and force me to implement my own typing system every time I need to upcast.
Basically, stick with C and leave C++ programmers alone. I haven't seen a less useful article about C++ in a long time, and as an HN reader, that's really saying something.
One thing I've noticed about a lot of these "strict C" developers is that quite often they actually refuse to learn C++. One of the most common complaints of C developers regarding C++ is "it does things behind the scenes/performs magic", often with regards to operator overloading. When they refuse to actually look at the implementation (y'know you can check if an operator has been overloaded) AND they refuse to acknowledge that a huge chunk of "pure C" does HEAPS of magic behind the scenes (that the developer has no idea about) unless they've actually studied the spec in detail. Malloc and memory allocation methods are at least 10k+ lines of code for instance.
> y'know you can check if an operator has been overloaded
And there lies the problem with C++: to be sure, you have to check. C++ code can't be taken at face value -- the most innocuous-looking code could be a ticking bomb.
Trust me, I know more C++ than most or all of my peers (working two jobs simultaneously), and I know a million ways that C++ features suck. Also standard library and containers. If you want I'll point out the ways in which std::deque, and even std::map, std::unordered_map, even std::vector (!) suck. IMO, just don't do it.
The standard library implements really do suck (in some cases), but this should be separated from C++ (the language). Even the standard splits the language grammar from the standard library cleanly.
You can't really separate the two, firstly because some parts of the standard library interact directly with the language's syntax (e.g. <initializer_list>), and secondly because the language standard dictates things about the behavior of the standard library that limit implementation options.
For example, the standard says that adding elements to an <unordered_map> is not allowed to invalidate references to keys or elements within the map. That makes it impossible for any standards-compliant C++ implementation to use a high-performance implementation in which keys and elements are stored contiguously in a flat array.
> Malloc and memory allocation methods are at least 10k+ lines of code for instance.
Only the really big ones, e.g. here is Emscripten's allocator that focuses on small binary size and is implemented in about 1.5 kloc (ignoring comments and whitespace it's actually under 1 kloc), and that allocator is perfectly fine for most use cases (especially C code bases which typically don't have a high allocation frequency):
Right tool for the job etc... big general-purpose allocators like jemalloc or mimalloc are usually a bandaid to somewhat salvage a failed memory management strategy.
A lot of us are too busy solving problems. Learning about the latest language features, which we often won't be able to use anyway due to the trouble of moving a large dev environment to a newer standard, feels like academic masturbation.
C++ folks are very much into their language, and can't seem to understand that most folks don't want to dedicate significant amounts of mental resources purely to language details.
Moving to new C++ is a non event, change the compiler / flags and done. Using the new features requires some learning but not a big deal since you can figure out what you need from a summary and learn what is useful for your problem.
The problems of the code I'm writing far exceeds the complexity of the language. Your complaint about complexity fall flat to me, unless you are working on a trivial program you need to deal with things far more complex than any language.
If we accept the maximum that "any sufficiently advanced technology is indistinguishable from magic", then c++ is indeed magic. It's so advanced that one of the worlds foremost experts in the language(herb sutter) has determined that the language is too complex and we need a whole new language(confront) which is simpler and can be converted to c++.
C++ is actually obscenely complex, I don't deny that. Just mastering object lifetime rules is crazy difficult due to all the edge cases, but it comes with the territory.
In the sense that C++ is complicated because it's C++, which is complicated? That's just a tautology. If you mean "the territory" in some other sense there's no reason to believe this.
This isn't because it requires RTTI, but dynamic_cast is also a typical code smell.
Orthodox C++ isn't generally against new C++ features, it only advices to wait about 5 years (or at least one C++ version) for stabilization and to apply some common sense before adopting them.
The notes about not using RTTI, exceptions and stdlib features that allocate under the hood are all justified by painful experience with those things in the context of game development.
In general, the restrictions outlined in the post make a lot of sense when considering that Branimir (of BGFX fame: https://github.com/bkaradzic/bgfx) is coming out of the game dev hemisphere, and from that PoV none of the restrictions are controversial - on the contrary, it would be highly controversial to suggest going all in on Modern C++ features ;)
LLVM uses a hand-rolled version of RTTI for the best performance (the parent constructor accepts its runtime type as an enum), and it seemed that the maintenance costs for it aren't that high. (See https://llvm.org/docs/HowToSetUpLLVMStyleRTTI.html)
Or if you're even lazier, you can easily make a more automatic RTTI system with some templates / macros that works much faster than dynamic_cast! (See https://github.com/royvandam/rtti)
Yep, the article is a old one and not particularly well written. As somebody who has been using C++ from the early 90s and not particularly a fan of (all of) "Modern C++", there is not much information here.
