[ghc-steering-committee] #536: Type-level literals as a separate language extension

[Joachim Breitner]

Hi,

thanks for the well thought through message and helpful
conceptualisation. I agree that it would be good if the Committee as a
whole would commit to one or the other paradigm.

Am Mittwoch, dem 18.10.2023 um 18:30 +0200 schrieb Vladislav Zavialov:
> (Pardon a digression, but this also calls into question whether 448
> is justified in splitting ScopedTypeVariables
> into ExtendedForAllScope, MethodTypeVariables, PatternSignatures.
> Maybe we don't want fine-grained extensions after all)

my understanding with splitting ScopedTypeVariables is not to let a
thousands (or 2³ in this case) flowers bloom, but to slice it into the
parts we want to keep and the parts we (or some of us) want to change.
Maybe not absolutely necessary, as we could have -XTheNewWay with some
overlapping functionality with ScopedTypeVariables, such as annotating
patterns with their types, but maybe even if it only serves to help
communicating the changes it maybe worth it.

Cheers,
Joachim

-- 
Joachim Breitner
  mail at joachim-breitner.de
  http://www.joachim-breitner.de/