your post literally doesnt make grammatical sense.

if you reworded that in actual english we might answer.

Here you go: Why does Rust failed on the retards?

>>25588503 ... did

Time to market is too slow it is better to iterate asap and suffer the segfault

>>25588503 >>25588506 you have to be trolling are you drunk?

>>25588510 >are you drunk Yes. The question remains.

>>25588514 the question doesnt make sense in the first place, what the fuck do you mean? what is "fail on the retards"?

>>25588519 They did not catch on to rust. It's mostly ignored.

>>25588521 because rust makes no practical sense for most enterprise programming use cases. there's Go, Java and C# for those.

Go in particular is what Rust wishes it were in terms of adoption and being the "cool new language" but that simply isn't happening without a garbage collector.

It got associated with trannies

>>25588548 no one who is not an imageboard addicted freak knows of the tranny association, that has no relevancy to real world adoption

>>25588533 So what rust claim is true, and by that I mean some memory safety doohickey, but it's a non-problem? I'm happy with GC, in the sense that I don't have to do anything, and the downside is some slowdown no one measures.

The public interface of an AI is easy to understand. You type your question in, you get an answer. It's the most natural interaction for a human. Rust is a low-level thing, you need a certain background to understand what's happening there. It's like, anyone can see the beauty of a piece of music or a painting, but not anyone can see the beauty of a mathematical theorem.

I think there are two camps. There are those who work with C and have dealt with all the insanity that optimizers, UB, and the memory model cause look at Rust like a necessary step forward if not pretty shit in itself in a lot of places. Then there are the zealots, who no one should pay attention to. But when actual systems guys look at it and say "meh, its better than C" for doing what they need to do, it probably is. The zealots immediately turn everyone off with these absurd claims and tranny-like behavior. I don't mean because they are trannies, but the same vicious, hard-headed attack stance they take is exactly similar. Probably if some normal developers were involved with Rust instead of the psychos at Mozilla from the beginning, nobody would think twice and adoption would still be right where it is, if not a little more.

>>25588572 So they understand neither of them.

>>25588566 >So what rust claim is true, and by that I mean some memory safety doohickey, but it's a non-problem? it's a problem only for non garbage collected languages. for GC languages it's a non problem, and has been a non problem for half a century. Rust might solve a problem for a few system programmers who program in pure C, but for the vast majority of programmers it doesnt matter in the slightest.

>>25588585 They don't understand the internals of AI, but they understand the results.

>>25588521 The concept of ownership and borrowing is old. It only works for some things and for others, you'll do the same shit as with C. It takes forever to compile and requires the programmer to write a lot of boilerplate code instead of having sane default assumptions. Rust doesn't go all that high like C++ and D which have object oriented but the standard library still seems to cater to application programming rather than systems programming and mostly assumes you're not working on a kernel

>>25588602 Anyone who worked on code more than one page long knows that "sane default assumptions" may suddenly not seem sane or default to someone else, or even to yourself one year in the future, and get broken all the time.

What y'all say is intredasting. Would it be possible to do weird distributed stuff in rust? Like ... dunno, erlang. Never seen erlang. Would it be easier to write some state machine exchanging messages left and right, tracking causality chains, and hosting it on bare metal, by the dozen, in rust? Easier than C?

>>25588624 >>25588602 give an example of what you mean by "lots of boilerplate instead of sane defaults"

>>25588624 Tell me when Firefox runs on Rust. They claim they rewrote 12% but I bet that's not counting any of the libraries it's using. https://docs.google.com/spreadsheets/d/1flUGg6Ut4bjtyWdyH_9emD9EAN01ljTAVft2S4Dq620/edit?gid=885787479#gid=885787479 If you add up Rust and C++ it's 7 million but Firefox has like 20 million lines of code total. So Rust is probably 4% of the total code that makes up Firefox.

>>25588639 most of "distributed" software is literally written in Java, take a look at Apache projects. this is to say, it doesnt require some niche specialized shiny language. writing "distributed" software is more about understanding very well things like time and concurrency and transactions than about some specific programming paradigm.

>>25588658 I'm writing distributed shit in java. And I don't understand rust and the shills. But java needs an OS and a JRE. Would rust help me track a billion records, and a billion jobs, and a billion request, all cross-referencing each other? And I'd like to use n machines, and m threads on each machine. In java, you just shrug. There must be some dozen mutant super-java-monkeys keeping this dream alive somewhere...

Radio Shack ?

>>25588733 its deda

>>25588652 One example is that they are very autistic about unsafe, so they'd rather write "unsafe" 10x in one line as specifier instead of using an unsafe block. >>25588718 You could try D. It uses GC by default, so you'd feel right at home. It can use LLVM like Rust does but also GCC and dmd (the reference compiler which produces worse binaries but even faster compilation). https://dlang.org/

>>25588743 >It uses GC I don't hate java. An usable alternative to the full blown gc experiance would be nice tho.If it's a silly application of some non-standard logic then I have no use for it. That's how I think about it, without really knowing anything.

>>25588777 D also allows manual memory management.

Otherwise you can do the ownership thing in Rust.

>>25588783 I'm still very confused. It's not the lack of you. Rust is a middle ground in memory management efforts needed by the monkey operating the keyboard. Is it a foothold to attack parallel problems? Is it a foothold to attack distributed problems?

