Add language about governing strategy of project

[NickleDave]

as suggested by @choldgraf (who is asleep, in France)

https://twitter.com/choldgraf/status/1520470381349462016

it would be great to see language about the governance strategy of the project (eg a goal of no org @ > 50% on steerco)

Would also like to see this so I'm raising the issue first!
Thank you for this great project

[choldgraf]

Hey all - thanks @NickleDave for opening the issue :-) here's some extra context from my thinking:

This package / framework seems like an exciting way to bring the Python ecosystem into the JS / web-component world. It seems like a goal of the project is to build a foundation that others may build on as well via extensions and new applications. For this reason I think it's particularly important to build trust and buy-in from the broader community.

There is a lot of good language here around defining the governance and scope of the project, so it seems like the authors are being attentive to transparency and governance design. This is a good sign if the goal is to "democratize Python" and empower more individuals.

However, I felt that it was missing the "strategy and goals" of the organization and how it relates to governance. I like how all of the decision-making etc is laid out in the CHARTER file. However, right now a majority of (actually, all of) the governing members of the project are employees at Anaconda (and moreover, one of them includes the CEO). To me this means that (at the moment) this is an Anaconda project, not a community-led project.

So I wonder - is the goal to move towards a collective governing model? If so, what kinds of commitments is Anaconda making to ensure that this happens? What is the end-state of this next phase of governance transformation? Does it look like a multi-stakeholder entity like the PSF, Apache, etc, or more like a "one main stakeholder that allows others to come to its table" project like Tensorflow? What kinds of oversight will it have to ensure that the project moves forward in-line with its mission and governing principles? Does Anaconda intend to move ownership over marks/licenses/etc into a neutral organization like NumFocus?

I am not advocating for one or another position here (though I obviously do have opinions), but I think it could be helpful to share Anaconda's vision, since right now that is the vision that drives this project forward.

[jezdez]

As you can imagine these are all very good questions, thank you for elaborating and spelling out the elephant in the room.

While I have an own opinion on the matter, I think it's best to let @fpliger, @kevingoldsmith or @pzwang respond to this to make sure it's not miscommunicated. Personally I'm interested in fully catering to openness and transparency while the project is taking shape.

[pzwang]

Great questions!

So I wonder - is the goal to move towards a collective governing model?

Yes we definitely intend for the project to be a community-governed OSS project.

If so, what kinds of commitments is Anaconda making to ensure that this happens?

Not sure... can you elaborate on what kind of commitment or insurance you have in mind? At the end of the day, if things really go sideways somehow, the community can always just fork the whole thing and go run with it, right?

What is the end-state of this next phase of governance transformation? Does it look like a multi-stakeholder entity like the PSF, Apache, etc, or more like a "one main stakeholder that allows others to come to its table" project like Tensorflow?

Multi-stakeholder.

What kinds of oversight will it have to ensure that the project moves forward in-line with its mission and governing principles?

I think that kind of detail would need to emerge from discussions among stakeholders as we get "real governance" (as opposed to placeholder governance) spun up.

Does Anaconda intend to move ownership over marks/licenses/etc into a neutral organization like NumFocus?

I think that would be the goal. We'll need to see whether the best home is NumFOCUS proper, or to do something like Django Foundation and Jupyter Foundation... I can see advantages to both approaches. In a similar vein to some of the discussions around conda governance that are happening right now, we'd want to ensure that Anaconda itself can make commercial use of these marks as long as it's a meaningful contributor and stakeholder.