Google to write all new projects in Rust

Google shares guidelinds on how to secure legacy C++ code which it will continue to maintain and estimates it to be impractical to rewrite everything.

In the meantime, they announced their commitment to write all new projects in a memory safe language like Rust.

https://security.googleblog.com/2024/03/secure-by-design-googles-perspective-on.html?m=1

Google is now about to join Amazon and Microsoft and start hiring Rust developers.

Homeless People Are Sexy Shirt $21.68

Ape Out Shirt $21.68

Homeless People Are Sexy Shirt $21.68

  1. 1 month ago
    Anonymous

    How does this personally affect me

    • 1 month ago
      Anonymous

      gayMAN is oxidizing and dragging a the industry in the same direction.

      • 1 month ago
        Anonymous

        >gayMAN is oxidizing
        Yes. gayman is failing.

    • 1 month ago
      Anonymous

      Google will be able to implement their new image formats without putting a remote code execution vulnerability in most of the world's computers. As of September 2023 that was still impossible.
      https://www.tarlogic.com/blog/cve-2023-4863/

    • 1 month ago
      Anonymous

      google will not start paying its employees properly for actually good code so now they will be producing secure code with shit employees getting paid shit money

    • 1 month ago
      Anonymous

      It means learning rust could get you a job

  2. 1 month ago
    Anonymous

    Something about Google using what Mozilla put out in the world just feels.. wrong.

    • 1 month ago
      Anonymous

      Mozilla won the open salsa war and lost the consumer war

    • 1 month ago
      Anonymous

      chrome was originally based off apple's webkit fork which was based off kde's browser engine

    • 1 month ago
      Anonymous

      big corporations like google use all kinds of software freetards code and maintain for them

  3. 1 month ago
    Anonymous

    why not go

    • 1 month ago
      Anonymous

      Go is very data race and race condition prone which can cause issues. Besides, i believe the introduction of arenas is on hold undefinitely which makes it unusable for some of the projects google works on.

    • 1 month ago
      Anonymous

      >being this new
      Go was invented to minimize the damage codemonkeys could do, resulting in a language where you need 10x the work to achieve 0.1x the results.
      It also didn't actually stop idiots from causing damage, so...

    • 1 month ago
      Anonymous

      Go has a runtime and GC by design. Rust is more like C++ in that regard. Go can't replace all use-cases for C/C++/Rust, since it will always have that overhead.

      Go is very data race and race condition prone which can cause issues. Besides, i believe the introduction of arenas is on hold undefinitely which makes it unusable for some of the projects google works on.

      Arenas were dropped because they would introduce the possibility of use-after-free, and could cause Go to fall into the "non-memory-safe" category. Race conditions are a skill issue in any language. There's nothing wrong with Go, it does the job well, but it's not a tool for everything.

      • 1 month ago
        Anonymous

        >Race conditions are a skill issue in any language
        Rust does more than most to handle them, it enforces thread safety in the type system. Can't prevent all race conditions of course but it does prevent unsynchronized access.
        I've barely used Go so my opinion isn't very valuable but none of these mistakes would even compile in Rust: https://www.uber.com/blog/data-race-patterns-in-go/

      • 1 month ago
        Anonymous

        >Race conditions are a skill issue in any language
        Rust does more than most to handle them, it enforces thread safety in the type system. Can't prevent all race conditions of course but it does prevent unsynchronized access.
        I've barely used Go so my opinion isn't very valuable but none of these mistakes would even compile in Rust: https://www.uber.com/blog/data-race-patterns-in-go/

        Data races are a specific type of race condition that can actually be prevented automatically
        Go just doesn't bother because it's simpler to ignore the problem

        • 1 month ago
          Anonymous

          >can actually be prevented automatically
          actually, the data race detector is embedded in the Go compiler. but it's does detection at runtime.

          • 1 month ago
            Anonymous

            Prevention at compile time is good
            Detection at runtime (what Go does) is also good, but not as good as prevention at compile time
            Both are better than the C/C++ approach of telling you "simply choose not to write the bugs"

          • 1 month ago
            Anonymous

            >Prevention at compile time is good
            >Detection at runtime (what Go does) is also good, but not as good as prevention at compile time
            I know. the data race detector isn't really too more different than the race detectors in C/C++, in the sense that you have to enable them through flags instead of being enabled by default at compile time:
            https://go.dev/doc/articles/race_detector

          • 1 month ago
            Anonymous

            It's also similar in the sense that it slows everything way the frick down, whereas if you can prevent data races at compile time there's no runtime check necessary

  4. 1 month ago
    Anonymous

    WH once again destroying their precious market leader. This feels like In-Q-Tel pulling the rug under them and giving everything OpenAI or some other new shadow corp from Israel.

