Clojure Turns 15 panel discussion video

I see some comments in here with the sentiment, "Clojure—What am I missing?" Here's an attempt to answer that question. I am just a hobbyist and enthusiast programmer with no formal background in programming or CS, and no real experience using Clojure (or any other language, for that matter) "in anger" in a serious production environment. These are just some nobody's two cents, take or leave them for whatever they're worth to you.

From what I understand, there are some applications and domains that the JVM is highly optimized for—especially long-lived, highly concurrent server processes or stream processing applications where microseconds matter in order to keep up with real-time. The fly in the ointment (for some) is that Java itself is Object-oriented, compiled, statically typed, and uses mutable objects and data structures by default. Not everybody loves those design choices. And even of those who do, not everybody finds Java ergonomic (hence Scala and Kotlin, among others).

So, much like Scala and Kotlin do, vanilla Clojure just gives access to the JVM (and its highly mature library ecosystem) in different trappings. It's a functional (rather than object-oriented), REPL-driven (rather than compiler-first), dynamically-typed, and immutable data-oriented language, with Lisp's minimal syntax and structural editing experience. Plus, the designers have invested a lot of time and their cumulative decades of programming experience into making a really solid standard library that comes with a lot of smart affordances out of the box. That combo really works for some people.

Then, folks who appreciated the conveniences and good design in vanilla Clojure started looking for that dev experience in non-JVM contexts, which is how you get ClojureScript, ClojureDart, ClojErl, Babashka, Scittle, etc. And then, other Clojure-inspired languages pop up that don't purely adhere to Clojure conventions, but co-opt significant portions of its well-designed syntax and standard library—languages like Janet (compiles to C) and Fennel (compiles to LUA) and Hy (compiles to Python).

And then, script kiddies like me come along and discover, Hey Wow, I can learn one syntax and suddenly be +80% dangerous in a dozen different environments, without having to start learning a new language from scratch in each one—from the JVM to the browser to mobile apps to native executables, I've got SOMETHING to start building on because I already know how to think in Clojure.

At this point, "Clojure" is not just "a language," it's a category of languages—just like Scheme is "a Lisp," Janet is "a Clojure." I don't think that happens with uninteresting, poorly-designed, or irrelevant languages. Clojure scratches a specific itch, and not everybody has that itch to scratch. But whether Clojure is your cup of tea or not (and obviously for many people in this comment section, it's not), I don't think anyone can debate that it's a significant project that justifies its own existence. That doesn't obligate anybody to use it or even look at it twice, but I think it self-refutes a lot of the negativity that gets thrown its way.

[dead]

https://www.hugsql.org/ seems pretty nice. Not exactly an ORM, but probably does what you need.

There's this:

https://github.com/metabase/toucan

But beware what you do with it. Avoid DB side effects from business logic.

I think the bottom line with Clojure is it's not an ecosystem well-suited for non-veteran programmers. For as simple as the language is, effectively using Paredit, navigating partially documented libraries, diving into source code to see how things interact—it's tough as a new developer. I don't believe Clojure is overtly hostile to newcomers; it's just crafted by veterans, for veterans. And this is the result.