If you're in a market that requires using C++, many of these decisions are made for you by the platform above you, and you're screwed. Turn on RTTI, build a fort to deflect the random exceptions they'll throw at you, and may the gods allow you to recoup your R&D before some well-intentioned yokel in some media or game vertical changes everything and requires you to change everything.
On the other hand, if you control your own destiny and care about velocity and code quality, many of these choices eventually become self-evident.
If you are messing around with the latest and greatest esoteric C++ stuff in 2026, bless you, you beautiful nerd. But it may be time to start evaluating where you are in life, and how you got here. (And if you're on a C++ committee, I revoke those blessings.)
For those who remain: if you have a C++ code base yet somehow have enough time and energy to write opinionated blog posts, it's really hard to imagine why you think you'd have a better take on this than Google.
> build a fort to deflect the random exceptions they'll throw at you
Sounds like you hate exceptions, right? In which case why do you handle them at all? Just leave them all unhandled and suddenly every exception is a crash. Which is really no different from someone choosing to terminate. Which you have to worry about even without exceptions.
> if you have a C++ code base yet somehow have enough time and energy to write opinionated blog posts, it's really hard to imagine why you think you'd have a better take on this than Google.
"Given that Google's existing code is not exception-tolerant [...] Our advice against using exceptions is not predicated on philosophical or moral grounds, but practical ones. [...] Things would probably be different if we had to do it all over again from scratch."
> Which is really no different from someone choosing to terminate.
If you std::abort(), you'll get a useful stack trace in the core dump. If you crash from an unhandled exception, you don't. That's a pretty huge difference and is one of the reasons exceptions suck.
I've developed a style that I legitimately call Heterodox C++ (mainly due to the popularity of Orthodox C++), it is effectively a purely functional & metaprogramming heavy style of C++. Quite the opposite of this, not everyones cup of tea, and it won't fit into every codebase but it is incredibly powerful. The template metaprogramming C++ offers is the most powerful of any imperative language, and (subjective opinion) is second only to Lisp, but few people make use of it. With some of C++26 features you can almost even replicate most of Rusts safety features in pure C++ (via function tagging + reflection)
> The template metaprogramming C++ offers is the most powerful of any imperative language
I'm curious what languages you're comparing to here. Feels like it's only slightly more expressive than pure generics, but I admittedly haven't done much template metaprogramming myself. How does it compare to, say, Zig's comptime?
C++ template metaprogramming is in some ways more powerful than Common Lisp macros, because it works at the type level: you can generate new types and dispatch into separate implementations by type. In contrast, Common Lisp type declarations are not available at macro expansion time unless you implement a full source-to-source translator in macros.
The point on exceptions I think is also misleading. Compilers typically make throwing an exception the expensive part, and the happy path inexpensive (not more expensive than a branch checking for errors, which should be the baseline for comparison, not an implementation with zero error checking.) So to say that they are "expensive" doesn't really make a useful argument.
And there are more things that could be done in this camp, like proposing a set of compiler flags, and a linter to enforce the subset you are subscribing to. Unfortunately the post offers none of that.
> <deque>: Needs a major performance overhaul", acknowledging that the standard's mandated block size is too small and the design needs to be rebuilt at the next ABI break
Except of course the standard does not mandate a block size. That's purely msvc picking a wrong block size and being stuck with it.
stringstreams and fmt are a godsend compared to printf/scanf, which have historically led to most memory bugs in the first place!
Printf/scanf are implemented as variadic functions without type checking and rely on the compiler to perform its own internal metaprogramming to inspect and warn about format mismatches.
Anyone advocating the use of the old cstdio as a primary design decision about which C++ language features to use is not serious.
