[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] [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.


[Advertise on 4chan]


File: Tor Rust.png (62 KB, 905x402)
62 KB
62 KB PNG
https://itsfoss.com/news/tor-rust-rewrite-progress/
>>
>>107527035
What is the chinese Ministry of State Security's opinion on rust?
>>
File: hmm.jpg (6 KB, 225x224)
6 KB
6 KB JPG
>>107527035
>puts its trust in
What are C#, Go, Java, and Ruby written in, again?
>>
File: 1760196378661619.png (305 KB, 1031x1085)
305 KB
305 KB PNG
was memory unsafety so bad that we need to rewrite literally everything? i don't remember ever hearing about memory issues all that much, they're treating this like it's the bug of the millennium.
>>
>>107527035
Should have switched to Hare instead https://www.youtube.com/watch?v=6wQ7pl7DVHw
>>
>>107527253
>was memory unsafety so bad
if you check here https://app.opencve.io/cve/?vendor=tor
they are as incompetent as it gets.
>>
>>107527253
this just means someone was paid to do it otherwise why would you
>>
Rust is proof that it's not a language problem, see: cloudflare incident.
>>
Rust developers aught to take care to make their software bootstrappable and not require long chains of version upgrade compilations.
>>
>>107528187
According to rust devs, doing so with gcc-rs is a very bad idea.
>>
>>107527035
Is it a bad time to start learning C++
>>
File: 1762528273895060.png (112 KB, 2400x2000)
112 KB
112 KB PNG
>>107527253
Memory safety is just so good that you need to rewrite everything,
>>
>>107528449
It's funny how sane people think that improvements should be done in small parts, on most important, and more importantly, newest code only.
Rustroons write no new code and their npm clone only enforces that.
>>
File: pepe33.png (236 KB, 528x438)
236 KB
236 KB PNG
>>107527035
Why is the TOR project constantly making very sus decisions lately?
This is just one of many in the recent years.
>>
>>107528514
You need complete memory safety to get the win of memory safety: Eliminating process context switching costs.
>>
>>107528681
Last I checked, rustroons rave about async and async is all about generating as many context switches as possible.
>>
>>107527253
Memory safety vulnerabilities tend to be worse than other kinds of vulnerabilities because you can escalate them into e.g. arbitrary code execution, while logic bugs tend to be more situational and less interchangeable.
They're a big deal in browsers and such. Those systems apply lots of different mitigations to prevent memory safety issues and limit the damage they can do but it's not enough.

>>107528857
Isn't the point of async to reduce the cost of the context switches you have to make anyway? You won't do better with threads.
>>
>>107529427
All rust async runtimes already use threads, simply put, it's misdesigned pile of crap.
>>
>>107527035
Oh no, it's sad news.
>>
File: 1765396358959230.png (608 KB, 1893x806)
608 KB
608 KB PNG
>>107527035
Good
Now spread this so that a new forks will reborn
>>
Rust has a backdoor, same as Java. C compilers can have them too but C is simple enough that there are from-scratch compilers written in assembly and so on.

Rust, Java, these "safe" troon languages all emanate from one big globohomo company or organization. There is only one implementation which is usable and the language will be arbitrarily updated over time to enforce this, making it impossible for anybody to catch up to Rust from five years or ten years ago let alone Rust as it is today or tomorrow.

>inb4 point to the backdoor
Go look at the CVE list for Rust and its cargo crates. New ones are inserted regularly as the older ones are discovered, as with Java and .NET.
>>
>>107527253
Memory unsafety wasn't as bad in the past but as C and C++ became more popular and replaced safer languages like BASIC, FORTRAN, COBOL, ALGOL, PL/I, Pascal, and Ada, there are more and more memory bugs. Even Lisp, APL, ML, and Prolog used to be more popular than C.
>>
>>107531022
>C compilers can have them too but C is simple enough that there are from-scratch compilers written in assembly and so on.
You're confusing C with Pascal, Oberon, or FORTRAN. C is such a complicated language that most C compilers are written in C++ because writing a C compiler in C is too hard. Nobody even knows what a line of C does because the standard is so ambiguous.
>>
>>107531096
And yet, it's a common fun practice for people to write their own little C compilers in assembly.

Nobody writes compilers for Oberon except that one guy who decided BASIC must be taken away from the common man, he's a very stupid haughty faggot.
>>
>>107531115
>And yet, it's a common fun practice for people to write their own little C compilers in assembly.
Not at all. TCC is written in C and it's only a toy compiler. The rest are written in C++.

>Nobody writes compilers for Oberon except that one guy who decided BASIC must be taken away from the common man, he's a very stupid haughty faggot.
I didn't realize Linus Torvalds wrote an Oberon compiler. Maybe I should have more respect for him.
>>
>>107527035
How does this solve the problem of 99.6% of all Tor nodes being run by the US Government?
>>
>>107527153
none of those are memory safe, they don't protect against race conditions
>>
>>107531482
Niggga, learn to write multithreaded code.
>>
>>107531482
>they don't protect against race conditions
Are you accessing shared memory without fences and mutexes?

I'm sorry to tell you this but no matter how good your compiler is it can't make you not a retard.
>>
>>107531060
>Memory unsafety wasn't as bad in the past but as C and C++ became more popular and replaced safer languages like BASIC, FORTRAN,
Most apps were written in Assembly, *maybe* pascal later on (when basic was popular many pascal runtimes were so fucking huge they need dual floppy drives and would swap while the app was running.)
>>
So this means only child rapists can audit the code.

It's over.
>>
>>107532033
Yes, but those other languages were still more popular than C and C++. Assembly isn't a high-level language.
>>
>>107527035
>hey i heard you like pozz
>so i put pozz in your pozz
cargo is like sharing needles you found under a park bench
but yeah, sure, go ahead
better fail often
bc better fail early
>>
>>107532333 cont
btw
i said this was gonna happen
it then hapenned
and then it hapenned again
how many more fucking times is it supposed to happen before crabNIGGERS fucking listen?

or is it like the schizos say
the goal of rust is to make everything vulnerable, not to make it secure
>>
>>107532022
>>107531994
not the point, rust won't let you do it in the first place, if you want mutability it'll force you to either use a synchronisation mechanism (ie Mutex, RwLock, or channels for message passing), or to use unsafe.
>>
>>107532411
shut the fuck up crab
unironically.
fucking kill yourself
>>
>>107532416
cnile seething lol
>>
>>107532416
Nice lack of argument. Concession accepted.
>>
>>107532421
no
its normal behaviour
youre peddling poison
doeznt matter what i program in

your thing is to promote trash
like leaking fucking garbage. with flies and shit
>>
>>107527336
for almost 20 years, that really doesn't look like a lot. I guarantee you, once they rewrite it in rust, they will fuck up the program logic and include some stupid deanon vulns. Like they did in the rust sudo, it was memory safe, but followed psychotic troon logic and allowed everyone to become root.
>>
>>107527253
I was at work today looking at logs and you'd see shit like no such function: function_#(*#_@_'!4+binarygibberish because it's some shitty pascal garbage that probably used a freed string.

This shit happens every day. Shit basically works purely on happy paths and strong copium. Rust basically doesn't have this issue according to Google and I believe them.
>>
>>107532432
the goal of rust is not safety
its vulnerability
>>
If I can’t trust the Tor project to write safe C, why should I trust them to roll their own crypto in any language?
>>
>>107532427
>doeznt matter what i program in
i never told you to use rust, i was just making a point that the compiler will make it hard for you to make a race conditions, you pm have to want one to exist.

> like leaking fucking garbage. with flies and shit
nice argument dude
>>
>>107527035
just stop update NOW
>>
>>107532446
i didnt give an argument doe
i just told you to shut the fuck up
>>
>>107532444
> If I can’t trust the Tor project to write safe C
99%+ of C project that have any userbase have had memory vulnerabilities in the past.

seems to me like no one can.
even the best engineers, someday they'll get distracted, or be tired, or on a short schedule or have some life bs distracting them, mistakes WILL happen eventualy, it's not a question of if but when.
>>
>>107532456
> not a memory bug
> some library
why should i care exactly ?
not only is openssl not free of memory vuln, it had logic vulns in the past as well, no one claims rust to be perfect.

imo it's just much better and you are basicaly arguing. "muh it's not perfect therefore bad" when it's better than anything else we have
>>
>>107532460
>>107532472
bc cargo pulls in this fucktardation and yuou wont even know it

i mean
you know it perfectly well
rust is SUPPOSED to be vulnerable
>>
>>107532481
> bc cargo pulls in this fucktardation
not only is this a fixed issue, you won't be exposed by an aes issue if you don't use aes.
also it was a logic issue not a memory bug.
your point is retarded.

and you are not forced to use any dependencies, you can use just what you want, or nothing at all for that matter.
> rust is SUPPOSED to be vulnerable
what are you even talking about.
>>
>>107532481
> muh using libraries bad
t. cnile nocoder.
>>
>>107532499
>>107532506
>unwraps all over your face
>>
>>107528857
>>107529482
You are confusing Rust with C++.
Rust async does not require threads and does not involve context switching.
>>
>>107532512
> /g/eet nocoder filtered by EXIT_FAILURE
lmao we can stop talking here
>>
>>107532460
>99%+ of C project that have any userbase have had memory vulnerabilities in the past
If the Tor C client has had critical security bugs for the last 20 years and the Tor project either didn’t know or didn’t care, why should I trust the new client?

The preventive memory safety reasoning doesn’t make any sense, rust is not the first or only memory safe language. i2p has/had a Java client and dnscrypt-proxy is written in Go. They could have written a non-rust based memory safety argument client 10-15 years ago if memory safety was such a critical issue.
>>
File: sepples-ripley.png (2.53 MB, 1092x2695)
2.53 MB
2.53 MB PNG
>>107532513
both are flaming garbage
kill yourself and take your handicapped langs with you
its gonna be an act of mercy
>>
>>107532481
This newfag must be too young for heartbleed or why there are like 10 active forks of openssl.
This is babby tier CVEs and you're literally retarded and should be roped.
>>
>>107532519
yeah so why dont you shut the fuck up then
>b-but use rust anyways !!!!
lmao
>>
Why does every rust thread have the same mentally ill ESL who spams cnile and /g/eet at everyone?
>>
>>107532522
You should write your own then.
>you won't because you're a hysterical nocoder who doesn't know anything.
>>
>>107532530
>hey, look at this completely unrelated shit
no, im looking at rust now
squirm harder you whore
squirm for my entertainment
>>
>>107532522
> why should I trust the new client?
i agree with you but that's beside the point, the same can be sade about pm all C software.

