[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.




Is C/C++ on it's deathbed? Should I start taking HRT now and start using Rust?
>>
I love ru-ACK
>>
>>88766776
C# For Backend
TS for frontend
python/bash ok for scripts
rust for everything else
simple as
>>
>>88766776
why would i ever listen to any propaganda from microsoft?
>>
>>88766776
No i will just use cppfront
>>
>>88766776
Very good advice from a company that owns their own competing language/ecosystem that they use to funnel people in to their cloud services
>>
embrace
extend
extinguish <-- you are here
>>
File: rope.png (440 KB, 645x1260)
440 KB
440 KB PNG
>>88766911
Finally.
>>
well Mark, what language is windows written in?
>>
>>88766804
C# for backend
C# for frontend (winform, WPF, WinUI3, Blazor)
Powershell for scripts
C# for everything else
simple as
>>
I wish the "C is bad" and "rust is good" meme would just fucking die.

Rust in unnecessary. It's the wrong solution to the wrong problem.

People have misunderstood the nature of the beast. The problem is that "C is unsafe". The problem is that "people from India have low IQs".

That's right. That's all the problem ever was. Pajeets write shit unsafe code and that breaks your products.

However creating a "safe" language doesn't stop bad code from being written by idiots. In fact it makes the problem even worse. Why? Because by allowing code to be written mindlessly by idiots (brown people) you are essentially lowering the barrier of entry to programming and making people who can't even manage memory or manipulate pointers by hand (yes they exist, and /g/ is full of them) to think they are actual developers. These people then go forth and write the most worthless, buggy, unusable shit you've ever seen.

meanwhile rust is a bloated, ugly piece of shit. it doesn't do anything C can't. But makes many things C can do MUCH more difficult.

Rust is garbage.
Using it will not stop bad developers and bad code.
Using it WILL create more bad developers who can't code.
Using it WILL make simple tasks for competetent developers take longer, be slower and work less efficiently.
Using it WILL reduce the proliferation of people who actually know what they are doing.

Advocating for rust is bad move. It doesn't improve anything, it just make it easier for shitty people to be shitty.
>>
>>88767211
>b-b-b-but the dark skins
lel
>>
>>88767327
>t. doesn't work as a programmer

If you were a developer then just the thought of an Indian would fill you with such unbearable disgust and frustration that you wouldn't have the audacity to post that ridiculous reply.
>>
>>88766949
retardo
>>
>>88767211
The counter point is that because C is simple on the surface, pajeet is empowered to have a go at it anyways.

The problem is we have is that pajeet isn't going to stop programming, for as long as he's cheaper than a westerner. Now we have also seen that even the highest IQ aryans have trouble writing perfect software, so why not leverage a tool to your advantage? If you [the galaxy brain] and pajeet both write Rust, you're still going to develop something of better quality (higher performance etc), and for that you charge a premium. Pajeet was going for programming jobs you weren't interested in anyways, so the safer he does it, the better for everyone.
>>
>>88766776
"Microsoft drove home this point in 2019 after revealing 70% of its patches in the past 12 years were fixes for memory safety bugs due largely to Windows being written mostly in C and C++. Google's Chrome team weighed in with its own findings in 2020, revealing that 70% of all serious security bugs in the Chrome codebase were memory management and safety bugs. It's written mostly in C++."

Microsoft and Google must really think people are stupid. C and C++ don't create bugs by themselves. Are they seriously insinuating that the compilers create the bugs? I don't get it. Oh that's right, they're trying to deflect from the fact that the bugs exist because of shitty code.
>>
>>88767211
Pajeets don't program in C.
Anyone who thinks "just don't be dumb" solves memory issues is an actual retard who has never worked on a large project with a deadline. The PM has better things to do than listen to you grovel about how the project needs another month because you're still trying to hunt down a leak and the guy who caused it quit his job.
>>
>>88767360
Yes, complex code is hard, and I'm sure you can at least admit that the Google Chrome team and those working on Windows are upper echelon developers.
>>
>>88767211
Finding non-idiot programmers is not only harder, it's directly counterproductive to the corpohomo goals. Corpohomo wants programmers to be silent cogs, as cheap and interchangeable (== disposable) as possible.
First it was Java and C#. Now it's Go and Rust, the new wave of langs actively pushed by corpos to create numerous, cheap, easily wrangled workforce.
>>
>>88767398
>the Google Chrome team and those working on Windows are upper echelon developers.
lol
lmao
the only devs with functioning brains at MS are the NT kernel devs
>>
>>88767420
You have to be 23 to post here.
>>
>>88767211
>filtered by rust

just get good. It's that easy. Or keep writing your header file idk
>>
>>88767360
It's not the fault of the language that Microsoft devs are too retarded to use reference counting. C++ has had reference counting in the standard library since 2011 and well tested libraries have existed for reference counting for a long time before that (ie. as seen in the Linux kernel)
>>
>>88767334
wrong. i'm an sde at a company you know. i've seen just as many absolutely puzzling CRs from (more senior) white engineers as h1bs.
i never said i fully disagree, though.
>>
>>88767100
You sure love the Microcock™
>>
Why is this place so obsessed with rust? Never heard anyone talk about it in the real world.
>>
>>88767482
brainlets coping that they can't complete advent of code day 1 with rust so they have to hate it
>>
>>88767482
>It's the hot new thing to learn [#1 most loved on SO]
>It is a fairly complex language so it enrages people
>It wants to replace c++, which undermines the skillset many people have spent years honing.
>It's made by mozilla, so gets attached to their baggage
>It has a higher than normal ratio of trannys due to the autism requirement.
>>
>>88766805
It's not propaganda.
They can't keep using a language they won't be able to find programmers for.
>>
>>88766804
java For Backend
java swing for frontend
java/batch ok for scripts
java for everything else
simple as
>>
>>88767482
Schizos are obsessed with trannoids and Mozilla used to develop Rust.
>>
>>88767398
If there upper echelon, then they shouldn't writing buggy code.

Have you actually tried to clone and compile Chromium? They have a massive infrastructure tool just for getting all the dependencies and source together and compiled.

Microsoft and Google are known for overengineering everything. Most engineers are writing code with the aim of job security. "Make it as complex as reasonable possible so that they can't fire me" mentality.

Have you seen Group Policy lately? 100s of fucking policies in Windows 11 that have been deprecated in Windows XP. Go open the C:\Windows directory and then the System32 directory. It's a fucking shitshow in there. I would have no idea if one of those files were infected with a virus. Defender is absolute trash and unreliable.