>>25588891 >Rust is a middle ground Middle of what? The needs of system programming and application programming are vastly different. >Is it a foothold to attack parallel problems? Is it a foothold to attack distributed problems? I don't see how Rust enables you to do parallelism any better than other compiled languages. For fucks sake, just try Rust and see if it suits your needs.

rust doesn't actually solve any issues when a rustlet can't manage memory and causes memleaks its a "C++ design issue" but when someone complains that the borrow checker is unbelievably stupid and limiting it's a "skill issue" or "checked unsafe code is okay" i hate rusters so fucking much its unreal

>>25588955 >system programming and application programming One man's system is another's application. >any better than other compiled languages That's some real answer. >For fucks sake, just try Rust and see if it suits your needs I may should. The cost may be too high, without any result whatsoever.

>>25588955 This is more unsafe calls that I've seen in total in five years developing in Rust. I have a feeling none of them are necessary, and the code was written by some C brainlet.

>>25588997 >The cost may be too high What cost? It's free and you'll probably know after a while whether you love it or detest it. Just get sober.

>>25589017 It was written by a Rustacean very vocal about his love for Rust.

>>25589019 >What cost My precious time. The threda helped me a bit.

This video explains it perfectly.

have you watched the movie bubble boy? well it's actually about rust programmers

Did Bernd know that the C++ guys banned a contributor a few months ago because they felt "The Undefined Behavior Question" sounds like "The Jewish Question"? https://slashdot.org/submission/17330375/c-standards-contributor-expelled-for-the-undefined-behavior-question

>>25589051 >travolta This is more special than I can handle.

>>25588539 >Go in particular is what Rust wishes it were in terms of adoption and being the "cool new language" but that simply isn't happening without a garbage collector. Go was designed as a joke to make fun of terrible pajeet coders. They "FORGOT" to put enums in it, that only happens when your language was designed after some beers on a napkin as a joke. Go is terribly designed and ignores many lessons language designers should be aware of. The reason Go is popular is because even with the flaws it has, it is better than other slop language for pajeets since those have even worse design. There are dozens of easy to find blog posts of experienced people tearing Go's awful language design to shreds. Rust was designed to confidently write large, complex systems codebases with C/C++ tier performance but without their huge bug potential, and it's the only language which is actually delivering on it. Go is totally useless in that realm. >>25588589 >Rust might solve a problem for a few system programmers who program in pure C, but for the vast majority of programmers it doesnt matter in the slightest. The vast majority of programmers write unimportant code, yes. Rust is a niche language for the elite of systems programmers, yes. >>25588602 >The concept of ownership and borrowing is old. Rust is the first popular implementation of it. >It only works for some things It is the core memory model of Rust, so it HAS TO and DOES work for everything. >and for others, you'll do the same shit as with C. You're pretending Rust is 1 concept - borrowing and ownership. Rust has dozens of interesting concepts which solve important problems. >It takes forever to compile Many years ago, when it was new, yes. Now the compiler has been optimized and people learned to compartmentalize. My largest project compiles under 1 minute for a full recompile, something you do every couple of months tops. Incremental changes day-to-day are sub 1 second compiles. You also pretend that you do not get anything for this compile time. C compiles faster, sure, but you get NOTHING from C. Rust is famous for code also doing what you want most of the time once you get it to compile, with C you get random shit and undefined behaviour you have to debug. Rust's tradeoff is a few seconds / 1 minute of compile time vs hours of debugging a mess in C with terrible tooling. The Rust compiler does A LOT of work the C programmer does manually. >and requires the programmer to write a lot of boilerplate code That is fundamentally incorrect. Rust is a strongly typed language with a very rich typesystem which is essentially turing complete. While it might feel to a novice you're "wasting time" in defining your types well instead of having some garbage interpreter infer it and create a huge mess, the "boilerplate" you write in rust actually defines a lot of your program already. Well-defined types with methods and traits on them can be most of your program logic already. Boilerplate code is useless, but Rust has no useless code. The syntax is hyperdense, and everything you write is required for some fancy feature you don't seem to understand. >Rust doesn't go all that high like C++ and D Complete rubbish, there is nothing limiting higher level implementations in rust. The standard lib is relatively small, but there are many ecosystem crates which are basically considered standard libary tier crates offering various high level comfort features. The standard lib being small is due to rust devs having an insane focus on quality, which is far preferable to C++'s bloated garbage corpse standards. >>25588743 >One example is that they are very autistic about unsafe, so they'd rather write "unsafe" 10x in one line as specifier instead of using an unsafe block. First, let's pick apart unsafe. You're complaining about a feature 99% of rust code doesn't need. 99% of the time, you write SAFE rust, so this doesn't bother anyone. In C or C++, all your code is unsafe, and you have to opt-in into safety measures and tooling people developed externally to the C/C++ standards for the most part to ensure safety. NOBODY does this opt-in, hence why the C ecosystem is such a mess. So fundamentally, Rust forcing everyone to use SAFE code for 99% of the time is highly appropriate. Now, why is rust so anal about when you use unsafe, why don't they just slap a big unsafe block on everything? Because of granularity. Even when writing unsafe code, not all of your code needs to be unsafe. By using a keyword infront of specific functions, traits, etc instead of blanket blocks, you can pick exactly where you want to opt-out of safety. Since unsafe code is rare and the exception, this slight verbosity is a good tradeoff for the compiler safety gurantees provided. >>25588783 >D The only D anyone uses is in their pants, go ask your mom about it. >>25588891 >Rust is a middle ground in memory management efforts needed by the monkey operating the keyboard. >Is it a foothold to attack parallel problems? Is it a foothold to attack distributed problems? There are 3 memory management paradigms: 1 manual (think C) 2 automatic garbage collector (think Java) 3 borrowing / lifetimes If you do 1, you have to make sure you never fuck up. Easy on a hello world, but as soon as you write a large codebase fuckup is inevitable. C was designed on a PDP 11 with a few kb of memory, programs would be a few 1000 lines of code tops. C is a TOY LANGUAGE only suitable for the TINY computers from 50 years ago. It is unsuitable for today's computers which do so much that programs have grown to millions of lines of code. If you do 2, your performance is unpredictable, since GC kicks in during runtime and shits up your program execution. Just look at unity games. Either small, performance uncritical games or they stutter thanks to GC. If you do 3, you give the compiler EXTRA INFORMATION on how long the data you stuff into memory is gonna be needed by the program, and the compiler then figures out when it's safe to drop based on that information. If you fail to give the compiler enough information to do that, it nags you about it. That feels like it's terrible and restrictive, and it is compared to writing small cozy daredevil C programs, but once you work in a HUGE codebase you'll realise it's actually freeing and empowering. You pawn of a menial task you had to do yourself to an automatic assistant who never fucks up and only asks you for some extra information to do it's job properly on occasion. >>25588955 >I don't see how Rust enables you to do parallelism any better than other compiled languages. lol >>25588997 >>system programming and application programming >One man's system is another's application. I mean I suspected you didn't really get it but now you confirmed it. Sigh. A systems language cannot have unpredictable performance. When you write a filesystem, a network stack, a driver, you need to know how long it takes. You want to basically know what assembly code the compiler generates from the code you write for the entire performance critical path. The idea of having a garbage collector pause a file read from your disk or pause receiving network packages on the TCP/IP stack to clean up memory for a few ms is absurd. The reason you get away with a GC language for your applications is because they mostly don't do anything performant and you have 100x the computation you require under your ass so you can be wasteful. If you wrote an OS like that it would be a catastrophy. The Rust debates always boil down to the same shit. - People who don't actually know what programming is about being confused - C people who want the freedom of writing sloppy code and deny that large C codebases are a mess - Rust evangelists who realise how freeing and awesome Rust's restrictions are if you embrace them I would never try to convince anyone to learn Rust. The vast majority of programmers are too shit to learn it, no offense. They're just not smart enough. Rust is a language for the elite, not for everyone, unironically. The use you get out of rust as an average plebbian programmer is vastly better libaries, tools and software you get to interact with. The really good rust programmers took zero convincing. They looked at the language design, and thanks to being smart realised how amazing it is. There is a reason why rust is the most raved about language and the most admired language in dev surveys while also the most hated language in online debates. I would oppose any attempt to dumb down Rust to make it more accessable as well. The Rust ecosystem is in such an amazing state thanks to that elite filtration, the quality tooling and libaries available is utterly amazing. They manage to crank out documentation consistently ... absurd. Decades of coding, never seen decent documentation on anything cutting edge, here's rust, cutting edge as fuck, and it has GOOD to EXCELLENT documentation on pretty much EVERYTHING almost the second it comes out. Like wtf. I honestly think it's the trannies. Somehow flooding a male brain with estrogen creates peak coding ability. And I say this as someone who's still a mediocre programmer myself, especially in Rust. I don't even own thigh-highs yet.

