Super. That approach is lines up well with: propose a resolution, discuss resolution, vote/withdraw a resolution, if needed propose a new or amended resolution. That makes it clear what is being voted on and less likely for someone to misinterpret the thing that they are making a decision on.
We should follow this effort up with a clear definition of abstaining. What does this mean specifically and how does it impact quorum?
“Low-Quorum Voting” and its impact on velocity in Jupyter.
Low-Quorum could allow people to form smaller ‘ad-hoc sub-committees’ that can still move without formally being generated.
Working through existing issues in SC.
When joining the SC, different people come in with different values.
There’s a lack of definition of specific rules and even culture within the SC.
One example is discussions of conflict of interest (COI).
There is no canonical definition of COI.
Eg. Some people believe there is space to prioritize things that they are personally and professionally aligned with, others think that should never happen.
Another example is public messaging.
Examples:
The blog could be the voice of Jupyter, expressing jupyter interest and direction.
The twitter, on the other hand could be used to talk about what’s going on in the greater Jupyter ecosystem, even congratulating competitors on releases and just generally communicating excitement around the community.
I’m not sure what this means. Is this a “catch-all” phrase for all aspects of Steering Council composition and behaviors.
To ‘mutation test’ Jupyter Governance?
Not sure the meaning of a mutation test.
As for term limits, I find it healthy to have turnover in leadership. It adds new energy and ideas. Here’s a reference with Pros and Cons for board governance and term limits: https://boardsource.org/resources/term-limits/
Hi Carol! I think the “existing issues” topic was issues that had come up in the last handful of SC votes and not completely generic, i.e. the conversation in a couple recent voting threads.
The “mutation test” idea … I can’t recall with complete certainty but I think it meant something like swapping in a new member because of a term limit might create a stronger organization. If anyone from the call remembers the specific context more clearly, please update!
Thanks for the reference on term limits! I think this is valuable. We should talk about this in the next call.
The weekly hourly office hours call is open to the community members who care about governance issues. The call is held 9-10AM PST on Tuesdays.
Meeting Minutes – June 2, 2020
Attendees
Tim G.
Darian
Sharan
Chris H.
Brian
Jason
We opened the current draft of the Board of Director language
What are the commonalities between working groups that are formed by the BoD? The process is analogous to the JEP in the technical space, so it should probably be defined in governance.
Does the Organizational Director in the BoD have a vote? Is this person a “member” of the board?
One central remaining question is: how do we select the members of the Board of Directors?
One approach is a nominating committee to find BoD member replacements.
The Technical Steering Committee (TSC) already has a “natural” process for who would populate it, but the BoD is a harder problem.
Merely bootstrapping by seeding an initial BoD is insufficient.
In comparison to the way Apache Foundation works: what is the soundness of members on the BoD who have contributed funding in order to have a seat?
How would we prevent the external market competition between corporate members entering the BoD affecting the internal decision making?
The central question of paying members of the BoD is: how do prevent both the reality and the appearance of quid pro quo? How do we ensure that the members have the community interests at heart?
What other incentives can we give entities that donate to Jupyter other than direct governance votes?
On June 16 2020 we adjourned the meeting early because there were only three of us and we decided that the time might be spent better thinking and drafting. We will meet again next week at the usual time.
The weekly hourly office hours call is open to the community members who care about governance issues. The call is held 9-10AM PST on Tuesdays.
Meeting Minutes – June 23, 2020
Attendees
Fernando
Tim G.
Chris H.
Sharan
Darian
Reviewing work done up to date, including update on current priorities
Voting Procedures: Blocking on Board of Directors discussion
Board of Directors: Brian and Fernando need to chat.
Darian’s Recap:
Old understanding - Executive body was going to be distinct from organizational body and we had charted the primary leadership to exist in the executive body and the organizational body was a community facing body primarily events community considerations.
New understanding - The organizational body is now the Board of Directors, this is the primary leadership body, we don’t know how the structure is, but this is where the current idea for the main leadership body has landed.
Talking over the existing situation vs the envisioned system in mind for the board of directors.
Oversight vs operational management.
Where does the work and management happen?
In projects and working groups
BoD is in charge of creation and oversight of working groups and projects.
“The key to driving innovation in open source is to keep the people doing the work daily need to be the one’s driving the project’s direction.”
The weekly hourly office hours call is open to the community members who care about governance issues. The call is held 9-10AM PST on Tuesdays.
Meeting Minutes – June 30, 2020
Attendees
Fernando
Tim G.
Darian
Chris H.
Steve
Sharan
Recalibration conversation over the purviews of the Technical Steering Committee (TSC) and the Board of Directors (BoD)
When an organization is represented on the BoD, that individual’s involvement in the project may well be a pure function of employment
We discussed the presence of “delegates” to the TSC coming from working groups, which led to:
Maybe the TSC is misnamed: it is not a purely technical body.