  5. 1 month ago
    Anonymous

    I'd be very curious to see timescales on this. I somehow doubt that even if they dropped all C++ now and only started new projects in rust, I still think they'd be writing new C++ code decades from now for existing systems.
    These empty commitments don't mean anything because when there's actual development going on, it's still all C++ and rust is limited to very small scale and isolated things at Google (Android Bluetooth stack, chrome qr code module, etc.), vs new products heavily integrated into their existing infrastructure.
    I guess time will tell to see how much this actually does. I think at least at Google, they might even use more carbon than rust assuming they don't drop that.

  6. 1 month ago
    Anonymous

    Google is a sinking chip, having said that: nice. I just with Rust had more competition.

    • 1 month ago
      Anonymous

      I guess they sink because all that money is too heavy?

      • 1 month ago
        Anonymous

        The future is overglorified Markov chains and Google doesn't seem able to get them right. They are fricked.

      • 1 month ago
        Anonymous

        The pace of growth has declined. People have noticed. Google reacted by making huge job cuts. But you've cut something off to become a rust programmer. So go work for google. You probably won't get cut again.

      • 1 month ago
        Anonymous

        >Jeetified dying company choose parasitic dying language
        I'm shocked.

        Literally doesn't mean anything. Google is fricked long term, everyone knows this.

      • 1 month ago
        piggyanon

        what about profit?

    • 1 month ago
      Anonymous

      are those 30B monopoly dollars adjusted for inflation?

  7. 1 month ago
    Anonymous

    What happened to Carbon?

    • 1 month ago
      Anonymous

      That's what they'll eventually use to secure theor legacy code but they're still writing new projects in rust.

  8. 1 month ago
    Anonymous

    Google is indeed joining other tech giants like Amazon and Microsoft in the adoption of Rust and the hiring of developers skilled in this language. This shift reflects a broader industry focus on security and efficiency in software development.

  9. 1 month ago
    Anonymous

    New language to fill the graveyard

  10. 1 month ago
    Anonymous

    Still no jobs which is cheeks

  11. 1 month ago
    Anonymous

    I know nothing about programming. In what way is Rust better? Less resources and space used?

    • 1 month ago
      Anonymous

      It has something called the borrow checker. Basically, any allocated memory has to be freed when the variable goes out of scope otherwise the BC will shit and piss itself.

      • 1 month ago
        Anonymous

        I suppose this additional checks occupy some space in the final program.
        How much more space does Rust needs than c++?

        • 1 month ago
          Anonymous

          Rust is compiled.

          • 1 month ago
            Anonymous

            Yes, and? How much more space a program (i.e. the executable) written in rust occupy than one written in c++? I know it's compiled man.

          • 1 month ago
            Anonymous

            These additional checks don't occupy space in the final program.

          • 1 month ago
            Anonymous

            Bullshit. There must be some sort of compromise, otherwise Rust would just be a magnitude of order better than C++, and it won't explain why only now is starting to be widely adopted.

          • 1 month ago
            Anonymous

            OK.

          • 1 month ago
            Anonymous

            So you don't know shit. Ok.
            I was actually opened to discussion but it seems you're throwing a fit for some reason.

          • 1 month ago
            Anonymous

            It's not that hard to grasp, anon. It's a compile check. If you write code that doesn't obey Rust's lifetime rules, the compiler refuses to compile it. If the code is fine, it gets compiled normally without adding runtime checks or anything.

          • 1 month ago
            Anonymous

            So the compiler ALWAYS knows if a pointer will be dangling or that will be allocated forever and things like that?

          • 1 month ago
            Anonymous

            Yes, as long as you follow the "Rust lifetime rules", then it's possible for the compiler to always keep track of all those things.

          • 1 month ago
            Anonymous

            >So the compiler ALWAYS knows if a pointer will be dangling
            It doesn't know if a pointer will be dangling, but it maintains rules that guarantee it will never be dangling.

            >that will be allocated forever and things like that?
            Memory leaks are not prevented by Rust(or by any system, even garbage collectors, see Minecraft). Rust only prevents you from doing things that would let you observe undefined behavior.

