Guile and Racket are the two schemey languages I think are most "production ready". I'm still very interested in "whatever Scheme-in-the-browser" seems most viable first.
I'm hoping Guile does that soon, but I'd also take any r7rs scheme.
Scheme.Wasm and Schism are promising.
@cwebber Andy Wingo has been working a lot on Wasm lately, and he has dropped in the idea that Guile should work in the browser. So fingers crossed.
(Btw he just fired off three very interesting blogposts garbage collection - another part of Guile he's working on. They went waaay over the top of my head.)
@cwebber Oh, thanks! That might have been the source I was talking about, but I didn't really get deep into it. I'll check it out - every Wingo post or talk is worth taking long notes and thinking a lot about what's at stake.
@cwebber @acousticmirror would that make wasm guile's first compile target?
@meena @acousticmirror I'm not sure what you mean? Guile is a compiled language.
@cwebber i think that opinion formed based on answers from this question https://cathode.church/@meena/108401370988536048
@meena Ah. Well, it's still a compiled language... though it doesn't produce standalone static binaries on its own, that's true. (It does use the ELF format though...!)
However, I think Wingo's blogpost "Webassembly: the new Kubernetes?" is exactly what you mean https://wingolog.org/archives/2021/12/13/webassembly-the-new-kubernetes
@technomancy @cwebber I think i did the old mistake of conflating a single executable with compilation
@technomancy @meena I'm shipping the SICP metacircular evaluator as-is THANK YOU
@acousticmirror Yup, Wingo and I have talked about this. He agrees the WASM stuff is important but is more personally excited about the GC stuff atm. But the GC stuff might help the WASM compilation effort, is my understanding.
Wingo's talk "Compiling Webassembly: Notes from the Trenches" is a spectacular watch, btw. https://archive.fosdem.org/2021/schedule/event/webassembly/