Title: bDIP-03: Dynamic forum quorum for bDIPs
Draft Author: jengajojo
BDIP SUMMARY
The static forum quorum requirements set in the constitution do not reflect the activity of active contributors in the DAO which consequently leads to governance frustration. Instead, the suggestion is to adopt a formula based approach which can evolve as the activity in the DAO evolves.
BDIP BACKGROUND
Recently we saw in bDIP-02: Coordinated timings across bDAO that in times when there are less active contributors, it was really difficult to reach quorum for something relatively simple. The quorum thresholds were defined in the constitution in a time when there were many more active contributors. The challenge with a static threshold is that it cannot reflect the number of active contributors in the DAO at any given time, this is especially relevant for an industry such as crypto where the attention can vary significantly between the bull and the bear phases.
BDIP SPECIFICATION
How can we address this challenge?
Option A: Update thresholds at regular intervals
This option allows contributors to suggest new thresholds based on manually sensing activity in the DAO. The con is the bDIP needs to reach a threshold each time for an update and there may be several updates required each year which costs a lot of admin overhead. Hence I do not recommend this option
Option B: Update thresholds based on formula
This option requires less changes as compared to option A if the formula allows for thresholds to be set dynamically based on activity within the DAO. Overall this reduces the admin overhead.
Current Form
Governance update
Forum quorum
Approval %
Timeline
Major
63
70%
1 week
Minor
51
70%
1 week
Patch
40
70%
1 week
Proposal
Modify the quorum table
Governance update
Forum quorum
Approval %
Timeline
Major
30 Day Average + (0.2 x 30 Day Average)
70%
1 week
Minor
30 Day Average
70%
1 week
Patch
30 Day Average - (0.2 x 30 Day Average)
70%
1 week
Add the following items are criteria for the above in the constitution
- The dataset for 30D avg starts 1 day prior to the date of posting a proposal.
- All forum quorum calculations are rounded to the nearest whole number
- Only polls which have been live for atleast 7 days should be taken into account
- If a forum post includes more than one poll, each poll should be considered a separate datapoint
- 30D Average formula = Sum of votes / number of independent polls
Modify the bDIP proposal template and add the following:
Date posted
Post title
#votes
ddmmyyyy
Title
n
Title
Value
Major
m + (0.2 x m)
Minor
m = 30D avg
Patch
m - (0.2 x m)
EXPECTED IMPACT
Makes forum quorum more dynamic and reflective of the latest activity in the DAO
ANALYTICS
Here is an example of this formula applied for the current proposal.
Date posted
Post title
#votes
02-12-2022
Snapshot voting - Implement Shielded Voting
53
14-12-2022
Weekly rollup recap video series
38
12-12-2022
Purpose Driven Guild Funding
29
06-12-2022
Proposal to re-create the Education Guild Organizational Unit within the DAO
29
25-11-2022
Marketing Department Season 6 - Content Coordinator Re-election
58
Title
Value
Major
50
Minor
41
Patch
33
NEXT STEPS
Update Constitution
Update bDIP Format
WORKING GROUP BACKGROUND
jengajojo: bDAO contributor involved in governance and project management
Polls
1. What type of update is this?
Major
Minor
Patch
0voters
2. Modify forum quorum to a formula based approach as detailed in this proposal
Being the FFOG here, I am not in love with constant change but this proposal is so web 3. Very relevant andI described it as a Major change. It might seem like a minor adjustment but I anticipate Major relevant impacts if/when implemented. Appreciate the thought that goes in to these improvement proposals. Good processes and Goverance helps maintain the alignment with MVV and helps with talent retention to this wonderful community.
I like this proposal a lot, I think it is extremely necessary to adapt our quorums based on the activity of the DAO. 2 thoughts I have, 1) if this is implemented I would love to see a forum dashboard somewhere that holds relevant data for us to keep making this crucial adjustments( bar graph of proposals split by size to show the average amount of votes they are getting, maybe a similar display but over time to see the trajectory of our engagement, and some metrics on failed proposals) so we can track the success of each of our proposals and how it relates to this new rule. 2) it would be interesting to see how these requirements would be affected if we adjusted them based on guest passes, L1s, L2s, etc.
It would be nice if Analytics guild can help track this. I believe they aggregate and track certain data in the DAO. might be interesting to have them add this and create a dashboard for it.
Thanks for the feedback. Don’t you think defining a minimum defeats the purpose of a dynamic approach? Also, could you please help us understand why you think there should be a minimum?
A formula with no floor is vulnerable to low participation governance attacks. Quorum could be reached with as little as 1 or 2 votes. Potentially 0 if the inequality is defined as greater than or equal to.
Defining a reasonable minimum prevents this - I would suggest at least 15 votes.
The current forum settings do not allow for a lot of these privileges such as specific data exports or token gating etc, the long term solution is to host our own forum. Meanwhile, I am working with @Icedcool to find and update this info as best s we can.
And a discord<>discourse integration(which may be best!).
The big challenge is our hosting, which right now is basic, and for us to get additional plugins we would need to move to either business or self hosted. @jameswmontgomery.eth and infosec are looking at self hosted, so expect more on that soon.
Hello I am late to this discussion but I am seeking clarification because the formula and example application are not clearly articulated here.
Date posted
Post title
#votes
ddmmyyyy
Title
n
Title
Value
Major
m + (0.2 x m)
Minor
m = 30D avg
Patch
m - (0.2 x m)
What is ‘m’ in this formula? Is it the 30-day average?
I assume it is, although that’s not stated apart from in the ‘Minor’ line.
Is ‘m’ rounded before it’s used to calculate the quorum, or is only the final answer rounded?
I’m aware that rounding usually only occurs at the end of a calculation, but seeing as it’s not possible to have a fraction of a vote, I think it should be rounded before the calculation and stand alone as the 30-day average.
In the example, the 30-day average is 207/5 which equals 41.4.
That means the quorum set for Minor update is correct at 41.
If Major = m + (0.2 x m) that is 41 + (0.2*41) or 41 + 8.2 which rounded down = 49, not 50.
It’s only rounded up to 50 if you don’t round down ‘m’ first.
Also: All forum quorum calculations are rounded to the nearest whole number
I assume the rules of rounding would be followed (.5 and above are rounded up) but good to make explicit for everyone.
The dataset for 30D avg starts 1 day prior to the date of posting a proposal.
Does this mean that the last day of 30-day calculation period is the day prior to the day that the proposal the quorum will apply to is posted?
If so I’d suggest this wording: The dataset for the 30D avg counts back commencing the day prior to the date of posting a proposal.
It will be part of the bdip proposal format as stated in point 3. So each proposer has to do it for their own proposal
I envision chaos if each proposal poster has to calculate their own quorum. I just know there will be disagreements about calculations and inclusion - for example what happens if the votes on some of the polls used as a basis for the calculation are still open? Won’t that change the quorum? We’d need to specify that it’s a fixed calculation on the day of posting, I suppose.
So all in all I think dynamic quorums would be helpful but I’m not sure a formula, or this presentation of the formula is the solution.