No exceptions or RTTI make sense in an embedded system that needs to ensure determinism, but are arbitrary and unnecessarily hobbling for high-level systems and application programming. How do you do runtime method dispatch without creating vtables and RTTI from first principles? How do you propagate a runtime error deep from the bowels of a component all the way to some top-level event loop? The "orthodox" approach would be a mess of integer return codes with associated enums (and none of that enum class nonsense!). No Thanks. It's clear the author has no idea what he's talking about.
I've been doing embedded systems in C++ since rocks were young, and this is a great summary of what to avoid.
I would sure love a good coroutine runtime, and first-class support for defer. You can do these manually, but language/toolchain/debugger support is nice to have.
(Pragmatically, I will be retired by the time they would be useful)
It is still a bit amazing to me that it was significantly easier to do coroutines in Sigma 5 assembly and likely most any assembly than in C or C++. Two languages supposedly close to the machine.
I have seen a pure C/C++ implementation of coroutines (it used setjmp/longjmp, and memcpy to copy stacks in and out of the native arena). Not the most portable of constructions, but it worked absurdly well.
Being able to write "async" code essentially in-line is a superpower.
In embedded, I like wrapping a class around a set of registers. Nobody else gets to write to that piece of hardware except that class.
As far as avoiding things... avoid basically everything you don't need. Don't add language features that don't actually help you, just because they're there. Keep the subset you use small. But pick that subset to match your problem well, rather than out of dogma.
Nothing surprising here. People who view C++ as just a better C always outnumbered those who view it as another language.
That's exactly how democratic governments make their decisions… you might think it's stupid, and you'd be right, but that's democracy. It's the majority that counts, not what's right. At least you can have a little fun with their arguments, they're pretty inventive you know.
typedef is a little bit of a hassle but you can do it, even in a very strict mechanical way if writing plain C. But it's a hassle.
And namespaces suck too, so much noise for little gain. You know what, a big part of programming is naming. You just have to come up with good names. Namespaces don't magically make names better, if anything, they make them worse. And they add a lot of syntax noise.
Sometimes I actually want objects that are transparent, fully public, and 'struct' is perfect for that. But if I then go and put methods into those structs, does that make me unorthodox?
That's simply because we live in a world where UFCS is restricted to niche languages and we're stuck with "methods" instead. At least Rust/Kotlin/Swift support type extensions (with a thousand papercuts, i. e orphan rules)
No, it is perfectly valid if your design demands it.
For example, the Windows MFC framework had classes whose data members were all public. Some of reasons were;
1) MFC was a wrapper over lower-level Windows API/structs and therefore unnecessary getter/setter methods were avoided. The C++ class could be a simple wrapper over the underlying C-style POD.
2) The framework classes were supposed to be used by implementation inheritance to build one's skeleton application which led to tight coupling between base and derived classes. Hence the designers decided to make all members public which meant that users were not limited by any omissions in the basic design.
I actually used these ideas in a project where i implemented a C++ api over C state machines for H.323 protocols.
This is gonna be a long critique, I'll try to keep it concise.
> C-like C++ is good start, if code doesn’t require more complexity don’t add unnecessary C++ complexities.
C is almost obsolete nowadays. Not to mention that C++ is effectively a strict superset of C (nearly 99% of the C standard is in C++) and the few features that aren't are included as compiler extensions (VLA, restrict keyword, nested functions). There are a handful of C features that aren't in C++, and for very good reason (most of them suck).
When was the last time you ran into a C library that a pure C++ compiler couldn't compile? Only if someone decided to spam the new keyword all over the codebase (or something similar).
> In general case code should be readable to anyone who is familiar with C language.
Most C++ already is? Even very template heavy C++.
> Don’t do this, the end of “design rationale” in Orthodox C++ should be immedately after “Quite simple, and it is usable. EOF”.
A lot of the methods in that document are necessary to make C++ shine, especially template metaprogramming.
> Don’t use exceptions.
Optional but irrelevant.
> Don’t use RTTI.
.. Why? Reimplementing RTTI in C will give you almost the same overhead.
> Don’t use C++ runtime wrapper for C runtime includes (<cstdio>, <cmath>, etc.), use C runtime instead (<stdio.h>, <math.h>, etc.)
.. Why? Those wrappers all include the "raw C runtime" under the hood (literally they do #include <stdio.h|xx>. Near 0 compiletime overhead?
> Don’t use stream (<iostream>, <stringstream>, etc.), use printf style functions instead.