> rust is not the first or only memory safe language
true, however it is the only one that's 1. compiled to machine code and can be as performant as C and target nearly as many platforms.
2. is not a pain in the ass to work with.

also go is not memory safe.
> 10-15 years ago
i don't think you realise how little choice we had 15 years ago, especialy if you wanted something with performance that's not dogshit
>>
>>107532526
why so mad bro
>>
>>107532542
>im the victim actually
youre right
youre thie victim.
now stop defending yourself while i beat you
its pointless, and its gonna make me beat you even harder
>>
>>107532532
you think you have some kind of point but you are too retarded to realise that it's not one.
see: >>107532530
>>
>>107532499
>fixed issue
yes the point of the image is that the same backdoor showed up despite being “fixed” before.
>logic issue
You mean like the vast majority of real bugs, including malicious ones?
>>
>>107532556
>flagrant backdoor
>damage control from crabs
as expected. youre a bunch of corporate whores.
>>
>>107532542
because cniles will shit threads like this guy : >>107532555
thinking it's some kind of own or something not even realising their own incompetance.
same thing with people posting "muh unwrap" anyone that thinks exiting a program with an error should never happen is a nocoder.
>>
>>107532567
ywbaw though
doesnt matter how many ack-doors you put in your code
>>
>>107532560
Why aren't schizos banned on site for low quality posting? This is the same retardation of people claiming systemd is compromised by the CIA niggers and full of backdoors and yet none of those retards can present any actual evidence. It's obvious you're mentally ill and should be immediately executed for the betterment of society, but alas...
>>
>>107532542
Sorry left out nocoder here
>>107532543
>nocoder
That’s you, Patel. You spam that in every rust thread.
>>
>>107532575
*ywnbaw
worst typo in history of typos
>>
>>107532560
just because it involves the same pieces of code does not mean the issue was the same, and even if it was, regression are not unprecedented in sw dev, but only a nocoder would not understand that.
>>
>>107532578
rust is a jeetlang though
go write some fukken react app faggot
>>
>>107532575
>>107532590
> ywnbaw
thank god, i'd not want to shlomo.
>>
File: obsessed-rust.png (49 KB, 173x330)
49 KB
49 KB PNG
>>107532596
if only that was the case...
>>
File: file.png (161 KB, 991x753)
161 KB
161 KB PNG
>>107532593
>>
>>107532603
> is it possible to use rust with just the std
yes, what's the issue exactly?
heck i've written no_std code in the past, not even the std is necessary.
>>
>>107532604
tldr
ur a faggot
>>
>>107532606
the issue is that youre a retard
just a retarded fucking nigger, nothing more, nothing less
>hurr durr whats wrong with manifest ack-doors
that they get declared only every two years?

whats your bets in 2027 were gonna see another cve in the same package?
>>
>>107532545
>agree with you but that's beside the point, the same can be sade about pm all C software.
This is literally cult mentality. Everything that isn’t X is inherently evil. If all C code is insecure by default, then (1) all cryptography is broken and almost all will continue to be broken for decades including Tor because of legacy C and (2) the people selling solutions can’t be trusted because they’re promoting stuff like rust without identifying and fixing the issues. Why didn’t the Tor project put the Arti money and manpower into auditing their C client?
>performant
That’s irrelevant, tor isn’t OpenSSL where speed matters. Tor’s greatest risk other than traffic analysis is tor itself getting broken so traffic analysis isn’t necessary.
>choice
There are literally hundreds of languages that don’t let you sleepwalk into CVEs like C does. They might not be fully memory safe, but they’re all better than C. If memory safety was a huge security issue for decades, why is it only being addressed now?
>>
>>107532611
> feb 22, 2010
lmao, what even is that sad reply.
rust can be as (and sometime more) performant than c / c++.
>>
>>107532567
>one guy is spamming, therefore me doing it is okay too
>>
>>107532624
> obsessing over some aes-gcm implementation you don't even have to use.
what's your point exactly, nothing is preventing any dev from writting bad logic / code.
and nothing is preventing you to use openssl or libsodium.

also this is literaly a nothingburger compared to many issues openssl has had in the past.
>>
>>107532576
>systemd
yeah just ignore that the xz backdoor only worked on systemd distros
>>
>>107532629
never more
barely competitive with the utmost effort
and completely usuitable for RTOS'es unless you write c in rust

and even then theres things you cant do in rust you can in c
your language sucks cocks through the ass
everything for fuck all in the end
no, back doors at the os level
thank you, rust
how fucking thoughtful
>>
>>107532593
Stop thinking I’m a rusttard you fucking retard.
>>
>>107532625
>If all C code is insecure by default
that's not what i'm saying, C can be written with no bugs and error, however, practicaly speaking, most codebases have memory errors in them, some have been found, some are still to be.

even if code has a memory vulnerability, 1. it takes someone to find it, 2. it needs to be an exploitable one.
for it to be an issue.
but yes, there have been tons of RCE on C software in the past and that's kind of what rust is trying to solve without the performance cost of a garbage collector.
>>
>>107532636
>go back to sleep, goy
no. fuck you
give me a big tiddy jewess
and ill fuck her too
>>
File: crab-roast.jpg (86 KB, 683x459)
86 KB
86 KB JPG
>>107532647
>t. damage control
>now deny youre even a rustnigger so you still have shreds of authority
no
i eat crab soup. with curry. and chili
extra noodles
fucking coco based broth
>>
>>107532625
>performant
hey dude, i agree with you, but i didn't make the decision, i wasn't there when they chose to use C, they must have had their own reasons, probably familiarity.
> but they’re all better than C
we can agree on that.

> why is it only being addressed now
it isn't, we had other attempts in the past, ie ada SPARK, Coq, and other formal verification, the thing is, it has always been a PITA to work with.

rust tries to be a high performance language that fix those issues without being a PITA to use.
and honestly i think they are succeeding at it, i honestly don't care that much about those issues, i use rust just because it's imo a LOT nicer than c++ to write stuff with and i don't have to pay a performance penalty for that.
>>
>>107532661
You have got to be false flagging. /v/ isn’t the only board that needs IDs.
>>
File: 1754037542940762.png (638 KB, 1240x783)
638 KB
638 KB PNG
>>107528673
A real mystery
>>
>>107532669
man i'd love threadid on /g/
>>
File: c++-vs-rust.png (437 KB, 726x1207)
437 KB
437 KB PNG
>>107532665
>> but they’re all better than C
>we can agree on that.
thats because youre a bitch
im sorry but theres no two ways around that

also thread theme:
https://www.youtube.com/watch?v=rDBbaGCCIhk
>>
>>107532676
i2p is better anyway
>>
>>107532669
I still think /v/ is less bad as /g/, but ya, thread ids would be a boon.
>>
>>107532681
GC languages are not fully memory safe but moreso than c, this isn't an opinion just a fact.
>>
>>107532686
>>107532681
good music btw
>>
File: rust-jobs.png (73 KB, 442x822)
73 KB
73 KB PNG
>>107532669
>>107532683
group cope therapy
go dilate youre sore ass where you came from
>>
>>107532693
> muh ad populum
how is that relevant exactly?
also rust jobs tend to have much higher turnover, they don't stay open long because people actualy want them, C++ jobs can stay open for over a year and a lot of them are fake anyway.
>>
>>107532686
>>107532692
you can graft an intermediary representation onto c and achieve the same rust set out to do
but better in every way
less work
more efficient
easier integration
full spectrum domination

happy you enjoy the muzzak
since you like this im sure we can see eye to eye
>>
>>107532693
let's see in 10 years how that evolved, rust is growing whether you like it or not.
>>
>>107532706
dont try to maintain semblants of adult discussion, whore
you lost that privilege the second you ignored the flagrant backdoors facilitated by cargo
>>
>>107532706
I wouldn't be surprised if most of those C++ jobs are:
>Help! Our boomers have retired and we need to replace them so we can shove all these features customers want in our shitty mloc app that barely works!
I've worked a few and they're always the worst jobs I've ever had. I don't even think the most jeeted Java shit I've done was as bad.
>>
>>107532708
sounds like cope
c-ultists dont need to do anything, only watch you stumble into self-destruction
just like on g
janny's on your side?
ok, we'll ban jany then
>>
>>107532707
Ok, so build it then. You won't, because you can't. You're a delusional retard.
>>
File: 1765337151573156.png (681 KB, 1334x750)
681 KB
681 KB PNG
>>107527153
C
>>
>>107532723
Which c utils work on Windows without mingw garbage?
Fun fact: uutils works on Windows.
>>
>>107532728
other people are already on it
and its actually not a big problem if you reformulate your code to be data centric, just had a discussion about that in another thread
just fucking watch lamao
fellow lisptranies are gonna cut your throat and dump you in a ditch
>>
>>107532741
>Meme language no one cares about with no useful software (killer app)
You're fucking retarded.
>>
>>107532745
>forefront of language development
>point where both mathbois and trench-diggers meet
seethe harder
>>
File: 1765231783140497.png (1.56 MB, 1280x1707)
1.56 MB
1.56 MB PNG
Sorry rustrannies but femboys don't want to learn rust and cute boys hate rust also
>>
>>107532711
>flagrant
literaly a nothingburger, openssl had stuff 100x worse than that.
>>
>>107532733
> what is musl
>>
>>107532758
>but what about
wree not talking about the whatabouts doe
i thought you wanted a fresh start?
you got a freshly baked critical cve
hows that for a fresh start
>>
>>107532650
>that's kind of what rust is trying to solve without the performance cost of a garbage collector.
I understand that argument for something like OpenSSL, a kernel, maybe a reverse proxy but not an anonymity protocol implementation. This screams trend following to “attract new devs” because rust is in and malicious politics to starve the C client of resources to cause issues and say I told you so, never mind the human damage. There are more realistic threats Tor could fight, like making traffic analysis more difficult by inserting random packet delays or expanding their snowflake network. Speaking of which, apparently their Go backend is being replaced by rust too.
>nothing to see here.jpg
>>
>>107532765
> but what about
my point is you are obssessed with the most trivial thing made with rust when there is stuff 100x worse literaly everywhere else you look.
and again, what one random shit lib is doing is irrelevant to the discussion.

nothing is gonna prevent you from making a bad crate.
> critical cve
1. it's not realy critical
2. no programming language is gonna prevent you from writting broken logic.
the point of rust is memory safety, it can't prevent you from being a retard altough it can prevent retards from making anything with it, which is a win in my book.
>>
>>107532777
>apparently their Go backend is being replaced by rust too.
>unironical cancer spreading
lamao
>>
File: jeff-bezos-vs-evil.jpg (93 KB, 1280x800)
93 KB
93 KB JPG
>>107532784
>plaintext exposure in the context of aes encryption is the most trivial thing
do you proof read what you post?
because you should
>>
>>107532777
> but not an anonymity protocol implementation
well today it makes sense imo, rust is memory safe and comfy to work with, i like to pick it for most of my project just because i'm confortable with it and enjoy using it.
> There are more realistic threats
true, but i'd abandon tor altogether, i2p is better imo.
> their Go backend is being replaced by rust too.
maybe they wanted to unify their codebase who knows.