          • 1 month ago
            Anonymous

            >any system
            makes me wonder why there isn't a language or a feature in a language that forces you to allocate all the memory your program will ever use at the startup, like what they do at tigerbeetle (their custom NASA style programming rules https://github.com/tigerbeetle/tigerbeetle/blob/main/docs/TIGER_STYLE.md )

          • 1 month ago
            Anonymous

            Because, outside of very specific use cases, that's very stupid.
            Also it doesn't prevent leaks, you can still lose track of what chunks of your initial allocation you're actually using.

          • 1 month ago
            Anonymous

            Yes, you can get rig of memory leaks if you get rig of dynamic memory allocation altogether. But that's not very practical and memory leaks are not really something all that common if you use RAII and your brain.
            There are also various runtime analyzers that can help you find memory leaks too.
            Also a single buffer overflow or something can be catastrophic for the security of your application, while small memory leaks might not even matter if your program doesn't run 24/7.

          • 1 month ago
            Anonymous

            It is. Insecure brainlets are slow to accept change and fear their time was wasted learning languages they know

          • 1 month ago
            Anonymous

            I don't see how learning a new language somehow erases the knowledge of a different language

            >So the compiler ALWAYS knows if a pointer will be dangling
            It doesn't know if a pointer will be dangling, but it maintains rules that guarantee it will never be dangling.

            >that will be allocated forever and things like that?
            Memory leaks are not prevented by Rust(or by any system, even garbage collectors, see Minecraft). Rust only prevents you from doing things that would let you observe undefined behavior.

            Ah I see now, thanks. I might consider learning Rust. I'm a c/++ senior embedded programmer but it won't hurt knowing other things.

          • 1 month ago
            Anonymous

            >anon finds out what static analysis is

          • 1 month ago
            Anonymous

            compiled code is just CPU instructions after processing the source. All the extra "safety" shit of rust is a bunch of checks that occur during the process of compiling the source.
            You can write C/C++ source that will produce the exact same machine code.

          • 1 month ago
            Anonymous

            >otherwise Rust would just be a magnitude of order better than C++
            This is literally what people have been saying, but people who have been burned before by previous language wars are so fricking skeptical that they insist it's all lies and the hype is undeserved
            I don't think anyone posting on IQfy will change your mind but maybe a few more years of steadily climbing adoption will do the trick

          • 1 month ago
            Anonymous

            >There must be some sort of compromise
            There is.
            Sometimes a bound check can't be optimized away and has to be checked in runtime.
            Sometimes taking utf-8 into account adds some overhead, even though your program could just ignore it.
            A lot of the time, Rust adds quite a bit of cognitive overhead. Especially for new programmers it's often significantly harder to write Rust than C++.
            However neither of these add a significant performance or space overhead. In practice Rust is as performant as C/C++ while being much safer, at cost of being somewhat harder to learn.
            Like other anon said, it's really not that different from using a static analyzer. Except that Rust was designed for such analysis from grounds up, so it's more ergonomic and can have a little better guarantees sometimes.

          • 1 month ago
            Anonymous

            >There must be some sort of compromise
            There is. It's called inflated compile times. You can achieve similar levels of "safety" with C or C++ if you use the right tooling and the features which come with GCC or clang, that are not enabled by default. People don't use them is because 1) people are lazy and 2) many don't know about them.

          • 1 month ago
            Anonymous

            Compile time isn't slow, it's linking time
            >You can achieve similar levels of "safety" with C or C++
            False, or we wouldn't be in this situation.

            >In our experience, it is not sufficient to merely make safe abstractions available to developers on an optional basis (e.g. suggested by a style guide) as too many unsafe constructs, and hence too much risk of bugs, tend to remain. Rather, to achieve a high degree of assurance that a codebase is free of vulnerabilities, we have found it necessary to adopt a model where unsafe constructs are used only by exception, enforced by the compiler.
            >Attempts to mitigate the risk of memory safety vulnerabilities through developer education and reactive approaches (including static/dynamic analysis to find and fix bugs, and various exploit mitigations) have failed to lower the incidence of these bugs to a tolerable level.

          • 1 month ago
            Anonymous

            >Compile time isn't slow
            It is slow but it's getting better because it's "just" a matter of doing a shitload of work to optimize the compiler, not some deep design flaw in Rust

            >There must be some sort of compromise
            There is. It's called inflated compile times. You can achieve similar levels of "safety" with C or C++ if you use the right tooling and the features which come with GCC or clang, that are not enabled by default. People don't use them is because 1) people are lazy and 2) many don't know about them.