>>25589110 If you typed this I'm not gonna read it.

>>25589110 jesus fucking christ dude

>>25589122 tl;dr - You're too dumb and got filtered by Rust before even trying it

>>25589145 dude you're missing one simple things and all dumb rust evangelists miss it too - if you can code good rust you can code good C++ (stdc++17 and above) and then you just dont have the memory problems because the language gives you the security you need

>>25589110 Go is popular because of Google. I read that some people were masturbating over it before it was even released. If it wasn't made by Google no one would care. >It is the core memory model of Rust, so it HAS TO and DOES work for everything. No, it doesn't. Hence UNSAFE UNSAFE UNWRAP UNSAFE UNWRAP UNSAFE It's a situational alternative to manual memory management in Rust. >99% of the time, you write SAFE rust Is it really SAFE if you shove unsafe code in the standard library and call it TRUSTED?

btw 99.99% of the time you write safe C++ too rust dumblets are so fucking stupid

>>25589110 Dude, I'm drunk. I just write code at work and I try to avoid bullshit tasks. So far it is going fine but eventually people will pick it up and bully me into doing bullshit. Fuck this shit. Why won't other companies adopt me? Why must I suffer. I once applied to Google but I fucked up the coding interview and I felt so ashamed that I never reapplied. It has been over three years. I'm such a fuck up.

>>25589155 I'll have to confess that I ignore C++ as much as rust. It just started way earlier. Billions and billions and characters spent, and still, you fuck up a little thing and the crystal palace is coming down on your neck full force. I'm not even a gc apologist. You fuck up the same way there too, you just don't have to face it for a while.

i never had to malloc or free or whatever in the last 5 years of writing c++ any single fucking time because everything is managed by smart pointers if i need it and by destructors and move semantics otherwise and it JUST WORKS if you're too dumb to understand how memory management in C++ should be done than go use Java or some shit because you're too dumb to use rust anyway fuck rust evangelist

>>25588485 Rust is a tranny language

Calm down guys. Everything is a tradeoff.

just evagelist propaganda without background history, no real success, rustfags should pay more respect to oldies. if you need force it on every C/c++ project this is a huge red flag

>>25589155 I specifically elaborated on how the opt-in psychology of C++ for safety doesn't work while the opt-out psychology of Rust does work. How many C++ projects opt-in to a level of safety tooling close to the level of Rust? Basically none. Meanwhile basically all Rust projects are written in safe rust. C++'s safety culture is broken by design and thus fails to work.

>>25589281 >How many C++ projects opt-in to a level of safety tooling close to the level of Rust? you don't need to opt in to anything if you use modern c++. all is given to you via move semantics, smart pointers and so on. that's the thing rust people don't understand.

>>25589293 Ah, so you have no idea of what Rust actually provides. That explains your position. Another filtered C-nile gatekeept from Rust paradise.

