Closing the loop on PyPA Governance / BDFRN

A joint reply to @pf_moore @ncoghlan @bernatgabor and @pradyunsg and anyone else who’s wondering about:

You may have seen the December 2018 post on pypa-dev in which I mentioned that Changeset would be doing communication/facilitation/roadmap work for PyPA. I’ve now posted an update (belatedly for sure!) about what’s happened since then and what Changeset’s work has gone into so far.

I’m glad @dustin started this thread, which I think is pretty related to Remove the "Authority" from Packaging I do think that the PyPA needs to, at the least, make it clearer how a person or project joins the PyPA, what the criteria are, what powers it believes itself to hold, under what circumstances the PyPA would stop including a person or project (perhaps because of inactivity), and whether we want some kind of unified roadmap that we’re all working towards. And deciding all of those things are easier if we have someone who is sort of like the Debian Project Leader or a clerk in a Quaker meeting, and that’s what I’d particularly appreciate in a BDFRN role.

Yes, if we decide we want to stick with the PEP process, I think this would be reasonable to do.

Agreed. I think this discussion is pretty focused on making technical decisions, but there are a number of other decisions that occasionally need to be made that are currently very ambiguous (especially to “outsiders”) and would benefit from a governance model.

Essentially, we could re-invision this discussion as an alternative to Remove the "Authority" from Packaging, and put actual authority in the organization.

1 Like


Let’s do this then – we can switch away from the PEP process at a later date if we deem that as necessary; a model that changes as little from status quo as possible would likely be easier to work with right now.

Is anyone working on the updated governance model, that uses the PEP process?

governance model, that uses the PEP process?

Well… I found some time today (while procrastinating on my assignments) and came up with one. :slight_smile:

I created a new topic for discussing that model: PyPA Governance - A "Status-Quo" Model.

I’ve taken Pradyun’s modifications to my original proposal and turned it into a PEP draft (with some very light editing):

10 posts were split to a new topic: Should PyPA opt out of the PEP process?

The PEP is now published:

Per request I’ve also started a discussion thread in the “PEPs” topic:

1 Like

A post was merged into an existing topic: Should PyPA opt out of the PEP process?