Hacker News Reader

Comment by berkut | Hacker News Reader

> Take for example, `auto`, added in C++11. If you don't have to support C++98 or C++03, the use of `auto` significantly simplifies many workflows and is often strictly better with less mental overhead.

When writing code, yes, but in my experience, it can also make interpreting what code that overuses auto is doing (especially what variables are and what functions return) an awful lot harder in many situations, and you end up jumping all over the place to work out that "auto result = ..." is actually "uint32_t result = ..."

That's one of my personal pet peeves about more modern C++ (and other languages to some extent like Rust): they seem to be more optimised for writing code quickly (which generally only happens once), not understanding it later (i.e. you didn't write the code someone else did) and maintaining / altering it in the future, which generally happens a lot more over code's lifetime.

View on HN

This is one of my worries about type inference (which C++ auto is not), but so far, the Rust code I've read has definitely been much easier to understand than the equivalent C code, and it's often much more generic. I like how Rust forces refuses to infer function signatures, which means you still always know the types of the inputs and outputs. I don't use IDEs, but I'd assume that they also make this whole thing a non-issue.

Relying on IDEs to do this type of thing really doesn't help for situations like merge conflicts (i.e. which merge tools understand languages fully?), and it doesn't completely solve the problem of easily understanding code just at a glance - you still need to go round mouse-overing or clicking on stuff to see extra detail.

Rust actually deliberately requires types to be declared, expect inside function bodies. So function return types have to be explicit. The main reason is exactly your readability argument. (Of course it also makes type inference inside the function body easier).

Yes, to be clear I wasn't accusing Rust code of having this particular problem, I was just using it as an example of a modern language where in my opinion there's a bit of an over-emphasis on brevity in terms of writing the code, as opposed to reading/understanding it and maintaining it, which I feel for long-term large projects, I'm not convinced is a total win.

I could have easily used Go or Swift as examples instead.

And to be clear, these new languages are in general improving things a lot, but I just worry there's an over-emphasis on "being able to do a lot of complex stuff which very few syntax / characters", which I don't really fully agree with (at least for large complex long-term projects).

That's not to say I'm right, but in my experience of programming over 15 years in everything from ADA, C, C++ to Java and Python, at least on large projects, the speed at which code was created was very rarely that important. Getting it right, bug-free and performant was generally much more important. In some cases newer more-condensed syntax can definitely help, but in others, it can cause trouble.

I definitely agree with most of your points.

Code is read much more often than written, and most newer languages optimize for convenience rather than readability.

In my view Rust is actually pretty verbose compared to other new languages, and feels a bit cumbersome to write. Technically it could be made quite a bit leaner syntax-wise.

I just wanted to point out that Rust and the Rust community often share your view and want to optimize for readability as well. There actually was a lot of heated discussion last year around proposals that suggested reducing readability for convenience.

Most pre-C++11 code I've seen that gets iterators from `begin()`, `end()`, and friends just typedef out their actual type anyways, so you the experience there is not that different from `auto`.

But those typedefs almost always (in my experience) contain some indication of at the very least an abbreviation of the type and some hints as to whether it's a pointer or not.

That's a lot better to go on than "auto".

You already know the type if you're calling .begin() or .end() on it. You don't care about the type of the iterator because you know it can iterate over the collection you're working with.

That's what auto fixes. It avoids coupling the concept of iterating from the specific type of the iterator which isn't important.

Similarly auto helps you achieve DRY. 'auto myFoo = std::make_unique<Foo>();'. The type wasn't removed, it just wasn't repeated twice.

But at this point auto, or things like auto, exist in nearly every major language. So you'll need to teach best practices for working with it at some point. That's a general thing that's everywhere.

is there a specific case that the uint32_t helps understanding the logic of the code?

That was a contrived example (although I've often seen auto used instead of that or even bool which I personally think are total over-uses of auto).

But knowing if something is a base type, a reference, a (smart) pointer, or an expensive-to-copy class/struct is very important when writing high-performance efficient code. Being able to see this at-a-glance by the type in the code in my experience helps tremendously with understanding what the code's doing and the implications in terms of data passing / transfer and understanding how that section of code interacts with other parts or could be changed to do other things.

I also think it allows people to be a bit sloppy and not care what's going on (i.e. with regards to whether it's by-value or by reference, etc) as they don't fully need to understand the code, and again, in high-performance computing where you seriously care about processor cycles and memory allocations, this can make a big difference if you're not careful.

Reply on news.ycombinator.com