anyway, as much as i like using rust, and i think it's can be a realy good pick for new project, trying to rewrite everything is mental illness.
>>
>>107532799
>i like using rust
its insecure garbage though
the interfacing alone is reason enough to pass on the lang
its multi-dogshit
its dogshit on all the levels
>>
>>107532794
> plaintext exposure
did you even read what it was exposed to?
to exploit it you'd need to have access to the machine anyway, there's a reason it's rated 5.6 on cvss.
also the plaintext content is returned in the buffer, which you'd more than likely discard anyway, i'd be very doubtful it has any real world effect.
>>
>>107532752
So basically nocoders? Seems like yet another masterbatory language with a lot of nocoders and no actual projects. If you're going to use a meme language, at least use something with proper dev tools and an actual userbase, like lean4.
>>
>>107527035
Sepples /b/ros the trannies are destroying our bussy
>>
>>107532810
> no real criticism
alright, if you say so.
i don't agree.
>>
>>107527153
it's all LLVM even C.
>>
File: anakin-econmics.png (678 KB, 800x804)
678 KB
678 KB PNG
>>107532819
yeah but that is not supposed to happen
its something that should get solved with lifetime control, right? ROIGHT?
>>
>>107532832
none needed at this point
>>107532825
no, just the people who make the languages you end up using
no biggie
>>
>>107532810
>the interfacing alone is reason enough to pass on the lang
traits are a realy elegant pattern, even non rust people tend to agree on that.
implementation being separated from definitions means.
1. you can implement your own trait for external types (ie library types).
2. you can implement library traits for your own types.
3. this means library can just write functions that implements a type implementing the trait.

this means you can make a library like serde.
do derive(Serialize, Deserialize) over your struct definition.
and then another library like serde-json can convert any type that implements the trait to and from json.

because of that with just that single line of code, you can convert your struct to and from pm any type known to man, or implements your own serializer / deserializer for custom type, make it avalaible as a library, and now everyone can just use it without rewriting anything.
>>
>>107527035
reminder that software fundamentally cannot be secure as long as you do not have control over the hardware
it will never ever ever matter what you do with your software as long as the hardware is backdoored
software based security will always be a meme
>>
>>107532854
yeah i agree with that
in fact i will be implementing something of the sort for c
i will opensores the shit bc i think it can become something generally useful

but that doesnt mean i need ot have dogshit ergonomics
or even have ergonomics at all for 99% of the info because it can be derived from the code itself
>>
>>107532854
>>107532864
you know what?
here, i say it
me, anon from the ass end of the internet i think i can do better than rust
here
fuck you
>>
>>107532864
You can't build anything like type classes in C. Even in C++ you still have to wait for something like compile time reflections in C++26.
>>
>>107532899
o yeah?
i dont need to build them in c
i only need to ***express them in c
fkn nuances you grand autismo you
>>
>>107527035
People are switching to rust because it's a retard proof AI language and they don't want to do any real work.
>>
>>107532920
its more token intensive than c
another metric of how this language fails-
its higher level
but still more complex
what the fuck are we even doing here?
>>
>>107532887
> i think i can do better than rust
then stop talking and please do
talk is cheap.
>>
>>107532915
Uh ok retard. So how do you generalize it enough you could do something like println!("{xyz}") or std::println("{}", xyz) ? You can't and this is the most basic "sell" of these features which you literally cannot create in C.
>>
File: recycle-rust.png (351 KB, 600x600)
351 KB
351 KB PNG
>>107532942
in absolute you create a runtime in c
the question youre asking is:
is c turing complete. i think.
bc c is more than tooring who is a mere faggot btw

>>107532941
youre all goading me into doing that
i fucking will
i will post my progress
everything takes a fuckign year with me, but i decided
i will fucking spank you
i will make crab stew
and we will all eat it and say how delicious it is
>>
>>107532929
you are a retard if you think that.

in rust you can do stuff like stream.map().filter().whatever().for_each_concurent(concurency_limit, <your function).

streams and iter are also lazy, meaning they only only get executed when consumed, which mean you can have a function taking a stream / iter in, apply things to it, and return it, and still no computation will have been done or data will have been copied around until it's consumed, this is useful for making libraries.
there are cases where you have 100 lines of boilerplate in c that's just a oneliner in rust.

also just being able to use derive() over struct definition is great, you can have cli parsing, or any serializable / deserializable format just by defining a normal struct with derive() over it.
>>
>>107532974
>youre all goading me into doing that
>i fucking will
please do, i honestly would not mind, more competition is good.
i doubt you will succeed though, but good luck !
if you think this would annoy me you are wrong, as much as i like rust i do think it has issues, i'd not mind something better than it, there just isn't currently for what i'm doing.
>>
>>107532976
>t. angry retard noises
youve just given me the perfect example as to why is it that rust is more token intensive than c

youre such a complete nigger
it would have been sad if i were your mother, but its funny since im not
>>
>>107532994
> more token intensive
you can literaly do in 3 lines what would take C a whole fucking file, and it'd not even do it as well.
heck, even for parallel computation using multithreading we can do it as oneliners.
and c doesn't even have proper metaprogramming, they have some shitty preprocessor that ends up being cursed garbage.
>>
>>107532920
Since when was being retard proof a bad thing? Even if Rust didn't end up being as fast or faster than C++, wouldn't you want having faster compilations, faster and nicer standard library, no segfaults, better build system, real generics, better tooling and language servers? Or would you trade that all to feel smart for making a linked list with self-referential pointers without having to add unsafe{}?
Do you not feel like a retard chud every time you wait 30 seconds to compile your code with a collapsing standard library that has 10%-20%+ performance tax due to "ABI compatibility"?
>>
>>107532990
like i said
i wont be dropping everything just to prove g wrong
but proving g wrong is on my bucket list
idk, eta 1 month or something like that until first prototype
i will post it
i will call it cide
bc its a c ide
just search desu for cide and you will find me
>>
>>107533009
also i'd argue it's not retard proof, the retards are filtered by the compiler, get frustrated and don't go any futher with the language.
once you master it however yes it avoids a lot of pitfalls but retards never get to that point.
>>
>>107532994
Tokens mean what in this context? You'll use less "AI tokens" for rust because it probably won't generate complete dog shit that you'll need to follow-up with your shitty AI to "fix" your segfaults.
>>
>>107533026
>bc its a c ide
lmao, so you will be locked in to a shitty ide to have a fraction of the features a language that's ide agnostic have ?
doesn't sounds nice just from hearing about it.
>>
>>107533027
that sounds like just doing c
>>
>>107533039
NTA but i understood tokens as language tokens, ie the AST.
but yea with hindsight he was probably thinking as a nocoder llm token lmao.
>>
>>107533026
printf and qsort are both shit, so if you want your meme typeclasses for C library to go anywhere, you might want to actually show alternatives to those.
>>
>>107533045
i use the same infra to make the ide i will be using to do the stat analysis im gonna graft onto it
it all about frames of reference
im honestly surprised maffbois didnt think of that one, but it wont be the first time i think of something original
>>
>>107533049
C is much easier to learn than rust, it's realy simple but also easy to shoot yourself in the foot.
rust is a language with one of the hardest learning curve out there, most people agree on that.
though, if you pass it, it'll be much harder to shoot yourself in the foot.

you can still have logic errors obviously, but to mess up memory safety stuff you pm have to want to be doing it in 99%+ of cases, it doesn't just happen by mistake.
>>
>>107533066
>>107533049
all to say, any jeet will be able to write broken C and compile it and have a segfault.
jeets will generaly not get past the compilation phase for any software that grows too much in complexity.

i've literaly never met a single jeet in my career since i used rust and it has been nice.
>>
>>107533045
>>107533062
its like, dead fucking simple
you rearrnge the code to be data centric
and then its retardedly simple to verify type correctness, boundaries, etc
its halfway to formal verification
CRAFTSMASHIP, BITCH
AEONS OF EXPERIENCE THAT YIELD A LAYOUT THAT JUST WORKS, BECAUSE IT MIMICKS MY OWN WORKFLOW
>>
Troons won
>>
Historically, the NSA has influenced the development of standards to maintain a technological able to exploit them. Wouldn't it make sense to have a focused front on making Rust out to be unnecessary tranny tech since organisations like the NSA benefit from stockpiling mountains of run-of-the-mill, buffer-overflows? Not to mention they would by now have incredibly sophisticated tooling to discover more of these exploits. If software were to migrate entirely to Rust, an entire capability disappears. Modern cryptography is already good enough that the only way to attack it is indirectly, through side-channels. Having an implementation of Tor in Rust would be a nightmare for the NSA.
>>
>>107533066
>>107533073
c is exactly as complex as the machine youre building
its the true measure of a man because you have nothing to hold your hand, to ease the pain of failure
no fucking parachute

but if you have the work ethics
the methods
not even the iq
then c is your best ally because it doesnt get in the way (too much. fuck you comitee btw. fuck em with a stick then set them on fire)
>>
C#, Java, and Ruby are largely written in C++ and C, while Go is written in C and Go itself (self-compiled), allowing for high performance and control, with C# and Java leveraging their respective runtimes
(CLR, JVM) but built on foundational C/C++ components for speed.
>>
>>107533088
> but if you have the work ethics
> the methods

that works well in theory and for small project, but in any real world codebase things end up becoming a mess.
people get on tight schedules, they can get distracted by life, sometime be tired etc.
and honestly, very often they won't even care anymore because it's a waggie job.
99% of C codebases will have memory bugs eventualy, it just always happen, even with the best engineers, because even the best engineers are not always at their best.
with C you have to be careful for memory issues with every single line you write.
with rust you don't have to spread your attention over the whole codebase and can realy focus on the few parts where you did use unsafe (which you don't even need in most projects).
that's less mental load, and that reduced mental load allow you to focus on more things that you'd probably not otherwise.
>>
pozzed beyond belief
>>
>>107533142
>but in any real world codebase things end up becoming a mess.
yeah, except in things that cant fail like your router. or satellites. planes. pacemakers

>99% of C codebases will have memory bugs eventualy, it just always happen,
now what if someone would build a smart ide that could basically give the same guarantees as rust, if not even better ones
and only by virtue of a fucking ide
the most drop in of drop ins that ever dropped in a toolchain

if i pull this off i am gonna smother rust
im gonna fucking strangle it
let it breathe
then strangle it again for hours until it dies of exhaustion

and at the root of it is a refactoring tool that i actually wanted for myself
formal verification is only an emerging property of the frame of reference i wanna be working in (thence the ide-a of making an ide that provides this feature)
>>
and this is the truly genius part of the idea
its not something new, or a modification of something that already exists
its just a different way of arranging the data youre working with
and a different way of dealing with code
its just a different frame of reference
but in a purely logical sense, no maths involved
maffbois couldnt conceive of something like that
only a fukken weirdboy like myself
a straddler of all disciplines, master of lateral thinking
>>
>>107533189
> yeah, except in things that cant fail like your router. or satellites. planes. pacemakers
yea, but those are not the same kind of project.
also many of those use formal verification, ie ada SPARK.
and it's not like you couldn't do them in rust if you wanted to as well, nothing that can be done in C cannot be done in rust (and generaly in less time / effort).

> a smart ide
no one will switch ide for that, if it's an ide i don't want it, i have my own editor.
also beside the guarantees, rust is a MUCH more productive language than cpp, let alone C.
> the most drop in of drop ins that ever dropped in a toolchain
the ide isn't part of the toolchain, no one wants to have to change their editor for a language, it's annoying, some people are happy with vs code, some people are happy with vim, or emacs.
you are not gonna get people to use your new language if they have to stop using the editor they are confortable with.

