Comments by "Anony Mousse" (@anon_y_mousse) on "Fixing HORRIBLE Embedded Rust So You Don’t Have To" video.

Since hundreds of libraries exist already for C it would've been even easier. Instead he's writing it from scratch in a language which treats him like a baby and loses most of its proclaimed advantages at that level.
3
@felixst-gelais6722 Thanks for the compliment, but I'm not God.
2
@justinwatson1510 Interesting thought, considering both yours and Deline's condescension while simultaneously being wrong.
2
@AlFredo-sx2yy At this point it almost seems like they're trying to start their own religion.
2
@justinwatson1510 I'm not crying foul over your rudeness. I'm pointing out your hypocrisy of castigating me for doing that which you are accusing me of. And in this situation, it would be me who is humoring you.
2
@d3line So basically, Rust is safer than C at this low level because??? It prevents you from dereferencing null pointers? If you write code properly that won't be a problem. 10 times out of 10, dereferencing a null pointer is programmer error. The same is true for using something after you've freed it. And at this low level, type mismatches won't occur because you're not dealing with complex types. So type checking in macros is just pointless abstraction and convolution here. Bounds checking isn't a real concern if you're not a dingus and just don't have any off by one errors and of course check user input. If that bounds checking isn't done at runtime it won't be of real benefit anyway and if it is it'll slow your program down which on a 200mhz processor is a bad thing. If you think any of what Rust provides is both real and of benefit then you must not be a very good programmer in the first place. Tell me which programs you've contributed to so that I may excise them from my system if I'm unfortunate enough to have them?
1
@d3line Nearly all programmers are bad programmers, this is a known fact. So instead of making it easier for them to be programmers and make far worse mistakes than common ones, let's instead shut them out. Linux has bugs because it has thousands of developers who probably shouldn't be allowed to contribute code to it. How many of those bugs are for the common core of Linux as opposed to drivers for hardware that not everyone uses? Do you know? Do you even understand the difference? A Raspberry Pi may not have the limitations of an Arduino or some other niche piece of hardware, but you still shouldn't be cavalier with resources. There are many applications where you need smaller hardware and using Java isn't an option, but what's more is like Rust, many of the features are unimplemented on such hardware and all you get is the syntax. Proving code correctness is impossible for a program to do, at best you can prove partial correctness, which is not good enough. If bounds checking is only done at compile time, then when your improperly written code encounters an out of bounds condition at runtime the check will have proven worthless. So it's slowing down the compile time for little potential benefit to prevent errors which don't happen with good programmers and no benefit for the runtime portion. So it won't slow down the running code, sure, but it won't catch the more pernicious errors. In fact, Rust won't catch the majority of the more pernicious errors which mostly occur because idiots slip through the cracks by learning languages like Java, Python, Go, C# and yes Rust. As for what Microsoft thinks, I don't care, I don't use Windows and won't and they can suck a big D.
1
@d3line Let's see, false equivalency, as flying airplanes is not the same as writing code. As well errors made there are more immediate unless you're using bad code. Oh wait, that's happened and they had to ground a huge swath of planes which had slow building problems because of poorly written code. For kernel vulnerabilities, do you have any idea how to trigger any of those? How long did it take to find them? On the RPi you just made another argument for me because a dingus like most new developers are these days won't understand algorithmic complexity and won't have any clue how to actually implement them. So good luck getting one of them to write good code for embedded systems and if they don't learn it early, good luck getting them in later years to do it as well. You mistakenly assumed I was talking about hardware features when I was talking about software features, as in what the language provides, or in this case fails to. And the point still stands that at that level, Rust and C are on equal footing except for a few points which as I said only apply to bad programmers. Bounds checking can be as simple as "you're using `i` to index the array and checking at the top against a.len()", but if you have a mutating index, as in based on data that may be external to not just the module you're writing but your entire program, it would have to either do runtime bounds checking for each access or none at all. For work on embedded systems you'll encounter that a lot. Airbags actually do cause more harm than good, so that's an invalid point right there. Microsoft still can't write a good OS, but you trust their expertise on writing good code? Talking to beginners is, hopefully, how I influence the world to be better. If I can dissuade ones like you from writing code the world is that much better, and if I can teach better ones than you to write better code by them learning better techniques, better still.
1
@d3line I've stopped caring enough to read your whole post, so I'll just read the last paragraph and address that. I'm anonymous, I don't provide proof because it would deanonymize me. I don't entirely mind people becoming programmers, if they're taught the right things. If I convince no one to get on the right path, then too bad for all of you. Bad code can kill especially in embedded settings such as airplane firmware or medical equipment. Duck that if you don't believe it because there's plenty of proof on that. I don't even care if his videos reach billions, as long as he doesn't promote Rust. I used to believe in humanity, but people like you have really disappointed me. The more D-bags like you that I encounter the less sad I'm becoming at the thought of humanity dying out. Keep contributing to that, please.
1
@d3line Unimportant until a bad programmer kills someone with their bad code written for a piece of medical equipment. Or downs an entire fleet of airplanes which already resulted in hundreds of people dying. You would undoubtedly blame the language for these problems, and you'll undoubtedly claim that your language of choice would've prevented them, but you're wrong. That absolves the programmer of blame for being negligent, not merely not knowing enough, but not caring enough to check their code, and the error in question wouldn't have been prevented by any language's compiler.
1
@d3line In the case of your radiation example, no, Rust would not have prevented those errors. Blame is important in this instance to address where the failure occurred and to correct it for the future. And Rust also, still, would not have prevented what happened in the case of the 737. What's even more important yet, is to not teach children that they can do whatever they want and still be safe, that's how we have Darwin awards. Teach children when they're doing something wrong and how to do it right and you'll get better outcomes.
1
@AlFredo-sx2yy True, and I would add that they fail to understand that logical errors that can lead to very bad code can be written in any language.
1
@d3line You can still have race conditions in Rust. One, by using a library which modifies your variables or which you access the variables of. Two, at the low level they were writing the code for the PDP, if it had been written in Rust it would've required the use of "unsafe" and would not have prevented the error anyway. Did you not look up the 737 or the accidents that occurred? It was due to the auto-pilot software adjusting for a perception of what was happening based on a faulty sensor reading. A logical error that no language would've solved, but a good programmer would not have allowed. Figures that you'd find the Rust compiler aggravating. Do you know how to code?
1
@AlFredo-sx2yy Kind of seems like they think time travel is possible with it too.
1
@d3line Libraries most definitely are excepted. You can't know what a library is doing unless you've compiled the source yourself, and if it's closed source, you sure as heck can't do that. It could modify your data in an unexpected way that you must account for, and which must be manually handled in any language as even Rust won't handle it for you. Are you agreeing with my point about low level hardware usage or do you just not understand the convolutions that must occur to abstract it away and make your code look superficially clean to attempt safety you could still have with the same method in C? Surely you're not claiming that in unsafe code Rust is somehow safer than C, because that would be idiotic. And you're blaming the pilots and/or the training they received for the software being faulty? Sure the pilots could've handled it mistake free and no one would have died, but if the software had been written correctly then the pilots would not have factored into it. As to finding good programmers, I know of precious few. Sure, there are bugs in the Linux kernel, but they're found and fixed. How many bugs still haven't been found in Windows because it's closed source and can't be inspected by users. How many of those bugs are actively being exploited by malicious actors because they can only be discovered by Microsoft who doesn't even care anymore. These are rhetorical questions as no one can know except for Microsoft and the people using their bugs. Hopefully you're not going to claim they don't exist because you can't see them.
1
@d3line Doesn't really help people reading your comments to constantly quote things. For clarity's sake perhaps don't. As for the argument about safety. Are you saying that using Rust won't prevent all errors? Are you saying that there are common errors that it won't prevent? If you can admit both of those then I think we're done here because it can't prevent all errors and there are plenty of common errors that it won't prevent. If you can't admit that then you shouldn't be a teacher because you're indoctrinating students with magical thinking.
1
@d3line You're not making a whole lot of sense here. You don't support closed source, fine, I don't either, but to act as though no one else will, that's just stupid. If you're using a library you wrote yourself or one that someone wrote who is into open source, then yeah you'll see the potential failure points, but there will be multitudes who will be using closed source binary blobs and literally can not see into it without reverse engineering it. If you were claiming that Rust can't do that, and we both know you're not, then you'd be really idiotic. Instead you're obfuscating by alluding to one thing and saying the opposite, and don't tell me that's not intentional. With regards to low level code, you seem to be under the delusion that Rust will prevent all errors while still stating that it won't, or maybe it's just your unwillingness to acknowledge that you agree with me on this point, but at the low level some of us write code at, Rust offers no advantages that C doesn't also offer over and above assembly. Further, there are far more compilers and target system code generators for C infrastructure than for Rust. From the sounds of things, you seem to be confused about a lot. At least I hope it's confusion because the alternative is gaslighting. If you're writing code at a high level, as in not system level programming, then use Rust if you want, even if it's still not the right choice for any job thus far. It won't make your code better than anyone else's code, and if you're a bad programmer you won't compare favorably to a good programmer no matter what language you're using.
1
@d3line Is English your first language and what kind of grades did you get in any English classes you took?
1
@d3line This explains so much. No wonder you can't understand me and keep saying things weirdly, you don't have a full grasp on English. I was beginning to think you were attempting to gaslight me, but nope, you just don't understand.
1