Fork me on GitHub

You can do fine using some js interop and some Google Closure stuff. Unless you use some kind of virtual dom, figwheel probably won't work, that was my biggest miss.


@dmarjenburgh: > Svelte can bring in a lot of dependencies and leave out the modules you don’t use in the build - The Closure advanced compilation will optimize it better ClojureScript will always bring with it the runtime, won’t it?


Regardless, I would love to be able to use something like Svelte from ClojureScript. Even if we wouldn’t be able to get all the way in terms of ergonomics, that performance…


If you want performance it may be time to have a cljs that compiles to WASM.

David Pham07:10:01

I thought WASM would only help for numerical heavy tasks

👍 4

is it appropriate to think of row polymorphism as a sub-set of Clojure's brand of dynamic typing?


it seems closest to associative destructuring (at compile time?), with Clojure you get full on do-what-you-want dynamism, and sequential destructuring as well


I’ve heard that javascript actually performs better than wasm for small array sizes and typical business logic code execution paths. Partly because it’s very new still no doubt.


Yes and no, WASM can do more and more 'on it's own' dom manipulation goes through js interop, is probably a bit slower because of that. In rust there are some frameworks to build the entire page with WASM.


@michael.e.loughlin row polymorphism is a typing discipline. Clojure has no typing discipline.


What Svelte does is dealing with the root cause, stop doing unnecessary work. Offloading the unnecessary work to WASM doesn't feel right to me. 😄 The Svelte approach + WASM, OTH...


Sounds like someone should open a "research repo" for people to toss their ideas and discoveries into the ring and see which one people like the most. Like the most open of open source software, or competitive open source... Though I'm pretty sure I'm reinventing a wheel here.

👍 8