Sean Leather wrote:
That's interesting. So, maybe there should be some language extension or
warning (with associated -fno-warn) for this in GHC.
Personally, I prefer the way it's done now. (I guess that's obvious,
considering I'm developing a library that will take advantage of it. ;) )
But it makes sense that instances are only looked up when needed, instead of
globally tracked.
see: http://hackage.haskell.org/trac/ghc/ticket/2356
_______________________________________________
Glasgow-haskell-users mailing list
Glasgow-haskell-users@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
