Clojurians Log v2

or in the end how sci implements it is the same somehow?

it's not a dumb question, but I don't think it matters, there is no interpreter overhead for lazy seqs vs transducers I think, since the only part that's interpreted is the higher order function argument and this is executed the same number of times with lazy seqs vs transducers

oh interesting

I built a clojure-lsp custom linter that is using sci, and did some perf improvements using transducers, and noticed is the same perf, so I was wondering if it was worth it

you should just optimize for transducers for the same reasons as in clojure, not because of SCI

makes sense

that was the idea indeed, optimize the coll iterations