The only logical conclusion that anybody can come to by looking at the mess that is Windows is that it is intentionally a disorganized mess to hide the slimy shit MS is doing in the background.
>>
>>88767482
>Why is the dedicated shill containment board shilling a thing?
Bro, are you actually retarded?
>>
>>88766776
I'm willing to bet that a bunch companies are just gonna keep using it. Especially game dev companies.
>>
>>88767482
Rust is unironically hard.
It immediately filters out C/C++ fizz-buzz "programmers".

It's kinda a problem, cause it makes them believe that it's Rust (and trannies) fault, when in reality they're still to dumb to be writing any real-world C/C++ in the first place.
>>
>>88767471
Yea cause it gave me a 6 figure salary per month
>>
>>88767599
Show us on the doll where Ken and DMR has touched you.
>>
Is there any alternative for embedded stuff? Doesn't C still have a gigantic grip on that field?
>>
>>88767211
>pajeets
>writing C
>>
>>88767696
Only C and Forth are any non assembly languages that matter there
>>
>>88767707
In what language is Windows programmed?
>>
>>88767563
With their resources, they can certainly hire the best. The problem is that even the very best are not perfect. Obviously a project with the complexity of windows will have bugs, and probably a lower bug per line ratio than pure pajeetware, but of the products Google and Microsoft build matter far more.

Ideally, you hire 140+ IQ programmers and you do everything in Ada, and the hard parts in Spark. But, I just see Rust as a step into the right direction. It seems to have taken the best parts from Haskell and OCaml but has made it easier to reason about performance.
>>
>>88767211
Ewww. I don't want to waste my time and effort on this shit anymore.
>>
>>88767420
We're literally in a fucking Mark Russoinovich thread, one of the few functioning brains at Microsoft who knows as much about the NT Kernel as anybody, and he's saying to use Rust.

Appearantly people are so hellbent on parroting the same stale opinion about how c/c++ aren't issues if you are a skilled programmer they aren't even paying attention to who is saying what. The issue with this dumb opinion is the people who actually know what the fuck they're talking about like Mark Russovinich say that's a stupid idea and the only people who actually endorse this idea are incompetent and unaccomplished programmers.
>>
>>88767471
I only have to learn one technology and get good at it
not 5 different
>>
If you're not a retard and write reasonable modern C++ without too much direct pointer manipulation, you're not going to have memory bugs.
Rust only makes sense in an environment where you're forced to hire retards.
>>
>>88767372
Rust has zero use case.
If you can't handle memory management - you use a Garbage Collected language.
Rust's solution is actually braindamage, it's not actually zero cost, it turns a register machine into a stack machine, it forces a shitty concurrency paradigm, etc.
>>
>>88767748
>they aren't even paying attention to who is saying what
If Mark Russinovich jumped off a bridge, would you also jump off a bridge?
>>
>>88767821
And now do all of that but multithreaded.
>>
>>88767835
so much this
>>
What language do people actually learn for AAA games. I don't code at all but I've always been curious about how that stuff works. Like is UE5 based on some other engine?
>>
>>88767851
>And now do all of that but multithreaded.
Yes? And? Do you think multithreading is hard?
>>
>>88767835
>can't handle
Or, just a thought: productivity. Just because you CAN handle fully manual memory management doesn't mean it's a worthwhile use of your time.
I want my computer to work for me, not the other way around.
>you use a Garbage Collected language.
GC performance impact is not always acceptable, and some languages have shitty GCs that only make things worse (Go is an example).
There's also the fact that the borrow checker isn't Rust's only feature. Basic ADTs, pattern matching, iterators, type-based error handling, etc, make it much more pleasant to use than C++ for pretty much anything.
>it turns a register machine into a stack machine, it forces a shitty concurrency paradigm
It sounds to me like you're mindlessly regurgitating things other people say.
>>
>>88767696
No. C++ has never been viable in embedded, and neither will Rust.
>>
>>88767918
Only asm, Forth and C matters in embedded.
>>
>>88767933
pssh
nothing personnel kid
>>
>>88767914
>GC performance impact is not always acceptable
There are langauges with GC implementation that can be faster than Rust, or even C.
LuaJIT is one of those. Julia also comes pretty close.
99% of the time, GC isn't the reason your program is running slow.
If GC is in fact the problem, then you DO need the control over memory, and Rust getting in the way is not the solution.
Use ASM or C.
>Or, just a thought: productivity
If you need productivity, use a GC'd language.
If you are in the 1% of cases where a GC is getting in the way, Rust's memory model will actually lower productivity compared to C.
>It sounds to me like you're mindlessly regurgitating things other people say.
No. Both RAII in C++ and Lifetimes in Rust are using heap like a stack.
Pushing and Popping a heap object into a function stack is still fundamentally a stack.
If you actually want to emulate registers on heap, you have to use malloc().
>>
>>88767964
I wrote
>matter
>>
>>88767211
these are your supreme amerimutt coders
>>
>>88767970
>using heap like a stack

wtf are you on about nigger, you wanna use stack as much as possible, and if you need to use reference counting or allocate heap memory for object of unknown size you go for heap.

This is copium on poot poot level
>>
>>88766804
Nim For Backend
Nim for frontend
Nim for scripts
Nim for everything else
simple as
>>
>>88768004
>you wanna use stack as much as possible
If you use only the stack in C, memory management is 100% automatic btw.
No use for Rust.
>>
>>88766804
Bash is not OK for scripts. If you need to use an interpreted language use node
>>
>>88767970
>There are langauges with GC implementation that can be faster than Rust, or even C.
In very specific cases, not generally.
>Use ASM or C.
I value my time.
>Rust's memory model will actually lower productivity compared to C.
Why? You can't just say "it sucks because it just does". Actually prove it hampers the user, and that you are not just a coping retard.
>No. Both RAII in C++ and Lifetimes in Rust are using heap like a stack.
Rust does not allocate on the heap any more than C does. What are you talking about?
>>
File: 1659651970224022.jpg (223 KB, 914x1200)
223 KB
223 KB JPG
>>88766804
Go for backend
Go for telnet frontend
Go for scripts
Go for nothing else
Simple as
>>
>>88768024
As much as possible doesn't mean exclusively. There's still plenty of uses for fucking vectors, hash tables, and BSTs, which typically will use the heap due to using data of unknown size.
>>
>>88768053
>There's still plenty of uses for fucking vectors, hash tables, and BSTs, which typically will use the heap due to using data of unknown size.

what kind of brain damage is this sentence?
>>
>>88768035
>Bash is not OK for scripts
based and redpi-
>node
neck yourself
>>
>>88768067
He's not the retarded one

your original objection doesn't make sense, so don't be surprised when actual replies are trying to guess what is even your problem.