>>25589304 i have a complete overview of rust, I actually learned it and used it for a couple of projects to be able to discuss with absolute dumbfucks like you what YOU don't understand is that the "memory safety" guarantees of rust are NOT NEEDED provided you know how to write C++. That's as simple as that!

rust niggers spent thousands of hours to figure out how to prevent dereferencing inexisting stack variables without realisng that such behaviour is just an effect of bad design! they're such dumb idiots that they need a compiler to tell them that their code is super dumb and they need to correct it, or better yet, call .clone() every other line like half rust libraries out there do

rust niggers also spend hours of their time figuring out how to compare floats instead of "relying on undefined behaviour" which is actually super defined if you are smart enough to know how u2 encoding works like. detecting float overflows is something you learn on coding 101 but of course rust niggers don't know that because they are jobless dumbfucks who only learn coding out of pure boredom while leeching welfare

Both Rust and C++ are tedious compared to D.

Bjarne never wrote a compiler. He transpiled his fantasy language to C. Rust only has one implementation on top of LLVM. Meanwhile D has three compilers. One written by the language creator.

>>25589404 Writing a compiler is not in any sense an achievement or in fact anything hard. In most Computer Science university courses students commonly write their own compiler during year 2.

>>25589110 Take a break

>>25589412 Write a C++ compiler if it's so easy. I bet it's just as hard now as creating a web browser that runs modern web apps.

>>25589326 >i know all of rust >rust is only about memory safety it's time to stop posting

>>25589454 Where did I write "only"? Why don't you fuck off and stop teasing me? You have reddit tier ways of discussing

>>25589464 >Why don't you fuck off Sorry, I am unable to fullfill that request. I derive immense pleasure from filtered midwits seething about rust.

>>25589404 The only place where D is used is your mom's bedroom buddy.

>>25589496 My mom can take a lot of D, buddy.

>>25589491 I don't see any midwits in the thread except for you. After all it takes a midwit to not know how to handle C++

Calm your tits, boys. The truth has not emerged yet. We can fling shit afterwards.

The lifespan of a rust nigger is as follows >Be too stupid to write anything bigger than one .cop file without it crashing in c++ >Switch to some niggeresque language like java or whatever >Some time later hear that this rust thing does the job for you while spawning a ton of other problems with the only purpose of preventing you from shitting all over your code >Proclaim rust the C++ killer because you obviously can't handle memory There are GOOD things that rust has Those are: >crates >borrowing vs unclear c++ move semantics >Proper rust being much easier to learn than proper C++ >ensuring minimal code quality so basically even dumb 80 IQ idiots can contribute once they bash their head against the compiler enough times However if you compare c++ GOD vs rust GOD, the c++ programmer will write better, faster and Less complicated code. Simple as.

I get paid $200k+ to code and have zero clue what Rust is and constantly hear about this shit on chans on Reddit and never seen Rust what is Rust

>>25589607 >what is Rust I think no one seen rust in this thread.

>>25589607 Rust is like the dark souls of languages People spend hours to learn it just to flex on people who didn't "git gud" just yet without realizing that the thing they have actually learnt isn't the best tool for anything. It's a relatively good tool for general purpose coding but it's not the best there is for any specific task

>>25589624 What is the usecase lel. The wikipedia sounds esoteric

>>25589615 You refused to read the one post from an actual rust dev due to length, which is a strong indication of how likely you are to actually read anything of substance, which explains why you know nothing about programming.

>>25589634 Keep going!

>>25589604 >However if you compare c++ GOD vs rust GOD, the c++ programmer will write better, faster and Less complicated code. Simple as. C++ has so much baggage and problems. Let's just get to the basics: What's a string? C++ has no answer to that and you'll find C strings C++ standard library strings and 10000 other string implementations that are all incompatible.

>>25589630 The use case is memory safe systems programming. That's an umbrella term for everything. Whatever you usually use c++ for you would use rust for, that's why they're most often compared

>>25589639 I'm not the Rust guy btw.

>>25589639 If you code in c++ you use STL std::strings. C doesn't have strings, it has char pointers. They might be just called strings due to some agreement about naming. That's one of the better points you made though, since this is an issue, the baggage thing.

>>25589656 >C doesn't have strings, it has char pointers. Char pointers that are null terminated. Clearly a string. And C++ recognizes them so strongly as strings that it requires constant time for getting a C string from a C++ string which in turn means that every C++ std string needs to be 0 terminated internally which slows down everything and requires copying whenever interoperating with strings, that are not null-terminated, is required.

And C++ strings suck so badly, every large project has its own string implementation. Qt has QString, all kinds of video game engines have own strings or even entire custom "standard libraries". Simply because the C++ standard guys are such fuckups, not because they are really keen on going through all this extra effort.

There are very, very few good arguments against rust: 1. your coders are dumb -> need easy slop lang - my response to that would be, get rid of your dumb coders and hire good ones instead 2. you're interfacing with something that has a fundamentally different paradigm which is hard to interface with in rust, Ladybird browser devs claim the OOP structured APIs they have to interface with are a bad fit for rust, but none of them has any rust experience and they tried it for 2 weeks, so it's probably just the typical skill issue. They now use Swift, so I struggle to take them too serious 3. you're an oldfart very used to C++, your codebase is not that large, or you have done monumental efforts using very strict coding standards enforced by a dozen bespoke tools half of which you coded yourself to turn C++ into a worse, duct-taped together pseudo-rust which is even more restrictive, less performant, but you are so invested it would be suicidal for you to abandon it - but that's a YOU decision and a YOU problem. Anyone knew coming into this will either have to be stockholmsyndrom'ed into your suffering, or run away screaming and go to a sane place - So you're in a dead project at that point. All other criticism is nonce shit from idiots who don't know how to code and never faced the struggles of a serious codebase. If all you ever do is dick around in small programs as a hobbyist, or if you write sloplang code with no performance expectations like web shit or business logic slop - yeah ofc the C++/Rust tradeoffs are obtuse to you. For any new project without baggage, especially if it requires a true system language, there's no competition. Rust is the clear superior pick. There isn't even really a debate about it anymore. All the new exciting things in the systems space happen in Rust.

