felko 2017-03-08 08:47:40
Tuplanolla: ok it doesn't matter anyway, i can do what i want without, thanks
Tuplanolla 2017-03-08 08:49:42
I recall it's there to keep type-level lambdas out of the system.
Tuplanolla 2017-03-08 08:50:09
Those would have troublesome computability properties.
felko 2017-03-08 08:52:03
Tuplanolla: why ?
Tuplanolla 2017-03-08 08:52:29
Lambda calculus is Turing complete.
felko 2017-03-08 08:53:27
i agree that a type-level lambda calculus could lead to haskell black magic but is it problematic ?
Tuplanolla 2017-03-08 08:54:40
That depends on your view of problematic. It certainly is if you want to keep compilation decidable.
felko 2017-03-08 08:55:42
ok i understand, thanks
unclechu 2017-03-08 09:05:05
lyxia: thanks a lot for the help! now everything is green and i'm happy because i have completed everything i'd planned for today: https://hackage.haskell.org/package/qm-interpolated-string
lyxia 2017-03-08 09:05:43
unclechu: nice! you're welcome!
lyxia 2017-03-08 09:07:33
I'm pretty sure we already have Turing-completeness in types.
monochrom 2017-03-08 09:07:48
What does "qm" stand for there?
lyxia 2017-03-08 09:07:49
At least we have UndecidableInstances
byorgey 2017-03-08 09:19:44
monochrom: from the readme: 'm' in 'qm' means 'multiline'.
byorgey 2017-03-08 09:19:54
presumably the 'q' stands for 'quote' or 'quasiquote' or something like that
