[DISCUSSION] æternity Governance and Future Community Votes


Hello everyone,

Please use this topic, for any suggestions, recommendations, and feedback you have on future community votes and governance in general.



Please adopt preferential voting next time (http://australianpolitics.com/voting/preferential).

The results will be calculated by an iterative algorithm. It would be really interesting to see it in action.
For example in terms of BRI a voter will have one or two votes. In each vote s/he can vote for:

  • against BRI
  • x% payout to BRI, where x will be <0, 20>

It could be expressed on the slider how it was already done, but there will be 0% option additionally.

So the voter can have:

  • 1st preference to reject BRI
  • 2nd preference to give x% of payout to BRI if BRI will be incorporated


I like this idea!!
I didn’t know this existed but this preferential voting makes perfect sense instead of having multiple votings


If the AE team had just tweaked their voting algorithm a little, they could have gotten a more fair result for the BRI vote.

All they had to do was averaged in the 0% votes, with the caveat that if more than 51% of the respondents voted 0%, BRI will not be implemented. I mean, functionally, we’re assuming that the people voting 0% BRI will want a 0% BRI payout if BRI is implemented.

As it stands now, all the people who voted 0% BRI did not have their votes counted in setting the BRI payout rate.


@aocypher If you will have only one vote, the outcome will be slightly different I believe if you will have another preference vote. But maybe I am wrong.


Are you saying that the people who voted against BRI would not want a BRI payout of 0%?

But I do agree that preferential voting should be the way to go in the future. It’s just that the BRI vote was somewhat unfair for the people who voted 0%.


Hmm, this is the assumption and in terms of BRI voting, maybe that is true. This is an option that many voters wanted to express and couldn’t. The BRI voting is a specific one because in fact, the first question is boolean (pro/con BRI) and then if you are pro, how big the BRI should be as a % of PoW reward, but if you are against it is quite obvious that you want to minimize payout to BRI, thus 0% BRI assumption.


Ya. I agree. Although I, personally, am happy with the result (since I’m in favor of a BRI of 10%), the way the voting was tallied was unfair to the 0% voters.

The BRI vote could have just as easily be expressed as:
If 0% BRI voters > 51% of respondents, do not implement BRI
average all BRI payout percentages.


Anyway preferential voting should be a standard. You can analyze the voting paths and outcomes, design the voting paths and give user limited and reasonable voting options, but if you want to quickly prepare voting poll and doesn’t block any voting preferences, then just give voters the options. Even if it will let them to vote against BRI as a 1st preference and 5% as a 2nd preference (instead of quite obvious 0%).


For the votes,We really need a quorum.


I am super happy to see the discussion is already happening! This is really valuable.

The final BRI voting results were just published. Have a look:



Exactly! One question:

If I understand correctly, in the preferential vote system -> if you reply “no” to the first question, you get to vote “0%” or any other positive percent on the second? So if it passes, your “0%” vote (or whatever number) is counted? It is a bit counter-intuitive to me - if you vote “No”, isn’t it certain that you will also vote “0%” on the second? I guess the whole idea here is to have your 0% vote counted, in case the vote passes.

Thank you for the interesting comment!



Generally you can have just one question. But the BRI voting is a little bit more complicated and there is a simpler example in the second part of this post. Anyway, the poll can looks like this:

  • I am against BRI
  • I am for 0% BRI
  • I am for 1% BRI

Then voters may have two votes. Then the algorithm that counts votes first should determine if majority of voters are for or against BRI. When the majority want BRI we need to consider the second preference (second vote) of voters that were against BRI (because we don’t want to reject their vote even if they are minority, we want to calculate consensus - that is the reason of preferential voting). Then it will calculate weighted average for BRI payout.

The other example could be voting for new monetary policy/inflation characteristic.

Question: Do you want to change Aeternity’s monetary policy?

  1. No - don’t change inflation curve thus PoW reward.
  2. Yes - divide the current inflation curve by 3.
  3. Yes - set the constant PoW reward (100 AE for example).
  4. Yes - set the constant 1% inflation.

Then for example my preference might be to vote for 2. and then 3. and then 1. If majority will vote for 2., I am happy. If not, my vote should belong to 3., but if that option won’t win, then my vote should belong to 1. The correct calculation might be a little bit tricky, but there are solutions on github for that.

There are many algorithms to count votes and one of them is:

but I propose a modification to iteratively review another preferences of rejected votes:



@alibaba_yin Unfortunately not a lot of people votes on blockchain. Also, some tokens might be lost over time so they can’t be used anymore. The best solution would be to calculate quorum based on past “elections” results. So the quorum might be 50% of average of tokens used in the last n “elections”:

quorum = 0.5 * avg(tokens_used_in_election_1, tokens_used_in_election_2, ...)


I’d like to suggest votes happen in multiple rounds or in a single modular round.
Per example:

First round: discus and vote on the rules for the proposal vote. Example, how to weight BRI votes.
Second round: vote on accepting or rejecting the proposal.
Third round: If accepted, set parameters for said proposal if applicable. Example, % of BRI


Hello everyone,

We just created a governance-dedicated sub-category. You can start topics there.