> if i pull this off i am gonna smother rust
i realy doubt you'll get anywhere if you think the place to start is an IDE.

also i think you should learn rust and build some things with it just to have an idea of what you are trying to compete against is like.
i initialy learnt it to be able to criticize it properly and i unexpectedly ended up enjoying working with it, it has pretty much entirely replaced my use of c++.
>>
>>107533189
>the most drop in of drop ins that ever dropped in a toolchain
> requires you to switch ide / editor
that's the futherest thing from a drop in you can make.
an actual drop in would be using bcc (ie better c compiler) instead of gcc and not have to change anything else in your toolchain.
>>
>>107527153
more over how come how come rust news never was about original project but which well respected legacy projectit managed to infect
>>
File: 1765445133673625.png (194 KB, 1055x607)
194 KB
194 KB PNG
>>
>>107533242
that's completly false, there are tons of amazing project written in rust, thing is, you probably use some of them without even knowing what they were made with, because people generaly don't care abotu that kind of things.

i think rust is a realy good pick for writting new shit, but i do agree that trying to rewrite everything with it is just mental illness, the only thing that should justify a rewrite is if you want to add features that can't be added in the original project in less time that it'd take you to rewrite the whole thing and add the features.
>>
>>107533223
>doesnt see the picture
and thence the ide
im gonna shove it into your face once i have a prototype, then you wont need to imagine anymore
even without any verification function
merely rearranging code to be data centric is gonna eliminate the possibility of a good portion of bugs like double frees or use after free
because you have a fukken timeline of all the code that concerns a certain variable/resource
verifying clean release is just checking the end of the timeline if it says "free()"
and if it doesnt, or if theres other shit that happens after free you have a problem

no automation whatsoever, just a different frame of reference, achieved through an ide
im gonna shove my vision in your face and you will start to write correct c whether you like it or not
unfuckingironically this is how i operate
and i need a tool that facilitates that
the idea isnt even born of formal verificationshit
its just an emerging property of a data-centric frame of reference
thats the magic
its fucking natural
without any further processing code suddenly gains properties you can leverage to verify correctness
>>
>>107533247
> muh tranny allegations
nice try shlomo.
nearly all comunities and organizations are fucking gay (see linux fundation), says nothing about their userbase which are almost always 99%+ straight white men.
that is the case for rust as well.

troons infecting organizations and comunities because they want muh attention and validation says nothing about the actual userbase of the tech.
>>
>>107528228
As of today, Guix builds 12 versions of rust, which is a lot less than I understood to be. Rust 1.74 lists llvm 17.0.6 as its' compiler dependency, for which the later versions of rust are self-built.
>>
>>107533261
> and thence the ide
no one is gonna change ide for a shitty lang.
if your lang isn't ide agnostic, no one will want to use it.

the fact that you can't imagine a solution that doesn't involve writting a whole fucking ide makes me realy doubtful that you'll able to achieve anything regarding that specific problem.

> you will start to write correct c
i've used C much longer than i've used rust, almost a decade.
any c++ dev that got to master rust ends up prefering it over c++
C still has some uses though, but even then rust ate a lot of them for me.
>>
File: 1765472111343710.jpg (94 KB, 800x800)
94 KB
94 KB JPG
>>107533263
>straight white men.
Usecase for straight white men?
>>
>>107533261
> IDEjeet
lol
are you using codeblocks too ?
only jeets thinks the IDE is a relevant part of any toolchain.
have you ever built a C program from the command line ?

no programming languages should be ide dependent, no one will use it if it is, especialy if you want to be able to build without a gui lmao.
>>
>>107533287
usecase for kikes ?
>>
File: togif-7.gif (336 KB, 738x1320)
336 KB
336 KB GIF
>>107533294
>He called me a kike
>>
>>107533278
its not a lang
its an ide, at its root
im not inventing anything new, im just presenting code in a different way, originally

originally, its not even an actual ide bc the tool actually works once you refactor
it always pisses me off to need to re-learn my code when i come back to an old project
so i want to rearrange my code to be data centric

but again, once you do that the code gains certain properties you can leverage to check for correctness. like with resource release
but also constraints like in refinement types you check against the loops youre running
which, rolling drums, are two variables you check against eachother
so that was an easy solve
>>
File: chud-shrug.png (23 KB, 964x851)
23 KB
23 KB PNG
>>107533289
>t.fucking retard
ive explained it three times
if you dont get it until now
youre thoroughly fucking retarded
>>
>>107533325
>its an ide, at its root
yea, which is why no one will want to use your shitty tool, if you can make it an ide, you could make it an lsp, in that case maybe people would use it, but an ide, yea it's DOA.

also, i'm sorry but if you think you can fix C and not make new lang you are realy delusional.
the only way to "fix" it is to make it a subset, which will make it near unusable.
also nothing that hasn't been done already, we got formal verification, but guess what, no one uses it because it's a pain in the ass to do anything complex with.
>>
>>107533331
no one wants to use a jeeted ide.
the closest thing you can make to that that people would agree to use is an LSP.
your project is dead on arrival if you think people are gonna change setup to use your ide.

also, IDE themselves as a concept are just about the most jeeted thing there is (and can come up with).
almost no real software engineer use an IDE, and none of the good ones do.
>>
>>107533384
>no one
i will
and i cant be unique in my weirdness, am i?
bc that would make you say im special <3
>>
>>107533392
i said none of the good ones.
only jeets are aesthetically attracted to IDE.
have you ever built something from command line ?
if you did you'd realise it's a much better way to build anything than having to try and load it into some shitty ide.
>>
>>107533399
>thinks its about aesthetics
catastrophic reading comprehension failure
are you a trany? bc tranies tend to be angry at jeets for taking their it janitor jobs
>>
>>107533392
anyway, an IDE is just about the worse way you can solve the problem.

make it a LSP or a compiler, otherwise it's pm a dead project that no real engineer is gonna want to use.
>>
>>107533404
> doesn't understand the meaning of aesthetics.
it's not about looks, it was a spiritual statement.
>>
File: kernighan-on-dijkstra.jpg (86 KB, 850x400)
86 KB
86 KB JPG
>>107533407
>make it a compiler
sure
>just duplicate work thats already been made for fuck all reason bc were doing something entirely else
>t. autist
>>
What does rust's memory safety even fix in 2025? Rust users talk as if every overflow is a ACE exploit but that hasn't been true since Intel borrowed NX segment locks from better CPUs. Can't do a stack return exploit into your network buffer when the data segment is flagged NX. So most exploits these days involve finding OS functions that a program can be tricked into feeding bad data. Rust only partially mitigates fuckups like that. Hell, most of the CVEs out there in recent years are logic bugs, not some schoolboy use after free or bounds checking issue. Some internal scripting language incorrectly interprets certain invalid token sequences and lets people do bad things. Or multiple processes get into a race condition and the authentication token persists when it should have been invalidated. Rust only fixes concurrency within its own process, which isn't really how modern programs are written.
Rust just seems like a language stuck in the past. Java already went there and failed, we don't really need to revisit that mistake.
>>
>>107533418
>>just duplicate work thats already been made for fuck all reason bc were doing something entirely else
> meanwhile wants to write yet another shitty ide

dude, if i have to stop using nvim to code i'll just not use your shitty technology.
most devs will say the same about their editor of choice, be it emacs, vscode, helix whatever.
you are not gonne get anyone to change a setup that's confortable for a programming language.
>>
File: antioxidant manifesto.png (140 KB, 825x3574)
140 KB
140 KB PNG
>>
>>107533412
its gonna look nice
at its core each line of code will be its own individual object
ill be using sdl
and i intend to do a good job with interfacing so if anyone wants fancier graphics its a matter of drop-in, recompile
im gonna do things right, the code will be for public consumption
>>
>>107533423
you will write a frontend for nvim
which is turbo trash for autistic retards btw
what the fuck are you even thinking lamao
just use a modern ide like a fucking human being
>but muh servers
what servers? youre unemployed
>>
>>107533420
> NX segment locks
have you ever heard of ROP (returned oriented programming), even with NX you can write and execute your own logic, many modern exploit works this way.
and even not so modern ones, heck even the 3DS which is fucking old at that point was hackable thanks to a rop chain exploit.

> finding OS functions
GOT override is also trivial, but only one of many techniques.
> Rust only partially mitigates fuckups like that
it mitigates memory errors, that includes those.
> most of the CVEs out there in recent years are logic bugs
that's false.
> race condition
rust also mitigate against those, you can't have shared mutability without either using a synchronisation mechanism or unsafe.

> Rust only fixes concurrency within its own process
fair enough, but you couldn't expect anything more from a programming language.

> language stuck in the past
it has tons of nice features you don't find anywhere else, again, even if you removed all security features i'd still use it because it's that comfy and productive to work with.
>>
>>107527253
>i don't remember ever hearing about memory issues all that much
Are you serious? Like half of CVEs are buffer overflows
>>
File: 1757626058833002.jpg (126 KB, 768x1024)
126 KB
126 KB JPG
>>107531482
lol don't protect against race conditions, retard alert
>>
>>107533440
>you will write a frontend for nvim
then that's not an ide.
> like a fucking human being
did you mean jeets?

i don't need a bloated mess when all i need is a text editor and cli.
why would i want to click todler icons to do my work when i can work a whole fucking day without needing the mouse, which not only makes it funnier to work, but also avoid wrist pains due to having to do back and forths from keyboard and mouse.
also i can have a powerful editor i'm confortable with over ssh, which is sometime realy fucking handy.

> what servers
> doesn't even know what a LSP is and pretends he's not a jeet.
i didn't even mentioned servers.
lol you are never gonna make something good out of C with your skill level, you are either a jeet or 12yo, you don't even know what a LSP fucking is, go look it up retard.
>>
>>107533479
was that an english sentence.
also it shows you don't know shit about rust, because yes, it mitigates race conditions as well, it'll not let you do shared mutability without either using synchronisation primitives (mutex, rwlock etc) or unsafe.
>>
>>107533491
if you dont do servershit you have zero reason to use vim + associates
>b-but soul
yeah yeah, enjoy ricing your de for the 100.000th consecutive hour i guess
>>
>>107533479
>>107533499
also, race conditions can lead to corrupted memory UB, and even arbitrary code execution in some cases.
there has been Linux kernel exploits relying on race conditions to get priviledge escalation.
>>
>>107533501
>if you dont do servershit you have zero reason to use vim + associates
false again, if you click icons like a toddler you are gonna be less productive than someone that can just use keybinds, that's just a fact.
anyway, i also enjoy macros and many other things that you can only do in vim / emacs.
and i don't want the bloat that emacs is.

enjoy writting an IDE that no one will uses because IDE's are a retarded concept in the first place and no real engineer in the real world uses them anymore.
they only are a thing given to people that still learn programming and that jeets didn't grow out of.

if you tell me you use an ide for programming there's a 99% chance you either are still learning or are a jeet.
>>
>>107533531
>hurr durr i feel better because i do things the retarded way
window layering means more code accesible at the reach of a click
modern ides have powerful refactoring features
you have zero reason to use vim&co
except for being retarded and worthless in general
>>
>>107533531
>>107533547
or working remote but then you use vim nano and such only as last resort
because who in their right mind would use fukken vim to do anything serious?
>>
>>107533547
> window layering
can do in nvim retard, also you can access infinitely more things with commands and shortcuts than any GUI will ever allow.
there wouldn't be enough space on the display to give an icon to every command there is even if they were 1 by 1px.
> modern ides have powerful refactoring features
vim / neovim too you know, just attach a lsp and there you go.
> you have zero reason to use vim&co
you just don't know a single shit about what you are talking about point being, you didn't know we can have refactoring tools within nvim.

heck, if i want a feature that's not in my editor i can write a pluggin for it in less than 20m if one doesn't already exists at that point, you should realy look it up because you seem to have no idea what you are talking about.
>>
>>107533559
>because who in their right mind would use fukken vim to do anything serious?
half of the best engineers i've ever known use vim as their editor.
you've never been in the real world.

imagine needing your IDE to support something whenever you want to do anything kek.
>>
>>107533574
>but i can actually do window layering
you can do that in all of them but then youre using the mouse with an interface that isnt designed for that
>but i can code it
to bring it to the level of a modern ide?
how is that fucking smart?
>>
god people honestly think that the solution to memory safety is a fucking IDE.

i'm out of /g/ this board has been infested by jeets and other brown streetshitters.
>>
>>107533582
>but big corpo
yeah? they use agile/scrum/etc which is the equivalent of living in an iron lung
what makes you think its any example to follow?
>they have money
pokimane has money too lamao
>>
>>107533589
>ad populum
yeah get the fuck back to r*ddit, faggot
youre retarded, nothing of value will be lost
>>
>>107533585
>to bring it to the level of a modern ide?
there are a gazilion things i can do in vim you just can't do in any of your shitty ide lmao.
also, you have no idea the amount of plugins there is.

anyway, i'll tell you why.