>it turns a register machine into a stack machine, it forces a shitty concurrency paradigm

this is the whole origin of the brain aneurysm. Lad is just pointing out that you use stack for data of known size and heap for data of unknown size.
>>
>>88768138
1. heap is a software concept, hardware knows no nothing about heap
2. "unknown size" is just a buzzword
>>
>>88767360
Fucking idiot
>>
>>88768191
Okay are you trolling ? Stack is on CPU cache, heap is allocated on RAM, the difference is massive speed difference.
>>
>>88768257
I'm afraid you're the retard here and he's right.
Stack and heap are software concepts (with some support in commonly used CPUs).
>Stack is on CPU cache, heap is allocated on RAM
That's just plain wrong.
>>
>>88768257
>Stack is on CPU cache
Go back to school.
>>
>>88768257
>Okay are you trolling ?
Nope

> Stack is on CPU cache,
Not all CPU have caches, Stack is just a data structure, it can be located everywhere. It can fit in cache, it can be much bigger than cache.

>heap is allocated on RAM, the difference is massive speed difference.
I know no CPU architecture aware of thing called "heap". It's level of abstraction higher than CPU or memory.
>>
>>88767466
>well tested libraries have existed for reference counting for a long time before that (ie. as seen in the Linux kernel)
Manual cleanup bros...
>>
>>88766776
for me it's crystal. i would still use c occasionally for parts which need to be really fast and without much overhead.
>>
>>88767970
Lua and Julia are only faster when you don't actually manage any memory because all your data is in one or two arrays. Look critically at those benchmarks.
"This language with a GC is fast when I don't use the GC" is an uninteresting statement.
>>
>>88766776
I'm not touching rust until a GCC implementation is complete and the GNU project starts using it.
>>88767748
Drew Devault's opinion matters more
https://drewdevault.com/2021/02/09/Rust-move-fast-and-break-things.html
>>
>Rust on bare metal

If that phrase is not a warning in itself you are probably a solid candidate for next Darwin Award
>>
>>88768414
Why do you care so much about feet-fungi eating kike?
>>
>>88767211
>Using it WILL make simple tasks for competetent developers take longer, be slower and work less efficiently.
>Using it WILL reduce the proliferation of people who actually know what they are doing.
Thats the goal dumbass
>>
>>88766776
Sorry, but nu-Microshart is retarded - just look at their JEETSCRIPT GUI. That's why I will continue using Windows 7, which contains no Rust code.
>>
>>88767372
memory issues is solved by writing & testing allocators and never malloc on the fly, YWNBAW
>>
>>88768584
>2022
>using Windoze
>>
>>88768347
>>88768332
>>88768315
https://www.sciencedirect.com/topics/engineering/stack-memory

https://en.wikipedia.org/wiki/Static_random-access_memory

most of the stack is cached on CPU cache
>>
>>88768656
Which still changes nothing in case of what that n00b said
>>
>>88768626
>2022
>Linux still has no thumbnails in the file picker
oh no no no
>>
>>88768727
Linux is a kernel.
>>
>uh, ackchyually, lincucks is...
dont care
linux is trash frfr
>>
>>88767100
>Powershell for scripts
>his scripts don't consist of a single powershell call to execute an inline c# class that does everything
pleb
>>
>>88767835
>>88767970
>Rust's solution is actually braindamage, it's not actually zero cost, it turns a register machine into a stack machine
>No. Both RAII in C++ and Lifetimes in Rust are using heap like a stack.
These are the "expert C programmers" telling you not to use Rust.
>>
>>88766776
wait, isn't azure discontinued?
>>
>>88768772
but linux is a kernel, wtf are you smoking?
>>
>>88768772
Uh oh, the Microsoft schizo is here
>>
>>88768414
https://drewdevault.com/2021/02/02/Anti-Wayland-horseshit.html
>>
>>88768414
>>88769415
>Rust is for fascist nazi bigot antivaxxers who have a hardon for authoritarian compiler checks and think memory unsafety is a conspiracy against them
>>
>>88766804
>c#
Useless trash, stopped reading
>>
>>88768656
ulimit -S -s 4000000
now it isn't
>>
>>88767398
We are really fucked in the next few years.
>>
>>88768191

>"unknown size" is just a buzzword
Okay, now I KNOW you're retarded. Have you never had a program that needed to process some data, including storing it in memory for some time, that you didn't know ahead of time how big it would be? Fuck's sake, even just basic string building tends to use vectors under the hood.
>>
>>88769940
Not him, but you can do dynamic allocation on the stack, a few languages use this to avoid the need to resort to a heap allocator.
>>
>>88767211
>People have misunderstood the nature of the beast. The problem is that "C is unsafe". The problem is that "people from India have low IQs".
>That's right. That's all the problem ever was. Pajeets write shit unsafe code and that breaks your products.
Your whole post is gaslighting. C code from Bell Labs and UNIX companies in the 70s and 80s and Microsoft in the 90s was even worse than it is today. It was full of buffer overflows, gets(), use after frees, and all the bugs that created the bad reputation Windows and UNIX had. C was like this from the beginning.

>In fact it makes the problem even worse. Why? Because by allowing code to be written mindlessly by idiots (brown people) you are essentially lowering the barrier of entry to programming and making people who can't even manage memory or manipulate pointers by hand (yes they exist, and /g/ is full of them) to think they are actual developers. These people then go forth and write the most worthless, buggy, unusable shit you've ever seen.
This is more gaslighting. The kind of programmers Rust was made for have decades of C and C++ experience. Mozilla programmers made Rust because they were using C++ for years. People who have many years of C and C++ experience like this Microsoft programmer and Firefox programmers can appreciate Rust because they know the problems it solves. The "the most worthless, buggy, unusable shit you've ever seen" is usually C code. The C standards committee could adopt that as a slogan. C: the most worthless, buggy, unusable shit you've ever seen.

>Advocating for rust is bad move. It doesn't improve anything, it just make it easier for shitty people to be shitty.
Rust improves things for the best programmers and filters the bad C/C++ programmers.
>>
>>88769940
And here's a page showing some examples.
https://en.wikipedia.org/wiki/Variable-length_array and of course there is alloca
>>
>>88770005
>Rust improves things for the best programmers and filters the bad C/C++ programmers.
This is the key here, because most professional rust programmers also tend to know C or C++ extremely well, so these aren't dumb rubes ignorant to the enlightenment of C.
>>
>>88767211
>People have misunderstood the nature of the beast. The problem is that "C is unsafe". The problem is that "people from India have lo
Jeets are almost non existent in C/C++ positions and i can assure you that the people working on MS and Google C++ codebases are either nords, wasp americans or asians. YET, nonetheless, throughout a period of 12 years, 70% of all the issues identified in chrome and in windows were memory safety issues.

