Hello @johnwright. This JSON library looks interesting! I'll read its paper, but I was wondering about its tradeoffs too (perf, size, theoretical blockers, etc.)?
The idea of having a free decoder from an encoder or vice-versa sounds pretty attractive, and if possible, I'd like the official Reason one to be free of ppx (which is needed mostly because of the usual encoding/decoding boilerplate).