johnshearing
commented
10 months ago Greetings @JaredCorduan, Thank you for taking the time to read Beemocracy and moreover, thank you for all your work on CIP-1694
...network likely could not handle it...Of course, perhaps participation would always be low,<
That delegation will be low is almost a certainty and is intentional. This will be enforced by making the delegation threshold so high that no one will delegate to the BReps unless the proposal really really matters to the common folks who wish prevent a BRep decision at a jury trial. So there is no network traffic except to register a BRep's solicitation for or against a proposal and to lock their vote.
A high delegation threshold will force almost all governance proposals to a jury trial of randomly selected BReps who are anyone that has written at least 10 solicitations within the last 365 days. These 10 digitally signed solicitations with locked votes are used by the prosecution and the defense to evaluate the character of the BReps during jury selection. Prosecution and defense will alternate selections from a random pool of eligible BReps based on previous solicitations and some standard questions asked by the judge. If a BRep is selected to serve on the jury and has written a solicitation on the matter with an associated locked vote then the locked vote is not binding. The BRep is free to listen to all arguments and decide anew. Locked votes are only binding if the delegation threshold is reached in which case the matter has already been decided by the locked votes as weighted by delegation of ADA.
So the only reasons for putting an issue up for vote are to:
- Get a body of solicitations for or against the governance proposal that the BRep jury can use in deciding.
- Think scout bees examining various sites for their new hive.
- Think scouts coming back to the hive for a waggle dance debate with first hand information to use in making a decision.
- Create a large educated jury pool of BReps who are invested by their anonymous reputation earned in the solicitation process.
- Think a guaranteed quorum of interested and invested decision makers with first hand information to work with.
I am personally convinced by this article that what the bees are doing is score (range) voting.<
So this is definitely not score or range voting. This is governance by well informed, highly invested, randomly, selected juries. But I claim that Beemocracy abstracts the very qualities of collective decision making that evolution always selects for and which is demonstrated by honey bees. The article you linked mentions this quality so I will paste it below for convenience.
7. Honeybees have [29] their own version of the commonly adopted “quorum” rule preventing candidates from winning RV elections with too few genuine numerical votes: They refuse to terminate an election until a quorum of at least 10-20 scouts simultaneously appear at the winning site (implying ≈ 150 had inspected it)
The quality that honey bees exhibit in their collective decision making is that decision makers always have good first hand information to work with and each and every decision maker must explain their decision in a vigorous debate.
Please listen to Tom Seeley in this video which is queued up to the correct moment when he explains the difference between bee democracy and human democracy. In the video he says that no decision maker advocates for a site they have not seen for themselves whereas in human democracy the decision makers often cast votes based on affiliation without ever investigating the matter for themselves. This is the quality that Beemocracy abstracts from the collective decision making process that evolution has selected for in the honey bee colony.
The following is repeated from Beemocracy to highlight the quality that Beemocracy abstracts from the collective decision making process found in nature.
Unlike our DReps...
- The Bee Democracy structure enforces that BReps must earn delegation through honest debate using information they verify for themselves. The result is good decisions.
- The Human Democracy structure has no means to enforce that DReps debate nor is there any means to ensure they make decisions based on information they have verified for themselves. The result is echo chambers, information bubbles, ignorance, hysteria, cronyism, and ultimately very bad decisions.
- BReps only receive delegation for the proposal they are currently soliciting for - Then delegation ends
- This is key to a functioning democracy - DReps in the bee's world have to earn their delegation for each proposal
- In human democracy the DReps only their earn delegation once and then learn to exploit their power for personal gain during the rest of their careers
Please @JaredCorduan, I am sorry that my presentation was not clear. I will be reworking the document. But given that this is not range voting but rather governance by well informed, highly invested, randomly, selected juries and that network traffic is only required to register BRep solicitations and lock their votes, would you please comment further on Beemocracy in as it applies to Cardano governance?
Much thanks for your input here and for all the work you have done on Cardano governance.
First off, thank you @johnshearing for putting all this effort into writing this up, it's a fascinating topic.
I am personally convinced by this article that what the bees are doing is score (range) voting. I am also convinced (though I'm not an expert), that score voting is one of the best systems we've found.
How range voting could be used with the plutocratic elements of a proof of stake system (and one without an identity system) is interesting, but I think not straight forward. If I am understanding correctly, your proposal is associating the score (or length and intensity of the bees' dance) with how much ADA a given governance proposal can attract. In normal score voting, individuals are all equal in their ability to score the candidates, but in this PoS setting, it is as though the variability of stamina to dance between individual bees is enormous. In reality, the bees really do have a fair amount of identity working in their favor (one "bee" is a pretty uniform and discrete unit, even though the bees themselves are quite fungible). My concern then is that the voting system described in beemocracy is very different than score voting, and so the merits of it cannot be assumed to follow from the success of bees.
On a more depressing and practical note, there's an engineering concern that this proposal runs into. Having every ada holder delegate to every BRep solicitation would involve so much traffic that the network likely could not handle it (or have room for nothing else). Of course, perhaps participation would always be low, but that's a separate problem. The DReps were a compromise between giving everyone a vote on everything, and limiting traffic. A compromise so that we could have something ready sooner rather than later. CIP-1694 could try to emulate Beemocracy by encouraging folks to be their own DRep and voting on a per-governance-action basis according to experts/scouts they trust.
Let me know if I've misunderstood beemocracy!