This is a design decision.
> Don’t use metaprogramming excessively for academic masturbation. Use it in moderation, only where necessary, and where it reduces code complexity.
There are many programs that are _impossible_ to write in a finite time without metaprogramming. How will you (with zero runtime overhead) dispatch a function with a variable arity of random types to a handler that requires exactly that type of function? Arbitrarily? In C++ it's possible, in C it isn't.
But, one of the annoying habits of WG21 (the C++ committee) is sending stuff to WG14 (the C committee) to have them make it part of their language rather than accept that it's a C++ problem. Even the stupid type qualifiers are actually C++'s fault, K&R doesn't have this abomination but the pre-standard C++ did so too bad now it's in C89.
> When was the last time you ran into a C library that a pure C++ compiler couldn't compile? Only if someone decided to spam the new keyword all over the codebase (or something similar).
In C, you can use goto to jump over a variable declaration, and you can't in C++. I understand why this is, but it's the thing I see the most often that makes C code not compile as C++.
Man, all of the confusion and gnashing of teeth in the C++ world really makes me grateful for my job. Smaller company, I solo develop a central module on the product stack, and I was able to evaluate languages for the project.
Nim became the obvious choice, and I wasn't a fanboy before. Simple semantics, in a very functional style oriented around data's value. References and identity have to be trapdoored. Everything is single-owner unique lifetimes by default, no annotations or best-practices required. You end up writing extraordinarily functional/procedural code that produces very fast and memory-safe binaries, it fits right into C++'s niche.
The only objection I could steel man was that the standard library and most packages are composed of relatively pure functions that return new values, so allocations are happening there. But when types as complex as data frames can be semantically used as just values, and you know they have scoped lifetimes by default, the benefits are obvious.
With all of C++'s insanely specific, subtle, implicit, compiler- and platform-dependent behaviors, I've often wondered when the industry will finally consider its dominance an artifact of first-mover inertia and simply move on. There are vastly better ways to do all of the things it does, while easily exposing levers for the the things it's considered to do exceptionally well.
You can take
from my cold dead hands.Also, you can fight me if you want to take
and force me to implement my own typing system every time I need to upcast.Basically, stick with C and leave C++ programmers alone. I haven't seen a less useful article about C++ in a long time, and as an HN reader, that's really saying something.
One thing I've noticed about a lot of these "strict C" developers is that quite often they actually refuse to learn C++. One of the most common complaints of C developers regarding C++ is "it does things behind the scenes/performs magic", often with regards to operator overloading. When they refuse to actually look at the implementation (y'know you can check if an operator has been overloaded) AND they refuse to acknowledge that a huge chunk of "pure C" does HEAPS of magic behind the scenes (that the developer has no idea about) unless they've actually studied the spec in detail. Malloc and memory allocation methods are at least 10k+ lines of code for instance.
> y'know you can check if an operator has been overloaded
And there lies the problem with C++: to be sure, you have to check. C++ code can't be taken at face value -- the most innocuous-looking code could be a ticking bomb.
Just like any C function without looking into the translation unit, don't say you blindly believe on the function name.
But isn't this a problem with all code? Looking at a Rust function signature how can you be sure that it does what it says it does? Or python?
> C++ code can't be taken at face value -- the most innocuous-looking code could be a ticking bomb.
You can't take C code at face value either. The name of a method or type doesn't tell you what it does. It could longjmp for all you know.
Trust me, I know more C++ than most or all of my peers (working two jobs simultaneously), and I know a million ways that C++ features suck. Also standard library and containers. If you want I'll point out the ways in which std::deque, and even std::map, std::unordered_map, even std::vector (!) suck. IMO, just don't do it.
The standard library implements really do suck (in some cases), but this should be separated from C++ (the language). Even the standard splits the language grammar from the standard library cleanly.
You can't really separate the two, firstly because some parts of the standard library interact directly with the language's syntax (e.g. <initializer_list>), and secondly because the language standard dictates things about the behavior of the standard library that limit implementation options.
For example, the standard says that adding elements to an <unordered_map> is not allowed to invalidate references to keys or elements within the map. That makes it impossible for any standards-compliant C++ implementation to use a high-performance implementation in which keys and elements are stored contiguously in a flat array.
Your map example only concerns the standard library, not the language.
