3 Oct
2012
3 Oct
'12
5:25 p.m.
On Wed, Oct 3, 2012 at 9:26 PM, Robin KAY
On 03/10/12 14:20, Edward Kmett wrote:
I don't think anyone has proposed weakening parametricity in this way.
[snip]
I don't think so either, but is there any reason it shouldn't be done?
Uh, because it weakens parametricity? :) Free theorems would promptly vanish, since we would have effectively added typecase to the language: it would no longer be possible to conclude from a type signature a -> a that the function was the identity, for example.