do with what code is generated

>>105858029
That's legacy thing of Java and C#. If you work in Typescript, Scala or Kotlin you don't need to do that. No they're not going to change it. I'm not sure about C# but I think they do the same.

>>105858199
how can people like you spout such incredible nonsense with so much confidence

>>105858240
I don't know. You tell me since you wasted your time with a pointless reply.

>>105858199
>with the complexity of nowadays games you need guaranteed and thread safety
C++ isn't thread safe though

>>105858228
C# now supports top-level statements, but no project more complex than FizzBuzz is likely to use the feature.

>>105858219
It's a human design paradigm that makes you write code in a different way that will impact the actual code generated (negatively). You are just sperging about definitions now.

>>105858279
>that will impact the actual code generated (negatively)
It won't actually, OOP translates to assembly pretty well, that's one of the reasons it's a thing

>>105858262
It's not. C++ uses the same mechanisms as all other imperative languages, so locks, high level structures, and maybe async which is just a fork-and-join thing. I don't know what makes you think it's safe compared to something designed for concurrency like the many functional programming languages out there.

>>105858208
There is nothing more OOPbrained than taking the object model and applying it literally everywhere, even places where it makes absolutely no sense. I would say the hallmark of a true cultist is applying it especially in places it makes no sense. Only someone suffering such delusions would dream up something as retarded as dumping functions into a class that has no data to begin with because they want to encapsulate their entire program in an object. You literally cannot get deeper into the OOP mindset than that.

>>105858294
The post says "with the complexity of nowadays games you need guaranteed and thread safety" yet C++ isn't thread safe and that's the language people use to make games so the statement is false

>>105858287
Things like polymorphism make you use heap memory all the time (at least in C++).
>>105858295
>dumping functions into a class that has no data to begin with because they want to encapsulate their entire program in an object.
The functions are STATIC in the Beef example. "Program" is literally just a namespace here. "Program" is not an object.
>taking the object model and applying it literally everywhere
If you call your source file program.cpp or entity.cpp, you are already applying the object model in your head. Checkmate atheists.

>>105858287
nta but 1 potential negative (in terms of speed, rather than mapping) are additional tables for dynamic dispatch

>>105858364
>Things like polymorphism make you use heap memory all the time
Polymorphism is optional, method calls can be determined static at compile time and turned into regular calls

>>105858372
Yeah, that's literally the only runtime cost of OOP, and like I said it's optional and easy to optimize out

>>105858313
That's your retarded interpretation. Ideally you don't want to deal with that at all when developing a game. That's why complex games use engines like unreal, so they don't have to deal with concurrency at all. It's one way to fix the problem, let the engine developers deal with that, hiding the problem. Another solution, if you want to make a game from scratch without an engine, which seems to be the case here, you use a programming language that can actually deal with concurrency properly, and that's not C++. It's very difficult to have thread-safety with C++ and it's not going to be that performant. This programming language doesn't fix that problem.

You are confusing what's a good idea with what people use. C++ is used in gaming because of momentum and predictability, not because it's the ideal programming language. It's not. Takes a lot of effort and time to get things right. The advantage is predictability (pointless in concurrency), potential low memory footprint (we have at least 16GB in PCs), and micro-optimizations (this is a real advantage).

>>105858107
You're somewhat right. Nearly all language constructs in Beef are internally represented as objects, but that is what makes the language extensible.
Like it says on the website, Beef is multiparadigm. The language itself is expressive enough that you rarely need to use Gang of Four's design patterns. Inheritance is usually discouraged.

I almost ever have to write a class. The POD struct is good enough.

>>105858378
You sound like you haven't made any games at all

>>105858374
Function overloading isn't really the polymorphism most people think about.

>>105858394
I did and that's how it is. If you don't like it than go pick Jin and try to make it concurrent and see how it goes.

