"Rust or Go?" is not the question

Part 2: (But Rust is definitely the answer)

Part 3: Rust is definitely production ready

-> part 2 include coworker conversation tidbits draft notes:

These are two very different languages
These two languages are solving two very different problems
What attracted me to Rust is the error handling and borrow checker. You don't need a runtime and you don't need to worry about a variety of memory pitfalls.
Rust is not just a systems-level language, and Go is not just a server-side language.
Rust isn't *really* about speed or performance. It's about safety.
Rust is just plainly a more powerful language.
Go is for Python developers who need speed. Rust is for C++ developers who need safety.
Rust has opt-in concurrent runtimes and opt-in garbage collection. These are standard in Go, there's no getting out of them.

Go has great concurrency. Goroutines are high-performance, parallel green threads. Rust's concurrency is provably-correct.

Why is the immediate question when someone says they wrote something in Go, "why not rust?". The inverse is true. When I tell a dev I wrote something in Rust, the immediate response is "you should have used Go, it's better." This is false.

What does suck about Rust? The compiler is slow. It will probably always be a degree of magnitude slower than another compiler for a similar target.

It's not Rust vs Go, it's when to use Rust and when to use Go. And the number one argument I get for why Go should be used is it's simpler and faster to learn and work with. There's the answer! The answer is use whichever one works best for you. There's no better or worse, or superiority. Redditors will say otherwise.

sources

Go vs Rust discussions are ridiculous. It should be more like: When to use Go. When to use Rust. When to use X… — Inanc Gumus (@inancgumus) September 19, 2019 source

Is Rust in Trouble After Big Mozilla Layoffs? Hello World! (Rust Foundation) Killed by Google The Dart Programming Language Discord swapped Go for Rust

Both Microsoft and Amazon have just recently announced and released their new officially supported Rust libraries for interacting with Windows and AWS. Official first party support for these massive APIs helps make Rust people's first choice when deciding what to use for their project.

Source Dart -- apples to oranges? I'm not trying to say that go is going the way of dart, I'm trying to say that industry-leading companies aren't always stewards of their creations. take FreeBSD. It's a thriving, excellent operating system capable of "industry-leading company" usage. See the usuals (Netflix, Sony, etc). Look at the FreeBSD foundation. Now look at RedHat and IBM. Again, apples to oranges? No, just a bad argument to make in the first place.

quotables

Take a look Go as well. I think you will find Go much faster to program in. The other aspect is threading. They have very different threading models. Not sure if you had the chance to research that yet or not

Yes, a big thing is threading. Unfortunately, Rust uses a similar model as Java for threads :(. Go is based on Fibers approach which so much faster for temporary, lightweight requests. Go is definitely superior for HTTP REST API apps. Rust can be better for a single-thread app or general "systems" programming.

Rust is not a "systems programming" language. Systems programming is not a genre of languages. It's not like saying Italian is a "Romantic language". Systems programming is a specific, targeted programming *application*. It's the destination, the use-case that a language is being applied to. Rust is a general-purpose programming language. I have used it to write a variety of tools, low- and high-level, server-side and client-side, graphical and CLI. Yes, I used it for some systems programming. Also used it to make a very simple and robust web service digested by a variety of other developers at our company.

The only way is to learn and try both. That's what I did. Most of the info from both sides is biased...Go is definitely very fast and [garbage collection] is not the issue people make it out to be. I started last month porting [a chess] engine to Rust. I recently took a break from it because the syntax and borrow checking were getting insane to deal with. Once I learned about the threading issues in Rust, I have put it on the shelf for now. Rust is still evolving which is good and bad. It needs better IDE and Debugging support than current levels. Hopefully that will continue to improve. There was a big Mozilla shakeup (Nov 2020) where they let go of the Rust developers and cancelled the project. AWS hired them. So honestly, I am not sure which direction the language is going in. Meaning, now that AWS owns the braintrust, I don't know where they are headed. My guess is that AWS is using Rust for some behind the scenes script-like stuff. Not sure. Will be important in the next year or two on which direction things end up going. For Rust to benefit long-term, it needs the support of a corporate backer

Rust is absolutely ready for production use.

Anti-Rust zealotry is just as strong as pro-Rust zealotry. The hype goes both ways. No, your talking points shouldn't come from Reddit.