How to design efficient and resilient DAO

[Samupaha]

Maybe I've failed to understand the voting process of bitshares; but why is low voting turnout in DPOS a bad thing? For instance I don't vote, so don't my shares simply go from an active to passive state as you talk about in the article which doesn't affect active voters?

It's a problem because the decisions become too random. If only a minority is actively taking part in discussions, they don't have enough power to make everything happen that they want. Uninformed voters might come any day and vote against consensus because they don't understand why certain decisions are made and are too lazy to find out. This makes the decision making process too chaotic to be efficient.

The main argument you made was that limiting the amount of decision makers produces a more efficient DAO, and that could certainly be the case. So along with annual delegate voting, would you want to decrease the number of delegates which a DPOS system uses from 51 down to 10 for example?

I don't propose annual voting (it was just an example how companies currently limit the decision making power of their shareholders).

In my model there is no role that is exactly a delegate. Of course block producers are delegates in a way that the power to produce blocks is delegated to them. But there are no delegates like Bitshares 0.x had.

The executive team has power to decide how many block producers there will be and how much they are paid. If there is too few, resiliency will suffer because the existence of DAO is not guaranteed anymore. DAO could also incentivize node maintaining by paying for them.

My understanding is that the benefit of a smaller number of delegates would mean more money from block rewards to fund development for each delegate rather than having to share those rewards with a higher number of delegates and possibly less disagreement between delegates about how to best move the DAO forward. But I'm sure bitshares made the number 51 to prevent help prevent collusion as lower numbers could more easily allow 51% attacks. How do you defend against that if there's only annual voting and possibly with less delegates?

My model uses a reserve pool which is also used in Bitshares 2.0. "Delegates" won't earn money from block rewards, instead there are workers who are paid from the reserve pool.

Read more here: Stakeholder-Approved Project Funding

[bitAndy]

Maybe I've failed to understand the voting process of bitshares; but why is low voting turnout in DPOS a bad thing? For instance I don't vote, so don't my shares simply go from an active to passive state as you talk about in the article which doesn't affect active voters?

The main argument you made was that limiting the amount of decision makers produces a more efficient DAO, and that could certainly be the case. So along with annual delegate voting, would you want to decrease the number of delegates which a DPOS system uses from 51 down to 10 for example? My understanding is that the benefit of a smaller number of delegates would mean more money from block rewards to fund development for each delegate rather than having to share those rewards with a higher number of delegates and possibly less disagreement between delegates about how to best move the DAO forward. But I'm sure bitshares made the number 51 to prevent help prevent collusion as lower numbers could more easily allow 51% attacks. How do you defend against that if there's only annual voting and possibly with less delegates?

[Samupaha]

You might be interested in my post about DAO design: How to design efficient and resilient DAO.

I describe a model for an efficient DAO that's based on Bitshares 2.0. It could be used to make new and better version of Bitshares because it removes the problems that Bitshares currently has.