Mario

That's not my decision, but I don't expect so.

I like `-XIncompleteRecords` the best, as far as naming goes. But if there is any name, it's for a command-line option and LANGUAGE pragma extension, right? I'm against breaking backward...

The problem appears to be in function `datatypeEqualW3C` in module [DataTypeLibW3C](http://hackage.haskell.org/package/hxt-relaxng-9.1.5.5/docs/Text-XML-HXT-RelaxNG-XMLSchema-DataTypeLibW3C.html). The data-type map `norm` in there is either used in a wrong way, or it should map all missing...

> Btw, here's the original proposal from the Prime wiki with which may save us some time later: https://prime.haskell.org/wiki/ExplicitForall Thank you. The present proposal can be seen as a condensed...

> The reason instance forall exists is that it gives a canonical place to put kind signatures on type variables. Ah. That did not occur to me. I'm not sure...

I don't think Haskell2020 necessarily has to include `-XAutoDeriveSuperclasses` together with AMP. For teaching purposes, would it not be enough for GHC to support the former feature?

The canary RFC https://github.com/haskell/rfcs/pull/17 is on its way to become the first accepted and merged proposal. Once it flies through the complete process, I think this one could be next....

I know, but we shouldn't let the perfect be the enemy of the good. I think we can agree that bringing the report up to sync with GHC 8.0 would...

> What’s the challenge to having return = pure and leaving that report language as is? Just the proof reading ? That would certainly be the easiest way to do...

I'm glad that the passions have proven still alive, but can we please narrow the scope of this discussion a bit? *This* particular proposal is merely to add the `Applicative`...