It's easy to pretend it's a problem with the workforce when in reality it is a problem inherently and deeply rooted in the fact that C++ is a broken language. You will never be capable of writing a production software in that language that is free of memory safety related issues even if you're too dumb to identify them.
>>
>>88770113
>Jeets are almost non existent in C/C++
You should see what they try and create in VHDL. Monstrosities that you could never believe. They do not, and will not ever, understand how FSMs work. Fortunately these projects never make it to production (because they can't) but it's usually some first draft that pajeet implemented. I'd take a nignog that says "idunnoo maan" anyday over a pajeet that thinks he does. 100% sir I know java so I know VHDL and VERILOG thank you kindly.
>>
>>88769347
unfortunatelly it is not
>>
>>88770208
then fuck me
>>
>>88770052
Well, if someone was professionally programming in C/C++ for more than 3 years and is still filtered by Rust then he may be retarded.
>>
>>88769940
>Okay, now I KNOW you're retarded.

No, you don't.

>Have you never had a program that needed to process some data, including storing it in memory for some time, that you didn't know ahead of time how big it would be?

Yes I heard.

>Fuck's sake, even just basic string building tends to use vectors under the hood.

The point being? You are literally shitposting for sake of shitposting and disprove nothing. Only thing you prove however is that you have no fucking idea what you are talking about. Sorry not sorry

>>88769980
You can allocate memory whenever you want. Even on a floppy disc. Allocation is just abstraction over access to memory. It's most commonly done nowadays on OS level. But could be done on software level like virtual machines, or in application itself - like it was common in older games like for example in Doom or Wolfenstein3D that have own allocation management.

Stack is just data structure. CPUs commonly reserve for stack purposes some areas of memory, but it's still the very same memory that the rest of memory on the machine.
>>
>>88770466
tl;dr; version
>anons in Rust thread have no idea how memory managment works and memory in general
color me surprised
>>
>>88767748
NT kernel, just like windows, is shite
>>
>>88767868
C++
>>
>>88767914
>Says as he is getting pegged by the cooompiler
>>
>>88767964
I hate C# shills. Your language is slow and ugly, stfu
>>
>>88767100
>Blazor
one of the slowest frameworks on the web
>>
>>88767539
BASED DURGA SIR NAGOOR BABU SIR
>>
File: 1660513460459394.jpg (48 KB, 645x729)
48 KB
48 KB JPG
>>88767372
>filtered by C
>>
>>88768410
Can't believe there are retards who believe GC'd languages can reach the speed of non-GC'd langs
>>
>>88768414
Rust is a disaster as much as C++ is
>>
>>88768727
Works on my machine. I use KDE.
>>
File: 1627743917511.jpg (216 KB, 660x720)
216 KB
216 KB JPG
>>88771776
>it works on my ma-KRASH!
>>
>>88771813
>He fell for the krash meme from 2008
Works on my machine
>>
>>88771760
nocoder myth. GC is more efficient than manual memory management. It avoids needless copying, the application stops to collect memory less often and it allows for better heap compaction.

Manual memory management only looks better on poorly written microbenchmarks.
>>
>>88769484
Real
>>
>>88768414
>Drew Devault's opinion matters more
Hare is literally the solution to the Rust/C++ war
>>
>>88768656
>most of the stack is cached on CPU cache
You're mentally retarded, if you suddenly have a heap heavy workload, your stack will get pushed out of cache.
The only way this wouldn't be the case is if you used NonTemporal reads/writes on heap.
The only other way this could be the case was if the cpu had an option to mark memory to not be cached out of memory, which could be set by the operating system, and we aren't living your fantasy world with on stack cache.
>>
>>88771861
>Stopping the program and checking for every value in the heap and decide to clean it or not is as fast as not doing any of this at all

You're a fucking retard
>>
>>88766804
Rust + Embedded is very cringe
>>
>>88771861
Do you have benchmarks showing this?
How does it impact memory usage?
>>
>>88770017
alloca(3)
The alloca() function returns a pointer to the beginning of the
allocated space. If the allocation causes stack overflow,
program behavior is undefined.

Can't really use alloca() without eventually blowing your foot away
Iirc Varriable Length Arrays have a size 1-8MB limit compared to Static Length Arrays which go far past 8MB.
>>
>>88771897
Do you think that freeing up memory manually somehow does not do "any of that" at all? Do you think memory magically frees itself?
>>
>>88771861
Unless you code like a retard, that's false.
HOWEVER, the vast majority of programs will run just fine on a GC language. Kernel, embedded, window managers, DEs should still be written in compiled non-GC languages however
>>
>>88767970
>Use ASM or C.
>Rust's memory model will actually lower productivity compared to C.
Maybe if you just do NeetCode all day, but for real developers, Rust is awesome, and the more you use Rust, the simpler it becomes.
The only ones complaining are idiots who get filtered by the borrow checker and gave up cause "Why can't I do this? I can do it in muh C!".
>>
>>88771956
Yes, you fucking retard. Manual memory management means I don't have to stop the execution of the program every fucking ms or some shit to clean up the mess that the retarded Pajeet developer sharted. It means you fucking clean variable whenever you finish using it you dumbfuck. Have you ever programmed in a non-GC'd programming language? Do you think people who make AAA games and operating systems are just retards for using langs without a fucking GC and lose muh productivity? This definitely has to be a bait, can't believe people are this dense
>>
>>88771949
/STACKSIZE:1024M
there you go, problem solved.
note that 1GB is reserved, but not committed, so you don't actually waste 1GB of real memory. On 32 bit systems you might not want to go full hog wasting entire 1GB of your address space unless the vast majority of allocations are stack allocations, though.
>>88771956
If you create private heaps for parts of your program, allocate with HeapAlloc, and then delete the entire heap when you're done, it can be very very fast compared to fucking around with malloc and free (ew). Because deleting a heap can blow away thousands and thousands of allocated objects without having to loop through them all.
>>
>>88766776
The only time I ever heard about Rust was on this place, in the real world everyone uses Java, C++ and C#
>>
>>88772045
>Yes, you fucking retard.
Stopped reading there, no point arguing with someone who thinks manual memory management is more automatic than actual automatic memory management.