>>105858397
we're talking about virtual methods

>>105858364
if it was a namespace it would be a namespace, not a class. its a class. why is it in a class called Program in the first place? that’s just nonsense

>>105858400
wtf is Jin?
When it comes to threading you can have safe or you can have fast, you can't have both
Games pick fast so they use C++

>>105858376
>that's literally the only runtime cost of OOP
the main cost of OOP is horrendous cache coherency

>>105858410
I meant Jai, not Jin. You don't know much about concurrency. It's very obvious. You can only really have safe and fast. Your "fast" means you're program will eventually crash because it's not safe. Game developers are not very good at concurrency. Desktop and web developers have a lot more experience in that department, especially with distributed systems. Game developers usually make it single thread, fuck up concurrency, or use an engine to deal with the problem.

>>105858420
OOP doesn't determine your memory layout

>>105858435
You don't know anything about game development but somehow you think you're an expert on the topic
Fast means sharing data between threads, safe means not sharing data between threads, that's never going to change

>>105858458
Sharing data between threads is inherently unsafe and slow because now you need to use locks, and locks are not fast.

>>105858474
>now you need to use locks
No you don't
Who's the one who doesn't know anything about concurrency exactly?

>>105858486
So what do you use when you're using data everyone is writing and reading? Black magic? Or your application is simply not safe?

>>105858497
You MAY need locks when you share data, but you design your systems to need as little locking as possible to make them faster

>>105858506
No shit but the moment you use locks you already fucked up performance. It's not faster to share data, unless it's read only. It's also a lot harder to make it correct. There are ways that don't use locks, but C++ is not very good at it.

>>105858404
It's just a keyword bro.
>>105858403
I don't program my games this way but you need the heap in order to be able to do things like dynamic_cast Entity to EnemyEntity or call a virtual method at runtime because the size is unknown. You would need a union with all the possible entity classes in order to avoid heap allocation for every entity.
>>105858438
it mostly does

>>105858372
Not true. Watch this https://www.youtube.com/watch?v=i5MAXAxp_Tw

>>105858522
The alternative to sharing data is copying data, and sharing data is faster than coping data
It's hard to make it correct, but that's the price you pay for speed
Writing lock free code has absolutely nothing to do with the language you're using, unless it's forcing you to use a particular concurrency model
You have peak "dunning kruger web dev" energy

>>105858527
>SHACHAR SHEMESH
Oh no no no C++ bros

>>105858524
>You would need a union with all the possible entity classes in order to avoid heap allocation for every entity.
This isn't true

>>105858532
It does have to do with the language because for example in Haskell you guarantee functions are pure, in C++ you don't. It also has mechanisms for parallelism with streams with green threads. In C++ you don't have such a thing, you have to do it explicitly. Haskell also has STM to deal with it like a database. C++ also has that but in a very cumbersome way to program. Haskell also has a thing borrowed from Erlang to deal with distributed systems, which is called today reactive programming, and only works properly with immutable data.

The problem I see here is you only know imperative programming concurrency, which is cumbersome and explicit.

>>105858573
I should have known you were a functional weenie with no actual programming experience
Your argument boils down to "it's hard". Yes, it is hard, but that's the price you pay for efficiency
You can make concurrency safe and you can make it easy, but that makes it slower or less powerful, this is just an unavoidable fact

>>105858582
No, it's not about being hard. It's hard, unsafe and inefficient. You have to compromise with C++. Also even Java is better at concurrency than C++. C++ is kind of awful in that regard. Yes Java has many other problems, and it's not the best in concurrency even in the imperative world.

>>105858594
I don't think you understand what's efficient and what isn't
C++ is hard, unsafe and efficient
You can make things easy and safe by introducing inefficient abstractions like immutable data

>>105858615
Okay I'll leave it at that. I should go back to work anyway.

>>105858625
You're a fucking idiot

>>105858537
How would you do this otherwise? I will watch the cpp talk later because I am curious and need to stay ahead of the jeets.

>>105858637
With inheritance you can't know the size of an object at compile time, but that doesn't mean everything has to be on the heap, it doesn't stop you using arena, pool, bump allocators

>>105858648
Oh okay, makes sense. Unfortunately the C++ STL sucks for this, but I probably just need to grow up.

