#13408: Consider inferring a higher-rank kind for type synonyms -------------------------------------+------------------------------------- Reporter: goldfire | Owner: (none) Type: feature request | Status: closed Priority: normal | Milestone: Component: Compiler | Version: 8.0.1 Resolution: wontfix | Keywords: TypeInType Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Changes (by goldfire): * status: new => closed * resolution: => wontfix Comment: Actually, this is all bogus. In `type G = F`, `F` is undersaturated, which is disallowed. And once we write `type G z = F z`, when we're inferring a higher-rank type for `z`, we've gone further off the map than I'm comfortable with. Additionally, getting this right would require tracking recursiveness in type declarations more than we do already, and it's not worth the pain. So I'm closing as wontfix. Anyone who wants this can always add a kind signature. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/13408#comment:2 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler