4 ms·

> Maybe not yet, but it is heading in that direction; It's definitely getting more complex, but C++ has a huge lead haha. C++ is like a fractal in that you can

> Maybe not yet, but it is heading in that direction; It's definitely getting more complex, but C++ has a huge lead haha. C++ is like a fractal in that you can look at almost any feature closer and closer to reveal more and more complexity, and there are a lot of features... Here's a page on just one dark corner of the language: https://isocpp.org/wiki/faq/pointers-to-members https://isocpp.org/wiki/faq/pointers-to-members and it interacts in interesting ways with all the other corners (like virtual vs non-virtual inheritance) in fun and exciting ways... Also, there are far more ways to cause UB in C++. Rust has a big lead on formalizing what constitutes UB, and even those rules you only need to learn if you are using "unsafe", whilst in C++ you don't have that luxury.

tialaramex 2y ago
> Also, there are far more ways to cause UB in C++. As well as lots of Undefined Behaviour, C++ also has what its own experts call "False positives for the question is this a C++ program" the Ill-Formed No Diagnostic Required features, nothing like these exist in Rust, they're cases where you can write what appears to be C++ but actually although there are is no error or warning from the compiler your entire program has no meaning and might do absolutely anything from the outset. I've seen guesses that most or even all non-trivial C++ invokes IFNDR. So that's categorically worse than Undefined Behaviour. Finally, C++ has cases where the standard just chooses not to explain how something works because doing so would mean actually deciding and that's controversial so in fact all C++ where this matters also has no defined meaning and no way for you to discover what happens except to read the machine code emitted by your compiler, which entirely misses the point of a high level programming language. One of the things happening in Rust's stabilization process is solving those tough issues, for example Aria's "Strict Provenance experiment" is likely being stabilized, formally granting Rust a pointer provenance model, something C++ does not have and C23 had to fork into a separate technical document to study.
- googh 2y ago
  Most (if not all) of your posts here on HN boil down to "C/C++ bad, Rust good". I wonder what you are trying to achieve by this, but I assure you that this does not do Rust any favor other than giving the impression that the Rust community is obnoxious.
  - tialaramex 2y ago
    > Most (if not all) of your posts here on HN boil down to "C/C++ bad, Rust good". I haven't measured but it's easy to say categorically that it's not "all" unless somehow my posts about network protocols, aeroplanes, security and psychology among others fall into this vague category. And yes, like Ignaz Semmelweis I can see an obvious improvement to how my profession does what it does and it's infuriating that the response from many other practitioners is "No, I don't like change, therefore you're crazy for explaining why I should change" Ignaz Semmelweis died in an asylum. But on the other hand while Ignaz was correct and his proposals would have worked he couldn't explain why because germ theory was only confirmed after he died. Rust isn't in that situation, we know already exactly what the problems are with C++. So that means I can tell you not just that using C++ is a bad idea, but why it's a bad idea.
    - googh 2y ago
      Most HN users already use languages with GC which are more memory safe than Rust. People still use C++ either because they are maintaining existing code, or they work in domains where memory safety is not really necessary (games, HFT, ML). Apart from these, C++ is rarely used in the real world. > practitioners is "No, I don't like change, therefore you're crazy for explaining why I should change" Who exactly are you referring to here? Your co-workers? LLVM maintainers? or the Linux kernel developers? Please be more precise.
      - hyperbrainer 2y ago
        What is "more memory-safe than rust" supposed to mean? Rust is completely memory-safe.
        tialaramex 2y ago
        Only safe Rust can guarantee this, and only as a consequence of any unsafe Rust being correct. Most of the popular Garbage Collected languages of course also have a way to escape, in some cases via an "unsafe" keyword or magic unsafe package to a language where the same safety rules do not exist, in this sense the difference in Rust is that it's the same language. I'd actually say the more memory safe option would be a language like WUFFS where it pays a high price (generality) to deliver categorically better safety and performance. Most software could not be written in WUFFS but also most of the software which could be written in WUFFS isn't.

tialaramex 2y ago

> Also, there are far more ways to cause UB in C++. As well as lots of Undefined Behaviour, C++ also has what its own experts call "False positives for the question is this a C++ program" the Ill-Formed No Diagnostic Required features, nothing like these exist in Rust, they're cases where you can write what appears to be C++ but actually although there are is no error or warning from the compiler your entire program has no meaning and might do absolutely anything from the outset. I've seen guesses that most or even all non-trivial C++ invokes IFNDR. So that's categorically worse than Undefined Behaviour. Finally, C++ has cases where the standard just chooses not to explain how something works because doing so would mean actually deciding and that's controversial so in fact all C++ where this matters also has no defined meaning and no way for you to discover what happens except to read the machine code emitted by your compiler, which entirely misses the point of a high level programming language. One of the things happening in Rust's stabilization process is solving those tough issues, for example Aria's "Strict Provenance experiment" is likely being stabilized, formally granting Rust a pointer provenance model, something C++ does not have and C23 had to fork into a separate technical document to study.

googh 2y ago
Most (if not all) of your posts here on HN boil down to "C/C++ bad, Rust good". I wonder what you are trying to achieve by this, but I assure you that this does not do Rust any favor other than giving the impression that the Rust community is obnoxious.
- tialaramex 2y ago
  > Most (if not all) of your posts here on HN boil down to "C/C++ bad, Rust good". I haven't measured but it's easy to say categorically that it's not "all" unless somehow my posts about network protocols, aeroplanes, security and psychology among others fall into this vague category. And yes, like Ignaz Semmelweis I can see an obvious improvement to how my profession does what it does and it's infuriating that the response from many other practitioners is "No, I don't like change, therefore you're crazy for explaining why I should change" Ignaz Semmelweis died in an asylum. But on the other hand while Ignaz was correct and his proposals would have worked he couldn't explain why because germ theory was only confirmed after he died. Rust isn't in that situation, we know already exactly what the problems are with C++. So that means I can tell you not just that using C++ is a bad idea, but why it's a bad idea.
  - googh 2y ago
    Most HN users already use languages with GC which are more memory safe than Rust. People still use C++ either because they are maintaining existing code, or they work in domains where memory safety is not really necessary (games, HFT, ML). Apart from these, C++ is rarely used in the real world. > practitioners is "No, I don't like change, therefore you're crazy for explaining why I should change" Who exactly are you referring to here? Your co-workers? LLVM maintainers? or the Linux kernel developers? Please be more precise.
    - hyperbrainer 2y ago
      What is "more memory-safe than rust" supposed to mean? Rust is completely memory-safe.
      - tialaramex 2y ago
        Only safe Rust can guarantee this, and only as a consequence of any unsafe Rust being correct. Most of the popular Garbage Collected languages of course also have a way to escape, in some cases via an "unsafe" keyword or magic unsafe package to a language where the same safety rules do not exist, in this sense the difference in Rust is that it's the same language. I'd actually say the more memory safe option would be a language like WUFFS where it pays a high price (generality) to deliver categorically better safety and performance. Most software could not be written in WUFFS but also most of the software which could be written in WUFFS isn't.