archive this group & redirect conversation elsewhere?

Skip to first unread message

Sumana Harihareswara

Apr 13, 2020, 10:18:47 PM4/13/20
to pypa-dev
TL;DR: ok to archive this Google group? Reply by May 12th.

Below: Context and proposal, reasoning, and timeline.

Context & proposal:

People talk about Python packaging problems, work, and plans in many
different media: , distutils-sig, blogs,
Twitter, conference talks, IRC, ,
individual GitHub issues on several different repositories, Stack
Overflow, and more. So people frequently ask me: where should I go to
keep up, or to announce something or ask for feedback? It's hard to
guide them, because of this proliferation and fragmentation. And people
have commented on that before, both senior folks like Donald[0], and
people who are earlier in the learning curve[1].

We can't and shouldn't stop people from talking about Python packaging
on social media, at conferences, and so on. But three mailing
lists/forums on nearly identical topics strikes me as more than we need.

So I suggest that, one month from now, we stop posting to this list
( and essentially archive it.

Reasoning (why close THIS one?):

We now have three mailing list-type places to talk about Python
packaging tools and progress. All of them allow both reading and posting
from the web or from an email client, and all of them have web archives
with built-in search. Generally, the people who want to talk about one
of these topics want to hear about the same topics (things happening in
PyPA and about related things in Python that will affect PyPA) no matter
what venue they're in.

1. pypa-dev (here). Started in 2013. About 5 posts in the past month,
mostly cross-posted to other places as well. Hosted by Google in a
closed-source application that doesn't seem to get much love from
Google's product folks.

2. The distutils-sig mailing list[2] which has expanded in its scope.
It's a place to discuss and resolve problems that cut across different
parts of the Python packaging ecosystem, and to announce new releases or
in-progress work. You can log in an account, or with Facebook, GitHub,
GitLab, or Google authentication. About 12 threads in the past month.
Hosted by Python Software Foundation with an open source application
that's under active development.

3. The Packaging category on Python's Discourse forum , which started about a year and
a half ago[3]. Very wide scope. You can log in with an account, or with
Facebook or GitHub or via email. About 21 posts per month. Hosted by PSF
with an open source application that's under active development.

Maybe I'm missing something. Maybe there is a function being served by
having a mailing list that is specifically labelled "PyPA" (for
instance, we could add "get on the Google Group and that makes you a
member of PyPA" to the docs[4]). Maybe there are people actively
reading/posting here who feel unwelcome on the other two lists/forums,
because of atmosphere or user interface. As a person doing a bunch of
work on PyPA stuff over the past ~2.5 years, I haven't noticed either of
those conditions, so please speak up if I'm wrong, or if there's some
other reason to keep this Google group going.

Timeline and methods:

Here's what I suggest, and what I will carry out if there is no objection.

In one month, on May 13th, I would verify that no one has argued here
for why this Google group should continue to be open for posting. Or,
even if a few people have objected to closing the list, I would check
for rough consensus, especially of people who are doing SOMETHING
productive having to do with PyPA (teaching, answering questions online
or in person, running key infrastructure, writing documentation, making
or fixing software, etc.).

I would post a final message to this list, marking its close and
suggesting that people use distutils-sig or instead.

Then, I would stop members from posting to this Google group. That is, I
would stop members from creating new posts, but leave past posts up at
their current URLs, so links, browsing and search would work.

And then I would look through relevant documentation within PyPA
repositories to see what needs updating (READMEs and so on pointing to
the old list), and submit pull requests.

I appreciate the work folks here have done to carry forward Python
packaging over the past several years. I don't mean to diminish that or
to insult anyone here. I want to help us out, and I think closing this
list will help focus that energy better. But I am open to hearing that I
am wrong.

Sumana Harihareswara
Changeset Consulting






Jason R. Coombs

Apr 14, 2020, 9:20:37 PM4/14/20
to Sumana Harihareswara, pypa-dev
My initial reaction was that I _need_ this list, but after a moment’s consideration, I think you’re right. +1
> --
> You received this message because you are subscribed to the Google Groups "pypa-dev" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to
> To view this discussion on the web visit

Sumana Harihareswara

May 1, 2020, 6:02:16 PM5/1/20
to Jason R. Coombs, pypa-dev
Thanks, Jason.

Nudge to the group; 11 more days to comment.

Sumana Harihareswara

May 10, 2020, 6:45:55 AM5/10/20
to pypa-dev
Hey all -- I've heard no opposition, onlist or offlist. As I asked in April: please speak up if I'm wrong, or if there's some other reason to keep this Google group going. And please reply if you agree with the idea - Jason's the only one who's replied so far.

Reply by May 12th (2 days from now).


Paul Moore

May 10, 2020, 8:17:25 AM5/10/20
to Sumana Harihareswara, pypa-dev

Xavier Fernandez

May 10, 2020, 3:56:43 PM5/10/20
to Sumana Harihareswara, pypa-dev

Sumana Harihareswara

May 28, 2020, 3:23:49 PM5/28/20
to Xavier Fernandez, Sumana Harihareswara, pypa-dev
The group has spoken. I'm decommissioning this list now; I'll send a
final closeout email and then stop the ability to post. Thanks.

Reply all
Reply to author
0 new messages