News for hackers. Stuff that matters.
Re:Ask HN: Is Erlang an albatross to Elixir adoption? by lukaszkorecki on Monday May 09, 2022 @11:08AM (#31315056)
In my experience (7+ years of using Clojure) this is caused by the misunderstanding of what Clojure actually is: it's a LISP designed for the JVM, you cannot escape the host and pretend it's not there. In return you get access to one of the biggest software ecosystems and pull any library from Maven Central etc.

Coming from Ruby/Go/JS to Clojure was definitely harder than I expected but rather than fighting "this stupid JVM" it made more sense to buy into the idea of how Clojure is built and used.

Re:Ask HN: Is Erlang an albatross to Elixir adoption? by ModernMech on Monday May 09, 2022 @11:35AM (#31315486)
> It's a LISP designed for the JVM

Honestly, this should be the Clojure tagline. Would solve a lot of the confusion. To be truthful, I never pictured it that way, but to be fair I think Clojure sells itself as more than that.

Re:Ask HN: Is Erlang an albatross to Elixir adoption? by lispm on Monday May 09, 2022 @11:42AM (#31315578)
But it's more than that. It has different data structures, operators and semantics from Lisp - stuff which goes beyond just a simple retargeting to the JVM. It also provides an integration into/on top of the JVM, but also stuff like persistent functional data structures, avoids OOP, ... It's opinionated.

Lisp and Scheme on the JVM exist, too.

Re:Ask HN: Is Erlang an albatross to Elixir adoption? by melony on Monday May 09, 2022 @05:49PM (#31320247)
I wish it would focus more on using technology provided by the underlying platform instead of its own half-baked and slow implementations. Clojure’s concurrency story is horrendous. There are multiple versions of shared memory concurrency on its JS and JVM runtimes but none of them scales to match the performance of the local APIs (Promises and Project Loom for Java). They are always playing catch-up five years after the original features have been shipped. Even today, trying to find an up to date library for features like WebWorkers or WebRTC that has seen a commit in the past year is a waste of time.