
Dear all, Richard has decided to stop working on the local module proposal [ https://github.com/ghc-proposals/ghc-proposals/pull/283#issuecomment-9743823... ]. Just as I was about to say that we ought to accept it unless a strong opposition was voiced within the next two weeks. What's left to do, as far as anyone can tell, is address two minor technical points (there may be other corner cases that have been missed: it's an ambitious proposal after all; but it's looking like it's standing on firm ground). I've stated previously that I care a lot about this proposal. Unfortunately, I currently don't have time to take care of the proposal myself. I may revisit this decision in 2022 (which, fortunately, is rather soon), depending on how my schedule changes. But until then, unless someone steps up, I'll mark the proposal as dormant. Best, Arnaud

In my mind the trouble is that it's a "big" proposal, with far-reaching consequences. I too am strongly attracted by it. But I am also a bit daunted by the engineering costs of adopting it. (E.g. we have not begun to talk about Backpack or HLS.)
So, much as like it, I'm not upset about it going to sleep for a bit.
If some sponsor wanted to adopt it, and resource the development of the proposal and its implementation, as Tweag did for linear types (another "big" proposal), that would shift the calculus a bit. We could wait to see if/when that happens.
Simon
PS: I am leaving Microsoft at the end of November 2021, at which point simonpj@microsoft.commailto:simonpj@microsoft.com will cease to work. Use simon.peytonjones@gmail.commailto:simon.peytonjones@gmail.com instead. (For now, it just forwards to simonpj@microsoft.com.)
From: ghc-steering-committee

On Nov 22, 2021, at 4:28 AM, Simon Peyton Jones via ghc-steering-committee
wrote: If some sponsor wanted to adopt it, and resource the development of the proposal and its implementation, as Tweag did for linear types (another “big” proposal), that would shift the calculus a bit. We could wait to see if/when that happens.
Yes, that seems about right. Though I originally thought I would get the proposal through the door, I knew I would not be the implementor, so the job was never going to be finished by me, anyway. Perhaps it makes sense for someone committed to do the implementation to finish it off. The key thing for me is that there were several other proposals in this space, and I just thought there was a better design. I do think the design there is mostly good, with details to be filled in later. Thanks, Richard
participants (3)
-
Richard Eisenberg
-
Simon Peyton Jones
-
Spiwack, Arnaud