Three stewards will be elected for each working group.
Each working group has a separate vote (see links to Snapshot above).
Voting is open from 9am UTC on June 10 to 9am UTC on June 15.
Elections are conducted by ranked choice voting.
At the conclusion of the voting for each working group, the three candidates ranked highest will be elected as stewards of that working group.
The results displayed in the Snapshot UI are not an accurate reflection of current results. Please consult this thread for periodic updates on the current rankings.
Notes on Ranked Choice Voting
To vote, rank candidates in your preferred order, with the candidate you most want to see elected as a steward ranked first.
Steward nomination statements
You can read through the nomination statements of each candidate in the relevant Working Group (WG) posts:
Where is it written how does the score of each candidate based on the ranked votes is calculated? I’m just curious, but I couldn’t find the details.
Edit: I don’t want to mess up the election updates thread, so I’m asking again here. There are many ways to calculate the results of ranked voting, which one is used in the stewards election?
Candidates are “eliminated” over rounds of counting, meaning it is no longer possible for them to be elected. Candidates with 0 votes have been eliminated from the vote counting, and votes for them are distributed amongst remaining candidates.
This is of course only for any given vote count, and doesn’t mean that they are actually currently eliminated from the election (until the election ends).
As far as I know, Meeks STV method decides who are the three winners, but doesn’t rank them as “1st, 2nd, 3rd”. The “votes” numbers specified in the results you post are an intermediary score the algorithm gives each candidate, but not really the number of votes they got.
So maybe in this case it’s better to just post who are the three winners, without ranking them or specifying their voting score in the Meeks STV algorithm? I have some fear it may cause confusion at some point in the future, where people would really think that nick.eth finished third…
The vote count was extremely confusing in term 1 and remains confusing for this term. I understand it’s because of “the mechanism” but it’s kinda lame. It would be nice if the votes displayed were the actual votes that occurred.
This is how it was originally formatted but there was concern that it would be confused with a finalised tally.
I suppose primary votes could be displayed instead of votes with a keep factor applied, but it wouldn’t be an accurate representation of what the numbers are.
Maybe it’s worth considering switching to another method? From what I remember Meek STV downside is that it’s very hard for humans to calculate the results themselves manually. There are other STV methods where it’s a bit easier.
There are some advantages to voting systems where people can verify the results themselves, even if they’re not as efficient. For one, it gives people more trust in the system. Another reason is that candidates can see how far they are from being elected, and know if it makes sense for them to make a strong last push to gain more votes.
The difference is not minor if it can technically change the outcome.
Could you please run the new code on Term 1 elections? The new code only does it for Term 2 and I cannot compare. Since Term 2 results are significantly different between two codes, I do not buy that Term 1 results remain unaffected. Can you please back your argument with some results? Thank you!
Slightly easier maybe, but the nature of any STV method implies that it has a level of complexity to it. I don’t think that should really be a focus when it comes to choosing a voting method.
The reason it’s harder to see a lead someone might have in votes is the exact reason we are using Meek STV in the first place. It’s much more efficient at making every vote count, meaning that it virtually eliminates strategic voting. I think the effectiveness of a given voting method is much more crucial than how it might be displayed.
That being said, we will definitely try to do more in terms of visualising the vote for the next election. Ideally the vote count can just be seen on the snapshot proposal pages.
The difference in the result could be minor, and it could technically change an outcome. They don’t need to be mutually exclusive.
From the looks of things, you might have been using much older code. None of the elected candidates should be undefined.
The term 1 results from the new code are exactly the same as the previous code, the results for that election can be found here.
I git-cloned it less than a week ago. I am aware where the results are, but I need to see them calculated with the code. Alternatively, can you enable calculations for Term 1 in the new code? Old code spat out results from all proposals, not just Q3/4. New one only shows Term 2 results. It will be nice if it can calculate for Term 1 as well. I am sure it is a minor change to re-enable that feature.
From what I remember it’s much easier. But let me dive into it later on and maybe start a discussion for the next elections. It’s been a long time since I read all the math of the voting algorithms
For reference, was there a discussion when Meek STV was chosen? I don’t want to repeat one if it already occured.
Thank you! I simply wanted reproducible results and the code itself. I ran the new code on Term 1 and the results are luckily similar but not the same. EP12 clearly states the importance of ranking in clause 5.3:
The top-ranked Nominees from each working group vote held during a First Term Election Window or a Second Term Election Window (each an ‘Election Window’), will fill any available positions for the role of Steward for those working groups for the Term immediately following an Election Window, based on the order in which they are ranked in each working group vote.
Vote counts and rankings are off compared to the official results in the Community, Ecosystem and Meta-Governance WG.
Would you consider it lucky that the results did not change as much as Term 2, i.e. only the rankings got swapped but the top 3 candidates did not change?
Did the reasults not change in Term 1 because the elections weren’t that close? Alternatively, if the elections were tighter in Term 1, would the results be significantly different like in Term 2? Things could have turned out bad if the results had changed a lot.
How will you propose going forward considering that now we know results from Term 1 could have been compromised? It’s a human error but could have been a costly one.
New Term 1 Results:
Q1/Q2 2022 Community Steward Election final standings [ 'spencecoin.eth', 'limes.eth', 'coltron.eth' ]
Q1/Q2 2022 Meta-Governance Steward Election final standings [ 'james.eth', 'jmj.eth', 'simona.eth' ]
Q1/Q2 2022 Public Goods Steward Election final standings [ 'sumedha.eth', 'ceresstation.eth', 'avsa.eth' ]
Q1/Q2 2022 ENS Ecosystem Steward Election final standings [ 'ginge.eth', 'slobo.eth', 'avsa.eth' ]
Ranking in that context means “winning candidates”, rather than meaning someone with x amount of votes will get x position. With Meek STV elected candidates shouldn’t be ranked by vote count, since it tries to distribute the votes as evenly as possible.
As for your other questions:
Human error is always a factor.
The code was open-source and easily accessible, meaning anyone could have identified the problem and disputed the result.
The nature of the bug meant that it was relatively unlikely for any visible change to the result.
I do not agree with this. EP12 clearly mentions order in which they are ranked; it is literal not contextual. When AvsA decided against taking up the Ecosystem WG Steward position in Q1/2, Bobjiang took his place as he was the next candidate in the order. So the order does matter. I am not trying to criticise you; the error was humane and luckily very minor. I just request you to communicate your position on the importance of ranking in STV, or otherwise, to the Elections Manager and self-appointed CEO of ENS DAO. The EP12 should not mention ordering and ranking if STV doesn’t depend on ranking or order. Considering EP12 has been voted on twice, I find it childish that people cannot agree on what’s in it.
