Establishing a Scheme registry Lassi Kortela (31 Jul 2020 08:14 UTC)
Re: Establishing a Scheme registry Marc Nieper-Wißkirchen (31 Jul 2020 08:39 UTC)
Re: Establishing a Scheme registry Lassi Kortela (31 Jul 2020 08:49 UTC)
Prior art: SRFI 97 Lassi Kortela (31 Jul 2020 08:59 UTC)
Re: Prior art: SRFI 97 Marc Nieper-Wißkirchen (31 Jul 2020 09:18 UTC)
Re: Prior art: SRFI 97 Marc Nieper-Wißkirchen (31 Jul 2020 09:20 UTC)
Re: Prior art: SRFI 97 Lassi Kortela (31 Jul 2020 09:39 UTC)
Re: Prior art: SRFI 97 Marc Nieper-Wißkirchen (31 Jul 2020 09:58 UTC)
Re: Prior art: SRFI 97 Lassi Kortela (31 Jul 2020 10:13 UTC)
Re: Prior art: SRFI 97 Marc Nieper-Wißkirchen (31 Jul 2020 10:18 UTC)
Python PEPs Lassi Kortela (31 Jul 2020 10:23 UTC)
Re: Python PEPs Marc Nieper-Wißkirchen (31 Jul 2020 11:12 UTC)
Re: Python PEPs Lassi Kortela (04 Aug 2020 07:04 UTC)
Re: Python PEPs hga@xxxxxx (04 Aug 2020 09:28 UTC)
Re: Prior art: SRFI 97 Marc Nieper-Wißkirchen (31 Jul 2020 13:31 UTC)
Re: Establishing a Scheme registry Marc Nieper-Wißkirchen (31 Jul 2020 09:13 UTC)
Re: Establishing a Scheme registry John Cowan (01 Aug 2020 03:49 UTC)
Re: Establishing a Scheme registry Marc Nieper-Wißkirchen (01 Aug 2020 06:29 UTC)
Re: Establishing a Scheme registry John Cowan (01 Aug 2020 13:19 UTC)
Re: Establishing a Scheme registry Marc Nieper-Wißkirchen (01 Aug 2020 13:48 UTC)
Re: Establishing a Scheme registry Amirouche Boubekki (01 Aug 2020 13:55 UTC)
Re: Establishing a Scheme registry Arthur A. Gleckler (31 Jul 2020 20:09 UTC)
Re: Establishing a Scheme registry hga@xxxxxx (31 Jul 2020 20:34 UTC)
Re: Establishing a Scheme registry John Cowan (01 Aug 2020 01:58 UTC)
Re: Establishing a Scheme registry Amirouche Boubekki (31 Jul 2020 09:04 UTC)
Re: Establishing a Scheme registry hga@xxxxxx (31 Jul 2020 20:52 UTC)
Re: Establishing a Scheme registry Lassi Kortela (01 Aug 2020 19:50 UTC)

Re: Establishing a Scheme registry hga@xxxxxx 31 Jul 2020 20:52 UTC

> From: Lassi Kortela <xxxxxx@lassi.io>
> Date: Friday, July 31, 2020 3:14 AM
>
> Earlier this week we discussed on the SRFI 198 mailing list that there
> should be a community-wide registry for foreign error types. The set of
> known error types is necessarily evolving so it can't be in the SRFI. It
> will surely be useful to define other evolving sets of things in future
> SRFIs.
>
> To that end, we should establish some kind of lightweight registry for
> identifiers and other such things. Harold suggested, and Arthur and I
> agreed, that it could be a Git repo + low-traffic mailing list.
> Submissions could be via pull request / mailing list, quite informal.

We'll probably end up with a "low-traffic mailing list", but a real
forum would likely be better for this sort of thing, but way too much
effort for us, at least today. But how about the following:

Submissions as pull requests to the current master repo would work
well, you just need discipline about putting discussions in the mailing
list ... or scrape the repo provider's pull comments when the
decisions is made about the pull request, and post that as a digest to
the mailing list.  That would help keep the traffic lower, not bothering
people about registries they have no interest in, except for the initial
announcement, the final digest, and whatever the participants decide
should go to the mailing list anyway.

For "quite informal", from experience elsewhere I recommend selecting
a small pool of gatekeepers, and OKing additions and changes that any
two of them agree upon.

> [ Mailing list specifics. ]
>
> The Git repo could be under....
>
> We can make subdomains under scheme.org, so we could also set up a
> simple static page at https://registry.scheme.org/

My idea was first that a page we control would direct people to the
canonical copy/copies of the registry data.  In this case, one or
more git repoes hosted by different organizations, or just switch as
needed if it comes to that.

That way SRFIs can refer to that page, and with the expansion in
ambition, a particular git managed file or subdirectory listed on it.

- Harold