Which sucks... unless you really need reference stability.
> Malloc and memory allocation methods are at least 10k+ lines of code for instance.
Only the really big ones, e.g. here is Emscripten's allocator that focuses on small binary size and is implemented in about 1.5 kloc (ignoring comments and whitespace it's actually under 1 kloc), and that allocator is perfectly fine for most use cases (especially C code bases which typically don't have a high allocation frequency):
https://github.com/emscripten-core/emscripten/blob/main/syst...
...and Seb Aaltonen's offset allocator (used for allocating GPU buffers in his Vulkan API wrapper) has under 500 lines of code:
https://github.com/sebbbi/OffsetAllocator/blob/main/offsetAl...
Right tool for the job etc... big general-purpose allocators like jemalloc or mimalloc are usually a bandaid to somewhat salvage a failed memory management strategy.
A lot of us are too busy solving problems. Learning about the latest language features, which we often won't be able to use anyway due to the trouble of moving a large dev environment to a newer standard, feels like academic masturbation.
C++ folks are very much into their language, and can't seem to understand that most folks don't want to dedicate significant amounts of mental resources purely to language details.
Moving to new C++ is a non event, change the compiler / flags and done. Using the new features requires some learning but not a big deal since you can figure out what you need from a summary and learn what is useful for your problem.
The problems of the code I'm writing far exceeds the complexity of the language. Your complaint about complexity fall flat to me, unless you are working on a trivial program you need to deal with things far more complex than any language.
Like implementing the compilers used by C devs.
If we accept the maximum that "any sufficiently advanced technology is indistinguishable from magic", then c++ is indeed magic. It's so advanced that one of the worlds foremost experts in the language(herb sutter) has determined that the language is too complex and we need a whole new language(confront) which is simpler and can be converted to c++.
C++ is actually obscenely complex, I don't deny that. Just mastering object lifetime rules is crazy difficult due to all the edge cases, but it comes with the territory.
> it comes with the territory
In the sense that C++ is complicated because it's C++, which is complicated? That's just a tautology. If you mean "the territory" in some other sense there's no reason to believe this.
> for (auto const & ess : esses) {
This is allowed by Orthodox C++
> dynamic_cast<Derived> (base_ptr)
This isn't because it requires RTTI, but dynamic_cast is also a typical code smell.
Orthodox C++ isn't generally against new C++ features, it only advices to wait about 5 years (or at least one C++ version) for stabilization and to apply some common sense before adopting them.
The notes about not using RTTI, exceptions and stdlib features that allocate under the hood are all justified by painful experience with those things in the context of game development.
In general, the restrictions outlined in the post make a lot of sense when considering that Branimir (of BGFX fame: https://github.com/bkaradzic/bgfx) is coming out of the game dev hemisphere, and from that PoV none of the restrictions are controversial - on the contrary, it would be highly controversial to suggest going all in on Modern C++ features ;)
LLVM uses a hand-rolled version of RTTI for the best performance (the parent constructor accepts its runtime type as an enum), and it seemed that the maintenance costs for it aren't that high. (See https://llvm.org/docs/HowToSetUpLLVMStyleRTTI.html)
Or if you're even lazier, you can easily make a more automatic RTTI system with some templates / macros that works much faster than dynamic_cast! (See https://github.com/royvandam/rtti)
Implementing STL iterators are a bloody PITA
You don't need iterators for the range-based for loop, just a .begin() and .end() method which returns a raw pointer is enough.
E.g.:
https://github.com/floooh/oryol/blob/eb08cffe1b1cb6b05ed14ec...
(don't use that project though because it's been archived, I've switched back to plain old C in the meantime)
Submitted a fair few times previously. HN's search turned up these submissions with some additional discussion:
https://news.ycombinator.com/item?id=40445536 (2 years ago, 63 points, 66 comments)
https://news.ycombinator.com/item?id=25554018 (5 years ago, 70 points, 102 comments)
https://news.ycombinator.com/item?id=13751244 (9 years ago, 29 points, 14 comments)
Looks like the page was moved from a GitHub gist to a github.io page in October of last year.
Yep, the article is a old one and not particularly well written. As somebody who has been using C++ from the early 90s and not particularly a fan of (all of) "Modern C++", there is not much information here.