>>88772050
>Because deleting a heap can blow away thousands and thousands of allocated objects without having to loop through them all.
That's true. It's part of the reason why modern concurrent GCs are so fast.
>>
>>88772006
>neetcode.io

Borrow checker sucks and causes a lot of time wasted and performance penalty
>Vec<Rc<RefCell<Box<Fuck<Shit>>>>>
>>
>>88772091
Don't use rust with any api that uses callbacks and don't write gui in rust (aka don't be retarded) and you should be fine.
For command line apps, rust is completely fine
>>
>>88772111
>rust can only be used for writing shitty linux terminal toy apps
jesus christ
can some rusttranny tell me if this is actually true? imagine a language that can't even do GUI. now imagine this language as the main languages of linux. LMFAO. KEKWell.
>>
>>88772111
I agree, it's nice for clis
>>
>>88770193
Really? I've never seen an fpgajeet, just embedded swjeets. Those are bad enough. Tell some more fpgajeet stories.
>>
>>88772148
It CAN do gui. but unless you are using a library that uses opengl like egui, you will have to use callbacks and that will make your life extremely hard, because chances are you will reference other widgets in the callbacks, and this means that since callbacks can run at any time, rust will demand a 'static lifetime, which is impossible to achieve in gui apps. So you will then have to wrap the widgets in Rc, or Rc<RefCell<>> if you use mutability as well. It can be done, but compared to C++ it is less efficient to execute and less efficient to write.
However you can still code the core functionality in rust, and then do gui in C++ and interop with the rust library you wrote previously. That's the recommended approach
>>
>>88772148
>areweguiyet.com
The ecosystem isn't as mature as older languages, but you can still do basic GUIs in Rust.
tauri (tauri.app) is promising. For frontend, yew.rs looks nice.
It's not that it can't do UI (there are plenty of examples of very nice UI done in Rust), it's just that none of them are mature and fleshed out yet. It's going to be a few more years before there are any contenders that can face libraries like GTK and Qt.
>>
>>88772221
Jon Blow was right. All these "big ideas" languages are so annoying. Maybe Zig will compete on the market share of C++. It supports OOP, right?
>>
>>88767100
Good morning, sir.
>>
>>88772371
>It supports OOP, right?
Very poorly. It doesn't have inheritance afaik.
C++ is much much better for OOP. Or even D, even though D does not support multiple inheritance which is just ridiculous. People should stop making C++ killers that remove features from C++
>>
>>88772429
Multiple inheritance is probably bad. Java works fine without it
>>
>>88772505
It has it's uses, but it can also be abused. still no argument for removing it
Also while java does not have multiple inheritance for extends it does have multiple inheritance for implements afaik
>>
Why not just use C++ with memory safety features?
>>
>>88772429
I think the market will improve with a less radical change than Rust. Like people would just need a decent build tools, cargo is pretty nice for example, and maybe a compiler option to check for out of bounds for tests and stuff. I guess it's why some people still shill Pascal.
>>
>>88772593
Well if you only want that and don't mind GC, then Go is pretty much this.
Also hare has bounds check, but ddevault does not want to add package manager, because he thinks it would be a security vulnerability
>>
>>88772429
That's weird. I guess they're aiming more at C then not C++
>>
>>88772638
Zig is Neo-C.
(Go is nu-C.)
>>
>>88766776
>Microsoft
Stop
>>
File: 1663943779005.gif (564 KB, 800x430)
564 KB
564 KB GIF
>Microsoft hires father of systemd
>We should listen to Microsoft Devs when it comes to big sweeping changes
>>
>>88767211
You need to stop posting immediately
>>
>>88767334
I'm a developer and you are a fool
>>
>>88767334
I am a developer and I outsource my job without my boss knowledge. Why don't you the same? It's really comfy
>>
>>88773245
And those of us that have been in the industry long enough will know that M$ will be telling developers to use something "different" in less than 12 months depending on how much of M$'s shit developers eat.
>>
>>88766776
no.
microsoft is dead. their employees are exclusively retarded trannies. they say bullshit all the time, and nobody cares about it anymore.
microsoft is on its deathbed. already dead, some people just didn't notice yet.
>>
microsoft is going to be AI managed by a team of 10 or so jews thats thefuture of corporate america
>>
>>88773766
Is Mark Russinovich a retarded tranny?
>>
>>88773816
he is a litterally who nobody MBA who never touched a computer
>>
File: 250w.jpg (36 KB, 250x328)
36 KB
36 KB JPG
>>88766776
What's the best way to learn rust? "The book"?

https://doc.rust-lang.org/book/

There are a few other resources here but not sure what to choose:
https://www.rust-lang.org/learn

Or should I just get a physical book? Worried it might be out of date, but I was looking at "Programming Rust, 2nd Edition" (pic rel)
>>
>>88773891
Absolutely wrong, educate yourself.
>>
File: 1660673596139-0.jpg (28 KB, 600x490)
28 KB
28 KB JPG
>no defined memory model
>shit generics, weak const fn
>no released specification
>immature ecosystem
>barely used in real world systems programming
>runs on a fraction of platforms supported by c++

give it a decade and I'll consider it.
>>
>>88773896
I learned from the official book. IMO it's fine, but some people don't like it and the O'Reilly book is supposed to be better.
I wouldn't say it's out of date. The only big thing it seems to be missing is async but you don't need that for everything.
It's typically pretty easy to bring your practice up to date because the fundamental way you write code hasn't changed much since 1.0 and clippy (the official linter) can give suggestions for modernization.
>>
>>88766804
OCaml for backend
Reason for frontend
OCaml for scripts
OCaml for everything else
>>
File: 1662919201234431.png (135 KB, 316x501)
135 KB
135 KB PNG
>>88766776
Remind us the reason that C++ got so fucking popular in the 90s and now they are flexing their language king making muscles again. Imagine a world where Pascal became the defacto systems language.
>>
>>88766776
just give me an alternative then, rabbis
>>
>>88774141
Microsoft can only do kingmaking if it's actually relevant
Use Linux and you'll never be infected with RUST
>>
>>88767211
Correct.
>>
>>88774159
>Use Linux and you'll never be infected with RUST
https://linux.slashdot.org/story/22/09/19/205239/linus-torvalds-rust-will-go-into-linux-61
There's no escape, chuds.
>>
>>88774159
Oh no no no no no
>>
>>88774190
Time for NetBSD and their lua kernel modules
>>
>>88766776
>no guys really you should just use C# for everything
>J++? N-No it... it has nothing to do with that I barely knew her we were just friends!!
>>
File: wojacked.jpg (434 KB, 1328x1488)
434 KB
434 KB JPG
>>88774190
>>88774199
God is dead
>>
>>88774226
Join the OpenBSD white men committee.
>>
File: 1662640407450070.png (144 KB, 500x543)
144 KB
144 KB PNG
>>88767360
>>88767372
Sounds like there's a basic flaw in hiring practices. If they focused on hiring experts in their tooling instead of the jeet and/or leetcode circlejerking, there would be fewer problems.
>>
>>88774226
https://cdn.netbsd.org/pub/pkgsrc/current/pkgsrc/lang/rust/index.html
No escape.
>>
File: images (2).jpg (16 KB, 360x360)
16 KB
16 KB JPG
>>88766804
PHP for backend
JS for frontend
Python for scripts
C/C++ for everything else
>>
>>88767372
Hello sir.

>>88767398
Good morning sir.

>>88767450
Sir.

