#15078: base: Customary type class laws (e.g. for Eq) and non-abiding instances (e.g. Float) should be documented -------------------------------------+------------------------------------- Reporter: sjakobi | Owner: Azel Type: feature request | Status: patch Priority: normal | Milestone: 8.6.1 Component: Core Libraries | Version: 8.4.2 Resolution: | Keywords: newcomer Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Phab:D4736 Wiki Page: | -------------------------------------+------------------------------------- Comment (by dfeuer): Expanding a bit in another direction, I think it probably makes sense to note that `Num` instances are typically expected to form rings. In fact, the `signum` and `abs` functions suggest even more structure, although I don't personally know what sort of structure that ''is''. I'd be very hesitant to document any expectations beyond the ring laws, because there are a number of exotic `Num` instances in the wild. The best-behaved law-breaker here is `Natural`, which does not have a sensible `negate` or `(-)`; it's a perfectly good semiring (more specifically, a ''rig''), but we don't have a class corresponding to that. As usual, `Float` and `Double` are truly flagrant law-breakers. It might also be worth mentioning that types that are well-behaved instances of both `Num` and `Ord` are ''not'' necessarily ordered rings. In particular, `Int`, `Word`, `Int8`, `Word8`, etc., are not ordered rings. As far as I know, the only `Num`+`Ord` instances in `base` that implement ordered rings are `Integer` and `Rational`. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/15078#comment:6 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler