> It seems to me like monads must be the logical conclusion to this style of programming, or else you wind up with a mess (or just abandoning this technique.)

> It seems to me like monads must be the logical conclusion to this style of programming, or else you wind up with a mess (or just abandoning this technique.) Monads are nice, but there's a lot of work on algebraic effects/effects in general which may pan out into something useful (and more general).