atomic VS proposal-command-syntax

I wanted ClojureScript, without transpilation, in the browser so badly I implemented much of it in plain JS. But JS still lacks first-class protocols (although there is a T39 proposal for them) so I have to implement protocols first. And I can tell you from having implemented much of Clojure myself in plain JS that it's protocols all the way down. 80% of my library are types and their protocols.

I've watched a lot of talks, but it was Rich Hickey's which most captivated me and, ultimately, inspired big change in how I coded. After discovering Clojure I was so desiring FP (i.e. ClojureScript) in the browser without a build step and hoard of dependencies that I wrote my own library.

I fell in love with Clojure and ported most of the tools to plain ol' JavaScript. I stole reactives and FRP from Elm. I write a functional core then wrap it with an imperative shell. It's been my mainstay for 10 years now.

What I did was looked past the languages at the ideas and saw they could be applied everywhere, including JavaScript. Take a look at my example programs (via prior link) and you can see Clojure applied to the JavaScript/browser realm.

I have found that the amount of code needed to support any page/app need not be massive. I have implemented modern board games with 250-300KB of js and the core lib I use on all my projects maxes out around 250KB, unminified. The lib itself is built on the premise of build avoidance.

I do JavaScript most of the time, but I was so inspired that I ported the Clojure mindset into the web development world so I could use it even in the absence of Clojure. I know there's ClojureScript, but I wanted the goodness without transpilation.

So I don't even do Clojure per se. I do Clojure in JavaScript without ClojureScript.

I use functional core, imperative shell. It would have you write a core which is a pure simulation. Write everything in the core using immutable data and pure functions. I elaborate on that here, along with examples.

And quick aside. Command-Query Separation. This is FP, right? Is it something which plays heavily into how you think about and write code? For me, it's a resounding yes! I made a proposal for JavaScript called Command Syntax which leans heavily on CQS and FP thinking, but at times in discussions with other JS devs I don't usually sense a deep FP connection. I feel like I'm speaking another language to another community with different cares.