Or, I picked a random, reasonably popular library to check on Clojars: http-kit. The most recent stable release, 2.8.0, which came out last year, has only been downloaded ~600k times. 2.7.0 from 2023 was downloaded ~1.4m times. 2.6.0 from 2022 was dled ~2m times. Ditto for 2.5.3 from 2021.
I would have used Clojure itself, but I can't find maven dl statistics.
https://clojars.org/http-kit/versions/2.8.0https://clojars.org/http-kit/versions/2.7.0https://clojars.org/http-kit/versions/2.6.0https://clojars.org/http-kit/versions/2.5.3
---
The thing is, I've been seeing little pieces of evidence all over that Clojure is waning, and not much that it's genuinely increasing in popularity. Any individual example doesn't weigh that much, true, but everything seems in the same direction.
If people want Clojure to grow, whether because they need job opportunities, a big employee pool, whatever, it starts with a clear assessment of where it's at.
I need a tool that helps with problem-solving and product development, and works reliably and effectively across a wide range of use cases, from basic mobile apps to high-performance computing.
Clojure delivers that better than any other language or ecosystem that I know of in a uniform, well-designed package, all the way from the core internals of the language to deps.edn.
Between Clojure and Racket, those have always been my two favorite Lisp / Scheme languages. I don't do a lot of Lisp but when I do its either in Clojure (thanks to Lein) or in Racket.
[0] - Technincally Objective-C and Swift could also be considered, but they lack the industry wide adoption, as many cool tools only exist in Apple land.
Swift is quite good. I assume the tooling is catching up to Objective-C. It has been a couple of years since I had to solve something with Swift.
Any reason, apart from outside requirements, to pick the .NET CLR over the JVM as a runtime in your experience? Outside of library/framework support, which I've heard is good with the CLR.
How's the garbage collector with the CLR?
The transition of core.async specifically to VirtualThreads is still WIP as far as I understand, but with minimal tweaks, 90% of the benefits are already there, even with the current latest version.
I do find that for about 5 years things seemed to be slowing down. Though I keep seeing it pop up, and new exciting projects seem to pop up from time to time.
Just today I saw an article about Dyna3, a relational programming language for AI and ML that was implemented on top of Clojure.
I miss the Strange Loop conference. I think a lot of Clojure buzz was generated there. Clojure West and a few others so a decent job, but the quality of the talks at Strange Loop were second to none. Not that it was a Clojure specific conference, but it had that focus on elegance that I don't see very often, and the organizer was a something like the Prince of Clojure, if I recall correctly.
I'm still enjoying the language, and all my projects still build and run just fine.
The major frustration I have with the platform is 3D graphics. That's a JVM issue overall though.
Link to demo @ timestamp: https://youtu.be/UVsevEdYSwI?t=653
My experience with 3D graphics is minimal, but I'm curious to know if these newer developments are significant in any way for 3D work.