>>88767539
Java means durgasoft.
Durgasoft means Java.
Sir!

>>88767599
Do not redeem, sir.

>>88767707
Good morning!
>>
>>88767100
Ah, I have someone like you in my team, man he sucks. Senior dev with several decades of experience in backend and dev ops and yet he writes such shitty code and relies on outdated concepts, it's not even funny. And of course he can't even properly write simple React components.
>>
>>88773245
Only luddite cniles who hate systemd also hate Rust.

Documented standardized interfaces are obsolete UNIX shit for PDP-11s. Move on, gramps.
>>
>>88774253
That's just the compiler in their package manager for third-party software.
>>
>>88774391
That's how the infection starts.
>>
>>88767211
>has to work for a company that outsources their code or hires incompetent devs
Guess you are such a huge insecure faggot that no decent company actually wants to hire you. Pathetic.
>>
>>88767211
So Rust is the new Java?
>>
>>88766776
>>88767360
>Windows is a shitshow
>Microsoft: It's the language that's the problem!
lel
>>
>>88774190
Who cares.
If it works and doesn't break shit then why does it matter. It's not like I touch much of the kernel anyway.
I can't imagine why anyone would care. If it gets his blessing (and isn't removed like C++ was) then whatever. Not my problem and I bet it isn't the problem of anyone complaining about it either.
>>
>>88772591
Which memory safety features?
>>
>>88774605
smart pointers, debugger, memory leak detector
>>
>>88774615
lol
>>
>>88766776
Sure, everybody should use Visual C++, no doubt.
>>
>>88774581
Rust doesn't support every architecture that Linux supports at the moment. So the more kernel relies on Rust the less portable it becomes. For you that means less hardware support.
Bootstrapping Rust is also hilariously hard in comparison to GCC.
Having two official kernel languages puts twice the pressure on kernel subsystem maintainers because now they have to be experts in both languages. For you that means having the kernel be sub-optimally managed.
Rust is atrociously slow at compilation due to its nature. It also requires way more resources. For you that means distributions will be limited in frequency of kernel updates.
All in all this might be all worth it, but might not. We'll see in the future.
>>
PLEASE SIRS ITS GETTING TOO EXPENSIVE TO HIRE ACTUAL PROGRAMMERS
>>
>>88774647
Don't "lol"
Modern C++ is fully memory safe if you don't purposely write bad code that is obviously wrong
>>
>>88774141
Is Visual Basic M$'s Pascal?
>>
>>88774232
J++? Are you making up new languages, anon?
>>
>>88774775
Those are all band-aids on a language that's full of undefined behavior.
Even if you artificially limit your compiler to only accept the "safe" side of C++ it's not going to be enough.
>>
>>88774737
except it's not being used for any of those things so your argument is already useless.
it's not (currently) replacing C. you're fearmongering with imaginary scenarios.
>>
>>88774792
Jay Plus Plus was an actual thing for a while - https://en.wikipedia.org/wiki/Visual_J%2B%2B
>>
>>88774402
Take your meds
>>
>>88774792
Has the school informed your parents they're moving you to the sped class yet?
>>
>>88774782
It's MS Delphi if anything
>>
>>88774445
Nah
>>
>>88774512
REAL SHIT
>>
>>88774802
The first rust drivers are a NVMe driver and re-implementation of the Intel e1000e driver which is one of the most ubiquitous NIC.
I am not fearmongering, you're projecting.
>>
>>88774810
Lol, Microshit picks the worst names ever
>>
>>88772111
>don't use callbacks
>don't use literally the only feature or progamming that is actually reusable
>your language is so garbage you can't actually use any useful paradigms
Can rustniggers defend this?

If we never left assembly and never went to modern multitasking operating systems tech would have none of these problems and 41% less suicides.
>>
>>88774885
Imagine a world with 100% suicide rate of rustroons uwu
>>
>>88772429
>zig has poor OOP
>it doesn't use the most useless and retarded feature of OOP

All a language needs to have good OOP is encapsulation.
>>
>>88774445
No, because Java was actually good
>>
>>88774042
thank u my negro
>>
>>88774914
I don't get it. No OOP, no function overloading, are people supposed just to repeat the same code for all different possible types? ._.
>>
>>88774885
People have been complaining about shitty OO/callback libraries for years before Rust came along.
>>
>>88774914
Inheritance is convenient and good. If I ever want less features in a language I will just use C
>>
>>88774997
Can you name a *native* *GUI* that doesn't use OOP?
>https://imgur.com/a/jDwl8n4
>>
>>88774032
>runs on a fraction of platforms supported by c++
which platforms actually matter?
>>
>>88766776
>Russinovich
>yes goyim use our C# + enterprise VS for much shekels
>>
>>88774997
Why would they be complaining about callbacks? It's literally the only thing that makes code actually reusable.
If there is something shitty about GUI libraries, it isn't the callbacks. Some of them go overboard with OOP, but I can't think of a more streamlined way to create GUIs than an event loop with callbacks.

Using callbacks with OOP without any kind of inheritance is the God Tier way of programming.
>>88774974
Zig has generics that work in comptime
>>88775064
>Inheritance is convenient and good
>have to go through an entire tree of objects just to understand some hidden behavior
pass

The only semi reasonable inheritance related idea is traits because you can sensibly reason about the type.
>>
>>88775170
>have to go through an entire tree of objects just to understand some hidden behavior
Sounds like a you problem. I never had trouble with understanding this. besides better this than code repeat
>>
File: 1505851339261.png (523 KB, 671x678)
523 KB
523 KB PNG
>>88766776
Haven't people been saying this for years? Also isn't Azure competing directly against C and C++? Why would you care what the competitor for something like this says, especially the fucking CTO apparently.
>>
>>88775186
>>88775170
>>88774997
>>88774885
>>88774905
>>88775075
>>88772111
>>88772148
>>88772221
>>88772371
>>88772429
>>88772359
Have you guys seen the new functional/reactive Rust GUI libraries?

https://github.com/fschutt/azul
https://github.com/redox-os/orbtk


use orbtk::prelude::*;

fn main() {
Application::new()
.window(|ctx| {
Window::new()
.title("OrbTk - minimal example")
.position((100.0, 100.0))
.size(420.0, 730.0)
.child(TextBlock::new().text("OrbTk").build(ctx))
.build(ctx)
})
.run();
}


use azul::prelude::*;
use azul::widgets::{button::Button, label::Label};

struct DataModel {
counter: usize,
}

extern "C"
fn render_dom(data: &mut RefAny, _: &mut LayoutInfo) -> StyledDom {

let data = data.downcast_ref::<DataModel>()?;

let label = Dom::text(format!("{}", data.counter))
.with_inline_style("font-size: 50px;");

let button = Button::new("Increment counter")
.onmouseup(increment_counter, data.clone());

