#14266: AllowAmbiguousTypes doesn't play well with default class methods -------------------------------------+------------------------------------- Reporter: chris-martin | Owner: (none) Type: bug | Status: new Priority: normal | Milestone: Component: Compiler | Version: 8.0.2 Resolution: | Keywords: Operating System: Unknown/Multiple | Architecture: | Unknown/Multiple Type of failure: None/Unknown | Test Case: Blocked By: | Blocking: Related Tickets: | Differential Rev(s): Wiki Page: | -------------------------------------+------------------------------------- Comment (by RyanGlScott): Replying to [comment:7 chris-martin]:
I suppose that's something I'm not clear on. What is the language designers' attitude toward the `TypeApplications` extension? Is it there to be an imperfect workaround for the occasional odd situation, or is Haskell now intended to be a language where programming with ambiguous types and explicit type applications is normal and fully supported? It still //feels// a bit like I'm abusing the type system.
I can only speak as a single GHC developer, but I would certainly hope that `TypeApplications` is just as "fully supported" as anything else you can do in GHC (and I suspect I'm not alone in this regard). The somewhat unfortunate reality is that this extension is fairly new in the Haskell ecosystem, and as a result, there's been less time to discover odd interactions between it and other language extensions (of which this bug is an example). With the help of intrepid programmers like yourself, I think we'll find a way to stamp out these oddities, and hopefully make `TypeApplications` feel less like "abusing" the type system. -- Ticket URL: http://ghc.haskell.org/trac/ghc/ticket/14266#comment:8 GHC http://www.haskell.org/ghc/ The Glasgow Haskell Compiler