Question: list whether an EC member was elected by the UoC or by the EC

[jasongrout]

In Jupyter, we have an unusual way to fill EC seats that tries to balance between appointing an EC to ensure it is balanced in its strengths and electing an EC by community vote. The current language from the governance docs is:

Each annual election will begin with the existing EC notifying the UoC of a call for nominations. Any member of the UoC may nominate a candidate, including themselves. The EC will confirm that a nominee accepts their nomination. The candidate shall write a statement of their interest, qualifications, experience, etc. that will be distributed to voters. Once the full set of candidates is assembled, the UoC and then the EC will take ranked choice votes to fill their allocated number of EC seats. After each election, the total number of seats filled by the UoC should be equal to, or one more than, the number filled by the EC.
If a member leaves the council before the end of their term, the EC may designate an eligible replacement who will serve for the rest of the term. This replacement will count as a seat filled by the EC if the replacement serves more than half of the term.

(FYI, this language is being updated to be more clear in #252)

A question has come up several times over the last few years (e.g., here, here): should it be public knowledge which of the EC members were elected by the community and which were elected by the EC, or should it only be public knowledge in a given election how many seats were selected by the community and how many seats were selected by the EC? Note that the calculation above depends only on knowing the number of seats that were allocated, not on which people were voted in from each process.

Arguments I've heard for making this public knowledge:

• Transparency is an important part of Jupyter governance, and so we should be transparent about the results of the election.

Arguments I've heard against making this public knowledge:

• A member of the EC should think about themselves as equal to other members of the EC and representing the entire community. We don't want EC members to start thinking about themselves as representing different constituencies based on who voted them onto the EC.

A rebuttal to the argument against making this public knowledge is that if it is necessary to fill an EC in the middle of a term, then the process of selecting the replacement EC member (the EC appoints the person) obviously discloses if the person was elected by the community (i.e., they were not). Whether the replacement counts as an EC-filled seat depends on if the replacement person serves more than half of the term.

Thoughts?

[jasongrout]

FYI, I have been elected twice now to the EC and have helped run a third EC election. I do not know in the first two elections if I was elected by the community or by the EC. Obviously I know the results for the third election (where I was not a candidate), but have refrained from making it public unless/until we have a community convention.

[jasongrout]

CC @ivanov, @mwcraig who have both raised this issue in the past. CC also @jupyter/executive-council who runs the EC elections.

[rpwagner]

• A member of the EC should think about themselves as equal to other members of the EC and representing the entire community. We don't want EC members to start thinking about themselves as representing different constituencies based on who voted them onto the EC.

I am strongly in favor of this argument, and that it is not only the EC member that should think of themselves as equal to other EC members, but that the community perception applies, as well. In other words, I would not want the community to think of any EC members differently because of how they were elected. Or for the community to perceive that because they elected an individual, that individual is their representative.

[rpwagner]

To say it a little more dramatically, the EC should be a group of equals, perceived equally by the community.

Drawing attention to how individual EC members are elected risks creating distinct classes of EC members. That could impact both how individual EC members act, and how the community engages with the EC.

[ivanov]

The nondisclosure of the pathway to EC membership creates information asymmetry between those who were part of the EC and the rest of the community.

The reality is that there are two different pathways to joining the EC. I think it's important for this fact to be disclosed, instead of obscured. Keeping the information of which candidates were elected by the UoC and which were elected by the EC only to the previous EC also impacts how individual members act, and how the community engages with the EC.

The latest governance model for the project:

1. provides two pathways to EC membership (UoC election and previous EC election)
2. includes a mechanism for balancing between those two pathways (as was previously discussed and debated)

As a member of the community who already feels like there is too little information shared between the EC and the rest of the community, I have to say that it is odd to have a model that places so much importance on these two separate pathways to have them simultaneously erased out of public view in the election outcome. You cannot mandate the perception of equality by the community. I respect that it's a wish you can have, and I respect that some may wish to not know how the EC was formed, but that does not mean that we should normalize hiding these facts. Elections of all sorts should be transparent, not only because it allows for the assurance and verification of election integrity, but also because the electorate is entitled to the aggregate totals, not just the outcome. In a single winner election, for example, a two candidate 51%-49% vote is different from a 90%-10% election. This matters not only for the electorate, but also for the candidates. Last year I published excerpts from my private appeals seeking the disclosure of results in NumFOCUS election of 2018, and this situation is similar, in that, again, I do not think it is appropriate to keep the community in the dark about the actual totals.

