This may be an infelicity in defaulting (which I'm not terribly familiar with)
[...]
I will close this ticket, but do please reopen if you can characterize
#8155: Defaulting bug or unfortunate error message with closed type families -------------------------------------+------------------------------------ Reporter: nh2 | Owner: Type: bug | Status: closed Priority: normal | Milestone: Component: Compiler | Version: 7.6.3 Resolution: invalid | Keywords: Operating System: Unknown/Multiple | Architecture: Unknown/Multiple Type of failure: None/Unknown | Difficulty: Unknown Test Case: | Blocked By: Blocking: | Related Tickets: -------------------------------------+------------------------------------ Comment (by nh2): Replying to [comment:1 goldfire]: the bug as a specific infelicity in the defaulting mechanism. I know too little about how defaulting works, we would need somebody who understands that.
This is all very subtle and can be unexpected, but after lots and lots of thinking, Simon PJ, Dimitrios V, and I could not come up with a better way here
I am not saying that I find this behaviour of closed type families wrong or so. Is it not possible to change the generated error message to give a hint why this is rejected, e.g. based on what you explained here? I understand your reasoning here, but had I not seen your example here and would I encounter this message in GHC with the current error message, chances are high I'd think it's a bug. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/8155#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler