Functor, Applicative, Monad, Foldable, Traversable instances for (, , ) a b

[Bryan Richter]

On Thu, Mar 30, 2017 at 08:30:32PM +0200, Henning Thielemann wrote:

> On Thu, 23 Mar 2017, Fumiaki Kinoshita wrote:
>
> >It's surprising that they are missing (forgive me, I'm not here to
> >make people grumpy).
>
> I am not surprised because it was discussed at length a year before. I
> still think all these instances on pairs, triples and other tuples are
> more dangerous than helpful. It is so easy and much more expressive to
> define custom data types for your particular application.

I can't help but feel the same way. I find it detrimental that pairs have
a Functor instance. I don't care that 'length' is a member of Foldable,
or that anything that is a Functor can/should be Foldable —— I just
don't know why pairs ever got to Functor. I fear this was an ancient
decision... no?

My intuition, based on observed usage in Haskell and elsewhere, is that
tuples are used as anonymous product types. Thus, I am interested in
the rationale for making them Functors because I feel it has denied me
a common, valuable tool. I no longer feel that I should use pairs as
product types. By making "(,) a" a Functor, it is no longer an anonymous
product type —— it is an anonymous Reader. Are anonymous Readers
really all that much more valuable than anonymous product types? I cannot
conceive it. I must be missing something.

And certainly somebody might argue, "Just keep using them as product
types. Nothing has changed." But of course something has changed. My
product now has a sharp edge. There is some extra information I have to
hold in my head: "Don't put this in a Functor context and expect to be
able to follow the compiler errors!"

If I wanted to hold extra information in my head, surely I would go back
to using Perl.

I fear that pairs became Functors a long, long time ago, and now we're
stuck with it. Is that true? If so, I believe that crucial fact should
preface any argument about 'length', and hopefully we could agree the
root cause of `length (a,b) == 1` —— if it is to be considered a
problem —— is the fact that `fmap f (a,b) == (a, f b)`.

For whatever my vote is worth as somebody posting to this list for the
first time, 

-1.

Thanks,

-Bryan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: Digital signature
URL: <http://mail.haskell.org/pipermail/libraries/attachments/20170330/443a7f73/attachment.sig>