Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why I love Rust

speakerdeck.com

Post reply on HN
Print view View original

141–150 of 159 posts

Re: Why I love Rust

#141

Post by fiedzia »

nickpsecurity wrote: post #133
pcwalton wrote: post #122

Earlier quoted context omitted.

Yes, 1.0 is stable and the language will not change in backwards incompatible ways. It's been that way for months (google "rust stability deliverable"). "All bugs fixed" rules out every language in existence except CompCert, so I don't think that's a very interesting metric.

Good to know on stability part. Far as bugs fixed, it's more on the lines of day-to-day use of the language in common scenarios. Are people getting unnecessary troubles out of the compiler? Typically don't for most languages as they work out most of them in common idioms and library functions in first, few years. If they're serious rather than 1-3 people's hobby, that is. I certainly won't expect it to be bug free. N…

I've seen rust compiler bug once and that was long before 1.0. The only issue now is that some important API's are not yet declared stable (RFC still being open), so many 3rd-party libraries target nightly and won't work in stable. This is however matter or 1 or 2 releases, and proves that rust developers take this matter seriously and don't wont to stabilize anything they aren't 100% sure. For all practical purposes rust is usable now, and more and more people are using it - certainly more then 3. Generally looking at rust projects on github, most of issues are feature requests. I don't see issues with crashes or panics, while I know plenty for similar projects written in C, at least in their early days.

Re: Why I love Rust

#142

Post by l1ambda »

bjz_ wrote: post #16
middleclick wrote: post #2

I want to spend at least 3-4 months learning a new language but just can't figure out if I should spent time learning Rust or Go. I am thinking more of it from a career perspective than interest.

One perspective could be that Rust is the more interesting talking point, even if you don't work on it in the job you are actually applying for. But if you purely want to get a job in one or the other, Go is currently much more widely used. I have hopes that this year will see a large jump in industry adoption of Rust, but that still remains to be seen. But as others have said - pick the one you think you'd enjoy the…

Interesting point there. All things being equal, I would favor someone who is learning Rust over someone who is learning Go--even if I were hiring for a job that involved writing Go!

Re: Why I love Rust

#143

Post by pcwalton »

hellofunk wrote: post #140
pcwalton wrote: post #107

Earlier quoted context omitted.

> I think that Rust simplifies some things that you must do yourself in C++, and learning how to do them is valuable. Given that you didn't answer the question, I think it's pretty clear at this point that you haven't used Rust. Rust is a low-level language with low-level control over memory and resources.

I certainly don't know Rust as well as C++. Some other fellows in this thread provided some features in C++ that you can't learn about in Rust, fyi.

> Some other fellows in this thread provided some features in C++ that you can't learn about in Rust, fyi.

Variadic generics and template templates are hardly interesting compared to concepts, pattern matching, and lifetimes. I use those every day in Rust, they simply don't exist in C++, and with the exception of lifetimes plenty of other languages have them too, so they're broadly useful to know. Variadic generics and template templates, on the other hand, are useful for a small subset of advanced library authors who have been programming in C++ for years.

Re: Why I love Rust

#144

Post by zvrba »

humanrebar wrote: post #55
zvrba wrote: post #40

Earlier quoted context omitted.

> If you start with C++, you're committed to a multi year project, that's what it takes to advance from novice to junior level. This is nonsense. Perhaps it's true to some degree if you try to "learn" C++ from tutorials of dubious quality rather than from a decent book.

You sound like someone who has never really explored C++ lookup rules.

Funnily I've been working on MLOC-sized C++ projects and almost never had a problem with lookup rules. On rare occasions where it was a problem, it was a compile-time problem and easily resolved.

In any case, these rules do not fall under what I'd call "junior-level" C++ programmer as the OP wrote.

Re: Why I love Rust

#145

Post by mike_hearn »

steveklabnik wrote: post #131
mike_hearn wrote: post #129

Earlier quoted context omitted.

C++ has a well defined ABI on many important platforms. On GCC/LLVM based platforms it's standardised in specification documents. On Windows, I think the Visual C++ ABI is the de-facto standard. Regardless, there are only a handful of top C++ compilers. Interop with even one of them would be of tremendous value.

Yeah, I mean, the basics of interop might be possible, it will just never be as robust as it could be if C++ had features better designed to play with other languages. (This of course also affects Rust as well; we don't have a defined ABI either.)