            #include <iostream>
            #include <string>

            std::string const& id(std::string const& s) {
            return s;
            }

            int main() {
            // OH NO!!!!!!!
            auto const& s = id("Hello, World!");
            std::cout << s << 'n';
            }

          • 1 month ago
            Anonymous

            >

            >There must be some sort of compromise


            There is. It's called inflated compile times. You can achieve similar levels of "safety" with C or C++ if you use the right tooling and the features which come with GCC or clang, that are not enabled by default. People don't use them is because 1) people are lazy and 2) many don't know about them.
            >#include <iostream>
            >#include <string>
            >std::string const& id(std::string const& s) {
            > return s;
            >}
            >int main() {
            > // OH NO!!!!!!!
            > auto const& s = id("Hello, World!");
            > std::cout << s << 'n';
            >}

            uhm:

            ```
            <source>: In function 'int main()':
            <source>:10:17: error: possibly dangling reference to a temporary [-Werror=dangling-reference]
            10 | auto const& s = id("Hello, World!");
            | ^
            <source>:10:23: note: the temporary was destroyed at the end of the full expression 'id(std::__cxx11::basic_string<char>~~*const char*)"Hello, World!"), std::allocator<char>(*~~'
            10 | auto const& s = id("Hello, World!");
            | ~~^~~~~~~~~~~~~~~~~
            ```

            https://godbolt.org/z/x98c9fYeG

            thanks for playing, gayget

          • 1 month ago
            Anonymous

            ><source>:10:17: error: possibly dangling reference to a temporary [-Werror=dangling-reference]

            >>B-but y-you must use r-rus-*ACK*

          • 1 month ago
            Anonymous

            notice how he hasn't had any screentime lately?

          • 1 month ago
            Anonymous

            >notice how he hasn't had any screentime lately?
            I don't watched lame tech tips anyway so dunno.
            I'm glad if that's the case.

          • 1 month ago
            Anonymous

            >```
            you need to go back

          • 1 month ago
            Anonymous

            >you need to go back
            modgays need to add introduction button that's easy to access. it's not even in the sticky post

          • 1 month ago
            Anonymous

            also go back to where, to md file? you think only gaydit uses tripple quote?

          • 1 month ago
            Anonymous

            >you need to go back
            modgays need to add introduction button that's easy to access. it's not even in the sticky post

            also go back to where, to md file? you think only gaydit uses tripple quote?

            #+BEGIN_SRC c++

          • 1 month ago
            Anonymous

            Pack it up, folx, rust is CANCELLED

          • 1 month ago
            Anonymous

            so do you want to just throw away th esource code? fricking lmao

            Bullshit. There must be some sort of compromise, otherwise Rust would just be a magnitude of order better than C++, and it won't explain why only now is starting to be widely adopted.

            learn2code

          • 1 month ago
            Anonymous

            It's like typechecking, it doesn't add any additional code to your executable to make sure you only pass ints to a function that only takes ints
            If your code doesn't follow the rules it doesn't compile, there's nothing more to it than that

      • 1 month ago
        Anonymous

        You are confusing borrow checker with RAII.
        Borrow checker only ensures that no reference exists to something when it goes out of scope and that there is never more than two mutable references to something.

    • 1 month ago
      Anonymous

      No, it basically has a ruleset for how to handle the written code that will throw errors if it doesn't satisfy these rules, that enforces a certain standard of safety not present with something like C++. In C++ it is the programmer who has all the agency, so potentially unsafe memory handling can happen as there are no rules that prevent it.

    • 1 month ago
      Anonymous

      In no meaningful way that isn't astroturfed. Its supported by the white house to allow more backdoors into software.

    • 1 month ago
      Anonymous

      It's relatively strict when it comes to disallowing things that are risky in other languages. This makes it extremely useful if you're mostly hiring pajeets and women, as it minimizes their ability to break shit.

    • 1 month ago
      Anonymous

      In languages like C and C++, if you want to write something in RAM memory you have to first allocate memory space, and free it when you no longer need it. If you don't manage this correctly, you get errors and vulnerabilities. Like 70% of all program errors and vulnerabilities have memory management frick-ups as the underlying reason.

      Languages like Java and Go, allocate and free memory for you automatically, but this comes at a performance cost, because the automation is done by a program, called garbage collector, that runs alongside the original program.

