Some months ago @Bodigrim posted this on the GHC GitLab instance:
my impression is that GHC team operates under an unwritten assumption that maintainers of core libraries closely follow GHC developments and react proactively. This might have been the case with the previous generation of maintainers, but no longer is true. For the record, I do not follow GHC GitLab, do not read ghc-devs maillist, do not use IRC.
It came to my attention today (out of luck, really). But I think this is a really good point, and one we should pay attention to. Speaking for myself, I have implicitly been under that assumption, though now reading it made explicit, it’s a bit silly.
Without question, GHC and its core libraries will make breaking changes from time to time. I think this is a Good Thing, because it’s a witness to the growth of new ideas. How should we communicate them?
One opportunity and challenge in the Haskell ecosystem is that our means of communication have become much, much more diverse lately. When I started Haskell in 2011, mailing lists were the place to be… indeed about the only place to be. However, since then, a great number of other places to be have sprung up, and this means that communication has become harder. (Of course, all the new places to be have some great benefits – chiefly, that different spaces appeal to different personalities, and thus their existence enriches our community with diversity.)
That said, we probably should agree on some least-common-denominator place for major announcements, etc. If a change requires posting on 10 different platforms, the announcement simply will be skipped. But I think we can be persuaded to choose a place other than the mailing lists – maybe even 2 such places, if there is significant added benefit to the cross-post.
Here, Discourse, seems like a reasonable candidate. (We would probably make a new Category… but maybe not.) It’s accessible to us curmudgeons who love email. I’ve seen impassioned arguments in favor of Discourse from folks who don’t like email, so that’s a point in its favor. It’s open source. But maybe there’s still something better?
Just to restate: my goal is to establish a place where people can subscribe to communications. These communications will include advance notice of big changes to central parts of our ecosystem. By having ideally one such place, both readers and writers of such communication can find each other and, well, communicate.