If you're in a market that requires using C++, many of these decisions are made for you by the platform above you, and you're screwed. Turn on RTTI, build a fort to deflect the random exceptions they'll throw at you, and may the gods allow you to recoup your R&D before some well-intentioned yokel in some media or game vertical changes everything and requires you to change everything.
On the other hand, if you control your own destiny and care about velocity and code quality, many of these choices eventually become self-evident.
If you are messing around with the latest and greatest esoteric C++ stuff in 2026, bless you, you beautiful nerd. But it may be time to start evaluating where you are in life, and how you got here. (And if you're on a C++ committee, I revoke those blessings.)
For those who remain: if you have a C++ code base yet somehow have enough time and energy to write opinionated blog posts, it's really hard to imagine why you think you'd have a better take on this than Google.
https://google.github.io/styleguide/cppguide.html
> build a fort to deflect the random exceptions they'll throw at you
Sounds like you hate exceptions, right? In which case why do you handle them at all? Just leave them all unhandled and suddenly every exception is a crash. Which is really no different from someone choosing to terminate. Which you have to worry about even without exceptions.
> if you have a C++ code base yet somehow have enough time and energy to write opinionated blog posts, it's really hard to imagine why you think you'd have a better take on this than Google.
"Given that Google's existing code is not exception-tolerant [...] Our advice against using exceptions is not predicated on philosophical or moral grounds, but practical ones. [...] Things would probably be different if we had to do it all over again from scratch."
> Which is really no different from someone choosing to terminate.
If you std::abort(), you'll get a useful stack trace in the core dump. If you crash from an unhandled exception, you don't. That's a pretty huge difference and is one of the reasons exceptions suck.
I interview C++ developers often, and here in 2026 it seems pretty much everyone is using modern (C++20 and up) language versions.
Maybe the tooling finally caught up.
I've developed a style that I legitimately call Heterodox C++ (mainly due to the popularity of Orthodox C++), it is effectively a purely functional & metaprogramming heavy style of C++. Quite the opposite of this, not everyones cup of tea, and it won't fit into every codebase but it is incredibly powerful. The template metaprogramming C++ offers is the most powerful of any imperative language, and (subjective opinion) is second only to Lisp, but few people make use of it. With some of C++26 features you can almost even replicate most of Rusts safety features in pure C++ (via function tagging + reflection)
> The template metaprogramming C++ offers is the most powerful of any imperative language
I'm curious what languages you're comparing to here. Feels like it's only slightly more expressive than pure generics, but I admittedly haven't done much template metaprogramming myself. How does it compare to, say, Zig's comptime?
The problem with metaprogramming-heavy C++ codebases is always compilation times and obtuse error messages...
Template metaprogramming is sometimes very useful to get around C++'s language restrictions, but I tend to use it sparingly.
> obtuse error messages
With concepts and constexpr-if and consteval it's increasingly less of a problem
have you written about this anywhere, or hosting any examples?
You might find Functional Programming in C++ by Ivan Cukic relevant.
Not yet, I might one day.
C++ template metaprogramming is in some ways more powerful than Common Lisp macros, because it works at the type level: you can generate new types and dispatch into separate implementations by type. In contrast, Common Lisp type declarations are not available at macro expansion time unless you implement a full source-to-source translator in macros.
My codebase uses a fairly dumbed down version of C++, but I would have liked to see more depth in this post. As it is, it is not very useful.
There are many more things to avoid than just iostream. HFT university has a good recap: https://hftuniversity.com/post/the-c-standard-library-has-be...
The point on exceptions I think is also misleading. Compilers typically make throwing an exception the expensive part, and the happy path inexpensive (not more expensive than a branch checking for errors, which should be the baseline for comparison, not an implementation with zero error checking.) So to say that they are "expensive" doesn't really make a useful argument.
And there are more things that could be done in this camp, like proposing a set of compiler flags, and a linter to enforce the subset you are subscribing to. Unfortunately the post offers none of that.
[from the linked article]
> <deque>: Needs a major performance overhaul", acknowledging that the standard's mandated block size is too small and the design needs to be rebuilt at the next ABI break
Except of course the standard does not mandate a block size. That's purely msvc picking a wrong block size and being stuck with it.
The rant about lists is also nonsense.