      Rust on the other hand, simply doesn't allow you frick up, by enforcing you to write code in a specific way.

      There's more, but this is the essence of it.

      • 1 month ago
        Anonymous

        the nocoderest of nocoders
        i kneel

    • 1 month ago
      Anonymous

      because handling the computer's space resources is complicated operations on this space are prone to errors. Rust has a mechanism that tells you that you are a fricking moron and need to fix your shit when you mishandle memory, C and such languages do not do that because the disk memory is the programmer's domain and the compiler has no business with it.

  12. 1 month ago
    Anonymous

    Do people still care about google in 2024?

    Search is broken
    Youtube is shit
    Gemini is moronic
    Gmail is bloated
    Chrome redesign is a step backwards
    Google cloud is a joke

    The only thing that hasnt been turned to complete shit is google maps.

    • 1 month ago
      Anonymous

      people are very slow and many refuse to admit google sucks it seems

    • 1 month ago
      Anonymous

      >Search is broken
      it's still the best free search engine despite being shit, and at least they're FINALLY working on fixing the shitty blogspam/SEO trickery
      >Youtube is shit
      there is no better alternative that's actually relevant
      >Gemini is moronic
      ignoring the shitty image generation, the new version of Gemini is pretty close to GPT-4 quality, that's impressive for such an obviously rushed project

      anyway it's definitely going downhill fast. They should've fired Sundar Pichai ages ago. But they could turn things around if they fix their dogshit culture. Everyone expected Meta to die after the metaverse shit, look at them now

      • 1 month ago
        Anonymous

        > Everyone expected Meta to die after the metaverse shit, look at them now

        Literally nobody, and I mean nobody except autists on here thought Meta would die because of a side-project.

        • 1 month ago
          Anonymous

          >side-project
          Meta invested billions on VR research with the metaverse as the end goal. They literally changed their name for this this multi-billionaire "side-project"

          • 1 month ago
            Anonymous

            So what, it doesn't change the fact that their business is thriving. How exactly did you imagine Facebook, Instagram and Whatsapp would fail because Metaverse didn't take off? Makes absolutely no sense.

          • 1 month ago
            Anonymous

            >How exactly did you imagine Facebook, Instagram and Whatsapp would fail because Metaverse didn't take off?
            I didn't. Others did, around 2020.
            >Makes absolutely no sense.
            it makes even less sense to think Google will fail cause you personally don't like their massive extremely successful products and services

          • 1 month ago
            Anonymous

            >They literally changed their name for this this multi-billionaire "side-project"
            No, they used it as an excuse to pivet away from their rotten Facebook brand

      • 1 month ago
        Anonymous

        the only reason meta turned things around was a massive ad spending spree by temu who already peaked in the west and is on its way down
        they're not going to be shoveling money into ad spend like they have been

    • 1 month ago
      Anonymous

      cope troony

      • 1 month ago
        Anonymous

        He is a troony for disliking products from corpo? Lmao

        • 1 month ago
          Anonymous

          Liking things is leftist
          Disliking things is also leftist
          There are two wolves inside you
          You are a meat popsicle

  13. 1 month ago
    Anonymous

    I guess Carbon's dead then eh?

    • 1 month ago
      Anonymous

      the purpose of Carbon is to make legacy C++ codebases easier to deal with, nothing else. The Carbon devs themselves recommend Rust or something else for new projects

  14. 1 month ago
    Anonymous

    Its a google project so nothing is going to change in the long term. Remember carbon? Remember flutter? Remember dart? Remember Fuchsia? Remember google RSS? Remember google+? Remember don't be evil? Remember broadcast yourself? Remember the dislike button? Remember nested comments on youtube? Remember subscription folders where you could organize your subscriptions? Remember star rankings before like/dislike? All of these changes were "progress". lol

    • 1 month ago
      Anonymous

      kek, there's even a site collecting all of Google's dead projects:
      https://killedbygoogle.com/

  15. 1 month ago
    Anonymous

    I still don't get this ~~*white house*~~ le every programming language except rust is le "~~*unsafe*~~"
    Will I get viruses if I use software not written in rust? Will I get hacked? What are these troon Black folk on about

    • 1 month ago
      Anonymous

      It doesnt have much to do with "safety" rather its the zoomer generation being in love with the word safe and they also hate anything that anyone older than them uses so they absolutely refuse to use Java for example when it would be a fine choice for 90% of cases where the use of Rust is forced instead. Java would be more safe and more cross-platform compatible even. Probably faster too in many cases

      • 1 month ago
        Anonymous

        good morning sir

    • 1 month ago
      Anonymous

      If you read the papers, it cites rust exactly ONCE and then cites interpreted languages multiple times. Like if it's better to rewrite everything in java or python, disregarding time-critical operations.

