More use of Rust is inevitable in open source software

More use of Rust is inevitable in open source software(utcc.utoronto.ca)

30 points by h1x 3 years ago | 36 comments

bsaul 3 years ago |

After trying rust only for a few hours i really wonder how come this language remain the most loved one on surveys. From the outside it looks like a niche language you'd only use for super constrained or critical environments.

Take backend dev for example: unless you're trying to squeeze the last drop of performance from your CPU, most concurrency issues are usually dealt with at the middleware level (DB, Queue, Nginx, etc), and most performance issues are usually I/O related.

I'm currently evaluating developing a cross-platform mobile native library using Rust, but i really don't see myself inflicting the pain of that language to the rest of my team.

Or maybe it's just a reflection on how active the community is ?

edit: i realize a lot of people are probably still coding in C because they have no other choice, and for them it's probably a good idea to switch. However, for all the rest i really don't see the point.

newpavlov 3 years ago | |

>After trying rust only for a few hours i really wonder how come this language remain the most loved one on surveys

Arguably, it's because you tried it "only for a few hours". Rust has a relatively steep learning curve and it usually takes several months to become comfortable with it. But once you get over it, it often becomes quite hard to return to your previous languages, be it Python or C/C++.

Yes, there are tasks which are ill-fit for Rust and your IO-bound backend example is a good one. Not only Rust ecosystem is somewhat underdeveloped in this area (and don't get me started on Rust async and the heap of troubles it causes...), but also you have business requirements which change unpredictably and Rust's strictness introduces a certain friction in such environment.

But I think your mobile library can be a good project to properly learn Rust. If it's properly encapsulated, I don't think it will cause much trouble to your team outside of build system changes and risks associated with you being its sole developer.

simplotek 3 years ago | | |

> Arguably, it's because you tried it "only for a few hours". Rust has a relatively steep learning curve and it usually takes several months to become comfortable with it. But once you get over it, it often becomes quite hard to return to your previous languages, be it Python or C/C++.

You failed to address OP's point.

OP wondered why would he, or anyone, subject anyone else to a steep learning curve if there are no meaningful advantages to be gained.

As OP mentioned:

> Take backend dev for example: unless you're trying to squeeze the last drop of performance from your CPU, most concurrency issues are usually dealt with at the middleware level (DB, Queue, Nginx, etc), and most performance issues are usually I/O related.

Why do you believe that doubling down on a whole new programming language and tech stack would help address the fact that you do not benefit from any of it's sales pitches?

Honestly, your comment reads like a boilerplate reply to any criticism of Rust that presumes Rust's only problem is it's developer experience caused by it's learning curve but once that's behind anyone it's somehow the ideal solution for any imaginable scenario.

eternityforest 3 years ago | |

It's competing with C/C++/Go/Nim/Zig, not Python and JS.

I don't use it because I don't have a need for low level coding on platforms with mature Rust support, but if I did I am pretty sure it would be my language of choice without hesitation.

Communitivity 3 years ago | |

Rust has a unique experience. Initially you hate it, right up to the point where you grok it, then you love it because you realize it save you from so many boneheaded errors that C/C++ allow you to do easily. You can still do them in Rust, but you can't do them unintentionally (generally). The borrow checker, the forced composition over inheritance, and the trait mechanisms are major wins.

xigoi 3 years ago | | |

I tried Rust once and found it pretty interesting. The moment I left was when I wanted to write a simple web scraper, so I found a library, installed it… and found out that it had downloaded a gigabyte of dependencies.

andrewjf 3 years ago | |

It’s not just about performance. It’s the maintainability and correctness that its amazing type system gives you. Refactoring and maintaining code is so much easier when the compiler tells you when you missed something.

hitekker 3 years ago |

This article is pretty shallow. Despite the interesting title, the author is lukewarm on Rust, hasn’t used Rust that much, and hasn’t explored the space of C replacement OSS. Not enough expertise for an insight; it feels half “go with the flow”, half FOMO.

xcdzvyn 3 years ago | |

I agree. Of course Rust is going to see "more use", it's popular right now. But that's really the only claim I can see the author making. Or maybe they're making a point about Zig I can't parse; It doesn't have strong memory safety guarantees, and so... something?

habitue 3 years ago | |

On the contrary, I think this outside view is a really good look at the forces driving rust adoption beyond developer enthusiasm.

hulitu 3 years ago |

Is Rust made by the CIA ? Because is sure has a lot of propaganda behind it.

OTOH is very slow to compile.

One advice for people wanting more adoption for Rust: don't write existing programs: shells, cat, ls, etc. We already have them. Write new programs which will be useful to a lot of people. And then the adoption will occur naturally.

pipeline_peak 3 years ago |

Can Rust actually replace C? Or just the areas where performance and resources are less important like desktop development.

Apparently it’s difficult to implement doubly linked lists https://news.ycombinator.com/item?id=16443688

bravetraveler 3 years ago |

Obviously, even D is likely to get used more. It exists so we'll probably have to contend with it. We shouldn't wax/wane over the toolchain.

If your distribution environment doesn't strive to simplify this (eg: fedpkg, mock, RPM macros), what are you using it for?

jbirer 3 years ago |

A language is only good when the zealots don't have to constantly push it and persuade people to adopt it, it happens on it's own due to it's usability and features.

pjmlp 3 years ago | |

A language that isn't pushed in some form, big corporation, zealots or killer framework, usually dies.

No famous mainstream language in use today, has ever been adopted based on "it happens on it's own due to it's usability and features".

Before you rush out to write something like C, remember that it only came to use thanks to AT&T, Bell Labs and UNIX.

moloch-hai 3 years ago | | |

Even a language pushed by corporations, zealots, and a framework usually dies. Fizzling is absolutely the normal fate of any new language, absent a miracle.

Even a language that seems to have everything going for it can fade out. Ruby's success used to be thought assured. Then it suddenly wasn't. Python has run the oddest course. It plodded along for two decades, almost imploded on the 2->3 transition, and then got its miracle.

Kukumber 3 years ago | | |

exactly, and so far rust is being pushed by evangelists and pseudo-journalists

rust wants to penetrate the industry by-proxy rather than with interesting tech/projects

Mozilla gave up rewriting firefox in rust

Embark gave up writing their new game engine in rust

itsautomatisch 3 years ago | |

If people aren't pushing then people won't use it. There are plenty of niche and esoteric languages that could be considered "good" (whatever that means) but aren't being actively used.