Dom::body()
.with_child(label)
.with_child(button.dom())
.style(Css::empty())
}

extern "C"
fn increment_counter(data: &mut RefAny, _: &mut CallbackInfo) -> Update {
let mut data = data.downcast_mut::<DataModel>()?;
data.counter += 1;
Update::RefreshDom // call render_dom() again
}

fn main() {
let initial_data = RefAny::new(DataModel { counter: 0 });
let app = App::new(initial_data, AppConfig::default());
app.run(WindowCreateOptions::new(render_dom));
}
>>
>>88775170
>Zig has generics that work in comptime
This is terrible. How would one only accept certain generics? Like the
constraints.Ordered
in golang?
>>
>>88775170
>The only semi reasonable inheritance related idea is traits

So does Zig have traits?
>>
Cto already? Best Windows hacker in the industry turned corpo
>>
File: 1663952643064.png (2.08 MB, 1925x1307)
2.08 MB
2.08 MB PNG
I just started learning C though.
>>
>>88774190
oh no rust sisters we got too cocky and spergy
https://lkml.org/lkml/2022/9/19/1105
we forgot there are no upboats on lkml!
>>
>>88766804
Bash for backend
Fuck frontend
Bash for scripts
C for bash tools
Bash for everything else
>>
languageFor :: LanguagePurpose -> Language
languageFor _ = Haskell
>>
>>88774375
>Documented standardized interfaces are obsolete UNIX shit for PDP-11s.
UNIX was not documented or standardized. That's why Solaris had 6 different versions of some commands.
>>
File: poo in loo.png (172 KB, 842x412)
172 KB
172 KB PNG
>>88767539
>>88774933
good morning sirs
>>
>>88775265
>Azul uses webrender (the rendering engine behind Firefox) to render your UI
Great, another fucking chromium is what the world needed.
>>
>>88775489
>https://www.youtube.com/c/KathrynHodgeblondiebytes
>>
>>88767211
Post credentials
>>
>>88775431
KEK
Linus will cause an increase in rustroons suicide rate with his emails
>>
>>88766776
Boomers want to destroy everything they did not make themselves.
>>
File: 1659966490857747.png (28 KB, 129x116)
28 KB
28 KB PNG
>>88774190
based
>>
>>88775667
>>88775431
>>
>>88775431
>inb4 linus gets cancelled
>>
>>88766804
This
>>
File: 1663197907946204.jpg (18 KB, 300x225)
18 KB
18 KB JPG
>>88775681
You're delusional if you think it won't happen
>>
>>88775750
What you're seeing there is the line in the sand that rusty-wusties don't know how not to cross
They have a rigid idea of safety, Linus is saying they need to compromise, they don't know how to do that so they WILL cross that line and there will be a big argument and then Linus will ban them from systems programming forever
And then everyone, the whole world, will give up on Rust and go back to C and my parents will love me again
>>
File: shill.png (77 KB, 357x500)
77 KB
77 KB PNG
>>88775824
It's not too late to take the Rust pill
>>
>>88775861
>>88775824
I don't get the seethe over rust. it's only going to be in the drivers and only in the *new* drivers, so nothing will get rewritten, and besides segfaults happen in drivers all the time and rust is perfect for the drivers
>inb4 muh troons
I heard troons also breathe air, so by that logic you should kill yourself
>>
>>88775824
Read the followup messages.
What you're actually seeing is a levelheaded discussion about exactly where the line should be drawn, with the Rust programmers concluding that
- Some functionality should be marked unsafe to signal that the programmer has to manually verify the rules
- Some guarantees can be made using clever type system tricks
- Some guarantees can be made using custom linting
- For the rest compromises must be made
>>
c++ is the problem but rust is not the solution
>>
>>88767719
The file manager in Win 11 is webshit
>>
>>88775910
If rusty-wusties can't compromise with the rest of us how can they compromise with Linus?
>>
>>88775936
Read the mailing list
>>
>>88767719
C++
I kid you not they tried to rewrite it in C#, but it didn't work
>>
>>88775944
can't do much of anything after stack ranking cleared out anyone with an ounce of sense. if anything it's a miracle that c# has come out as good as it has
>>
>>88775936
They can compromise with Linus because Linus generally knows what he is talking about.
>>88775944
Well, it did kind of "work". It just turned out to be a fundamentally stupid and pointless endeavor.
>>
>>88775699
what, again? he's already cucked
>>
File: 1662284645070.jpg (56 KB, 850x1063)
56 KB
56 KB JPG
>>88766776
Does he consider COM, Windows API and WinRT deprecated as well?
>>
>>88767624
>6 figs per month
you aren't earning 7 figs a year with C#
>>
>>88775431
Trannies on suicide watch yet again.
>>
>>88766776
t. Mark Pajeetovich
>>
>>88775431
The audacity of these fucking people asking for system-wide changes as soon as they implement a single line into the kernel.
The label "troon" is more than a shaming term, it's an accurate descriptor.
>>
>>88775940
I don't know how to find the replies
>>
File: file.png (13 KB, 177x120)
13 KB
13 KB PNG
>>88776092
Bolded is current, below the bolded are the followups
>>
>>88776108
Can you just link to them
>>
>>88775996
Yes, Microsoft considers all of those legacy. Win11's shell is partially webshit tech now. They will push it more and more until another shiny thing comes along and it'll be added to the 17 already present UI toolkits in Windows.
>>
>If you cannot get over the fact that the kernel may have other requirements that trump any language standards, we really can't work together.
>>
>>88775431
What exactly is this random instance of Linus getting heated on the mailing list supposed to prove? He's just saying that dynamic checking is bad in certain contexts, so in those contexts only unchecked (unsafe) or statically checked stuff should be used.
>>
>>88776193
mostly it shows that rust-"acians" are insufferable and demand a mile if you give them an inch
>>
>>88776215
How?
>>
>>88776193
He's also saying that in some contexts no checks can be done and that you can't just call it quits on failure in kernel code, ever.
>>
>>88776232
you're doing it right now
>>
>>88776136
He's too afraid to show the autistic replies of rustrannies who will make Linus kick Rust out of the kernel kekw
>>
>>88776245
LMFAO
>>
>>88776239
Yes, which is why the standard library introduced fallible allocations and there's no reason that other areas where fallible APIs are unavailable won't also be upgraded if they are desired in the kernel.