In addition to election transparency, I believe that under certain conditions, it is not possible to enforce the balancing mechanism without disclosing the path to EC. For example, the current governance docs say

If a member leaves the council before the end of their term, the EC may designate an eligible replacement who will serve for the rest of the term. This replacement will count as a seat filled by the EC if the replacement serves more than half of the term.

In the event of such an early resignation and replacement for more than half the term by the EC, it matters which pathway was taken for that original seat, because if it was originally filled by the EC, the running total balancing mechanism remains unchanged, since both before and after the seat was filled by the EC. But if the original seat was filled by the UoC, the running total for UoC vs EC should now reflect a +1 for total number of seats filled by the EC.

[mwcraig]

Heads up, this is a little ranty. I'll propose some solutions in the next comment

• A member of the EC should think about themselves as equal to other members of the EC and representing the entire community.

I don't see how the method by which you ended up on the EC would have any impact on whether EC members see themselves as equals. In other settings when I've been responsible for setting up committees there was an analogous split between committee members on occasion: some were selected by their colleagues, some were appoint by me. In day-to-day committee meeting no one cared how the appointment happened.

That said, their was no equivalent to the requirement that

We don't want EC members to start thinking about themselves as representing different constituencies based on who voted them onto the EC.

Then why have them elected by different constituencies?

The document describing the council itself says there are "two types of seats on the Executive Council.".

Given that wording I think it would be natural for a newly seated member to ask "which kind of seat am I" and "how is my seat supposed to be different than the other type?"

I think the root cause of any perceived difference in the seats stems directly from this kind of wording.

[mwcraig]

OK, some solutions.

1. Opened a PR on Jason's PR here: Minor wording change regarding seats on the council jasongrout/governance#1. It tries to emphasize that there are two election styles, not two different types of seats.
2. In a moment I'll leave several questions about mechanics of the elections in Clarify various aspects of EC elections and the Union of Councils. #252.
3. The election method is fairly complicated. If you really prefer not to list who elected each member, then you could maintain a separate documents that tracks
   a. How many seats are to be filled by UoC and EC in the upcoming election, since Clarify various aspects of EC elections and the Union of Councils. #252 says "the UoC and then the EC will take ranked choice votes to fill their allocated number of EC seats," which implies there is a pre-allocated number of seats.
   b. How many seats were filled through each election process in that election.
   c. Updated current total of how many of the EC were elected by UoC vs EC.

Implementing number 3 would promote transparency while separating how each candidate won their election.

[ellisonbg]

Hi everyone, as a (now) former EC member and someone who worked on the governance model, I wanted to give some context and history about the EC election process. I am not formally speaking for the EC, but this does represent the general consensus of what the EC has talked about in this area.

On transparency in the EC elections. The Jupyter governance model tries to balance difference principles. Transparency is certainly one of them, but there are many others (inclusive, collaborative, trust, personal privacy, sustainability, etc.) and these are often in tension with each other. Given this, here are some observations about different dimensions of the EC elections.

• More often than not, participation in the EC has been a factor in employment and promotion decisions of the EC members (people have been promoted and gotten jobs, in part due to their serving). While we want EC members to benefit professionally from their service, we don’t want to harm them professionally by publicizing full results if they run and not not elected.
• Generally, we need more people to run for the EC, especially given the rotating leadership model. To encourage people to run, we want it to be low key and not stressful emotionally or psychologically to run. Publicizing the results or how each person was elected has the potential to make it more stressful to run and to discourage people from running if they are not confident they will be elected.

I don't think this means we couldn't be more transparent - for example, given how the election works, we probably should have a location that records the overall number of EC/UoC elected seats over time, even if we don't name the individuals.

[Ruv7]

From my personal point of view, in my role as EC member I would rather not know whether I was elected by the community or by the EC. For some it may create a discrepancy in status or value that doesn't serve me. In general I see value in transparency but for this item I believe the drawbacks outweigh the benefits.