1. my nvim configuration is only textfiles in ~/.config/neovim
that means i can track and sync them with git, that means i have them accross all my computers and containers.
if i open vim with those config it'll just automatically pull all my pluggins and config, i don't have to set it up again each fucking time.
2. for work, i use a LXD container, it's separate from my host, which mean i don't have to infest my host system with work related bullshit, it stays in its container, using a terminal editor is realy handy for that.
also it pairs well with a window manager.
pm any features i want out of my ide i have in my editor and more, it is extremely customized to my preference and i don't have to set it up everytime, if i change something it's immediately reflected on all my computers because it's tracked with git.

then, there is Mason which is neat, if i want to open a programming language file, i don't have to go and install yet another shitty extension for it, i just open the file, and automaticaly the lsp will be installed and started for it, i now have autocompletion, jump definition, refactor and all the features you'd expect out of an ide.

also it's not fucking bloat, it literaly takes about 100ms to start, it's snappy, it never hangs, freeze, or loads, and i don't have to click toddler icons, it's just a nicer setup and there is a good reason i stopped using IDE's years ago.
>>
>>107533621
>still posting
i thought youre leaving
>>
File: file.png (53 KB, 765x143)
53 KB
53 KB PNG
>>107533621
>>107533585
and then there is all the terminal integration stuff.
i do vf which turns into vim -p $(fzf) and it'll fuzzy search open my file.

i can do vim -p $(file . -name *.whatever)

and also shell history, i can just do ctrl+r for recent command, type vim and it'll open the last files i opened.
i can also do CTRL+T to search with fzf my command, ie picrel.

inside my editor i can also just ctrl+j to show the fzf file and jump to file.
i mean i got many niceties you just don't have in an ide without toddler icons.
>>
>>107533642
>>107533621
>>107533585
also, i don't waste space with toddler icons, my screen is just filled with code, no distractions.

>>107533628
NTA
>>
>>107533642
im not gonna repeat myself (a third time)
im just gonna call you a retarded faggot bc thats what you are
for some fucking reason youre seething at me an my ideas
prolly because youre jealous
which in turn, is prolly because youre a tryhard autist, completely devoid of imagination
so you seethe at me just like you seethe at jeets because you percieve us as threats

but as far as im concerned i dont give a shit about you
dont stand in the way, you wont get walked over
ez, no?
>>
File: file.png (572 KB, 1550x312)
572 KB
572 KB PNG
>>107533628
anyway, you think IDE's are better because you haven't tried anything else.
i've tried both and i think IDE's are dogshit, your opinion is based on ignorance, mine on experience.

i can only recommend you to at least learn about it and make your own opinion, ie i can search files

i could also just do ctrl+n to search grep and have previews, anyway, the point is, any feature i want, i can add myself and bind to whatever keybind i want, and then it'll be available on all my devices without having to set them up again.
>>
>>107533646
read the reply chain
its something entirely different from 'just another jeetbrains'
if thats what it was
id just use that "just another jeetbrains" and call it a day
>>
>>107533659
and yes
you vill just drop my ide into your routine
and you will like it
i will fucking violate your workflow
and youre gonna like it
nay
youre gonna fucking become addicted to it
>>
>>107533656
>seething at me an my ideas
no, i'm just laughing that you think the solution to memory safety is a fucking IDE lmao, no one will ever take you seriously and only a jeet would come up with that.
i doubt you ever built anything of value to begin with.

> you seethe at jeets because you percieve us as threats
jeet confirmed lmao

> threats
lmao, you shit spaghetti code, only thing you can do is webshit and even that you are very poor at.
see how i guessed your ethnicity before you even confirmed it because of how retarded your ideas are, that's why we are tired of you guys on /g/, you have a shit brain that produce shit ideas, shit code and shit software, also your streets are filled with shit and garbage, that's just your way of being and we don't want that in our countries or software.
>>
>>107533667
>you vill just drop my ide into your routine
lmao, ok jeetbrain.
you'll probably not even be able to write something that's not slow as ass and doesn't crash every 5 minutes.
>>
>>107533662
dude he confirmed to be a jeet, see : >>107533668
though that was quite obvious from his replies and idea.

i wish i could either shut down all indians from the internet or at least have a filter that'd block all content made by them, they fucking shit the whole internet wherever you go.
>>
>>107533668
>also your streets are filled with shit and garbage
thats not laughing
thats seething
extrahard

you know full well im not a jeet
>>
File: file.png (169 KB, 800x450)
169 KB
169 KB PNG
>>107533667
>i will fucking violate your workflow
jeet talking about rape
how surprising

> noo saaaar, it's not gonna be just another jeetbrain i swear saaaaaar.
>>
>>107533679
im gonna fucking dilate your throughput
so that you will NEED my software to prevent it from totally collapsing once im done with you
bitch
>>
>>107533681
you literaly confirmed it, see : >>107533668
you said "you percieve us as threats" US, when refering to jeet.
you can't backpedal on your admission, we know.

and we didn't even need confirmation, i said from the start that only a jeet would come up with such shitbrained ideas.
>>
>>107533688
jeet seething
> REEEEEEEEE
lmao, you will ALWAYS be brown.
stay in your country and don't shit in our streets, and if possible never log to the internet again with your jeeted ideas.
>>
>>107533687
im gonna expand your productivity to an extent you never thought was possible
but all it takes is just some patience and adequate coaxing
>>
File: file.png (67 KB, 780x783)
67 KB
67 KB PNG
>>107533699
lmao, sure think ranjeet.
>>
>>107533692
>>107533698
youre gonna be left wanting and craving for more bitch
fear, fear the day ill release the first demo
or maybe you already do
and thats why youre seething with such intensity, such passion
>>
>>107533699
> t. jeet think he can build anything of value
kek
>>
>>107533705
>nn u bwawn
im not.
doesnt bode well for the rest of your predictions...
>>
>>107533711
>fear, fear the day ill release the first demo
>>107533712
i'm laughing at you
you are a megalomaniac brown that think you can solve the issue when you don't even understand type theory.

you are incapable of building the most basic things let alone your grand idea to which you though the best solution was just another jeetbrain IDE.
>>
>>107533719
quote fail?
dont break your keyboard, lamao
>>
>>107527035
is this why tor keeps crashing now?
>>
>>107533715
you literaly admited it, i don't care what you have to say, you can't recover from that one, see : >>107533692

it's fun how jeets are self aware enough that they suck that they don't even try to defend their ethnicity but instead immediatly pretend that they are not brown lol.

you guys will even pretend to be white supremacists when you are brown, what level of cuckery is that.
>>
>>107533728
your redditspacing hurts my eyes
lurk more, impolite nigger
>>
>>107533723
no, i replied to your comment with this >>107533712
ie
> t. jeet think he can build anything of value
are you filtered by links now?
>>
>>107527035
Trannyware
>>
>>107533734
oh, youre quoting yourself now?
sorry, i just didnt actually read your post

ure repetitive
>nn u bwawn
yeah
and what if i was
actually, yeah im bwawn
imagine being btfod by a brownie
im a jeet actually
a dalit from the brownest jeets from punjab

still whiter than you
still better than you
>>
>>107533734
>not b&
??
post still deleted
i dont get it

still
seethe harder faggot
i will write myt shit
and im gonna make you watch every single step
>>
>>107533743
> oh, youre quoting yourself now?
on top of being a jeet you are a new fag.
it's common on 4chan to reply to a post with a post you already made because the answer applies again.

> ure repetitive
unwelcome to 4chan ranjeet
> actually, yeah im bwawn
don't need to repeat it, it was quite obvious, only browns have such shitbrain idea and not only that, think they are good.

> imagine being btfod by a brownie
you haven't built anything yet, you live in fantasy, also see how half your replies you tried to denied it, until you slipped and there was no going back.

just the fact that you guys try so much to pretend that you are not brown is enough to know you are ashamed of it (and for good reasons).

> still whiter than you
brown think he's white
lmao post hands, you are not gonna be whiter than a native frenchman.
>>
>>107533766
im a jeet and a newfag and everything that you might think of
but im still better than you in every way
you sure you wanna go down that route?
>>
>>107533762
> i dont get it
typical jeet behavior.

> seethe harder faggot
who's seething now
YOU ARE BROWN

> i will write myt shit
it will indeed be shit, can't wait to have a good laugh lol
> and im gonna make you watch every single step
please do, i'm getting popcorn that's gonna be fun.
>>
File: 1664505366075.gif (44 KB, 226x176)
44 KB
44 KB GIF
>>107533772
>but im still better than you in every way saaar
lmao nice cope, post code
what age are you on top of being brown i'm curious.

