Rust Developers Planning For The Rust 1.0 Language

[RamblingMadMan]

Not a viable replacement (yet)

Here's to hoping it eventually replaces C++ for all intents and purposes...

Just because you don't like a language doesn't mean that its a bad language.

[MartinN]

Just because you don't like a language doesn't mean that its a bad language.

I just finished a small project that takes input from an stmicro ARM chip that processes a vital sign waveform (can't disclose too much) and sends it back over USART to my C++ app for further analysis and transforming the output into various formats, matlab, physionet, csv, etc... used Boost ASIO to do it.

I love C++ ... that is why I want to see it replaced.

[RamblingMadMan]

I just finished a small project that takes input from an stmicro ARM chip that processes a vital sign waveform (can't disclose too much) and sends it back over USART to my C++ app for further analysis and transforming the output into various formats, matlab, physionet, csv, etc... used Boost ASIO to do it.

I love C++ ... that is why I want to see it replaced.

Have you tried C++11/14? I really don't see why a new language is needed to address features that could already be added to an existing language.

[tga.d]

Have you tried C++11/14? I really don't see why a new language is needed to address features that could already be added to an existing language.

The issue isn't the feature set, it's the entire style. Don't get me wrong, I love C++, and it was the language in which I learned to program, but it's just too dangerous to have as the long term solution to any program where security or stability are a concern -- which should be all programs. Stuff like use after free makes up the largest chunk of security bugs in Firefox, and there's no reason to keep using a language that allows something like that to happen when a language like Rust exists (other than things like supporting legacy codebases, obviously).

[RamblingMadMan]

The issue isn't the feature set, it's the entire style. Don't get me wrong, I love C++, and it was the language in which I learned to program, but it's just too dangerous to have as the long term solution to any program where security or stability are a concern -- which should be all programs. Stuff like use after free makes up the largest chunk of security bugs in Firefox, and there's no reason to keep using a language that allows something like that to happen when a language like Rust exists (other than things like supporting legacy codebases, obviously).

free/malloc is a C idiom, and unique_ptr is the safe alternative; but I do see your point. The real difference safety wise is that the C++ developer must ALWAYS treat their code lile it could be unsafe, but in rust you have to put unsafe code in unsafe{} blocks for unsafe things to happen.

[mark45]

Preceded by "Rewriting big C++ application/library isn't usually option...."

Why would you want Rust if you already have qt?
Might replacing qt be a long Rust plan? Would that be worth doing?

Rust can't replace Qt because Rust is a language and Qt a toolkit.

Regardless, neither Qt nor other serious project is moving away from C/C++ to Rust just like the USA isn't going to move from English to French no matter how many (naive) people find the latter language any better.

[rdnetto]

I don't complain about C++ but Rust is definitely the best currently at the fields that Go, C++ and D programming languages aim.

I'm not so sure that those languages are aiming at the same fields. Go and D both have garbage collection - they seem to be focusing more on replacing Java/C++. Rust focuses on manual memory management (with compile-time guarantees of safety) - it feels much more low level, like a replacement for C.

Regardless, neither Qt nor other serious project is moving away from C/C++ to Rust just like the USA isn't going to move from English to French no matter how many (naive) people find the latter language any better.

I'm pretty sure the reason Mozilla is researching Rust is so that they can rewrite Firefox (or its successor) in it.

[Szzz]

I'm not so sure that those languages are aiming at the same fields. Go and D both have garbage collection - they seem to be focusing more on replacing Java/C++. Rust focuses on manual memory management (with compile-time guarantees of safety) - it feels much more low level, like a replacement for C.

C++ also has manual memory management in the sense that there is no garbage collector but you can use smart pointers (just like in Rust, although Rust has optional gc). Also Rust developers state that Rust should be at least as fast as idiomatic C++. I would say that it aims both at C and C++ use cases.

I'm pretty sure the reason Mozilla is researching Rust is so that they can rewrite Firefox (or its successor) in it.

I think it was known from the beginning that they make Rust to write Servo in it.

[mark45]

I think it was known from the beginning that they make Rust to write Servo in it.

But then again, why bother creating a new language for writing Servo in it? It has no feature that warrants the transition from C++11 with all the downsides of doing an open project in a language almost nobody uses.

Unless Mozilla gets Google etc to start coding in Rust - developing a new language for a project is insane, or at least naive, or somebody at Mozilla has too much free time and had that naive idea each developer has at one point of creating a new "killer" language.

D tried - failed, Vala tried and failed, others tried and failed, how does Rust address the reasons why its predecessors failed?

[Pajn]

This. Apps usually use alot of libraries, and while you can cope with binaries, header only libs is the thing you wont be able to use in another language. So stuff like Rust currently works only for 'Hello world' or if you develop all components/functionality along.

But so much hype.

What, what?!

But then again, why bother creating a new language for writing Servo in it? It has no feature that warrants the transition from C++11 with all the downsides of doing an open project in a language almost nobody uses.

Unless Mozilla gets Google etc to start coding in Rust - developing a new language for a project is insane, or at least naive, or somebody at Mozilla has too much free time and had that naive idea each developer has at one point of creating a new "killer" language.

D tried - failed, Vala tried and failed, others tried and failed, how does Rust address the reasons why its predecessors failed?

It's memory safe, which C++ isn't and also it have a better and easier multi thread model.