Tony Morris

@dibblego Anyway if we agree "pure" is meaningless here and it's a term used only informally without a formal treatment we have no quarrel

@takeoutweight it is a synonym for a formally described concept. No I don't agree. You are misusing way too many terms to solicit agreement.

@dibblego OK np. Just saying monads were intro'd to CS by Moggi precisely to give denotational semantics (+ eq reasoning) to mutable refs.

@dibblego Probably trying to fit too many points into a 140 char quip. I just want to suggest "impure" things can have good eq reasoning too

@takeoutweight Note again that "things which are (im)pure" are called expressions. Not state, not monads, not languages, not mutable refs.

@dibblego This equivocation is useless w/o a real treatment of the term "pure." I know the informal usage in FP; its fine--just not formal

@takeoutweight Pure is simply a property of expressions where equational reasoning is preserved. Many people, including you, abuse this.

@dibblego ? My straw man is "State is pure because it's in haskell." To you, pure is wrt expressions, thats fine. I offer no defn for "pure"