Sure, you can't get completely perfect interop. However a LOT of real C++ codebases don't use or explicitly ban some of the trickiest features, so even partial support can go a long way. For instance, supporting the basic class/vtable layout C++ uses (or even just adopting the C++ ABI for Rust itself here) along with char* strings, would immediately allow you to call into a lot of real C++ APIs, albeit older ones.

Supporting C++ value types, even without templates, would in turn be enough to support quite a lot more. For instance Unreal Engine defines its own string types, as do Qt and so on.

Re: Why I love Rust

#146

Post by krylon »

dozzie wrote: post #44
middleclick wrote: post #2

I want to spend at least 3-4 months learning a new language but just can't figure out if I should spent time learning Rust or Go. I am thinking more of it from a career perspective than interest.

If you don't have many languages under your belt (and wanting to spend 3-4 months and asking for advice what to choose suggests that), you don't really want to go with a language that popped semi-stable only several months ago, hence will be most probably a subject to deep changes (either in the language itself or in standard library). Rust is not (yet) the way to go.

> Rust is not (yet) the way to go.

Not sure if that was intended, but that was one epic pun!

Re: Why I love Rust

#147

Post by steveklabnik »

dozzie wrote: post #44
middleclick wrote: post #2

I want to spend at least 3-4 months learning a new language but just can't figure out if I should spent time learning Rust or Go. I am thinking more of it from a career perspective than interest.

If you don't have many languages under your belt (and wanting to spend 3-4 months and asking for advice what to choose suggests that), you don't really want to go with a language that popped semi-stable only several months ago, hence will be most probably a subject to deep changes (either in the language itself or in standard library). Rust is not (yet) the way to go.

Rust is more than semi-stable: we take backwards compatibility very seriously. It will not be subject to deep changes from here on out, or at least, those changes will not break existing things.

Re: Why I love Rust

#148

Post by nickpsecurity »

fiedzia wrote: post #141
nickpsecurity wrote: post #133

Earlier quoted context omitted.

Good to know on stability part. Far as bugs fixed, it's more on the lines of day-to-day use of the language in common scenarios. Are people getting unnecessary troubles out of the compiler? Typically don't for most languages as they work out most of them in common idioms and library functions in first, few years. If they're serious rather than 1-3 people's hobby, that is. I certainly won't expect it to be bug free. N…

I've seen rust compiler bug once and that was long before 1.0. The only issue now is that some important API's are not yet declared stable (RFC still being open), so many 3rd-party libraries target nightly and won't work in stable. This is however matter or 1 or 2 releases, and proves that rust developers take this matter seriously and don't wont to stabilize anything they aren't 100% sure. For all practical purposes…

Thanks for the detailed assessment. Sounds like a high quality, Beta to me. I also like the good things I keep hearing about the quality of libraries, etc.

Re: Why I love Rust

#149

Post by steveklabnik »

mike_hearn wrote: post #145
steveklabnik wrote: post #131

Earlier quoted context omitted.

Yeah, I mean, the basics of interop might be possible, it will just never be as robust as it could be if C++ had features better designed to play with other languages. (This of course also affects Rust as well; we don't have a defined ABI either.)

Sure, you can't get completely perfect interop. However a LOT of real C++ codebases don't use or explicitly ban some of the trickiest features, so even partial support can go a long way. For instance, supporting the basic class/vtable layout C++ uses (or even just adopting the C++ ABI for Rust itself here) along with char* strings, would immediately allow you to call into a lot of real C++ APIs, albeit older ones. Su…

Yeah, to be clear, I think such a thing would have value too. Just hard to get 100%, there are always gonna be caveats.

Re: Why I love Rust

#150

Post by dozzie »

steveklabnik wrote: post #147
dozzie wrote: post #44

Earlier quoted context omitted.

If you don't have many languages under your belt (and wanting to spend 3-4 months and asking for advice what to choose suggests that), you don't really want to go with a language that popped semi-stable only several months ago, hence will be most probably a subject to deep changes (either in the language itself or in standard library). Rust is not (yet) the way to go.

Rust is more than semi-stable: we take backwards compatibility very seriously. It will not be subject to deep changes from here on out, or at least, those changes will not break existing things.

> we take backwards compatibility very seriously.

You're yet to display that. Half a year is too short time to give that confidence.

Post reply on HN