The answer is in the sentences right before the one you quoted:

Relatedly, it's more explicit than almost any other language. If a line of code doesn't look like a function call, it's not calling anything. There is no hidden control flow.

Consider the use-cases for C: operating system kernels, hard real-time software, low-level libraries, databases, embedded software. What is a common desire among these? Predictable low-latency and high throughput.

It's much easier to achieve these features if your language does not allow "magic." Implicit allocations, RAII, exceptions, overloaded operators; these are all examples of features which allow a library-writer to inject hidden control flow into your code. This can make it very difficult to analyze why code runs slowly or with unexpected random pauses, not to mention making it much harder to step through in a debugger.

And whether something is called or run inline is always decided by the compiler. Modern C doesn't promise you any relation between the way you break down your functions on your code and the actual function calls on the assembly it generates.

So, I keep seeing people complaining about overloading; always with the same reasons; that are patently not valid unless there's some implicit assumption they keep not stating. What is that assumption that breaks the equivalence between user-defined functions and operators?

The implicit part is the question of whether an operator is built-in or overloaded. In C, every operator is built-in, so you can look at a block of code and see that there are NO function calls in it. With something like C++, you must treat every operator like a function call.

With C, if I write:

    a += b;

Furthermore, this may be the last line within a particular scope. With C I know that nothing else will happen, and that the control flow depends only on the surrounding scope. With C++, I don't know this! There may have been many objects created within this scope and now their destructors are firing and potentially very large trees of objects are being cleaned up and deallocated, and even slow IO operations running.

All programming requires people to follow reasonable conventions. In C++ if you make a dereference operator with non-constant time, or an equality operator which doesn't follow equality semantics, the programmer messed up. It's like giving a function a misleading name, like `doThis()` and it doesn't.

Note that Java is filled with these kinds of conventions, such as overloading `equals`. How can you be certain it actually obeys equality semantics? You have to trust the programmer.

In C, if I'm searching for how a certain thing may be called from a given function, I only have to look for /\w\(/ and don't have to ever think about anything else.

Honestly, operator overloading isn't really that bad (especially if an IDE can highlight which ones are), but it's still a thing that can affect how one has to go about reading code that might not even use it.

a = b + c;

Is the addition done by itself, so it costs 1 clock cycle? Is it merged into some complex operation so the net cost is less than 1 cycle? Is it completely optimized away at compile time, so it's infinitely faster?

Does the addition trigger some trap, that will run some distant code?

Is the addition by itself? Or are there store and load instructions that can stall for way more than 1000 cycles?

I doubt you can answer any of those questions. All you and everybody else keep repeating is you can micro-optimize C better because that line, that you expect to take something from 0 to 2000 cycles is certain to not do a call and return pair, that takes less than 10 cycles. All while the alternative is almost certain to do the exact same, but you would need to check it up.

Honestly, that argument doesn't make sense; and I keep understanding it as people complaining that they want to micro-optimize a program, but don't know if it's operating on native integers or 10-dimensional hypermatrices.

At the same time, every single person that is good at micro-optimizations look at the compiled binary as a first step, because C is a high-level language that has little relation to the code the compiler actually creates.

For a long time I did just shrug it away and file those complains as "those people don't even know the language they are using". But its universality forces me to consider that there is a reason for complaining, and maybe it's worthwhile to understand. Now, given that this is all the answer I get, it seems quite likely that even the ones complaining don't consciously know what the problem is... But one thing is certain here, the people repeating that execution time is well known didn't actually practice micro-optimizations based on that fact.

Those are generic register allocation/instruction selection questions, which are always gonna be *added* complexity to any and all operations everywhere. So there's still benefit in cutting down the complexity elsewhere.

> Is the addition by itself? Or are there store and load instructions that can stall for way more than 1000 cycles?

..those are questions about the loads & stores, not addition. On embedded, afaik loads & stores will be significantly closer in latency to arith too.

> At the same time, every single person that is good at micro-optimizations look at the compiled binary as a first step, because C is a high-level language that has little relation to the code the compiler actually creates.

Yes, but being able to have good intuition is still quite important, because one can think & read code much faster than compile & read assembly.

> the people repeating that execution time is well known didn't actually practice micro-optimizations based on that fact.

The question of operator overloading is mostly about reading code, not writing it. And it doesn't have to be micro-optimization either, any level of optimization will be affected by a call happening where you don't expect one (probably most importantly the kind where you scan over a piece of code to figure out if it does anything suspiciously bad (i.e. O(n^2) or excessive allocations or whatever thing may be expensive in the codebase in question) but it isn't worth the effort diving into assembly or figuring out how to get representative data for profiling the specific thing).

Or you could just be exploring a new codebase and wanting to track down where something happens, where it'd be beneficial to have to just scan through function calls and not operators.

every single person that is good at micro-optimizations look at the compiled binary as a first step

That isn't an option when you're writing portable code that runs on many different platforms, some of which may not even exist at the time you're writing it. Furthermore, micro-optimization isn't the only reason operator overloading is bad. The implicit flow control dramatically inflates the search space for what every single operation can do, making all code much more complicated to inspect at a glance. This carries over to debugging, where stepping through code is much more cumbersome when each operation can involve large amounts of indirection.

Of course, C functions can be passed as variables. Or in a wider scope they might be inline, macros, or ifdef'd to different functions. But those cases are _typically_ recognized as undesirable and avoided.

Java's a bit of a different story, which I can't figure out a good way to explain. It's hard to explain problems in large code bases, as a quick example rarely suffices. I've seen more than one bug caused because foo.bar(qux) called a different method of bar than the original programmer intended (both because foo's bar was overwritten and qux was a different type than expected).

Don't get me wrong, I would use operator overloading in a heartbeat if I was writing code for a math-y CS coding assignment. It's fine for code that will have a lifepsan measured in weeks / months with probably only 2 or 3 people ever looking at it.

Saying what you mean, as clearly and directly as possible, has it's perks in certain applications (large code bases, life critical code bases, code bases that will last for decades with dozens of programmers). Otherwise stated, cases where code is going to be read many times more than written.

To answer your question more directly: User definable functions aren't a problem. Re-definable functions are!