On Mon, Mar 26, 2012 at 10:18 AM, Jerzy Karczmarczuk <jerzy.karczmarc...@unicaen.fr> wrote: > So, * the addition* is not invertible, why did you introduce rings ...
My intent was to point out that the Num instance that someone suggested for Num a => Num [a] was a bad idea. I talked about rings because they are the uncontroversial part of the laws associated with Num: I think everyone would agree that the minimum you should expect of an instance of Num is that its elements form a ring. In any case, the original question has been thoroughly answered... the right answer is that zipWith is far simpler than the code in the question, and that defining a Num instance is possible, but a bad idea because there's not a canonical way to define a ring on lists. The rest of this seems to have devolved into quite a lot of bickering and one-ups-manship, so I'll back out now. -- Chris Smith _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