>>88776245
What am I demanding?
>>
>>88776193
Read through the mailing list. Rustrannies working on the kernel as we speak can't into debugging, and think Rust is the magic elixir that solves this problem, if only the system would change to accommodate it's needs.
>>
>>88776367
Feel free to cite some examples of this.
>>
>>88776376
>>88776311
>>88776232
>>88776193
rust "community" running damage control
>>
>>88776376
>spoonfeed me
Stop playing dumb, it's literally in the first anon's pic
>as you know, we're trying to guarantee the absence of undefined behavior
>>
>>88771625
The kernel is fine. They just never stopped using DOS/9x mentality when it came to userspace.

>>88774834
Pretty much this.
>>
>>88776409
Why aren't you answering my simple questions?

>>88776451
If you actually read the post that Linus is quoting, you'll see at the end that the author is literally asking for Linus's opinion on it. Not demanding that it should be done as he is proposing.
>>
File: BookPlan.jpg (461 KB, 816x1056)
461 KB
461 KB JPG
>>88775393
And keep on learning it. Resources abound and its still more widely used than Rust.
>>
>>88776501
When did I use the word demand?
>>
>>88776540
>if only the system would change to accommodate it's needs
I admit I used the word "demand" because the other anon did. But your post has the same meaning.
>mostly it shows that rust-"acians" are insufferable and demand a mile if you give them an inch
>>
>>88776578
My main point is that Linus is going to spend more time tard-wrangling these people, who think Rust is going to save them from memory errors and debugging, than he could possibly save by implementing the language in the first place.
This is a common mentality among Rust users riding the newfound wave of interest.
>>
>>88776683
Do you actually think Linus has never dealt with clueless people on the mailing list before Rust? This isn't even on the same level as the biggest retards that have contributed to the kernel.
I'm willing to bet he'd take a Rust troon over another Kay Sievers any day of the week.
>>
>>88776683
Rust trannies are just like any other group of leftists. Destroy, denigrate, hypermoralize, and implement solutionism.
>>
>>88774799
Why is it "not enough"? What more do you need?
>>
>>88776683
It's possible, although Linus has also said that one of his goals of Rust in Linux is to get new people into kernel development.
>>
>>88766776
he is just going through tranny rust phase, like everyone give him few days
>>
>>88776754
>It's possible, although Linus has also said that one of his goals of Rust in Linux is to get new people into kernel development.
Good God, they really neutered the old bastard, didn't they?
>>
>>88776774
who's going to take over after he kills himself though
>>
>>88776776
that's the thing about cucking out - there's no coming back from it
>>
>>88776727
Kay Sievers is one person, Rust is a steady supply of tards
>>
>>88772591
nooooo shut him down!
>>
>>88776504
oh boy - downloading that Algorithms copy. I had only heard of the latter editions that are in Java or some shit.
>>
>>88776788
There's still close to 60% chance he won't.
>>
>>88776735
So you admit your only actual qualm with Rust is political and based off of a /g/ meme?
>>88776812
There are also a steady supply of arrogant krauts who think "no" means "yes".
>>
>>88774615
>debuggers are memory safety features
The absolute state of /g/.
>>
>>88776850
>the absolute state of /g/
You know, you don't have to be here. You are free to go back to where you came from (which we all know is reddit)
>>
>>88776834
>So you admit your only actual qualm with Rust is political and based off of a /g/ meme?

My qualm is with a bunch of retards that blame the tool instead of their lack of knowledge. Instead of understanding C or C++, they jump on some self righteous circle jerk.
>>
>>88766804
TS for backend.
TS for frontend.
TS for scripts.
JS -> WASM for low level bullshit.
>>
>>88776735
>Destroy, denigrate, hypermoralize, and implement solutionism.
You're confusing Rust with C. This is exactly what the C community does whenever someone doesn't want to use C.
>>
>>88776917
Why are you all going on and on about web app crap?
I thought Rust is a serious low level systems programming language.
>>
>>88776893
>My qualm is with a bunch of retards that blame the tool instead of their lack of knowledge.
Like you blame Rust because of your lack of knowledge?
>Instead of understanding C or C++, they jump on some self righteous circle jerk.
The majority of Rust programmers have more years of C and C++ experience than you've been alive.
>>
>>88776928
Because Rust is a solution to a problem that has been answered by, " quit being a fucking retard and understand your tools".
But I do see where you're coming from. Both camps are obsessed with orthodoxy and dogma.
>>
>>88776937
The post you replied to didn't pick rust for anything are you fucking blind
>>
>>88776951
>The majority of Rust programmers have more years of C and C++ experience than you've been alive.

ACK- says otherwise
>>
>>88776745
Because core "features" of C++ contain undefined behavior. If you want to make C++ safe it's not going to be C++ in the end.
>>
>>88776987
Which core features of C++ contain undefined behavior and why can't you just not use those?
>>
>>88776976
The post argues about webshit with a million other posts in what is supposed to be a Rust vs C/C++ thread.
Why are webshits even here? They don't know shit about C. But apparently they care deeply about Rust.
>>
>>88776928
C is the Rust of the 70s.
>>
>>88776075
The kernel NEEDS to respect OUR pronouns and make a safe space for US. This will end up in a disaster, and the trannys will just start (attempting) to rewrite the kernel in rust...
>>
>>88776980
They hang on to rust in order keep from hanging off the ceiling.
>>
File: 1647653505622.png (392 KB, 480x640)
392 KB
392 KB PNG
>>88777064
>>
>>88766804
This. Jou just had to listen, /g/..
>>
>>88776178
my favorite part. that was a dangerous enough reply that another rust troon came in to prevent the other troon from fucking it all up.
>>
>>88777006
https://alexgaynor.net/2019/apr/21/modern-c++-wont-save-us/
>>
>>88777656
>gaynor
>>
>>88777656
>look here's a handful of C++ features that could result in problems in contrived scenarios
Then either don't use them or fix them.
>>
>>88777006
References, pointers, initialization, arithmetic
>>
>>88777786
literally just use smart pointers; what's the problem?
>>
>>88777834
You mean reference-counting everything? Why not just use a GC language at that point?
>>
>>88777704
>There are significant challenges to migrating existing, large, C and C++ codebases to a different language – no one can deny this. Nonetheless, the question simply must be how we can accomplish it, rather than if we should try
appropriate surname
>>
>>88777856
>everything
Not everything, just what is performance critical. In GUI apps almost zero smart pointers are needed. Take a look at FLTK for example. It is LITERALLY impossible to unwillingly create a memory leak or a segfault
>>
>>88777856
You don't need to reference count everything. Just use them instead of raw pointers. That's it. There's no actual downside.
>>
>>88766804
front & backend can both use TS and then benefit from simply typing the whole API to guarantee the contract without swagger or other good morning sirs
>>
>>88776917
>JS -> WASM for low level bullshit.
That does not mean what you think it means
>>
>>88779202
webdevs consider it low level when they're only using two frameworks on top of the browser
>>
>>88778374
How will you share the same types over the two apps? Copy and paste?



Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.