1. Google - Abseil (absl::string_view) 2. Facebook - Folly (folly::fbstring) 3. Mozilla - Gecko (nsString, etc.) 4. LLVM/Clang - llvm::StringRef 5. Qt Framework - QString 6. Unreal Engine - FString 7. Amazon (AWS SDK) - Aws::String 8. Microsoft Windows SDK - BSTR (used with COM) 9. Oracle Database - OCIString 10. PostgreSQL - custom text processing classes 11. Boost Libraries - boost::basic_string_ref 12. Chromium (Google Chrome) - base::StringPiece 13. Apache Arrow - arrow::util::string_view 14. OpenSSL - ASN1_STRING 15. Intel TBB - tbb::basic_string 16. SQLite - sqlite3StrAccum (internal string handling) 17. TensorFlow - tensorflow::tstring 18. RocksDB - custom internal string classes 19. MySQL - custom string utilities for SQL operations 20. Adobe - custom string utilities in its various products

>>25589686 How would strings not being null terminated prevent the copying issue? For operations like concatenation you need to reallocate memory anyway, unless you implement some kind of wrapper that holds pointers to multiple parts of the string. I just don't get how this is an issue.

>>25589747 You know, strings are not quite an unsolved problem. Keep harping!

>>25589751 In order to add the null terminator, you need to allocate memory and copy the data behind the pointer.

>>25589745 Rust IS the language for dumb coders because it offloads thinking about code structure to the compiler. It takes a good programmer to write good c++ code. Rust code is always at least mediocre by design.

>>25589765 It's obviously not solved in a clean manner.

>>25589780 Do you know that russia has strings!

How come Rust can't into embedded? inb4 someone links a toy crate for an esp32

>>25589767 you just allocate your strings to be 1 byte longer each time and you don't need to copy anything. imagine all strings just end with 0 by design, the STL string doesn't allocate string first then reallocate it again to add the 0. Or at least I hope it doesn't if it does it's super dumb

>>25589773 Man you're so retarded because you have to abandon reality to cope with your own inadequacy. The Rust IQ cutoff is like ~130. Below 130 you cannot really learn rust properly, and you'll write shit rust code. Nobody will use that. Your crate will never see a lot of usage. Above that, I have to agree, Rust is easier than C++. Even with 200 IQ you'll fuck up in C++ because eventually the context you need to keep in your brain gets too big. It's impossible for humans to write good C++ above a certain project size. In rust, the complexity is sort of capped, thanks to aggressive compiler automatisation. Good C++ programmers also have an easy time to learn Rust, and if they switch to Rust they'll be able to achieve a lot more with it. C++ limits a good programmer, Rust empowers them. This is what you fail to understand. Below that, however, it's possible to get by with lower IQ writing shitty C++ code since C++ lacks the architectural complexity of Rust which is MANDATORY to do stuff in the ecosystem. C++ has more complexity in total than rust, but most of it is avoidable, not mandatory.

>>25589799 Why do you think std::string_view exists?

>>25589787 >How come Rust can't into embedded? Embedded is typically exactly the small and limited environment C was designed for. A PDP 11 is pretty close to a shitty µC. This means embedded devs get away with keeping the entire context in their brain at all times more than other devs. But there is no obstacle to writing Rust on embedded and there is a pretty huge community around it >inb4 someone links a toy crate for an esp32 There's more than just "toy crates" https://github.com/esp-rs/awesome-esp-rust

>>25589858 Pls, chatgpt-chan, could you sperg me some more?

>>25589765 Low IQ dev can't imagine 3 of those projects interoperating and exchanging strings. Enjoy writing string conversion interfaces for all of them. This is ONE example out of 9001, mind you.

>>25589869 >mind you Oh, you have blown it already. Is there some more?

>>25589852 C++ does not provide an owning string view, so you'll still need to write your own string shit, if you actually wanna do anything with it.

>>25589866 Your spatial IQ is ~110, your verbal IQ is ~90.

>>25589874 Such an accurate assessment. Now, boy! Give me what I have asked for.

>>25589825 >C++ has more complexity in total than rust, but most of it is avoidable, not mandatory. Considering the baggage and the libraries that still carry it, I'd say it's the other way around.

>>25589825 >Even with 200 IQ you'll fuck up in C++ because eventually the context you need to keep in your brain gets too big. Indeed. That's why we have perfectly operational huge software packages written entirely with C++ by multiple contributors that somehow work without crashing What you fail to understand that modern C++ is easy to write and not limiting. Old c++ was somewhat problematic but it was used in times where people actually had proper knowledge about how computers work, so it helped alleviate the issue. The whole discussion with you basically boils down to this: >Good rust devs write good rust code >There are no bad rust devs because rust is too hard for dumb people (I don't agree with that but it's okay) >Good c++ devs write good c++ code (you don't agree that it's possible above a certain project size which I disagree withz maybe you coded not enough c++, doesn't matter) >Bad c++ devs exist and cause problems Which means >Good devs are good, bad are bad Which is pointless discussion. I will keep using c++ because I mostly do scientific computing and rust doesn't even come close to the speed of mangling raw pointers.