>>105858682
Don't use the STL

>>105858527
>first case, basic example virtual functions are 50% faster
>second case, optimal concrete functions were 20% faster (but ran slower on another machine)
>third case, cache exhaustion on arrays is inclusive and varied again between machines
In what way would that mean it being a potential negative is untrue? Do you mean in comparison to concrete functions that can be slower?

>>105858686
But I like having to not write shit from scratch and not segfault on raw pointers.

>>105857969
>Because calling your software "not done" indefinitely doesn't shield you from all criticism.
Of course, but I'm hardly seeing any legitimate criticism beyond "What's the point when we already have x, y, z languages?"
That doesn't really seem to criticise the creation itself.
>>105857993
From my perspective, from the little that I saw of him, he just seems to share his unpopular opinions and provides a reasoning for why he holds them. Doesn't really seem like something to feel bothered by.
The people here on the other hand either provide little to no reasoning or just yap away about something hardly relevant. It seems to make no sense. Is it just some visceral reaction? Like, do they just hate him? Is it trolling? Maybe I'm too autistic to get it

>>105858762
>I don't see why people would be bothered by unpopular opinions
are you dumb

>>105858762
The compile times are the only things I care about. Too bad it will probably just be as barebones as C or Go without any basic convenience features like Optional and Result types baked into the language. If Jon made a sub 1s C++ compiler, he would be swimming in money by now and the game industry blowing his dick for ages. But no he chose to make an obscure toy language instead. Language maturity beats everything else. C++ has thousands of libraries and decade old codebases that won't be replaced any time soon. Something like Kotlin or Google Carbon makes more sense to me. Hell, even zig can just include C headers without problems...

Unironically what is the usecase of Jai for the general game industry?

>>105858821
If the person provides a reasonable explanation for the unpopular opinion, why would you be bothered by it?
>>105858917
Whether he swims in money or not should be his problem, not ours. If language maturity is your metric, then of course time since release is going to be a relevant factor. Given that Jai isn't released yet, naturally then it's a worthless language in your eyes at the moment.
I guess I just disagree with that metric being the most important one.
As for the usecase, I think it's sometimes hard to predict what exactly will make a language take off. Sometimes things just somehow fall into place in a holistic sense and then people find themselves using this or that. Which is why I think it's best to just watch and see whether he'll make an impact or not. If not, it's mostly his loss anyway, not ours.
Like, imagine Jai flops. Basically decades of his life go down the drain for creating something useless. So yeah, I just don't get why people seem so bothered when none of us have anything to lose here.

>>105858974
>If the person provides a reasonable explanation for the unpopular opinion
He doesn't though, he generally just calls everything bad and leaves it at that

>>105858982
In the few videos I have personally watched, he always gave some explanation. I guess we must have watched different content.

>>105858988
His explanations are always half opinion and half "you wouldn't get it"

>>105858994
Not the impression I got, but okay?

>>105859000
You wanted an answer and you got it, no idea why you're so defensive about it
Man making big claims that he doesn't back up doesn't endear himself to others

>>105859013
You make universal, unsubstantiated claims and act surprised when people don't just mindlessly accept it.
>he generally calls everything bad and leaves it at that
>his explanations are ALWAYS half opinion and half "you wouldn't get it"
Like, you leave no room for nuance at all. It's an infantile way to communicate. Nothing defensive about just seeing no point in not entertaining your ramblings.

>>105859034
I thought you were actually curious as to why people don't like him so I gave you an answer but you really just want to defend him

>>105859055
You provided your "explanation", which was theatrically black and white, and I provided you my own experience, which you negated. And now you act like I defend him. Just odd behaviour.

>>105859034
>>105859074
You don't actually think that people like the one you are talking to are honest, right?
They are here as activists against popular figures that don't bend to them.

>>105859074
>theatrically black and white
Stop being a fucking retard, I never said he always does it 100% of the time, obviously he doesn't
He just talks a lot of shit and doesn't back it up at the end of the day
Constantly complains how everything is bad, has nothing to show for it

>>105858974
>If language maturity is your metric
My personal metric is only iteration speed and basic comfort. Language maturity is objectively important in the (AAA) game industry because of their decade old codebases. I assume Jon is aiming for this demographic? He should be promoting the interop aspect more of his language instead of just showing pointless visual tools like in the lambda talk. Most of those tools are already available in Visual Studio.
>none of us have anything to lose here
True. I just spew out my schizo thoughts onto an imageboard instead of working on my games. Fortunately I'm not addicted to Twitter yet, like Jon.
>Like, imagine Jai flops. Basically decades of his life go down the drain for creating something useless.
Happens with artists all the time, nothing new. No one plays the shit I make.

>>105859104
>I assume Jon is aiming for this demographic?
hes not

>>105859093
What do they get though for opposing popular figures? Like, are they being paid to act like this? But who in their right mind would pay for it?
>>105859096
>I never said he always does it 100% of the time
You literally did. Like, literally. And that was AFTER I already said how my experiences differ, which implicitly negates my own experience. Yet you seem surprised that I don't just accept what you say?
It's like me telling you that there's no screen in front of you as we speak, yet act surprised that you don't believe me. Just absurd stuff.
>>105859104
>iteration speed and basic comfort
More or less agreed desu. Do you think Jai will perform badly on those two? Or are you saying other languages already cover these two well for your needs?
>instead of just showing pointless visual tools like in the lambda talk.
Yeah, that's probably true. My personal guess is that he was just showing off his latest implementations or features. He should definitely bring more crucial features to the forefront when it comes to advertising.
>I just spew out my schizo thoughts onto an imageboard instead of working on my games
Kek same

>>105859129
I'm not surprised by any of your behaviour just don't pretend you're some unibased neutral party just asking questions and playing dumb when clearly you are
90% of his opinions are "it sucks because vague ancedotal reason / you wouldn't get it"
If you haven't picked up on that you haven't heard enough from him or maybe you don't understand how half-baked his reasoning is

>>105859107
I remember him saying he wants his language to be adopted by the industry.
>>105859129
>iteration speed and basic comfort
Personally I just want something like D programming language: C++ without headers and with fast compile times. D unfortunately scared off every game developer because of garbage collection. Mistakes were made.
I would also love a language that is interpreted in debug mode, allowing easy hot reloading, and compiled in release mode.

>>105859157
>he wants his language to be adopted by the industry.
That kind of contradicts his eltist "I'm only making a language for GOOD programmers" attitude

>>105859136
>don't pretend you're some unibased neutral party just asking questions
Literally what I am. I've been pretty charitable with you all the time despite you calling me retarded, dumb and whatnot. It's crazy to me how incapable of normal conversation you seem to be.
>90% of his opinions are "it sucks because vague ancedotal reason / you wouldn't get it"
Not my experience, but you won't accept that anyway. If I disagree with you, I'm some biased entity trying to do what exactly?
>If you haven't picked up on that you haven't heard enough from him or maybe you don't understand how half-baked his reasoning is
Right. If I disagree, I must not be intelligent enough to see it the way you see it. Or I just haven't experienced as much as you have. Convenient stuff. Not infantile at all.
>>105859157
I've never tried D-lang personally. Funnily enough, I didn't see the usecase for it kek
Maybe I should give it a shot though

>>105859129
>What do they get though for opposing popular figures? Like, are they being paid to act like this? But who in their right mind would pay for it?
It's ideological activism. Understand that it has absolutely nothing to do with what work JB has done and that's why they can't actually engage on it. It's not technical; it's a matter of his position as a... normal person. He has no interest in their ideology of cuckholdry, kalergyism and castration and was previously "cancelled" for this years ago by game dev cucks.
Same thing as with RMS, XLibre, LunJew, etc. Notice how threads like this get 20+ replies instantly with the shilling.

>>105859184
This smug "hmm yes I disagree but I won't elaborate further, I'm just asking questions" shit doesn't fool anyone. You could provide an example but you aren't

>>105859193
As wild as that sounds, I actually believe you kek
I have yet to hear any actual substantiated claim. It's just condescending, dismissive, and passive-aggressive rhetoric. And it's especially funny since I don't really give a shit about JB. Though, I do find it interesting to see new languages being made, new concepts being explored. Maybe one of them one day revolutionizes the programming world.
>>105859220
>You could provide an example but you aren't
Ironic. Anyway, you aren't really worth talking to, clearly.

