SWG 2023-01-09

Meeting URL: https://meet.jit.si/StableHaskellMeetBiWeekly

Previous meeting notes

Agenda & Notes

New items

CLC has open call for participation.
Safe Haskell
- Does anyone in the group see a particular reason to fight for it to continue? (no)
- The cost: libraries feel obligated to use it, in case a user might want it. This increases the cost, and safety inference is brittle.
- E.g. new contributors try to "do the right thing", and Safe Haskell annotations are a footgun
- Should we just announce not to use it, rather than get rid of it?
- Arguments for removal:
  - It is broken: https://gitlab.haskell.org/ghc/ghc/-/issues/21705, https://gitlab.haskell.org/ghc/ghc/-/issues/20078
  - It has poor usability: https://gitlab.haskell.org/ghc/ghc/-/issues/19590
  - It interacts badly with GHC2021: https://gitlab.haskell.org/ghc/ghc/-/issues/19605
  - It can break in non-obvious ways and incurs maintenance overhead: https://gitlab.haskell.org/ghc/ghc/-/issues/19002
  - Its claimed promises are broken by GNTD, a ubiquitous and useful extension: https://gitlab.haskell.org/ghc/ghc/-/issues/8827
  - It poses a small maintenance overhead on GHC and tooling developers: https://gitlab.haskell.org/ghc/ghc/-/issues/22728, https://gitlab.haskell.org/ghc/ghc/-/issues/19604
- Perhaps we can begin by socially making it less prominent, e.g. by having Haddock output not prominently present it?

Updates on in-progress work

GHC API discussed previously
- Previously holding the token: David
- Moving, but slowly
What is the stable API of base
- Schedule Zurihac project?
  - No longer needed
- Preliminary result of discussions: All of base is a stable API and will use CLC to make changes (that might consist of extracting guts from base - no complete conclusion is found, but a crowdsourced effort is premature)
Update on Language/compiler features to help stability, specifically #4879
- Previously holding the token: David + Simon
- Seeking volunteers for #4879 (deprecating exports); none yet; think about HF funding for this
  - Have a lead on a potential volunteer.
- Are there any other tasks for which modest HF funding would unlock real value?
  - Blocked on this currently.
Bulletin discussed previously
- Previously holding the token: Chris
- No progress currently
GHC warning policy document as discussed previously
- Previously holding the token: Chris
- No progress currently, deprioritized relative to bulletin.
Creating and maintaining a set of hlint rules to promote stability
- Previously holding the token: Trevis
- Opened issue asking if we can write rules for typeclasses.

Parked for expected absence of token holder

For next week

Look over John Ericson's base reform proposal: haskellfoundation/tech-proposals#47