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

Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

sites.google.com

Post reply on HN
Print view View original

1–10 of 145 posts

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#3

Post by antirez »

What's great about the actor model is that you can kinda apply it in languages not having such native support. Even if it will not be a strict model, with some education from the programmer it can do wonders and can be backed by lock free queues. What I see as a problem is always they for a thread to wait on new items a syscall must be executed which is costly. But one could use a spinlock a few cycles and later degradate to syscall.

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#4

Post by readittwice »

antirez wrote: post #3

What's great about the actor model is that you can kinda apply it in languages not having such native support. Even if it will not be a strict model, with some education from the programmer it can do wonders and can be backed by lock free queues. What I see as a problem is always they for a thread to wait on new items a syscall must be executed which is costly. But one could use a spinlock a few cycles and later degr…

> What's great about the actor model is that you can kinda apply it in languages not having such native support.

Yep, although not listed in the article, JS would be such an example for me. TBH I am not entirely sure what's the difference between Web Workers and Isolates but for me it's more or less the same: Both allow safe multi-threading simply because threads do have separate heaps.

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#5

Post by mpweiher »

antirez wrote: post #3

What's great about the actor model is that you can kinda apply it in languages not having such native support. Even if it will not be a strict model, with some education from the programmer it can do wonders and can be backed by lock free queues. What I see as a problem is always they for a thread to wait on new items a syscall must be executed which is costly. But one could use a spinlock a few cycles and later degr…

Yep, I have used Cocoa(Touch) runloops in this way. Very straightforward, particularly if you use a bit of Higher Order Messaging syntactic support:

   [[someObject async] doStuff];
This will schedule the doStuff message to run on the runloop of the actor (=object + runloop) in question.

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#7

Post by readittwice »

smallstepforman wrote: post #2

I was really excited to discover Pony a couple of years ago, sadly there is negative momentum with this project. So much potential, yet the world isn’t ready for it yet.

I looked into Pony some time ago and I always found the capabilities-thing to be quite complex although very interesting. It was hard to imagine using it in a project with developers unfamiliar to this concept. Especially since I was mostly working on large web applications where only tiny parts were actually using shared state between threads.

TBF I have never written Pony code, I was mostly reading documentation.

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#8

Post by eeZah7Ux »

"The problem is, that sometimes, unfortunately, these tools are just not sufficient, it's still easy to shoot your own foot and get lost in a sea of complexity."

As a non-native English speaker I found this use of commas very difficult to understand. Often I get the feeling that native speakers don't even notice it.

Re: Fearless Concurrency: Clojure, Rust, Pony, Erlang and Dart

#10

Post by rubiquity »

ahaferburg wrote: post #6

In case the author sees this: The indentation of the code samples is all over the place, making it quite hard to read. Even more so for a white-space sensitive language like Pony. Did you mix tabs and spaces?

Pony is not white space sensitive.
Post reply on HN