Re: [ghc-steering-committee] #285: -XNoImplicitForAll, recommendation: accept

Dear committee: Proposal #285 is currently under consideration: https://github.com/ghc-proposals/ghc-proposals/pull/285 I previously recommended acceptance, and there was some followup discussion on the GitHub thread where Simon PJ and Arnaud requested changes. The changes, namely clarifications and more examples, have all been implemented and those that requested changes have signed off. I think we can accept the proposal now. Any final concerns? Cheers Simon On Wed, 13 Nov 2019 at 01:18, Sandy Maguire wrote:

I'm largely in agreement with Arnaud here.

On Wed, Nov 13, 2019 at 1:59 AM Spiwack, Arnaud wrote:

...

I've left comments on the Github thread.

I'm rather sympathetic to this idea. I personally think that the implicit binding rules that we have in place are a bit messy, so simply turning them off may be of help.

My Github comments raise two points:

- The proposal is a tad challenging to read, and I'd like some improvements before acceptance be considered - I'm questioning whether we really want two extensions for these two very related behaviours

On Mon, Nov 11, 2019 at 10:44 AM Simon Marlow wrote:

...

Proposal #285:

https://github.com/ghc-proposals/ghc-proposals/pull/285

suggests two new LANGUAGE pragmas -XNoImplicitForall and -XNoPatternSignatureBinds.

They have the effect of making a subset of existing programs illegal, and in that sense I don't think these are problematic. The intention is that these would be used with ExplicitForalls to enforce that all identifiers have explicit binding sites. I'm generally supportive of that.

However I think we should be wary of the potential future direction, mentioned in the proposal:

...

As described in the motivation, this opens the door to other means to bind the previously implicitly bound variables.

i.e. an unbound name in a type could refer to a type variable bound in an enclosing scope. This would be a language *change*, not merely an addition or subtraction, and hence potentially fork-like. So we would want to consider very carefully whether that's the direction we want to take the language.

But since that isn't part of the current proposal, and the current proposal is merely an optional subtraction, I don't think it's controversial.

Cheers Simon

On Thu, 7 Nov 2019 at 09:29, Joachim Breitner wrote:

...

Dear Committee,

this is your secretary speaking:

-XNoImplicitForAll has been proposed by John Ericson https://github.com/ghc-proposals/ghc-proposals/pull/285

https://github.com/Ericson2314/ghc-proposals/blob/no-implicit-forall/proposa...

I propose Simon Marlow as the shepherd.

Please reach consensus as described in https://github.com/ghc-proposals/ghc-proposals#committee-process I suggest you make a recommendation, in a new e-mail thread with the proposal number in the subject, about the decision, maybe point out debatable points, and assume that anyone who stays quiet agrees with you.

Thanks, 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

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

-- I'm currently travelling the world, sleeping on people's couches and doing full-time collaboration on Haskell projects. If this seems interesting to you, please consider signing up as a host! https://isovector.github.io/erdos/