> you sure you wanna go down that route?
the route that'll maximize laughs ? absolutely !
>>
>>107533774
>who's seething now
(you)
>b-but i said youre brooooown
>>
>>107533780
>t.tranimeposter
how can you hope to be taken seriously
>>
>>107528449
Wish I could just handwave away low level issues by slapping a "language is safe" label on it and call it a day.
Do people actually believe this shit?
>>
File: file.png (199 KB, 474x370)
199 KB
199 KB PNG
>>107533783
> no i'm not seething saaar, y... you.. you are saaar !!! i'm gonna make jeetbrain IDE, very good software saaar, you will seeth saaar you'll seee saar, <insert jeet rape terminology>
>>
>>107533801
no, im saying that youre retarded
you have a very poor reading comprehension anon
maybe /lit/ would be more to your speed
>>
>>107533789
it's about the picture being drawn and the comedic timming.
also why would i need to be taken seriously, it's a fucking jeet, he already lost just by being brown lmao
>>
>>107533794
you don't have to slap anything, for similar rust and cpp function compiler output will be almost identical, you can try it idk do a matrix mul.

heck sometime the rust code ends up being better because 1. it's more expressive and can give more hints to the compiler and 2. we have builtin simd intrisics.
>>
File: pepe-newspepe.jpg (355 KB, 1391x1405)
355 KB
355 KB JPG
>>107533809
>comedic timming.
theres nothing comedic about being a timmy
>>
File: 1687441910451991.png (592 KB, 747x800)
592 KB
592 KB PNG
>>107533806
> i'm not coping saaar i swear waaaaah
>>
File: pepe-shrug.png (8 KB, 700x500)
8 KB
8 KB PNG
>>107533826
y you mad bro?
is it because youre a fuckup?
im a fuckup too, dw, its just that im smarter than you, is all
>>
>>107533806
>>107533826
i'm being honest though please get out of /g/
your kind is shitting the board without even realising, we do laugh at you but you do not realize your own inadequateness and it's kinda frustrating.
can't you guys make your own brown image board, why do you have to go to white men places and shit them like your own ?

you know, your country isn't that shitty due to a bad roll of the dice or malchance, a country is a reflection of its people and that's why games such as picking a random spot in your country and trying not to find trash exist, because it's literaly litter everywhre, it is in your blood, you smear your shit both physical and mental everywhere you set foot and we are kinda tired of the stink.
>>
>>107533841
> smarter than you
post a single notable achievment, a single line of code idc, what have you done in your life.
you don't just sound brown, but also 12.
>>
File: viper.png (447 KB, 750x750)
447 KB
447 KB PNG
>>107533844
>i'm being honest though please get out of /g/
no
why would i even listen to you
>>
>>107533850
i already posted code
lurk more, learn to read reply chains at least
>>
>>107532676
>(((levine)))
>>
>>107533855
> pwease let me stay with you guys, i know i smell of shit but pwease
why do you think the "jeet" derogatory term exist, because you guys shit the board, everywhere you set foot things get worse.
everytime a bad idea is thrown someone will say brown and there is a good reason.
why do you think i guessed that you were brown even before you admited it, because we can smell the shit, and it's tiresome.
you guys would still be in poophut throwing sticks if it wasn't for the brits colonizing your country, none of your achievments are your own.
>>
>>107532366
schizos are always right
you should start listening
>>
>>107533867
i dont care about you
ill post and then you may seethe
but youltimately
ill still post regardless
>>
File: poointheloo.png (290 KB, 635x261)
290 KB
290 KB PNG
>>107533873
>youltimately
kek, you realy stoped trying once we guessed you were brown.
> ill still post regardless
yea, there is no stoping a jeet from shitting the streets, why did i even tried to tell you to do something that's against your nature, culture and genetic lineage.

heck even the west tried to tell you but to no avail, see :
https://www.youtube.com/watch?v=l01AMCBG0Wk

imagine being such genetic freaks that actually devloped countries have to keep telling you the most basic concept of not shitting your rivers.
>>
>>107533873
> pretends to be proud jeet
> post russian plane
what did he mean by that
>>
File: elektrokot-painting.jpg (47 KB, 661x603)
47 KB
47 KB JPG
>>107533893
>nn u bwawn
*checks skin*
nope.
ur still seething doe
>>
>>107533898
its a cool ass plane
>didnt notice how soulful the design is
go back to r*ddit
youre profoundly untechnological
>>
>>107533912
> nope
you literaly admitted it multiple times :
>>107533743
> actually, yeah im bwawn
and : >>107533692

>>107533920
>didnt notice how soulful the design is
i did, but my point is that nothing soulful comes from your kind.

> youre profoundly untechnological
say the guy comming from a nation that doesn't even have the basic technology of toilets to the point that literal NATO felled compeled to make a campaign to potty train your whole fucking country, but to no avail.
see :

https://www.youtube.com/watch?v=uI5WRxknOqA
https://www.youtube.com/watch?v=3JIkgW4pz_g

"about half of india's population shit in the street, the other half are blind to it".
>>
File: file.png (796 KB, 820x1275)
796 KB
796 KB PNG
>>107533920
>>107533938
> https://www.youtube.com/watch?v=3JIkgW4pz_g

they made a fucking POO APP to tell them where they can poop, lmao, now if that's not technology !
>>
>>107533938
>wall of text
>>
File: poointheloo.png (942 KB, 820x1275)
942 KB
942 KB PNG
>>107533961
> coping hard
lmao, keep shitting your streets >>107533959
>>
File: torvalds.jpg (24 KB, 487x362)
24 KB
24 KB JPG
>>107533967
>b-but my reddit
this is a wendy's, SAAAAR
>>
>>107533974
> trying to change the subject to divert from his streetshitting habbit
lmao the whole board laughs at your kind, do us a favor, crash your country's internet and never come back.
>>
File: tier-list.png (28 KB, 183x404)
28 KB
28 KB PNG
>>107533987
>enough about me, lets go back to streetshitting
you seem oddly fascinated with the subject
maybe you should see someone, anon
idk; a veterinarian, a mechanic, a priest
someone...
>>
>>107533995
>>107533987
i just want you, and all of your kind to never go on the site again.
>>
File: pepe-sus.jpg (11 KB, 474x245)
11 KB
11 KB JPG
>>107534002
yeah but youre an insignificant retard seething at others because they have ideas and rock your little curated hugbox of ritualposters and uninteresting tranies like yourself

why would anybody listen to you?
serious question
>>
>>107534012
> ideas
this was a brown idea, and i explained to you why multiple time.
the issue is that you guys come to the board and shit the threads with ideas that you think are good because you are too retarded to see that they aren't.
but to us it's just noise that divert from the actualy good discussions we can have when you are offline.
as soon as a jeet start participating in a thread it is the end of it.
no, an IDE will never be the solution to memory safety, it'd never get adopted and if you don't see why, i'm just sorry for you anon.

why do you keep participating in a forum where noticeably almost everyone hates you and your kind? honest question.
sure, if you want to write an IDE i encourage you, go for it, but if you think that it's gonna solve memory safety, that people are gonna quit their workflow for it, and on top have the hubris to say it with an arogant tone, don't be surprised people laugh at you, your ideas, and genetic lineage that tends to produce such ideas.

if you guys were constantly making good software, producing good idea and your country was an example to follow, we'd praise you, not be laughing at you or have such insults, the reason that we do is because you are not, do you at least recognize that ?
you yourself tend to pretend not to be brown until you are caught, why do you think that's the case ? if that wasn't a bad thing why would you need to lie about it ?

anyway, godspeed anon, i just wish the board stopped being shitted on like you shit in your rivers.
>>
>>107534050
whats the tldr?
>>
>>107534060
jeets have brown ideas, are too retarded to see it, no one likes them and as soon as they join a thread it's the end of it.
yet, they still come to shit threads like they shit their rivers.
>>
File: buffers-g.png (180 KB, 865x740)
180 KB
180 KB PNG
>>107534070
i still fail to see how that concerns me
>>
>>107527035
Stay out of my i2p, rusttrannies.
>>
>>107534084
because you are brown and admitted it many times.
and you have brown ideas that shit the threads, you don't realize it, and kill threads with your retardation.
>>
File: lay-thine-eyes-field.jpg (93 KB, 685x404)
93 KB
93 KB JPG
>>107534128
sure
whatever you want
bc youre insulting yourself when you think about it
>>
>>107534128
>>107534132
for anyone reading the thread,
proof here :
> actually, yeah im bwawn
>>107533743
and here :
>>107533692

godspeed
>>
File: paranoid-delusions-a.jpg (221 KB, 680x1215)
221 KB
221 KB JPG
>>107534185
>seething trany desperately believes he can spin a narrative
>>
File: paranoid-delusions-b.jpg (240 KB, 680x1215)
240 KB
240 KB JPG
>>107534185
>>107534198
oops, outdated picrel
this one has better annotations
>>
>>107533589
that poster is not asian lol, I don't believe it
>>
>>107534256
that trany broke and went on an insulting spree, thats what
you can see the preoccupations of that guy
if youre not a wagie and youre not into webshit you never encounter a jeet in your life

but hes an it janitor
so he hates em bc they take his job
he talks like a liberal
i bet he voted for it bc he thought hes gonna he jose do his lawn on the cheap but ended up being sacked and his job went to abduljeet
which left a bad chicken masala aftertaste in his mouth
and so now hes traumatized.

i guess
if rtraumatized-shaped, how not traumatized?
>>
>>107533427
I am not sure of all of the content in that image, but WTF is that SWAT harassment? Given the public harassment committed by Hector Martin and the many rustlings that helped him harass in public, and the harassment done by major parts of the Rust community at large, it'd fit entirely with the rustling modus operandi.
>>
File: images(13).jpg (84 KB, 451x443)
84 KB
84 KB JPG
>>107534278
I won't get too deep into the intersectionalism of trannies (transgender) and christians (transracial) and chris-chan, but yes if it queefs like a troon, walks like a troon, talks like a troon...
>>
The amount of controlled opposition and samefagging in this thread is off the charts.
>>
>>107534498
At least post the doctored screenshot showing the "samefagging"
>>
>>107534451
has paranoid delusions like a troon... >>107534498

i called it, kek
>>107534225
>>
>>107534544
Hello mentally ill, zero-IQ, wretched and utterly evil piece of shit and filth rustling troll, please commit suicide before you SWAT anyone else. >>107534447
>>
>>107527035
Nice, another botnet avoided. I will never run such a security vulnerability on my system.
>>
>>107534574
Refer to >>107534585

Are you going to call Linus Torvalds delusional next? https://lkml.org/lkml/2025/2/6/1292
>>
>>107534589
>good guys nsa avoided me vulnerabilities!
>>
>>107534591
>look ma
>im just like linus torvalds!
kek, no youre not
>>
i wish we could discuss Rust's good features and shortcomings without the thread turning into a jeet and tranny calling competition that leads nowhere.
why can't we just have a fucking normal conversation for once.
>>
>>107534622
>>107534585
>>
>>107534625
oh, now you want a normal discussion?
not ~70 posts ago when the trany started calling people jeets?
thats fucking curious
>>
File: kernighan-on-c-3.png (259 KB, 850x400)
259 KB
259 KB PNG
>>107534633
>n-no, ur mentally ill
>>
>>107534644
>>107534585
>>
>>107534656
>n-no, ur mentally ill
then how come you check all the boxes for paranoid delusions?
>>
>>107534681

>>107534585
>>107534591
>>
File: amnesty-troons.png (15 KB, 308x243)
15 KB
15 KB PNG
>>107534692
>t.
>>
>>107534697

>>107534585
>>107534591
>>
File: retard-chamber.jpg (33 KB, 638x680)
33 KB
33 KB JPG
>>107534709
>>
>>107534726

>>107534585
>>107534591
>>
File: pain-engine.jpg (142 KB, 1120x824)
142 KB
142 KB JPG
>>107534732
>>
>>107534732

>>107534585
>>107534591
>>
File: i-C-is.jpg (73 KB, 650x375)
73 KB
73 KB JPG
>>107534755
bismillah
>>
>>107534634
> more derailing
sigh
>>
>>107534692
>>107534697
>>107534709
>>107534726
samefag lol
>>
File: kot-pilled.jpg (106 KB, 750x924)
106 KB
106 KB JPG
>>107534768
dont you you me
he started
im trying to make it somewhat entertaining, at least
but ok, ill stop
>>
>>107527035
>tor
>rust
>nsa
Tells me all I need to know.
>>
>>107534778
>im trying to make it somewhat entertaining, at least
fair enough, fun picrel.
>>
File: three-letters.webm (3.47 MB, 960x540)
3.47 MB
3.47 MB WEBM
>>107534788
as for the op, i have basically the same opinion as this anon >>107534779
basically if nsa says we should do something
we actually should do the exact opposite
>>
>>107528681
>Eliminating process context switching costs.
are you fucking insane?
>>
>>107534819
> if x say we should do y we should do the opposite
i agree with the general sentiment but it isn't the case 100% of the case.
if they tell you not to drink arsenic / lead, you can look the data and make a fair assesment that it's probably not a bad recomendation.