  16. 1 month ago
    Anonymous

    Big tech is all talk, when it comes real business products they will keep doing whatever gives them new features quickly and thats not Rust.

    • 1 month ago
      Anonymous

      Meanwhile, Microshaft is rewriting NT kernel in Rust

      • 1 month ago
        Anonymous

        >rewriting NT kernel in Rust
        Source please? Or is it another very small nothingburger that you lying shills are pretending is the whole kernel?

  17. 1 month ago
    Anonymous

    Well noooo shit.

  18. 1 month ago
    Anonymous

    >lets rewrite millions of lines of code instead of just enabling shadow stack compiler flag

    • 1 month ago
      Anonymous

      Not to defend Rust, but do flags really protects from memory leaks/dangling pointers?

      • 1 month ago
        Anonymous

        They do not

  19. 1 month ago
    Anonymous

    Hmm I like that.
    Being a rust hobby developer for over 5 years might put me in quite an advantaged position.
    I wanted to move to write Rust professionally for quite a while but it always seemed as much less stable option than continue to do boring but well paid TypeScript jobs. Soon it might be the best moment to change my job and capture some nice position.

  20. 1 month ago
    Anonymous

    >We need to hecking rewrite everything!!!
    http://www.joelonsoftware.com/articles/fog0000000069.html

    • 1 month ago
      Anonymous

      >All new source code! As if source code rusted.
      Heh.

    • 1 month ago
      Anonymous

      Are you illiterate?
      >Joel ~~*Spolsky*~~
      Dumb homosexual

      • 1 month ago
        Anonymous

        Just a coincidence!

  21. 1 month ago
    Anonymous

    What was the point of creating GO then?

    • 1 month ago
      Anonymous

      So Jeet's can write concurrent application code.

    • 1 month ago
      Anonymous

      Certainly not making data races and race conditions hard to make

    • 1 month ago
      Anonymous

      Go is for morons. Competent devs have always written in C++ at israelitejle

  22. 1 month ago
    Anonymous

    google shouldn't be held up as some kind of thought leader in IT, anymore than IBM should be. they've become a large bureaucratic organisation like IBM/Oracle etc

  23. 1 month ago
    Anonymous

    Everything google has made besides gmail and their search engine has failed miserably. Who gives a shit.

    Everything successful from google outside of the afore mentioned items, they bought.

    • 1 month ago
      Anonymous

      There were a few good projects that they canceled because reasons.

    • 1 month ago
      Anonymous

      They bought maps too?

      • 1 month ago
        Anonymous

        Yes

  24. 1 month ago
    Anonymous

    Why, yes, Rust is to C++ what Scala3 is to fricking Java.

    But to understand this and
    what Rust actually does require some fricking proper education, which most of you lack.

    By just adopting the "at most one mutable reference" restriction from FP (where it is implicit with ref types, being explicitly passed along) it attains FP-like sanity.

    Move semantics by default is another win.

    Restricting and formalizing the behaviour of references by introducing explicit lifetime to every value is the only real innovation in the field in last 20 years or so (second only to the Wadler's typeclasses, on which traits are based).

    You have no idea how many things it does just right lmao.
    https://schiptsov.github.io/

    • 1 month ago
      Anonymous

      >https://schiptsov.github.io/
      who is this moron and why are you spamming the same copy-pasted message in every thread that mentions rust?

  25. 1 month ago
    Anonymous

    >going full Rust
    This is worse than making the Pope black in AI image generators

  26. 1 month ago
    Anonymous

    32-cogged wheel with R and a pentagram inside... yeah how about I just refuse to use anything that has interacted with google.
    My shovel and fishing rod already has Rust, no need to reinvent decay.

