Comments by "Anony Mousse" (@anon_y_mousse) on "my new wife" video.

Rust won't prevent all errors and triggers errors on code that would otherwise work 100% perfectly fine. It's training wheels for newbies, but no one who wants the language to get out of the way so they can do real work should use it. If you write your code in a way similar to how Rust wants you to write your code then literally any language would work equally as "safely", except that most others will let you get on with the task of actually writing the code. If you have to integrate with anything closed source, and I know Rust hates closed source, but you introduce a possible security hole and if you're writing real code you're going to use someone else's libraries instead of implementing everything from scratch yourself. If you want to write code that's safe, using a library written in an older language, and which has been debugged for decades, will give you the best path to success. Also, from a purely aesthetic point of view, Rust is uglier than C++, and from a usability point of view it's far, far slower to compile which makes iterative design harder to do.
7
@theninjascientist689 Invalid argument. Compare apples to apples. If you can write good C++, then you don't need Rust.
5
You can use C++ and never manually allocate or have "silly" errors that Rust claims they can stop. The fact is, Rust only stops one narrow point of failure, and not perfectly so, and all the Rustaceans seem to be under the mistaken impression that it stops all errors and perfectly. I'm sure someone will refute that impression or claim that the narrow line of errors it sometimes prevents are 70% of all errors, but they'll be lying if they do.
5
The majority of those older applications that were written in either C or C++ have been debugged to hell and back again. They work better than newer applications and have more features. So yeah, if memory corruption is discovered, it's usually in newer code, which can be written in any language, and is often the result of not understanding what you're doing with a given language or using an interface incorrectly. In other words, bad programmers write bad code in any language.
5
Yeah, if you're learning data structures then it is definitely easier to use a literal language like C or C++. Even Python would be a better choice and Python sucks.
5
@JaconSamsta Since it ate my reply I'll say it again, there is no C/C++, it's C and C++ which are two separate languages. C++ syntax is crap, but Rust is worse. C has better syntax than either and if you're having problems with C declarations then you're writing code that is far too complex for your fragile mind to comprehend. All the languages you mentioned also have worse syntax than C, and sometimes C++. Pure idiocy is calling the simple syntax of C bad while admiring anything about Rust.
4
@DissyFanart That's just propaganda. Stop it.
3
I was with you until you said "better than C". It's basically a warped and twisted Ocaml, albeit with stunted growth, mixed with C++ and given a worse syntax than either.
3
In truth you don't need to manually manage memory the way Rustaceans claim. Use a data structures library and use a GC library as your allocator and you never have to manually deallocate. Write your code defensively and even manual memory management is error free. If you use C++, you should never manually allocate and thus never need to manually deallocate, so a GC isn't needed there nor is it wanted, and as a bonus, C++ has all kinds of handy automation and even smart pointers, if you need them.
3
@JaconSamsta Why do people making these same braindead strawmen arguments. I don't use C because it prevents me from making mistakes that I would make if I were using assembly, because it doesn't prevent such things, I use C because it makes me more productive. If you're using a programming language because you expect it to fix your sloppily written code, then please get out of the industry before any code you write winds up in a medical device that kills someone. And if you can't be at least as productive, if not more so, in C than in Rust, then you're not a good programmer.
2
@emeraldbonsai I'd rather have fewer programmers than more code written by bad programmers.
2
@hottoniapalustris1541 That's complete bovine excrement and if you were wiser you'd know it.
2
@_colonial_ Odd, because C++ projects take an infinitely shorter period of time to compile for me even when you compare the speed of compilation for C++ on my Raspberry Pi to the speed of Rust on my desktop. And you're basically admitting that Rust is for awful programmers. Not a point in favor of its existence.
2
@_colonial_ Ah, never mind, apparently I did respond previously, but YouTube shadowed my previous response and the latest as well. I wouldn't have noticed if I hadn't double checked the latest notification which only showed two responses still. I'm going to guess that this will be shadowed as well and you didn't see either of the other responses I made to you and you won't see this one. Of course, I have to keep trying to get through to Rustaceans and continually counter their lies because the only thing necessary for evil to triumph is for good people to do nothing.
2
@johnterpack3940 When you're good, your screwups tend to be minor such as a typo. When you're average or below, you make logical errors that often aren't detected, even by the compiler, until after it's cost a company millions of dollars or a patient their life. If you need a programming language to make you good, then you're bad, and in such instances, it'd be better for everyone if you quit and found another job.
1
@johnterpack3940 Except that Rust doesn't prevent you from doing something you shouldn't be doing, and what it prevents you from doing isn't wrong in all circumstances, and that's even disregarding the false positives. The biggest problem with the arguments in favor of Rust is that they're propaganda. If a language is going to take an ideological design approach and prevent you from doing things that the designers deem to be bad, they better make damn sure that those things are universally bad as opposed to sometimes bad.
1
@johnterpack3940 And you don't see the fallacy in the logic. The very existence of unsafe makes the propaganda null and void. And the existence of Arc, Rc, Box and all the other cruft makes it more annoying to use even when you're doing the right thing. There is no winning with Rust, it's an uglier C++ with less utility.
1
@hehehehe_hah C++ isn't a superset of C, at least not anymore. While it may be easier to use than Rust and with a better syntax, it won't replace C either and isn't intended for the same spaces where C is still used. Although, if they keep adding to it the committee may kill it. I'd be fine with that if we had a decent successor in its space, but where C++ shines nothing else is as good. Not that that will stop people from espousing support for their favorite trend language.
1
@jonhdoe4119 Which is why it's better to have a programmer that doesn't need it, because C will work on 99% of those platforms while Rust, if it's lucky, will work on 10%. I'll say it again, there exist static analyzers for C. If you really need a borrow checker because you're a bad programmer then you can use one with C.
1
@nepp9574 I actually agree with all of that, and I don't personally use a GC library, but I still recommend the Boehm GC library which has existed for over 30 years to anyone that has problems managing memory. Of course, I also have my own data structures library that handles memory internally, so I have scant few locations where I need to worry about freeing memory.
1
@monad_tcp It's kind of hard for me to take this seriously when you misspell von Neumann, but it should be noted that the Harvard architecture doesn't have to mean a lack of self-modifying code. It just requires two separate address buses for code and data. Most Harvard architecture machines these days will have a ROM chip for the instructions, but sometimes it's shared data on that same chip, usually for static data, and sometimes shared both on that chip and on a RAM chip. However, it could just as easily mean a ROM chip and two RAM chips thereby allowing the program to be modified at runtime, and there are indeed long term setups like that with memory backed RAM.
1
@AlbatrossCommando Don't apply for work with the US government.
1
@AlbatrossCommando Would be nice, but knowing how to modify older projects would make you indispensable, and thinking that older code is inferior just because it's old is a red flag.
1
@AlbatrossCommando Can you give a valid reason or set of reasons to back up this hatred?
1
@AlbatrossCommando I don't guess you realize how stupid that sounds, but hating a language because it doesn't have a gimmick is as stupid as loving a language for any particular gimmick. Also, you can do OOP in any language, and C++ hasn't been a superset of C in decades along with the fact that numerous other languages have classes, and plenty of static analyzers exist for C that you don't need a borrow checker built into the compiler because it's a separate tool.
1
@_colonial_ The tools for checking correctness of programs written in C and C++ are far more mature than the tools for Rust, and have existed since before Rust did. If programmers don't use tools to ensure their code is correct, or better yet, learn how to write correct code from the start, then they probably shouldn't be a programmer anyway. You don't really need modules because PCH's exist. That's a feature that you can actually use, if you bother to learn how, whether you're using MSVC or GCC. Sorry for the late response, but YouTube loves to hinder users with viewpoints they don't like and it wasn't until the guy who wants to watch the world burn posted a response that I finally got a notification.
1