anyway, i'm also very warry of the glowies but i've not found anything in rust that seems to glow.
the worse would be a ken thompson hack style of compiler being compromised and compromising other builds of it.

but you can just bootstrap it yourself.
and if you still don't trust it then you may as well not trust c and cpp either.
>>
>>107534823
you can do that if you don't context switch at all.
that's why we can pin threads.
>>
>>107534844
>>107534823
oh and obviously tell the kernel not to use it (which you can also do).

you can make sure that an executable is in full control of a thread, you rarely would want that though.
>>
>>107534851
>>107534844
>>107534823
i meant core* not thread.
>>
>>107534839
i didnt think about the compiler too much but now that you mention it,
some things do seem fishy
like the lack of clear specification
also i heard something along the lines of the borrow checker being impossible to replicate
but thats hearsay

what im concerned about, however is cargo and the recursive dependencies
and im not the only one who thinks along these lines.
amazon originally didnt use cargo at all, and now they have a parallel package distribution method where they post packages they reviewed themselves
youre not gonna say amazon is paranoid, right?
if they do it that may mean theres something to it
they run their aws ship pretty tight
>>
>>107534839
>>107534870 correction:
actually i misread the pr blurb when i looked at it
aws actually offers private repos as a service
i wasnt attentive enough when i read it the first time

im 1000% sure amazon doesnt/didnt use cargo internally
it came up during one of our rust vs c discussions
>>
>>107534870
> like the lack of clear specification
i heard it has one now, but honestly i don't care so much, rust is already pretty well defined in the books and rfc's etc.
> however is cargo and the recursive dependencies
so what made npm so bad is that packages are not immutable and also jeeted.
you'll not find anything that uses "is_odd" unironically in rust and unlike node, you can't just change a deps's past, it's immutable.
though i agree, that applies for everything, if you use dependencies you need to trust them.

though, you don't have to use a ton of dependencies if any, you have to weigh the pros and cons i guess.
> where they post packages they reviewed themselves
makes sense.

in former company we'd write every line of the code, and we'd specify them in cargo.toml with git path, using cargo doesn't mean you have to use deps from crates.io.

> youre not gonna say amazon is paranoid, right?
i don't think they'd do it for all scales, but i could understand them doing so in the most critical stuff, no package manager is entirely protected from supply chain attacks.
even nixos where sources are verified, they often will verify that you are indeed pulling the right code, but they won't check the codebases themselves.
anyway, whenever you are gonna use software that you didn't write, you have to put some amount of trust in it, that's just how it goes.
and you couldn't use a computer if you couldn't trust any software you didn't write.

> they run their aws ship pretty tight
you answered your own question, they want to avoid the possibility of supply chain attacks because they are that big of a target.
though i don't think the solution to the supply chain attack issue is to make your language such a pain in the ass to use any libraries that you almost never do.

and even when i was using c++, i used a json library, i didn't bother to check all of the codebase, i trusted the project due to its popularity.
same when i use ffmpeg or libcurl.
>>
>>107534984
>>107534956
no worries.
there is no risk in using cargo if you only use it to download repositories you control.

but yea, the supply chain risk is universal and not specific to rust, i'd not be surprised if they did it with other lang that were in critical pieces of their infra.

and anyway, nothing stops you to use rust like C, or even use rust with nothing but C libraries you manually downloaded.
>>
>>107534994
>>107534984
>>107534956
also, if you code in C and use libs, you'll typically use dynamic linking against things from your package manager.

idk if i'd trust more a binary package manager than crates where you build from source and the code is audited by the comunity to some extent.
>>
>>107534994
>>107535002
on my end the trust issue boils down to the culture, personal bias, and hostile deign

lets start with the last one:
hostile deign in that importing code 30 libs without realizing it is way easier in rust than it is in c
my personal bias is that i only use the glibc in most of my code, most of the stuff i do is numbercrunching-related so i dont have the need for anything
and i expect most of c code is similar in being economical in imported libs.
i could use sdl 2 if i'd need a gui, i have all the sound related packages installed but this is something like 6 packages total?
and culture, when one programs in c theyre paranoid when rust users culture seems to be "if it compiles it means its correct"
its bias, feelings etc,
but what is trust then if not a bias, a feeling?
>>
>>107535076
yea fair enough.
still, i have a few no deps rust projects.
but most of them i just want to get things done to be fair.
though, the top 100 crates are so unanimous that if you don't trust them, especialy like top 10, you may as well not trust the compiler.

i'm not a fan of crates requiring a dozen crates which themselves uses another bunch though.
nodejs is much worse than rust on that aspect but still.
anyway, you can always be economical with crates, instead of tokio use a tiny runtime or your own etc.
for most popular crate there is often a smaller, simpler, self contained alternative.

and yea, you can write your own stuff or just use a C lib as well.
>>
>>107535116
>>107535076
also i find it odd how i keep seing that criticism with rust, but no one does it for python, go and literaly all other programming languages.
that's a universal issue, and for all langs you can choose not to use deps anyway.
>>
>>107535116
you sound like you have a good grasp on things
but many rust posters of g are downright scary
and an npm like package system possibly used in the linux kernel doesnt sound enticing at all

we even had a clash over that exact problem already with the bcachefs-tools story
rust dev drops a package with something like 50 deps iirc on the lap of debian maintainers, sees nothing wrong with that and even disses them on social media
this doesnt inspire confidence, at all
>>
>>107535122
Go is a really dumb thing to throw in there when go users are notorious for not straying from the standard library except for web server and router performance.
>rust
Because rust is heavily marketed on its safety, higher standards are applied to it, especially when it involves rewrites of working projects. It’s hard to trust an argument about safety when the original project has less than 10 steps and the rust rewrite has hundreds. E.g. librsvg that is installed on nearly every Linux desktop.
>>
>>107535181
>steps
*deps
>>
>>107535122
none of these languages are marketed as safe(tm).
and the unlikeability of rust evangelists only fuels the fires
plus connotations of lgbt etc etc
multiple factors are coming together, the result is a response that might seem unjust
and from a purely technical perspective maybe it is
but you cannot expect fair treatment from people who have been treated unfairly themselves
>>
>>107535181
> go
oh boy i've seen codebases
no, even go project can end up with a gagillion dependencies, it's mostly a dev decision.
> rust rewrite has hundreds
it needs not to, again, using deps is optional.
and again, that's an issue of trust, how much libraries do you have installed on your system right now ?
probably hundreds.

them being pulled from a package manager is no more trustworthy than them being built from source, in fact i much rather have them built from source.
>>
>>107535198
> marketed as safe
rust is marketed as memory safe, that's it.
> unlikeability of rust evangelists only fuels the fires
fair enough, confirmation bias though, you only hear about the most vocal ones, which tend to be mentally ill, sane people don't go on evangelist crusades.
> plus connotations of lgbt etc etc
what i just said, they crave attention, you have the same issue with the linux foundation, most linux users are not troons however.

> but you cannot expect fair treatment from people who have been treated unfairly themselves
fair enough.

though see my point of view, i'm not a troon, i've been using c and c++ professionaly for years, then i learnt rust (originaly to know enough to criticize it) and i've come to enjoy it.
it has over time become my main programming language, now i don't think it's perfect, far from that, but i do find it very comfy and the best thing i've worked with so far and that can also do the things i want.
and then, whenever i go online and want to have a discussion about it, all i see is muh troon allegations, missrepresenation of the language, attacks by omissions etc.
and extreme counter reaction from the retarded troons that i grant you, do ruin it for everyone else.
i think we should be able to call the troons out for what they are without having to insult any engineer that has had a good time with the language.

with all its flaws, i still think rust is a language with a lot going for it, and yes the community and organisation is insuferable, but so can be said about the linux foundation and many others, idk, i just don't interract with them and so do most rust devs.
>>
>>107535262
>even go project can end up with a gagillion dependencies, it's mostly a dev decision.
There are zero dep python projects, it’s a culture thing.
>needs not to
Yes, but they do. librsvg has nearly 300 nested dependencies from what I’ve seen. That’s on the upper end but not unusual for rust projects.
>package manager
I use gentoo and build almost everything on my system. I loathe rust and most other modern dependency management languages because the real number of dependencies is hidden from your package manager. I don’t build rust, know why? Because building rust-1.89.0 requires me to build llvm 20 AND llvm 21.
>>
>>107535322
>>needs not to
>Yes, but they do. librsvg has nearly 300 nested dependencies from what I’ve seen. That’s on the upper end but not unusual for rust projects.
This cope is so tiresome.
>yeah technically we could make software better, but now that we replaced you, we're free to make it worse, so what?
>>
>>107535322
> There are zero dep python projects
there are zero dep rust projects as well, even whole ass async runtimes.

also, there are many non zero deps projects but that are multiple crates of a bigger project, a project being split into multiple doesn't make it a sudden 10 deps, it can be considered into a single one since all parts of the split are in the same repository.

> librsvg
if you say so, idk about it.
> but not unusual for rust projects
you typically see it most for webshit.

> the real number of dependencies is hidden from your package manager
cargo tree
> Because building rust-1.89.0 requires me to build llvm 20 AND llvm 21
nope, you can build and use rust without llvm, ie cranelift.
you can build rustc using the cranelift backend instead of llvm.

but anyway, rustc is selfhosted, so you need rustc to build rustc unless you want to reply the whole boostraping process, which you can also do but i doubt gentoo is doing it.
>>
>>107535335
> 300 nested dependencies
> look inside
> it's actualy just 10 ish repos that define multiple crates within them.
>>
>>107535418
>I only have 10 dependencies for software that should have 0
>>
>>107535413
>>107535418
This is so disingenuous it’s unreal. Here’s the manifest, it’s well over 200 crates even with deduplication
https://github.com/gentoo/gentoo/blob/master/gnome-base/librsvg/Manifest
>>
>>107535413
Also
>cargo tree
is not my package manager fucker. Portage is. It’s not okay when a go project like kitty does it with like 40 deps and it’s not okay when librsvg does it with over 200. It’s appalling that hiding half the packages installed on your system got normalised by calling it modern.
>>
>>107535427
> let's reinvent the wheel everytime because there is no package managment in my language
typical cnile response, it can make sense to reuse code even for that kind of thing.

both for performance reasons, but also to integrate well with other things.
ie they use clap for cli parsing, pretty good choice.
chrono for time parsing, make sense, you don't want to write your own time system.
rayon for oneliner multithreading, makes sense.

i mean you can look at the deps list, nothing shock me too much.

yes they could make a tiny standalone implementation, however relying on dependencies allows a much bigger feature set and higher performance for the same amount of dev time.
so it's a fine tradeoff imo.
>>
File: file.png (22 KB, 526x103)
22 KB
22 KB PNG
>>107535468
my point is that many "separate" crates are defined within the same repositories, sometime not even having a similar name.
a better question is how many different repositories are those crates from.

