[ghc-steering-committee] Proposal #273: Local Types, Recommendation: needs revision

[Richard Eisenberg]

I mis-summarized my GitHub comment. I agree that the local modules proposals do not cover `reflection` and brought up a strawman counterproposal just to support `reflection` (and similar usages of classes); the local modules proposals cover the rest of the "would be nice" aspect of local types, in my opinion.

Richard

> On Nov 25, 2019, at 1:44 PM, Eric Seidel <eric at seidel.io> wrote:
> 
> On Mon, Nov 25, 2019, at 05:38, Richard Eisenberg wrote:
>> I have posted on the ticket. The proposal text as it now stands does 
>> not support the idea of "just in the renamer", and I find the 
>> motivation lacking. Do we need all this power just to write 
>> `reflection`? Seems like overkill.
>> 
>> Yes, I agree that local types would often be nice. But would that 
>> nicety be covered by local modules (either proposal)?
> 
> The motivating example here seems like it would require parameterized modules, ie something akin to ML functors. Neither local module proposal currently includes parameterized modules.
> _______________________________________________
> ghc-steering-committee mailing list
> ghc-steering-committee at haskell.org
> https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee