On Wed, 2011-09-21 at 00:04 +0200, Ketil Malde wrote:
If Haskell defined list syntax in terms of something that's not called Enum, that would be fine. Renaming is never all that big a deal. But the list sugar is a big deal, and I don't think there's any point at all in leaving the list sugar associated with something as minor as building a representation of the inaccuracy of your approximations.
I must admit I don't understand this comment. If the fixpoint library wants to provide the functionality (producing all values between two points), and can't/shouldn't use Enum, surely it must provide a different function, and let go of the list sugar?
Sorry to be unclear. I mean that instead of removing a useful instance, if people find the use of Enum for Float to be objectionable, then perhaps (via language extensions, deprecation, all the usual backward compatibility slow-change stuff) the desugaring of list ranges should be changed to not use something with a name you'd object to, rather than just removing the feature. In any case, as long as Enum *is* the backing for list desugaring, it seems like a mistake to define instances that are completely unuseful for list desugaring. -- Chris