also, my guess is that a lot of those are optional deps that you can enable / disable.
>>
>>107535489
you know, you can then just not use those projects that use a lot of deps.
and if you use rust, you don't have to use any deps either.

you can also use deps that have no deps themselves, for any big deps there is often a smaller standalone alternative.
>>
>>107535468
>>107535539
if you actualy look at the real list, i doubt there is actualy more than 40 separate crate.

and also it's a fucking gnome project, what do you expect.
>>
>>107535530
>lets drive using stone wheels because making better ones is too complex for my midwit brain
>>
>>107535564
>>107535539
also lots of build dependencies that will not be in the executable, many are target dependent too.
i mean yes it's more than it probably needs but who cares, i'm not using it anyway.
>>
>>107535564
>and also it's a fucking gnome project
so all crates on crates.io have same quality as gnome projects?
thanks for the laugh
>>
>>107535572
> stone wheels
retard, if you think you can make a better wheel than one that has 100's of hour of engineering behind it you are delusional.
and even in the rare case that you can, practicaly you will not do it, i've seen tons of cniles codebases where they half ass something and move on.
also don't even pretend like some c/c++ project don't use hundreds of deps as well, i've seen things worse than that in c / cpp.
>>
>>107535582
>so all crates on crates.io have same quality as gnome projects?
no.
there are tons of standalone crates.
>>
>>107535587
It takes me 5 minutes to make a better wheel than a stone wheel for the usecases where a wooden wheel would be superior to a fucking slab of stone
>>
>>107535539
Nice cherry picking but the over 200 number already took that into account.
>guess
librsvg has three USE flags that effect code, x32 introspection and vala.
>>107535554
I don’t have a choice in using librsvg because it’s a hard dependency of gtk3+ you retard.
>>107535564
>real list
That includes nested dependencies, not just what librsvg declares in cargo.toml. cat + grep name + wc gives me 357 lines from their cargo.lock kek.

I’m not responding to the rest of the cope you keep posting, get a real job ESL.
>>
>>107535599
my point is that it's not a stone wheel.
it's generaly a very well made wheel.
and practicaly speaking you are the one that's gonna make a stone wheel because you are in a rush and want to move on to something other than the wheel.

> muh code reuse is bad
what a retarded mentality.
>>
>>107535608
It seems like you don't understand wheels and should stop using them in your own midwitted analogies.
>>
>>107535607
>I don’t have a choice in using librsvg because it’s a hard dependency of gtk3+ you retard.
You actually do, if you are willing to lock yourself to oldest possible gtk3+ library that works and patch it enough until it stops requiring librsrvg because svg has literally 0 usecases, I never saw a svg image online in past 10 years and I wouldn't care even if I did, plenty of online conversion hosts that will run rust so I don't have to.
>>
File: tt.png (862 KB, 2000x2000)
862 KB
862 KB PNG
code reuse is bad
>>
>>107535607
>200 number already took that into account
no because your number would be closer to 50 then.
> librsvg has three USE flags that effect code
gentoo flags and rust flags are not the same, there are flags that don't concern gentoo.

> I don’t have a choice in using librsvg because it’s a hard dependency of gtk3+ you retard.
> rust is isignificant.
> can't have a choice to use it
i'm pretty sure you've said both in your life, pick one.

> grep name + wc gives me 357 lines
this is possibly the worse way you could have counted that.
do you understand that this is a wrokspace, ie there a many different project under it, they all share the same .lock but they don't all use the whole dependency list.

also platform dependent deps.
+ conditional compilation.
+ many "different" deps are hosted in the same repo and part of the same project.
+ many of the "deps" are self references to the project itself.

also many are build deps and don't end up in the program
and even then, only used functions will end in your executable.

you just don't know how to count.
>>
>>107532625
>There are literally hundreds of languages that don’t let you sleepwalk into CVEs like C does. They might not be fully memory safe, but they’re all better than C. If memory safety was a huge security issue for decades, why is it only being addressed now?
Because the rise of C made memory unsafety a huge security issue. C was less popular decades ago and more software was written in those hundreds of more reliable languages before C gradually replaced them.
>>
>>107535719
>>107535607
kek you didn't even filter out duplicates ...
>>
>>107535729
>Because the rise of C made memory unsafety a huge security issue.
No it didn't, all bugs are caused by midwits and nothing else.
>>
>>107535719
>>107535732
Neither did rust compiler which had to compile all that crap, recompiling same code 10's of times, very good design saaar!
fuck off retard.
>>
>>107535750
you are a retard if you think that's true...
also you do understand that you can have different version of the same dependencies because different deps rely on a different version ?
>>
>>107535761
That's another argument to avoid dependencies.
>>
>>107533066
>C is much easier to learn than rust
>Learning is hard
Learning rust isn't hard. It will take you a week to go through the book. If you want to read it then one evening will do.

Learning is part of life nigger. Sit the fuck down.
>>
>>107535763
nice cope.
imo deps are fine.
and you don't have to use them.
regardless your current system has hundreds of libraries.
have you ever seen what happens if you build pandoc ?

even that bloated librsvg, my bet is that the lib itself use a fraction of that list, of which there is less than 40 distinct repos.
which is not the end of the world when you consider that svg is pm webshit tech.
ie they use a lib for css parsing.
>>
>>107535793
Yup it's really nice to know that a bugfix doesn't matter because that one dependency is bug-free but your whole dependency graph has 9 more copies that aren't patched yet, very good design saaaaar.
>>
>>107535788
> If you want to read it then one evening will do
i agree, but most people report that they take a month or more to realy get good at it.
most people will have no productivity for at least a month.

also, even if you know the language, you still haven't learnt the ecosystem and commonly used crates.
> Learning is part of life nigger
no fucking shit, i have no issue with learning retard.
i never said that learning is hard, i just said that C is a lot simpler to learn than rust.
do you have any reading comprehention ?
"one of the hardest learning curve out there" doesn't mean hard, it just mean hardest.
>>
>>107533090
The new C# compiler is written in C#. It's fully boot strapped now.
>>
>>107535802
you are a retard if you can't understand why that's the case.
also no, it's not 9, the few that had duplicates had one other version and that's it.

and also we have that with C software as well, where you'll use 2 libs that both depends of different version of another lib, that's why not only you have hundreds of libs installed on your system, you more than likely have multiple version of many of them.
>>
>>107535805
Yes.
It takes 6 months to get good at any programming language. No ine gets good in 1 month. Not even Python. Python and Rust will take you the same amount of time to learn. The main difference is the curve.

My point if that you just need to find time to sit the fuck down and learn. Stop being a pussy.
>>
>>107535818
>where you'll use 2 libs that both depends of different version of another lib
this has never happened to me in C
>>
>>107535837
> My point if that you just need to find time to sit the fuck down and learn. Stop being a pussy.
uh anon, i already know rust, i've used it professionaly for years, what are you talking about...
>>
>>107535844
i read that as "i have never written anything complex".
>>
>>107535851
Okay, you have 5 minutes to name these libraries.
>>
>>107535844
>this has never happened to me in C
this happens but it's pretty rare, however this is a pain in the ass as the linker will scream at you and you have to do hacks around it.

C made it such a pita to use other libraries that most (but not all) libraries try to be self standing anyway.
this isn't the win you think it is.
>>
>>107535859
one notable pain in the ass is openssl version mismatch.
if you've ever used both libcurl and another library that needs openssl but a different version of, this can be a pain in the ass to deal with.
>>
>>107535920
I'm not a webshitter so I never had to depend on openssl.
>>
>>107535923
> openssl is only for webshit
what kind of retarded take is that, you could have said libcurl (which is literaly webshit) but you specificaly choose the one that was not related to web.

anyway, you get the point.
you could have the same issue if you use libssh and libpq (postgres).
>>
>>107535923
>>107535934
in such case C is realy bad at dealing with these issues, rust has no such issues as you can just have different version of the same lib at once.
>>
>>107535934
The only case where I had to deal with data that absolutely must not be leaked, it was classified and on an airgapped system.
The rest is webshit.
>>
>>107535946
> anything that's not what i did is webshit.
so writting kernel drivers is also webshit according to you ?
what about writting software for medical devices ?
>>
>>107535951
kernel drivers luckily don't have to depend on websihtssl
>>
>>107535966
still, there is more to non webshit than kernel drivers and what you did.
tons of things are not webshit.
webshit is anything related to http, html, css and jsshit.
>>
>>107535976
openssl has 0 usecases outside of webshit
>>
File: rust'd.jpg (116 KB, 1200x1000)
116 KB
116 KB JPG
>>107527035
Rustchuds won
>>
>>107535978
what a retarded cope, even postgresdb libs depends of openssl, and there are tons of reasons you may use a db not related to webshit
>>
>>107536018
omg, a webshit depends on webshit? woawie o mama
>>
>>107536020
bad faith.
just move on to something else idk.
besides, openssl can be used to encrypt shit, to sign shit, there are tons of uses not related to webshit.

and networking which may require a secure or at least signed tcp/udp connection doesn't mean webshit either.
>>
>>107536034
If I needed a secure network, I wouldn't be using webshit first and foremost.
>>
>>107536035
> I wouldn't be using webshit first and foremost.
openssl is not related to webshit and you know it, retard.
heck you may even use it to encrypt shit.
>>
>>107536044
It also isn't related to security.
>>
>>107536047
> muh cryptography isn't related to security
what kind of retarded take is that.
anyway, i generaly use libsodium instead but that was to give you an example.
just admit that it can happen and move on instead of calling everything you can't find an excuse for as webshit.

i could find you other example than openssl anyway.
>>
>>107536068
anyway, I accept your concession, and it can't happen, because C devs aren't mentally ill and don't include 300 dependencies just to parse some obsolete format
>>
File: file.png (14 KB, 246x191)
14 KB
14 KB PNG
>>107536078
> and it can't happen
i literaly showed you an example of where it can happen

> 300 dependencies just to parse some obsolete format
rust devs don't either, the project was literaly a ton of different programs under the same workspace.
the parsing stuff itself probably used a single crates which itself may have used an extra 5 total.
which are legitimate uses, see picrel :

serde, so that now anycodebase using serde are compatible with the format
phf is perfect hashmaps generated at compile time, a thing you can't even fucking do properly in C, so a performance reason.
smallvec, another performance thing.
i mean seriously....

>I accept your concession
huh so you are the no_std hello world and epoll filtered guy again, why am i even talking with a retard like you.
lmao goodbye
>>
>>107536104
I accept your concession.
>>
>>107536133
> can't write a hello world
> i accept your concession waaaah
lmao, your words can't touch me because i see you as an absolute retard.
>>
>>107536181
The reflection in the mirror isn't me.
>>
>>107536197
> can't write a hello world
> it's not me saars i swear
>>
>>107536238
huh?
>>
>>107536246
Don't play dumb, as soon as you said "muh concession" i knew who you were.
>>
>>107536310
Did a voice in your head tell you this or did you come up with it on your own?
>>
>>107528449
>"Simple"
>"Right"
>"if"
>"wrong"
Why so many quotations?



[Advertise on 4chan]

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.