Perhaps “working groups” are also misnamed. Any sub-project-level group that is entitled to send a delegate into the Council should maybe have the same designation: “sub-project”
We spent the bulk of the meeting white-boarding the two main bodies and re-assessing whether our drafts reflect the current state of thinking
Hi all - I have just gone through and cleaned up, added text, clarified, and otherwise tried to improve the proposed governance documents. I’ve tried to make each one follow the same pattern, and I’ve also highlighted where there is missing information that needs some discussion. I did not try to change the spirit of any content that was previously-discussed - only tried to improve and clarify the ideas that had already come out of these meetings.
I’d love to start making some tangible progress on these documents in the coming governance meetings so that we can get feedback and iteration from others. Please do chime in with your own edits, comments, etc. I look forward to discussing.
Update from today’s meeting - I’ve also created a “Jupyter Core Projects” document and populated it with some of the information discussed in the meeting (minutes are forthcoming). I updated the list above.
Note - I keep on getting requests to give access to people on a document-by-document basis. I don’t seem to have permission to give access for the whole folder. Can somebody with permission either:
Give access to the whole folder for anybody that wishes
Give me the ability to give access to the whole folder
Give access to the general public for the Drafts folder
The weekly hourly office hours call is open to the community members who care about governance issues. The call is held 9-10AM PST on Tuesdays.
Meeting Minutes – July 7, 2020
Attendees
Fernando
Tim G.
Darian
Layne
Brian
Sharan
Chris H.
What is left to discuss before we ‘write’ the governance documents?
Reporting
Meeting cadence
Communication
Example of Guido escalating pep to Python’s SC
In academia escalation is not welcome, in Amazon it’s a powerful tool to keep projects moving forward.
We need to have well defined protocols around escalation, so we can use it as a useful tool instead of a taboo thing to do once you’ve lost patience.
Definition of Core Projects
Repo organization (Projects/Subprojects)
Jupyter has:
Organizations
Repos
Software Projects
What is the mapping relationship between projects and orgs/repos?
Incubator
A place you can have a project, but no seat on the council.
What do you get / what is required / etc?
Minimal guaranteed level of support from the BoD?
How can the BoD guarantee something like this? Does it have its own funds and if so, how can it use them?
Fundraising for the Jupyter Project writ-large would be controlled by the BoD.
Some organization needs to be in-charge of ensuring that “core projects” are empowered to meet the minimal needs of being a core project
Might be a matter of $$$, or encouraging other individuals to support the project etc
Alternatively some core projects may need to lose core project status if they aren’t able to sustain themselves.
The specifics aren’t as important as the fact that this needs to be the problem of some specific group. Otherwise we have no guarantee that a Core Project will actually have the resources it needs.
Notebook project case study
No core notebook maintainer was working on it any more
over 100 open pull requests
team members from adjacent areas of Jupyter (server / lab) got together to work on the notebook repository
Spend bits of time in meetings for lab/server
But, basically only happening because their employers are letting them do this
How would situations like this be handled by the proposed model?
BoD’s job would be to notice the lack of resources
Find ways to mobilize resources (either $$$ or encouraging others to help out)
May also decide when to deprecate Core Projects
In the new structure, who owns the creation of new core technical projects, and who owns the removal of projects from “core” status?
It is a joint-decision from both bodies (both creation and removal) and requires a vote on both sides
There are a lot of small projects but that are really important. How should these projects get representation? e.g. jupyter_server or maybe nbgrader
Can we draw some conceptual boundaries around some repositories to naturally group them together and give them a representative?
Hey all - in the last meeting we agreed to brainstorm our own lists of “core projects” in the Jupyter ecosystem. Each “core” project would get a seat in the SSC and would also get a certain degree of priority when it comes to support and attention from the BoD.
We added a document to describe core projects (and a few similar pieces of info) here:
Hey all - I missed the last bit of the meeting today. I assume that notes/minutes will be posted, but in the meantime I wonder if somebody could summarize the main action items that need to be taken care of before the next meeting?
The weekly hourly office hours call is open to the community members who care about governance issues. The call is held 9-10AM PST on Tuesdays.
Meeting Minutes – July 21, 2020
Attendees
Fernando
Luciano
Chris
Tim G.
Layne
Steve
Darian
Do the conflict of interest guidelines need to be more specific when it comes to participation in forks and other open-source projects? (This came out of a conversation about nteract.)
We spent time considering the list of projects who send a delegate to the SSC.
Conceptual organization
Project organization
If we assign delegates by projects, smaller projects get proportionally higher representation.
We discussed the distinction between live services (nbviewer, mybinder.org, etc.) and software projects, the BoD is in charge of services and the SSC is in charge of software projects.
We spent most of the call working on compiling delegate-sending projects in the SSC
@choldgraf because it turned out to be a working meeting where we tried to solidify the list of delegate-sending projects, we didn’t have action items to prepare before the next meeting!