[ghc-steering-committee] Urgent: exension life cycle proposal

[Moritz Angermann]

Adam,

How can I support a proposal that cements a
already firmly established practice, I fundamentally disagree with?

Every extension in a stable GHC release by extension has to be stable,
maybe deprecated. But stable. This proposal basically says it is ok to have
unstable extensions in a stable GHC release. I do not believe that
-WXUnstable should exist in a stable compiler. And from an end user
perspective I think there can only be stable and deprecated extensions.

GHC is already a highly complex piece of software from an end user
perspective, let’s try to make it simpler not add even more complexity.

I have no intentions to turn this proposal into
a radical change to release practices. I have provided most of this for
context as to why I can not see myself supporting this proposal as is.

I can support a subset of the proposal. I do not see how I can support a
this proposal in its entirety?

Best,
  Moritz

On Sun, 3 Sep 2023 at 4:59 PM, Adam Gundry <adam at well-typed.com> wrote:

> On 03/09/2023 01:15, Moritz Angermann wrote:
> >
> > The practice that we put experimental/unstable features into our stable
> > releases is something this proposal seems to cement, which is why I’m so
> > much against it in this form.
>
> But that practice is already firmly established (de facto), it's not
> something being introduced by this proposal! So while I understand your
> opposition to *the practice*, I'm struggling to understand the strength
> of your opposition to *the proposal*.
>
> You're welcome to make the argument (in a subsequent proposal, perhaps)
> that GHC should more firmly gate access to unstable features. If
> anything, that would seem easier to specify once the proposal's approach
> clearly defines those features. And in the interim, at least the
> proposal makes it possible to switch off unstable features systematically.
>
> So I understand that this proposal doesn't get you everything you want.
> But it seems like a reasonable interim step, and one that is more likely
> to command broad consensus than a radical change to release practices?
>
> Best wishes,
>
> Adam
>
>
> --
> Adam Gundry, Haskell Consultant
> Well-Typed LLP, https://www.well-typed.com/
>
> Registered in England & Wales, OC335890
> 27 Old Gloucester Street, London WC1N 3AX, England
>
> _______________________________________________
> 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/20230903/7c71b23f/attachment.html>