I don’t believe the practical consequences will be severe. I know you do, but you have provided no proof.
I did an experiment and removed it, and recompiled base. And it hardly needed any changes.
From: Libraries [mailto:libraries-bounces@haskell.org]
On Behalf Of Edward Kmett
Sent: 21 February 2016 23:51
To: Jeremy
Cc: Haskell Libraries
Subject: Re: Haskell Foldable Wast
That leads to a ton of ever MORE nonsensical consequences like not being to weaken calls to mapM (which uses Traversable) to mapM_(which needs only Foldable) or doubling the number of combinators we have all over again for random prescriptive
reasons, right after we just starting finally healing the last source of needless duplication (Applicative not being a superclass of Monad).
-Edward
On Sun, Feb 21, 2016 at 7:57 AM, Jeremy <voldermort@hotmail.com> wrote:
Marcin Mrotek wrote
> I think that, as far as Foldable is concerned, a tuple is equivalent to
> Identity, so this instance is indeed useless. However, Foldable is a
> superclass of Traversable (and it wouldn't make much sense to make these
> classes unrelated, as one can always define folds with `traverse`), so
> I've
> always found it a necessary evil.
Perhaps the case of tuple is evidence that Foldable should *not* be a
superclass of Traversable?
--
View this message in context:
http://haskell.1045720.n5.nabble.com/Proposal-Add-conspicuously-missing-Functor-instances-for-tuples-tp5827530p5830710.html
Sent from the Haskell - Libraries mailing list archive at Nabble.com.
_______________________________________________
Libraries mailing list
Libraries@haskell.org
http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries