Create 2023-05-11.md

Author: david-christiansen

meetings/2023-05-11.md

@@ -0,0 +1,61 @@
+# TWG 2023-05-11
+
+[Last meeting's notes](https://github.com/haskellfoundation/tech-proposals/blob/main/meetings/2023-04-13.md)
+
+## Present
+
+* John Ericson
+* Laurent R. de C.
+* Jack
+* David Thrane Christiansen
+* Gershom
+
+## Updates on Projects
+
+### Error message index
+
+David: Status quo. Davean has not been contacted to move over the hosting. Has suggested [ANSI hyperlink escapes on error codes](https://gitlab.haskell.org/ghc/ghc/-/issues/23259) to GHC. David will follow up at Zurihac.
+
+### Advisory DB
+
+David: We now have a team to run it, meeting fortnightly and in the process of ramping up
+
+### [Maximally decoupling Haddock and GHC](https://github.com/haskellfoundation/tech-proposals/pull/44)
+
+A Summer of Haskell student has been selected and preparatory work is ongoing.
+
+## [Standard Library Reform](https://github.com/haskellfoundation/tech-proposals/pull/47)
+
+
+John: GHC developers seem very interested, CLC less so. Last time, we talked about planning a synchronous meeting with Bodigrim to address his concerns, but he wasn't interested in that way forward. Will be at Zurihac, perhaps we can fix this there.
+
+David: suggestion to gather volunteers at Zurihac to assemble a merge request. A concrete patch which leaves the `base` interface untouched might be enough to convince the CLC to move forward.
+
+Gershom: CLC 146 is moving toward a successful conclusion. It looks like we're also getting good side effects, like not exporting certain modules. Doesn't currently know where the CLC majority stands, but it seems that they're split on the issue. What if we make a list of each CLC member, go through the thread, figure out where each stands based on recent comments, and then move from there? After all, this is a majority decision, not a consensus one. It seems that PVP compliance is very important.
+
+John: Having an adversarial rhetorical stance is not a way forward. We should find a clear path that everyone can follow.
+
+Gershom: GHC and Cabal are implementing multiple libraries within a single package. There, you can version them separately but distribute them together. It seems to be a compelling argument that `ghc-base` + `base` is in alignment with the ecosystem as a whole.
+
+John: If Backpack ever happens, then we'll have similar issues.
+
+Laurent: Many of us will be at Zurihac, and things might get solved here. It will be worrying if we don't get this done by next meeting though. What about an approach where we just do the split, in a way that doesn't change the API of `base`, and then start sending proposals to CLC for removing re-exports form `base`?
+
+Gershom: That is basically what the proposal is, really. It would be socially unpleasant, but other than that OK.
+
+Laurent: At what point do we do this?
+
+Gershom: Let's do this at Zurihac if we can.
+
+John: will try to have more conversations.
+
+Gershom: Because it's a limited list of people, think about what you know about what each values, then ask about their concerns and return later with a tailored pitch that matches these concerns.
+
+
+
+## Other
+
+
+### Next Meeting (Zurihac/workshop overlap)
+
+We decided to move it by one week, to 15 June, so that we can follow up on Zurihac.