Such a missed opportunity to call it C <orthodox cross emoji>
Was thinking "Hesitance C" could work
that's one way to combine three plus signs!
stringstreams and fmt are a godsend compared to printf/scanf, which have historically led to most memory bugs in the first place!
Printf/scanf are implemented as variadic functions without type checking and rely on the compiler to perform its own internal metaprogramming to inspect and warn about format mismatches.
Anyone advocating the use of the old cstdio as a primary design decision about which C++ language features to use is not serious.
No exceptions or RTTI make sense in an embedded system that needs to ensure determinism, but are arbitrary and unnecessarily hobbling for high-level systems and application programming. How do you do runtime method dispatch without creating vtables and RTTI from first principles? How do you propagate a runtime error deep from the bowels of a component all the way to some top-level event loop? The "orthodox" approach would be a mess of integer return codes with associated enums (and none of that enum class nonsense!). No Thanks. It's clear the author has no idea what he's talking about.
I've been doing embedded systems in C++ since rocks were young, and this is a great summary of what to avoid.
I would sure love a good coroutine runtime, and first-class support for defer. You can do these manually, but language/toolchain/debugger support is nice to have.
(Pragmatically, I will be retired by the time they would be useful)
It is still a bit amazing to me that it was significantly easier to do coroutines in Sigma 5 assembly and likely most any assembly than in C or C++. Two languages supposedly close to the machine.
I have seen a pure C/C++ implementation of coroutines (it used setjmp/longjmp, and memcpy to copy stacks in and out of the native arena). Not the most portable of constructions, but it worked absurdly well.
Being able to write "async" code essentially in-line is a superpower.
A defer is just a dozen lines of code nowadays, if you really need it, but in most cases you don't if you're doing RAII
In embedded, I like wrapping a class around a set of registers. Nobody else gets to write to that piece of hardware except that class.
As far as avoiding things... avoid basically everything you don't need. Don't add language features that don't actually help you, just because they're there. Keep the subset you use small. But pick that subset to match your problem well, rather than out of dogma.
Nothing surprising here. People who view C++ as just a better C always outnumbered those who view it as another language.
That's exactly how democratic governments make their decisions… you might think it's stupid, and you'd be right, but that's democracy. It's the majority that counts, not what's right. At least you can have a little fun with their arguments, they're pretty inventive you know.
Somewhere within c++ there is a subset of c++ that is a great language.
The problem of course is that no one agrees on which subset that is.
At this point, if you want better C, just use Zig.
Except it is years away of being 1.0, with the industry support of C++.
Orthodox C++, to me, is C plus the one good feature of C++: you don't have to type struct all the time.
How about one of the C unorthodoxies that use typedef everywhere? (Namespaces seem suitable, too.)
typedef is a little bit of a hassle but you can do it, even in a very strict mechanical way if writing plain C. But it's a hassle.
And namespaces suck too, so much noise for little gain. You know what, a big part of programming is naming. You just have to come up with good names. Namespaces don't magically make names better, if anything, they make them worse. And they add a lot of syntax noise.
The criticisms of STL and allocation are fair, though move constructors improved the shallow vs deep copy problem on resize.
Smart pointers are good. People were doing them outside the standard in the late 90s.
Lambdas are a good feature.
Sometimes I actually want objects that are transparent, fully public, and 'struct' is perfect for that. But if I then go and put methods into those structs, does that make me unorthodox?
That's simply because we live in a world where UFCS is restricted to niche languages and we're stuck with "methods" instead. At least Rust/Kotlin/Swift support type extensions (with a thousand papercuts, i. e orphan rules)
No, it is perfectly valid if your design demands it.
For example, the Windows MFC framework had classes whose data members were all public. Some of reasons were;
1) MFC was a wrapper over lower-level Windows API/structs and therefore unnecessary getter/setter methods were avoided. The C++ class could be a simple wrapper over the underlying C-style POD.
2) The framework classes were supposed to be used by implementation inheritance to build one's skeleton application which led to tight coupling between base and derived classes. Hence the designers decided to make all members public which meant that users were not limited by any omissions in the basic design.
I actually used these ideas in a project where i implemented a C++ api over C state machines for H.323 protocols.
Holly bananas, that Boost Design Rationale post is, what's the word I'm looking for, intense.
Don't follow dogma.
This is gonna be a long critique, I'll try to keep it concise.
