3 ms·

“ Rust has been the "most loved programming language" in the Stack Overflow Developer Survey every year since 2016” https://en.m.wikipedia.org/wiki/Rust_(progr

“ Rust has been the "most loved programming language" in the Stack Overflow Developer Survey every year since 2016” https://en.m.wikipedia.org/wiki/Rust_(programming_language) https://en.m.wikipedia.org/wiki/Rust_(programming_language) 10 years is a long time. Swift is 5 years old and it seems old and baked. Considering the buzz Rust has had on HN, etc, I thought it was more popular. Sounds to me like it needs some sort of push to get over a tipping point. Often part of the problem isn’t technical. In business people don’t want to be first. Perhaps a few more high-profile users.

TheFiend7 7y ago
Though I would argue they are different demographics somewhat. Mobile app and mac developers I would argue keep up with the times much more aggressively compared to c/c++ developers I know.
simias 7y ago
Rust 1.0 was released in 2015, and before that you'd have breaking changes every month that effectively made it unusable for anything but toy projects. So it's not even 5 years old really. Besides a big target for Rust is the C and C++ world. On Hacker News in my experience you have a majority of commenters coming from the web world where no news in one year effectively means that the project is dead. For us in the low level world "stable for 4 years" means "maybe we can start considering using it in production" and the lack of big buzz every other month is more a pro than a con. I'll take boring and reliable over shiny and breaks-every-other-year. I've just started adding a new feature on a C project started in 2009. If I used Rust I'd want to know that my dev environment will still be usable in 2019. I think the commitment to stability will pay off eventually.
- stoicShell 7y ago
  Agreed, Rust's "popularity" in production projects is a topic to be revisited circa 2025-2030. My personal intuition is that it will have become a strong alternative to C++ by then, and Go will probably eat the other side of that (at the 'upwards' frontier of C++, before/underneath e.g. Python), which given a decade could result in maybe 25-30% of major C++ projects moving or with plans to Rust/Go. That would be a healthy balance of alternatives, a true victory for these mid/low-level contenders. It's not like the bottom of the stack can be won the likes of Python or Js at 80-90% within a decade. Structurally, it simply cannot.

TheFiend7 7y ago

Though I would argue they are different demographics somewhat. Mobile app and mac developers I would argue keep up with the times much more aggressively compared to c/c++ developers I know.

simias 7y ago

Rust 1.0 was released in 2015, and before that you'd have breaking changes every month that effectively made it unusable for anything but toy projects. So it's not even 5 years old really. Besides a big target for Rust is the C and C++ world. On Hacker News in my experience you have a majority of commenters coming from the web world where no news in one year effectively means that the project is dead. For us in the low level world "stable for 4 years" means "maybe we can start considering using it in production" and the lack of big buzz every other month is more a pro than a con. I'll take boring and reliable over shiny and breaks-every-other-year. I've just started adding a new feature on a C project started in 2009. If I used Rust I'd want to know that my dev environment will still be usable in 2019. I think the commitment to stability will pay off eventually.

stoicShell 7y ago
Agreed, Rust's "popularity" in production projects is a topic to be revisited circa 2025-2030. My personal intuition is that it will have become a strong alternative to C++ by then, and Go will probably eat the other side of that (at the 'upwards' frontier of C++, before/underneath e.g. Python), which given a decade could result in maybe 25-30% of major C++ projects moving or with plans to Rust/Go. That would be a healthy balance of alternatives, a true victory for these mid/low-level contenders. It's not like the bottom of the stack can be won the likes of Python or Js at 80-90% within a decade. Structurally, it simply cannot.