Understanding Ownership in Rust 

You are one of the chosen.

a good example from soft engg is to compare the ownership model with read-write lock semantics...

The basics of borrow checking isn't hard, but sometimes it can get tricky and very confusing in complicated code

This is the best concise explanation of borrowing I've seen yet

@@ddastoor because it's about the same problem

REAL

Wait until u learn traits

@@stardustbiscuits so crazy true

He doesn't derail ever, seamless additive commentary, and an enjoyable voice haha

+1

Most of what he said about stacks and heaps was misleading enough that I don't think he really understands it very well.

Basically just summarizing the book

@@-karter-4556 almost word for word plagiarism

Glad I could help!

25 years and this rank amateur explains things well? What? Are you one of those php script monkeys? It was awful.

this is how I sort of interpreted it.."stop doing these things unless you really need to because they're just problematic"

Thanks Gleb!

Most welcome!

True, you don't have to manage the memory yourself with raii data structures. Be aware these data structures are still doing dynamic allocation, but they are doing it in a safe way. You still take a potential performance hit but memory leaks shouldn't be an issue.

You can use a std::array if you want to avoid the free store and use automatic(commonly referred to as stack) storage. Std::array requires a size known at compile time.

I genuinely don't understand that claim. I might get that you'd 'appreciate it more' not having to deal with either garbage collection/erroneous manual memory management, but not having C/C++ being a prerequisite to learn rust

If you are already familiar with C#. I would actually recommend you learn a LITTLE bit of C++ before. The Syntax is kinda similar (They are oriented around OOP), but C# has a Garbage Collector, while C++ has only manual memory Management. So you'll learn about pointers, malloc and new, destructors, smart_ptrs, references (to lvalues and rvalues), copy constructor and move semantics. Those are things, C# does not have. So you can take like a day to learn the Syntax of C++ (How to do in C++ what you can do in C#) and then learn the memory management stuff which is the hard part. That way you can understand how hard and unreliable this stuff really is, and how rust, using the same abstractions (But with the added concept of borrowing (A type of move) that can be tracked by the borrow checker) can solve the same issues of C++, and a lot more, by default at compile time. And all the weird anoying part of the languages that you struggled with will probably now make sense as a logical next step, or a consequence of the things that you now know are happening under the hood!

You could also just keep going. Rust is s logical simplification of C++. But C++ is just a superset of C created to simplify reasoning about objects instead of just bytes in memory, and as such allowing things like RAII which are imposible in C. And C is just a logical next step above assembly. Hand compiling C is actually really easy. So you can stick to a subset of C (Just the atomic operations, the one that take a single CPU instructions) and have a very good idea of what's happening under the hood.

@@sebastiangudino9377 yeah thanks, I did learn C and C++ during my Computer Science degree in university years ago, so it's not that I don't know the fundamentals it's more about being familiar or effective in day to day use. What I meant by sticking with high level is because I actually don't want to really think about memory allocation, references etc and just like to work in higher level abstractions :P but ja, Rust started to feel like I needed to know more. So I still want to get into it again, but haven't :)

@@jamesbarrow Yeah! Going low level is just a performance thing. Since other ways there are just expensive allocations and copies EVERYWHERE. But to be fair. If you don't have performance needs you might just not need rust. I personally think Rust is beautifully elegant, but regular old C# can do the job really well

@@sebastiangudino9377 true. I do a lot of web dev, so the traction it's gaining along with Go in the JS tooling communities is making me interested at looking at it again ;)