[ghc-steering-committee] GHC stability discussions

[Simon Peyton Jones]

>
> Then maybe we can send #601 back to revision (or reject it), and
> just cherry-pick the ideas that we care about?
>

Let's ask Trevis and Chris, the other two co-authors.  There is significant
interaction with #620.  For example, #620 lets you warn on a use of GADTs,
whereas #601 lets you warn simply about the presence of -XGADTs flag.  (I
am pretty doubtful about the utility of having both.)

I am pretty keen on at least establishing a Stable/Experimental
distinction. I'm agnostic about how many other categories, if any, to
establish.

The “Stability Goals” document is a bit more than just goals, it’s
> actually a pretty clear policy.... But yes, we could ratify them and put
> them somewhere useful. But why not the GHC documentation?
>

I advocate for incorporating them in the GHC Steering Committee process,
not GHC documentation, because the rules GR1-3 primarily cover decisions of
the GHC Steering Committee.  That is, what changes do we accept?  For
example, we might be more willing to accept a breaking change to an
Experimental feature than a Stable one.

Simon


On Mon, 6 Nov 2023 at 08:57, Joachim Breitner <mail at joachim-breitner.de>
wrote:

> Hi,
>
> Am Sonntag, dem 05.11.2023 um 23:03 +0000 schrieb Simon Peyton Jones:
> > It would be good to get some of this decided.
>
> Agreed!
>
> >   Too much is in the air at the moment!  Personally I think:
> >  * We can agree at least the Stable/Experimental part of #601
> >  * The Stability Goals seem fairly uncontroversial.  Shall we just make
> it part of the GHC Proposal process documentation
> >  * #620 is new; your views would be welcome
> >  * The approach to #617 may depend on the outcome of #620.
>
> #601 has more categories than Stable/Experimental, and an enforcement
> mechanism, which seems to be subsumed by #617 and/or #620, is that
> right? Then maybe we can send #601 back to revision (or reject it), and
> just cherry-pick the ideas that we care about?
>
> The “Stability Goals” document is a bit more than just goals, it’s
> actually a pretty clear policy with rules that sound like fresh long
> distance hiking trails (GR…). But yes, we could ratify them and put
> them somewhere useful. But why not the GHC documentation? The
> guarantees written down there, and the definition of “stable package”
> are relevant to the users, not just those writing proposals.
>
> Agreed with parting #670 until #620 is done.
>
> Cheers,
> Joachim
>
>
> --
> Joachim Breitner
>   mail at joachim-breitner.de
>   http://www.joachim-breitner.de/
>
> _______________________________________________
> ghc-steering-committee mailing list
> ghc-steering-committee at haskell.org
> https://mail.haskell.org/cgi-bin/mailman/listinfo/ghc-steering-committee
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.haskell.org/pipermail/ghc-steering-committee/attachments/20231106/e61e0e86/attachment.html>