>>25589891 Rust has upfront complexity which you need to understand all at once, but which is very difficult to untangle to learn one-by-one without highly constructed unrealistic examples. Hence why rust is famously difficult to learn for midwits despite having the best learning ressources which are also overwhelmingly available for free. C++'s complexity can be compartmentalized and learned one-by-one to a far greater degree, making easier to comprehend for midwits. A libary you have to learn after learning C++ to work in some existing project is more complexity, yes, but the previously learned C++ was a cohesive mental model which works without the libary and thus was possible to compartmentalize. Smaller context window size requirement on the brain. If you're high IQ this dynamic is easily missed out on.

>>25589110 good read, I do not program myself but work quite a lot in that area and your opinion seems pretty valid. Thanks.

>>25589907 >That's why we have perfectly operational huge software packages written entirely with C++ by multiple contributors that somehow work without crashing You ignore the enourmous efforts in these projects, how rare they are, how many failures there have been for every successful one, the systematic failure of the industry to reliably reproduce them ... With C++ you need to get lucky and pick an exceptional project manager who navigates the dangerous pitfalls, and you'll still eventually end up with a project which will require rewrites. What rust offers it to RELIABLY produce high quality large complexity projects taking luck out of the equasion. >What you fail to understand that modern C++ is easy to write and not limiting. Old c++ was somewhat problematic but it was used in times where people actually had proper knowledge about how computers work, so it helped alleviate the issue. I remember you from the last thread, you only work in small projects. You clearly lack the insight of the horrors of actually large codebases. I don't want to sound too rude but your statement could simply never come from someone who has. >The whole discussion with you basically boils down to this No. Your logic when analysing my argument is flawed. Your summary is faulty. >rust doesn't even come close to the speed of mangling raw pointers. Rust does ~1.03x C speed without much trying on average, and if you run into an edgecase where that is not the case, you can absolutly mangle raw pointers in Rust. But everyone I ever met who claims to have to do that was just full of shit or too bad at rust, since it usually turns out you can do it better. I've seen maybe 2-3 exceptions to that in years.

>>25589799 Imagine all files just end with 0 each time.

>>25588551 everyone who uses discord knows the tranny and furry associations

>>25589799 No, you just mmap as much RAM as there physically is, that way the only overhead is in page faults.

>>25589974 >Rust does ~1.03x C speed would love to see the C code that causes this

0xcafebabe makes all rustacians seethe

>>25590065 if you git blame that, you notice that comitter is a tranny

>>25590071 As always.

>>25589921 I don't really understand why you think the concept of rust is difficult to grasp (or that rust has "complexity") I read the rust book on the rust website. It was very clear and understandable. Rust concepts were also very clear and I will give rust a point for being very well explained, c++ is very obfuscated compared to rust and you need to read a shit ton of stuff to understand what's happening under the hood. Rust is explicit and I will also give it a point for that. But I don't get how it would make it hard. Programmers these days can't read a book within a week to understand a language? That's what you mean by complexity? Still, for me it is very straightforward to write c++ code since I've been doing it since almost 15 years and I don't see myself having problems with navigating large codebaes that seems to be an issue you point out often. One last point: most c++ code you see online is some port/hooks of C libraries. And it's really not idiomatic C++. And theres no simpler way to get into memory issues than trying to write code that's both C-idomatic and C++C-idomatic at the same time >>25589974 I worked on large projects too. I'm just sharing insights of what's keeping me with c++ right now. The biggest thing I wrote with c++ is an RTOS for autonomous driving for a prototype electric vehicle and it was all coded in C++. From the sensors, through SLAM, up to the darknet YOLO (well this was C) and feedback mechanisms. Ran flawlessly with multiple interoperating processes without many issues. The secret of the project was that it was written in C++ and with C++ standards in mind from the beginning. As for the point about the effort. Do you claim that writing rust code requires no additional effort too? A good and experienced programmer knows to write good code from the start and take care of the codebaee. Yes rust enforces it, c++ doesnt, but it's not like rust is effortless. Adhering to the rules of good rust code requires effort. Just like in c++. Do you think that handling overflows whenever you compare floats is not additional effort? And I can list examples like that for multiple things that happen while you code. In c++ you write the happy path, in rust you take care about everything right here right now. And you know what? Not every system has to be extremely secure. You don't need rust for most things, because c++ is just safe enough. Often you know the overflow won't occur but you wouldn't slap unsafe over it in rust because it's not idiomatic and no one would use your crate. Rust is too hyper fixated about safety and this is my problem with it.

>>25590080 Rust is primarily tried by webshitters who got filtered by C or C++ who were always envious of systems programmers, since they now have a cope language that will fix their brain.

>>25590102 Its third world trifling. I literally only know R and make $200k+ how does that make you feel

>>25590107 I get paid $200k+ to only know R while you retards trifle for turd world $40k a year programming jobs in retard languages

>>25590107 Damn how many eggs does this salary get you? I bet you can get at least 3 dozens

>>25590107 Why would I care how much other people make or whatever? I'd be miserable if I had to program as a job, especially in R.

>>25590080 >>>25589921 >I don't really understand why you think the concept of rust is difficult to grasp (or that rust has "complexity") I don't know how often you want me to repeat the same stuff over and over. This is not my opinion, this is a well known famous and meme'd about fact in the rust community. You obviously didn't learn rust properly, otherwise you wouldn't have that opinion. >I read the rust book on the rust website. It was very clear and understandable. It's one of the best programming language learning ressources humanity has ever produced, yes. >Rust concepts were also very clear and I will give rust a point for being very well explained The language design is very good, yes, and the designers also took great care to document well, yes. >c++ is very obfuscated compared to rust and you need to read a shit ton of stuff to understand what's happening under the hood You haven't looked under Rust's hood yet. You only read the first introductory book to the language intended for beginners. >But I don't get how it would make it hard. Programmers these days can't read a book within a week to understand a language? That's what you mean by complexity? Go back to the rust book, and look at any simple primitive type from the standard libary it explains to you. Then go to the standard libary documentation and look up the source code of how that primitive type is implemented. It will have some traits or methods you used. Go look at the code. It is a horrowshow of nested macros, generic types, lifetimes, and you have to get it all at once, or you can't reason about the code. The complexity is there, you just haven't even looked at it yet. >I don't see myself having problems with navigating large codebaes that seems to be an issue you point out often. It's not about navigating, it's about bugs in the codebase nobody can reasonably find or fix. >>>25589974 >The secret of the project was that it was written in C++ and with C++ standards in mind from the beginning. And you don't see how the label "C++" has lost all meaning? When I use the label "Rust" you know it's an anally retentive language which forces you to do everything properly. When I say C++, it could be anything, from dumpsterfire to NASA certified spacecraft tier. I am sure somewhere, someone, made a C++ project where even the Rust crowd would nod in approval. The reality I see is more like UE5 microstuttering issues from 8 different AAA studios at the same time. >As for the point about the effort. Do you claim that writing rust code requires no additional effort too? >A good and experienced programmer knows to write good code from the start and take care of the codebaee. Yes rust enforces it, c++ doesnt, but it's not like rust is effortless. Adhering to the rules of good rust code requires effort. Just like in c++. Do you think that handling overflows whenever you compare floats is not additional effort? And I can list examples like that for multiple things that happen while you code. There are two mindsets here to me. One is the one where you refuse rust. Where you feel everything rust asks of you is a burden, extra steps you have to do which seem unnecessary. The other is where you embrace it. Where everytime rust asks you to do something extra, you know full well that rust has good intentions, wants to take care of you, and is helping the future you not to suffer. Then you stop questioning it. You embrace it as a new state of existing, where there is no alternative than to please the compiler. It's like breathing, you stop thinking about it and it doesn't evoke strong emotions anymore. >In c++ you write the happy path, in rust you take care about everything right here right now. And you know what? Not every system has to be extremely secure. You don't need rust for most things, because c++ is just safe enough. Often you know the overflow won't occur but you wouldn't slap unsafe over it in rust because it's not idiomatic and no one would use your crate. Rust is too hyper fixated about safety and this is my problem with it. I think the idea that rust is primarily about safety is not true. The philosophy behind rust is to do things properly, safety is a byproduct of that, but it is not the main goal. The fearless concurrency when programming multithreaded is a result of the same design ideas that lead to memory safety, but it is not a safety thing, it is primarily a convenience function for the programmer. I think of it this way: Every time you write a program, you have to figure out some solutions to some problems. Some problems occure so often that maybe you want to automate them away and have the language design/compiler solve them for you the optimal way automatically, without you having to do it manually.

>>25590173 >If a thing is simple to you you obviously didn't learn it properly why do you think that? >You only read the introductory book Again, why do you imply that? Can't you just accept that for people who have real applied programming experience learning new languages ISNT an issue? It's not my problem that redditors who do rust as a hobby find it hard. Anyway I'm out. I tried to discuss with you like normal people do but of course you stil have to insult my intelligence explicitly or implicitly with your every comment in a very reddit like fashion which I just don't find amusing anymore. I'm a dumb 89 IQ monkey who can't comprehend rust and that's why I perpetually shit on it. It was obvious from the start. Bye

I started programming in Gamemaker script out of all things, and tried literal dozens of languages later after I liked it... I was filtered by C and C++ for the longest time. But once I tried Rust, it all clicked into place. I used it for less than a year and got fed up with it being actual dumpster fire and went on to use C++, well long story short, I hate C++ like I hate Rust now and I am programming in C because I backtracked all the way from slop languages until there were 0 shit useless features left, just me, and compiler. I recommend everyone try Rust, knowing what stepping in shit feels like is necessary for character growth.

>>25590204 >>>25590173 >>If a thing is simple to you you obviously didn't learn it properly >why do you think that? >>You only read the introductory book >Again, why do you imply that? You never mentioned the advanced bits of rust, you said you spend 1 week reading the starter book which doesn't cover them. What other conclusion is there? >Can't you just accept that for people who have real applied programming experience learning new languages ISNT an issue? I have real applied programming experience and I dabbled in 2 dozen languages. And indeed, picking up most of them is rather trivial and not impressive. However, Rust isn't Moccascript.

>>25590254 I unironically approve of the concept of learning C and assembly first before anything else. I did as a kid when programming 8bit microcontrollers, but with hindsight i should have gone deeper on it.

>>25590300 I don't, I didn't know that C is best until I stepped in all the other shit.

>>25590306 If Dennis Ritchie was still alive he'd tell you to write BetterC

>>25591720 Well maybe that's why he's dead because C++ is trash

bump

>>25588485 >Wife to her Husband: While outside get bread! She never saw him again...

while ($outside == true) {
    getBread();
}

>>25593063 I wasn't refering to c++

Rust shills typically spew a lot of nonsense, then claim it solves the stuff they made up. See: >>25588583 >There are those who work with C and have dealt with all the insanity that optimizers, UB, and the memory model cause

This thread has even more LARPing than usual for KC, now that I read a bit more of it. What happened?

>>25600479 a bait thread that belongs on 4chan happened >>25600272 completely useless information since nobody refers to anything else as better C than C++ (and it's shittier C)

>why does Rust failed Ha, lol

Rust is for 4channers and troons. It's a meme.

>>25601338 > since nobody refers to anything else as better C Zig is commonly called that.

>>25601945 I don't hang around in shitholes where that happens

>>25601945 >>25601965 BetterC is commonly called that. Type it into a search engine. It's the first result from the top. Thank me later.

How much does this apply to Swift as a whole?

The explanation is that you're an homosexual who follows the news too much, language profit from the number of users, that's called the network effect, and from its duration which allows to use it for a long time.

>>25607282 Source: my ass

Btw h is pronouced in English, so it's a homosexual. Not an.

>>25607272 Swift is a dumpsterfire.

>>25608316 I never tried, so I have no strong opinion about it. What's so terrible?

Ladybird browser is moving to Swift. I'm still not sure what to think about it. The language has never been used for anything outside of apple before.

>>25609170 Seems like a terrible idea. Statements like that are always a bit dubious, but it seems to me that Swift is a "slower language" than C++. If they want to use C++ for lower layers and Swift for things close to the UI, it might be fine.

>>25609322 No, the exdruggie cancer patient wants to move the entire browser to Swift. He worked at Apple, uses Mac and is very Apple-patriotic.

>>25608929 Swift only has a robust ecosystem support for Apple, Linux/Windows support is still poor because duh, Apple doesn't care about it. They switch the syntax a lot and break compatibility all the time not giving any fucks. Rust editions show how to do it properly. They introduced optionals to prevent nullpointers, but the forced unwrapping defeats this half the time, verbose syntax for nothing. Haskell solves this well with algebraic data types, Rust with borrowing. The typesystem is a joke, their optionals/unwrap causes runtime crashes randomly due to loopholes, they badly copied rust's option type but it only works half the time. Protocol oriented programming is a neat idea, but their implementation of it leads to messy code which is hard to read, especially when using associated types and generic constraints. Their implementation of generics is poor, compiler messages mostly useless on errors. Many languages have good generics at this point, it's not that hard. Swift has mutable immutable variables. Yeah. Super verbose closure syntax, compare it to kotlin or scala which do it proper. Their actor model is inferior to Erlang or Elixir. The general language design is poor, whoever designed it has no solid programming language theory background. The only reasons why you would ever choose it for a project is either you're trapped in Apple ecosystem somehow and have to, or you have very little experience in languages in general and have never used a well designed one. The Ladybird homo guy said in an interview they tried languages for short periods of time, I think he said they tried "rust for 2 weeks" and also tried swift and "everyone liked swift". This is no surprise. Swift is approachable and easy in the beginning, it's easy to get started and write some code and feel productive. Long-term you'll create a lot of problems due to swifts poor language design, and at a certain codebase size you'll feel a lot of pain. Rust's language design is more well thought-out, solves a lot more problems for you, but also requires you to learn a lot more upfront. Small, simple tasks in rust feel overwhelmingly complex to beginners, so the first 2 weeks won't feel great. However, large rust codebases lack many of the problems many other languages will produce due to how many problems rusts solves on a language level and how much it asks of you as a programmer to bake into your code. I'm not a huge fan of C++, which is what Ladybird is written in right now, but switching to swift feels like switching footguns. Rust is the obvious choice for a browser, where you require a lot of the gurantees of rust and performance. Of course this is then a demanding project for a programmer, but at least you can be quite sure the end result will be what you want. Fast, stable, performant. I think really good devs can produce a quality browser in C++, but it is hard. With rust, it is easier. With swift, probably impossible.

>>25613505 Your description of Swift was good but have you tried D?

>>25588583 they ARE trannies tho

>>25588655 firefox isn't being developed anymore, mozilla is pivoting to be an "ai company", but they have no funding for that. their budget pays for five or six executives plus kickbacks to the dems, and they will run out of steam by this summer because us court banned google from funding them (more like laundering money for dem party politicians through tranny nonprofits). they haven't been investing in the browser for years now, they despise it, and they've been open about using chrome for their own browsing needs

i consider the IT sector a force for evil and i rejoice at any infighting

>>25615049 Good post. Corporate and its slaves need to burn.

>>25614798 >but have you tried D? It's a niche language, they're feature creeping almost as badly as C++ with just adding everything without discern. D has been around for 23 years, yet nobody uses it. There isn't a huge ecosystem, and what there is is fractured between GC and non-GC. What D lacked was a strong vision of what it wants to be.

>>25615239 Everything is fractured between high and low level. Not just th GC. You can't use abstraction relying on new/malloc, EH etc in a kernel. This devision is natural and not going to disappear so it was replicated fully inside D. You can write as low as C and as high as a scripting language.

>>25615239 >D has been around for 23 years It's only stable since 2007 and every other language has big corps shilling them and forcing their employees to use them.

>>25588583 >I think You do not.

>>25615262 >You can't use abstraction relying on new/malloc, EH etc in a kernel. C++ braindamage alert

>>25615262 It's insane to have it implemented with hard boundaries and lack of compatibility INSIDE a language ecosystem. Other languages show how to do it in a superior fashion and allow a seamless transition. Seriously deranged to think D is good.

>>25615324 You can use any C or BetterC libraries inside D. You just can't use D inside the others without a runtime. Same way you can't use C++ in C directly. >Other languages show how to do it in a superior fashion and allow a seamless transition. Other languages tell you to write another language when they suck at a certain use case.

>>25588485 It failed because it works. The secret service mafia wants to have back doors. Rust would close 90% of them. So Biden declared it to be the future, so the it would become political. Right now it is held in limbo, as is the world.

Rust was made to be "white programming" instead of being good for anything, it's identity politics made in a language, motivated by things like "the white man only does functional programming" and other things which cripples the actual usefulness of it.

>>25615434 the choice of language does not matter as long as developers are compromised, which they are