>>105859233
If you disagree you can say WHY you disagree, you know, if you're actually neutral like you pretend to be and not just defending his honor, what opinions does he have that you think he explains particularly well? I could give hundreds of counterexamples if I just pulled up one of his videos. It's usually someone in chat who asks him about something and he's like "yeah that's just not the right way to things... it just sucks, stop asking me about it"

>>105859233
>As wild as that sounds, I actually believe you kek
It started with /pol/ around 2016 and then happened in /g/, /his/, /sci/, /k/ and probably others years later. Don't confuse it with trolling and it is very much coordinated.
>I have yet to hear any actual substantiated claim.
This is how you identify them and it doesn't matter the board or the topic. They are waiting for whatever version of wrongthink they oppose so they can instantly start spamming and shitting up any discussion or questioning around it.
They always make assertions and never substantiate it anything with more than the mockery that gets repeated every thread. I suggest you ignore them as soon as you see them because they never contribute anything useful.

>>105859309
>Don't confuse it with trolling and it is very much coordinated.
Maybe. I'll be hesitant to draw too strong conclusions, but yeah, it's just odd to observe no matter how you flip it.
>They always make assertions and never substantiate it anything with more than the mockery that gets repeated every thread.
Yeah, seems exactly to be the case. And then you get accused of doing what they're doing. Just disingenuous and unpleasant to interact with.
>I suggest you ignore them as soon as you see them because they never contribute anything useful.
Makes sense. I guess I still do like to give people the benefit of the doubt, which is why I try to interact charitably. But when my good faith gets tossed aside countless times, I too lose interest eventually.

>>105859398
>I still do like to give people the benefit of the doubt,
No you don't
You already made up your mind on a topic then pretending you were looking for a geninue explanation but you weren't
You're acting in bad faith

>>105859157
>I remember him saying he wants his language to be adopted by the industry.
The industry is never gonna adopt a language that require 10 years to finnish a sokoban clone

>>105859165
do like a minimal amount of effort and actually listen to his jai proposal video series.
>>105859184
The usecase for D is/was being a more pleasant C++. Ideal for a C++ dev to pick up easily without having to change your existing workflow too much. Rust and Jai want to change the norm of programming too much. This makes them ironically both agenda languages imo.

>>105859442
He talks about programming languages for good programmers in the video in the OP
Elitism like that doesn't lead to adoption

>>105859398
>I guess I still do like to give people the benefit of the doubt, which is why I try to interact charitably
I used to be the same, but they ruin up so many threads where there could be useful discussion.
I just want sincere people on here to avoid them so every thread doesn't devolve almost entirely into shit-flinging garbage.

>>105859452
Then the goal of the language must have changed since its proposal.

>>105859463
No, his personal attitude just contradicts the goal for his language

>>105859470
kek. Yeah that’s true.

>>105859452
>>105859470
There are examples of elitist practices being adopted in the games industry (I'm not sure if it's elitism or a contrivance from wringing out MOER GRAHPIX), some consoles were infamous for having incomprehensible architecture so it even extends to hardware manufacturers

>>105859523
That's completely different
Nobody is going to adopt Jon's language if Jon says "you have to PROVE yourself to use it, I'm not letting just anyone in"
Languages that try to get everyone possible to use it still struggle with adoption, and it's not like there's anything amazing about Jai that warrants this attitude

>>105859537
I'm not sure whether the goal is mass-adoption or adoption in the game industry, in his first presentation the stated goal was the latter (could have changed by now tho) (it's literally titled "a language for games")

>>105859555
I think people who follow his cult of personality will use it and that's it

>>105859562
The same people that will call Rust users cultists. Ironic.

>>105859562
Sure, it's been in development hell for nearly a decade already, I don't know the exact objectives of the language the central question of parallelization (to me) appears unanswered by it

>>105859537
>if Jon says "you have to PROVE yourself to use it, I'm not letting just anyone in"
That's the closed beta, he said before he distributted it to a few hundred people. He just doesn't want random feedback from incompetent devs while he's still building the language.