> C-like C++ is good start, if code doesn’t require more complexity don’t add unnecessary C++ complexities.
C is almost obsolete nowadays. Not to mention that C++ is effectively a strict superset of C (nearly 99% of the C standard is in C++) and the few features that aren't are included as compiler extensions (VLA, restrict keyword, nested functions). There are a handful of C features that aren't in C++, and for very good reason (most of them suck). When was the last time you ran into a C library that a pure C++ compiler couldn't compile? Only if someone decided to spam the new keyword all over the codebase (or something similar).
> In general case code should be readable to anyone who is familiar with C language.
Most C++ already is? Even very template heavy C++.
> Don’t do this, the end of “design rationale” in Orthodox C++ should be immedately after “Quite simple, and it is usable. EOF”.
A lot of the methods in that document are necessary to make C++ shine, especially template metaprogramming.
> Don’t use exceptions.
Optional but irrelevant.
> Don’t use RTTI.
.. Why? Reimplementing RTTI in C will give you almost the same overhead.
> Don’t use C++ runtime wrapper for C runtime includes (<cstdio>, <cmath>, etc.), use C runtime instead (<stdio.h>, <math.h>, etc.)
.. Why? Those wrappers all include the "raw C runtime" under the hood (literally they do #include <stdio.h|xx>. Near 0 compiletime overhead?
> Don’t use stream (<iostream>, <stringstream>, etc.), use printf style functions instead.
This is a design decision.
> Don’t use metaprogramming excessively for academic masturbation. Use it in moderation, only where necessary, and where it reduces code complexity.
There are many programs that are _impossible_ to write in a finite time without metaprogramming. How will you (with zero runtime overhead) dispatch a function with a variable arity of random types to a handler that requires exactly that type of function? Arbitrarily? In C++ it's possible, in C it isn't.
> nearly 99% of the C standard is in C++
But, one of the annoying habits of WG21 (the C++ committee) is sending stuff to WG14 (the C committee) to have them make it part of their language rather than accept that it's a C++ problem. Even the stupid type qualifiers are actually C++'s fault, K&R doesn't have this abomination but the pre-standard C++ did so too bad now it's in C89.
> When was the last time you ran into a C library that a pure C++ compiler couldn't compile? Only if someone decided to spam the new keyword all over the codebase (or something similar).
In C, you can use goto to jump over a variable declaration, and you can't in C++. I understand why this is, but it's the thing I see the most often that makes C code not compile as C++.
For me it was already obsolete in 1992, when I was given a copy of Turbo C++ 1.0 for MS-DOS.
It was the next step from Turbo Pascal in terms of safety, with added benefits from cross platform.
Nowadays all C compilers that matter are written in C++ anyway.
> > Don’t do this, the end of “design rationale” in Orthodox C++ should be immedately after “Quite simple, and it is usable. EOF”.
> A lot of the methods in that document are necessary to make C++ shine, especially template metaprogramming.
So? Is your goal to make C++ shine, or is it to produce useful, understandable code? My goal is good code, not being a showoff.
I wish the site listed compiler flags for the most popular compilers.
And here I was thinking this was going to be about a schism from Holy C [1]
[1] https://en.wikipedia.org/wiki/TempleOS#HolyC
Man, all of the confusion and gnashing of teeth in the C++ world really makes me grateful for my job. Smaller company, I solo develop a central module on the product stack, and I was able to evaluate languages for the project.
Nim became the obvious choice, and I wasn't a fanboy before. Simple semantics, in a very functional style oriented around data's value. References and identity have to be trapdoored. Everything is single-owner unique lifetimes by default, no annotations or best-practices required. You end up writing extraordinarily functional/procedural code that produces very fast and memory-safe binaries, it fits right into C++'s niche.
The only objection I could steel man was that the standard library and most packages are composed of relatively pure functions that return new values, so allocations are happening there. But when types as complex as data frames can be semantically used as just values, and you know they have scoped lifetimes by default, the benefits are obvious.
With all of C++'s insanely specific, subtle, implicit, compiler- and platform-dependent behaviors, I've often wondered when the industry will finally consider its dominance an artifact of first-mover inertia and simply move on. There are vastly better ways to do all of the things it does, while easily exposing levers for the the things it's considered to do exceptionally well.