I would totally support that if PolyKinds was implemented without requiring TypeInType, but with the current design for me this is a solid `no`. On Thu, Dec 3, 2020 at 9:48 AM Joachim Breitner wrote:

Hi,

Am Donnerstag, den 03.12.2020, 16:40 +0000 schrieb Richard Eisenberg:

...

Vote for PolyKinds!

convinced.

Cheers, Joachim -- Joachim Breitner mail@joachim-breitner.de http://www.joachim-breitner.de/

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee

I find the distinction between types and kinds helps me understand programs better. On Thu, Dec 3, 2020 at 9:53 AM Simon Peyton Jones wrote:

I would totally support that if PolyKinds was implemented without requiring TypeInType, but with the current design for me this is a solid `no`.

Maybe you can say why? Solid sounds… solid.

Simon

*From:* ghc-steering-committee *On Behalf Of *Iavor Diatchki *Sent:* 03 December 2020 17:51 *To:* Joachim Breitner *Cc:* ghc-steering-committee *Subject:* Re: [ghc-steering-committee] a plea for PolyKinds

I would totally support that if PolyKinds was implemented without requiring TypeInType, but with the current design for me this is a solid `no`.

On Thu, Dec 3, 2020 at 9:48 AM Joachim Breitner wrote:

Hi,

Am Donnerstag, den 03.12.2020, 16:40 +0000 schrieb Richard Eisenberg:

...

Vote for PolyKinds!

convinced.

Cheers, Joachim -- Joachim Breitner mail@joachim-breitner.de http://www.joachim-breitner.de/ https://nam06.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.joachim-breitner.de%2F&data=04%7C01%7Csimonpj%40microsoft.com%7Ca7e57ca1aaa04682c10e08d897b416c6%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C1%7C637426147016852738%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=5BRa1m9VmGYrpfPd%2BCU6kuR2mRMwnGHL56rKmYDevag%3D&reserved=0

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.haskell.org%2Fcgi-bin%2Fmailman%2Flistinfo%2Fghc-steering-committee&data=04%7C01%7Csimonpj%40microsoft.com%7Ca7e57ca1aaa04682c10e08d897b416c6%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C1%7C637426147016862731%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=ajZ4O2lU0DtkM63hKuG6skeitXdkFtYZ6xiklAKzEms%3D&reserved=0

Oh? Perhaps I am confused, if we can have PolyKinds without TypeInType, I'd be happy to change my vote. My understanding was that basically `TypeInType` got renamed to `PolyKinds` but I might be confused. -Iavor On Thu, Dec 3, 2020 at 11:50 AM Alejandro Serrano Mena wrote:

But PolyKinds does not require TypeInType, according to the docs [ https://downloads.haskell.org/ghc/latest/docs/html/users_guide/glasgow_exts....] it is in fact the other way around: TypeInType implies PolyKinds. Of course, the fact that kinds and types are unified may leak somehow, but I have a hard time figuring out an example of such.

El jue, 3 dic 2020 a las 18:51, Iavor Diatchki () escribió:

...

I would totally support that if PolyKinds was implemented without requiring TypeInType, but with the current design for me this is a solid `no`.

On Thu, Dec 3, 2020 at 9:48 AM Joachim Breitner wrote:

...

Hi,

Am Donnerstag, den 03.12.2020, 16:40 +0000 schrieb Richard Eisenberg:

...

Vote for PolyKinds!

convinced.

Cheers, Joachim -- Joachim Breitner mail@joachim-breitner.de http://www.joachim-breitner.de/

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee

I agree with Richard as well on this point. The error messages GHC reports can be practically incomprehensible and even misleading when PolyKinds is accidentally turned off in a module that imports and attempts to use poly-kinded things. I could probably drum up an example if people want (I've run into this problem several times, but don't have one immediately on hand.) If it were just switched on by default, I'd feel a lot better about the extension, because I wouldn't have to worry that consumers of any poly-kinded things I wrote had to be aware of these stumbling blocks. Of course, if GHC could get better at suggesting that users should turn PolyKinds on in the cases where it's an issue, that would also be great, but I'm not quite sure how to approach that either. - Cale On Thu, 3 Dec 2020 at 12:48, Joachim Breitner wrote:

Hi,

Am Donnerstag, den 03.12.2020, 16:40 +0000 schrieb Richard Eisenberg:

...

Vote for PolyKinds!

convinced.

Cheers, Joachim -- Joachim Breitner mail@joachim-breitner.de http://www.joachim-breitner.de/

_______________________________________________ ghc-steering-committee mailing list ghc-steering-committee@haskell.org https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee

On Dec 4, 2020, at 3:18 AM, Spiwack, Arnaud wrote:

Is it even possible while still supporting Haskell2010?

My hunch is "yes". I can imagine trouble with PolyKinds, but each example requires one of NoPolyKinds (in another module), ScopedTypeVariables, or KindSignatures. And each case of trouble is obscure. Richard