    • 1 month ago
      Anonymous

      >32-cogged wheel with R and a pentagram inside
      stop noticing, goy

      anyways, it will become mandatory eventually. the foundation is being laid the US with the Biden regime's latest memo.
      recommendation now, required later. you might be able to use other languages as a hobby (until they inevitably outlaw and regulate compilers), but corps will be required to only hire rust coders (or will they be called "AI Instructors" in the future?)

      expect the usual
      >that's not gonna happen, stupid c.h.u.d.
      >it's happening and that's a good thing

      this stuff goes deeper than you think
      t. insider

      • 1 month ago
        Anonymous

        The memo just sez you should avoid C and C++, Rust is one of a bajillion languages they endorse and they don't mention it very prominently

        • 1 month ago
          Anonymous

          yes, for now. it's a slippery slope
          personally I'll enjoy my recreational zig programming while it's still legal

      • 1 month ago
        Anonymous

        >recommendation now, required later. you might be able to use other languages as a hobby (until they inevitably outlaw and regulate compilers), but corps will be required to only hire rust coders (or will they be called "AI Instructors" in the future?)
        This is what C/C++ programmers want to do. Rust programmers say people should be allowed to make new languages and replace code with something better when we learn better ways of doing things. Rust is not meant to be the eternal and last programming language, unlike what a lot of C/C++ programmers say about C/C++.

        • 1 month ago
          Anonymous

          >Rust programmers say people should be allowed to make new languages and replace code with something better when we learn better ways of doing things.

          Bullshit. I've only seen gatekeeping from the Rust community and powertrips. They're all against c/++ for the wrong reasons. It's all about political stances and has nothing to do with the actual flaws exposed by the languages

          • 1 month ago
            Anonymous

            >I've only seen gatekeeping from the Rust community and powertrips. They're all against c/++ for the wrong reasons. It's all about political stances and has nothing to do with the actual flaws exposed by the languages
            That's what C nocoders on IQfy say about Rust, not what people who actually use Rust say about Rust.

          • 1 month ago
            Anonymous

            What.
            Every single time someone moves to Rust we got dozens of threads full of ciniles seething over it.
            And what happens when some company moves from Rust to something else? We got dozens od threads full of ciniles saying "told you so" and then reposting it in every Rust thread for a year or more.

            The only people doing gatekeeping here is ciniles.

          • 1 month ago
            Anonymous

            It goes both ways. Every time a conversation about C pops up, Rust people can't stop themselves from proselytizing. Same in the inverse. They're both cults, one's just older, more entrenched, and more successful than the other (this isn't C's first bout of competition: C++, Java, Go all were meant to kill it, perhaps Rust will be the great C killer, but I doubt it).

          • 1 month ago
            Anonymous

            Even before C++ and Java, Pascal was a close rival to C
            I guess cniles must have been so thrilled at how they killed Pascal that it went to their heads and they became convinced they'd never lose a fight

          • 1 month ago
            Anonymous

            ah I forgot about Pascal, I remember reading one of Linus' rants about it and how it creates braindead programmers.

          • 1 month ago
            Anonymous

            >Every time a conversation about C pops up, Rust people can't stop themselves from proselytizing. Same in the inverse
            That's not gatekeeping. Gatekeeping is preventing new users from moving to/getting involved with something. Rust programmers do not care if you learn C. If you look into threads where people ask if they should learn C or Rust, the general consensus is to learn C first and then learn Rust because Rust might be too hard without prior systems programming knowledge. Rust also isn't a C replacement but C++ replacement.

      • 1 month ago
        Anonymous

        >recommendation now, required later. you might be able to use other languages as a hobby (until they inevitably outlaw and regulate compilers), but corps will be required to only hire rust coders (or will they be called "AI Instructors" in the future?)

      • 1 month ago
        Anonymous

        have a nice day schizo /misc/ moron

  27. 1 month ago
    Anonymous

    saar, do not redeem rust

  28. 1 month ago
    Anonymous

    obviously rust is hated here but it's the best language right now when the options are c++ or java

    • 1 month ago
      Anonymous

      Find me a compiler for ST microcontrollers that supports Rust.
      I'm an embedded programmer and there's literally no option for embedded shit to write in anything else but C/++, there's no IDE for embedded micros that supports Rust yet.
      All the biggest embedded IDEs (IAR, STM32CubeIDE, Keil etc.) only supports C/++.
      Please don't link me community projects, those are not obviously accepted in a professional environment and are only good for hobbyists.

      • 1 month ago
        Anonymous

        >I'm an embedded programmer and there's literally no option for embedded shit to write in anything else but C/++, there's no IDE for embedded micros that supports Rust yet.
        This is true, and a good argument for why Google shouldn't be using Rust for anything right now
        Unfortunately Google doesn't listen to reason
        Maybe send an open letter? Get Tim Sweeney and Theo de Raadt to sign? They hate Rust

        • 1 month ago
          Anonymous

          I guess they are only focusing on software for mainstream OSes, nobody really cares about microcontrollers despite being used for literally everything

      • 1 month ago
        Anonymous

        >Find me a compiler for ST microcontrollers that supports Rust.
        this means nothing to me, you're probably stuck with something even more archaic than c++98 as well, but you're not barking at the cpp community to fix their shit are you?

  29. 1 month ago
    Anonymous

    >went to install oxipng
    >not in repos, have to install cargo in order to install rust programs
    >cargo will use ~250MB, 4 dependencies
    >deny it
    >install optipng, 1 package, 193KB
    The future looks grim.

    • 1 month ago
      Anonymous

      >literally all software will be over a thousand times bigger
      I'm not sure if that's better or worse than Electron though

      • 1 month ago
        Anonymous

        Cargo is not a part of oxipng

        • 1 month ago
          Anonymous

          Oh, okay then, false alarm

        • 1 month ago
          Anonymous

          it's not, but having to install someting that uses 5 packages and 250MB just to install 1 program because I got nothing else in Rust is just moronic. everything is already bloated as it is and these morons want to make it worse

    • 1 month ago
      Anonymous

      GCC or clang is pretty heavy to install too, I don't know that this makes sense as a complaint
      oxipng itself ends up 1.4MB though, so that is still a lot bigger than optipng. Like half of that is the argument parsing library I think

      it's not, but having to install someting that uses 5 packages and 250MB just to install 1 program because I got nothing else in Rust is just moronic. everything is already bloated as it is and these morons want to make it worse

      If you don't want to build it yourself then you could grab a binary from github. If you do want to build it yourself then you need a toolchain yeah, what do you expect

      • 1 month ago
        Anonymous

        >If you don't want to build it yourself then you could grab a binary from github
        I tried. Need to install dependencies for it to work. I was using the .exe on a W7 VM but gave up, found optipng instead and it just works. oxipng is better, but only by less than 2% per image more often than not. some of them was literally a 10 bytes difference of optimization

        • 1 month ago
          Anonymous

          If you're on Linux then the musl binary should just werk. Dunno why they ship a glibc binary as well.
          Seems that the point of oxipng isn't better ratios but better performance through multithreading. If you want better ratios then tweaking flags is the way to go, I use optipng with -o7 -zm1-9 -strip all in certain cases. (This slows it down a ton.)
          Haven't tried oxipng before but I'm going to give it a spin.

          • 1 month ago
            Anonymous

            like I mentioned, I used oxipng on a W7 VM, and did optimize everything there. only got optipng now and used it on about 10 images, where I simply used -o7 because the rest is overkill for minuscule difference in my opinion
            I also wondered if the same for jpg is possible, and found jpegoptim
            then I was wondering if I can optimize everything an anything (mostly my giant collection of epub and pdf), and found minuimus. gonna give it a try tomorrow

          • 1 month ago
            Anonymous

            jpegoptim is nice yeah. There's also gifsicle, which can give enormous savings on poorly encoded GIFs but can also corrupt them on higher optimization levels.
            >like I mentioned, I used oxipng on a W7 VM
            I didn't understand whether you did everything in the VM or just ran oxipng there because of a dependency problem outside the VM

    • 1 month ago
      Anonymous

      Why not use zopflypng?

  30. 1 month ago
    Anonymous

    >google
    >new projects
    these project always eventually crash and burn and you cannot rely on them, unless they're made to opress the user like safetynet/play integrity. these will last forever and only get better at their job.

  31. 1 month ago
    Anonymous

    I read the link and the whitepaper. Nowhere does it say that they've mandated all new projects to be written in Rust. It's just a suggestion paper from their research division.
    You really should have a nice day, lying Rust shill. It's "people" like you who make me avoid the language due to its community.

    • 1 month ago
      Anonymous

      OP is ragebaiting
      YHBT
      HAND

  32. 1 month ago
    Anonymous

    >IBM to write all new projects in Java!
    >Yahoo to write all new projects in Perl!
    yawn

  33. 1 month ago
    Anonymous

    If you think C++ is bad, just read some of Googles C++ code, it makes you love standard C++. All their C++ codebases are also incredibly bloated, protobuf being one of the worst offenders, if you have to use protobuf, dont use google's runtime

Your email address will not be published. Required fields are marked *