Scala Native > Scala JVM
Resharing here on Nostr as I came across one or two Scala enthusiasts (we are a dying breed) around here:
> Are you an open source Scala maintainer? The Scala Center is preparing to participate in Google Summer of Code 2025 (GSoC), and we’re on the lookout for projects to include in this year’s program. Deadline is February 7. read on: https://www.scala-lang.org/blog/2025/01/28/gsoc-projects.html
-
@scala_lang@fosstodon.org 🔗 https://fosstodon.org/users/scala_lang/statuses/113908197271531256
#scala #devstr
Discussion
Scala on bare metal is awesome. By the way, I came across some of your older Scala projects as well! :)
I really want to see Scala Native graduate to 1.0 one day. It has been "almost" ready for an eternity already. The fact that Scala.js has been "stable" for so long and we still couldn't get enough folks behind Scala's LLVM backend to make it mainstream is sad. It could have been what Rust is now. Except better! 🤣
No hate for the JVM, though. It has paid my bills for a good couple of decades. The years when I could pay the bills using Scala were among the best.
I loved Scala when I was working on it, but I don't think anyone should do it because the way it handles dependencies is too awful, so unfortunately it has to die. Having to precompile them and publish targeting specific platforms and scalac versions is horrible. Why not just ship source code? I have no idea, but one reason might be that compile times are so awfully slow and they would be even slower.
I don't know, as a language it is great, but these aspects of the thing kill it for me.
I agree. Scalac compilation speed and SBT magic were two of the less enjoyable aspects of using Scala. And cross building was not only annoyingly slow and complex but also very expensive (likelh great for the likes of Sonatype and JFrog though 🤑). Tbjs may have been born out of naivety and a stubborn attempt to fit Scala into the existing Maven ecosystem at all costs, but the Scala community certainly didn’t do much to help itself back in the day.
The plan for Scala 3 was to organise and optimise the implicit/macro/typeclass/shapeless/advanced type-level programming stuff that slowed down the compiler, but the hype around the language (and the money to keep funding the ecosystem) just died before it got there. The jobs left, at least here in the UK, are kinda sad (maintaining old Spark stuff while the company migrates to Python, maintaining old, overcomplicated ZIO projects that the business is too afraid to touch, migrating away from formerly open source tools to forks, etc).
Still a pity. Despite all the problems, I really enjoyed the language. It hit a sweet spot for me. Powerful and elegant without fighting the real world like Haskell or, you know, being Rust 🤣.
