5 ms·

More use of Rust is inevitable in open source software

30 points by h1x 4y ago

hitekker 4y 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.
- habitue 4y ago
  On the contrary, I think this outside view is a really good look at the forces driving rust adoption beyond developer enthusiasm.
- xcdzvyn 4y 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?
jbirer 4y 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.
- waboremo 4y ago
  That isn't true and hasn't worked on numerous great languages, which are now often ignored entirely.
- ranger207 4y ago
  It seems contradictory that a good language wouldn't have proponents trying to get people to use it. I'm not sure I've ever heard of a language people think is good but don't think other people should use.
- jraph 4y ago
  Something (ideas, tools, languages) often needs to be pushed to spread.
- howinteresting 4y ago
  This is not true. You need a healthy balance of 10x improvements (which Rust has with the whole borrow checking business) and aggressive advocacy.
- itsautomatisch 4y 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.
- pjmlp 4y 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 4y 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.
    - andsoitis 4y ago
      > Fizzling is absolutely the normal fate of any new language, absent a miracle. There are more ways to be dead than being alive.
    - pjmlp 4y ago
      Ah the Python example from people that weren't there I guess. Zope in the early 2000's, Guido's employment at NIST, CNRI, Google, Dropbox and now Microsoft. CERN and Fermilab pushing for it during 2003 onwards. Ruby suffered the same fate as languages that are tied to a killer framework, then the killer framework is no longer seen as the new hottest thing to have on the CV. Yet, without Rails, no one would have cared enough for Ruby to matter.
  - Kukumber 4y 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
- habitue 4y ago
  I mean, this pretty much describes rust. Its merits speak for themselves, and people who were skeptical of it and try it very quickly find out the language is really well designed and pleasant to use (once you get used to the borrow checker). There are a ton of niche languages that have very enthusiastic fanbases, but they don't get the same snowball effect rust has gotten because it's a great language on so many dimensions.
bravetraveler 4y 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?
bsaul 4y 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 4y 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 4y 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.
pipeline_peak 4y 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 https://news.ycombinator.com/item?id=16443688
hulitu 4y 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.

30 points by h1x 4y 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.

habitue 4y ago
On the contrary, I think this outside view is a really good look at the forces driving rust adoption beyond developer enthusiasm.
xcdzvyn 4y 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?

jbirer 4y 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.

waboremo 4y ago
That isn't true and hasn't worked on numerous great languages, which are now often ignored entirely.
ranger207 4y ago
It seems contradictory that a good language wouldn't have proponents trying to get people to use it. I'm not sure I've ever heard of a language people think is good but don't think other people should use.
jraph 4y ago
Something (ideas, tools, languages) often needs to be pushed to spread.
howinteresting 4y ago
This is not true. You need a healthy balance of 10x improvements (which Rust has with the whole borrow checking business) and aggressive advocacy.
itsautomatisch 4y 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.
pjmlp 4y 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 4y 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.
  - andsoitis 4y ago
    > Fizzling is absolutely the normal fate of any new language, absent a miracle. There are more ways to be dead than being alive.
  - pjmlp 4y ago
    Ah the Python example from people that weren't there I guess. Zope in the early 2000's, Guido's employment at NIST, CNRI, Google, Dropbox and now Microsoft. CERN and Fermilab pushing for it during 2003 onwards. Ruby suffered the same fate as languages that are tied to a killer framework, then the killer framework is no longer seen as the new hottest thing to have on the CV. Yet, without Rails, no one would have cared enough for Ruby to matter.
- Kukumber 4y 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
habitue 4y ago
I mean, this pretty much describes rust. Its merits speak for themselves, and people who were skeptical of it and try it very quickly find out the language is really well designed and pleasant to use (once you get used to the borrow checker). There are a ton of niche languages that have very enthusiastic fanbases, but they don't get the same snowball effect rust has gotten because it's a great language on so many dimensions.

bravetraveler 4y 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?

bsaul 4y 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 4y 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 4y 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.

pipeline_peak 4y 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 https://news.ycombinator.com/item?id=16443688

hulitu 4y 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.