Thoughts on voting mechanisms and how they transfer in DAOs/DOs

If you are in any way related or part to the blockchain world you probably have heard the acronym DAO at least a thousand times this year. DAOs are the new hot thing that brings a lot of imagination, hopes, dreams and more than everything — confusion (being in the blockchain space I feel confused 🤷‍♂️ at least once a day, so this might just be me).

I won’t go in the details what DAOs are (you can check this post). A Google search will do magic for you but still, a TL;DR expert would be: Decentralized Autonomous Organizations or DAOs for short are organizations which reside on-chain, are governed using smart contracts and might or might not be connected to a legal entity. Like a normal organization there are members that are part of it, and depending on the structure, types of the voting system, and levels of participation, different rules and voting mechanisms may apply. Exactly in this last part, I’m interested in.

This post is nothing more than notes on the research I’ve done and some thoughts I had on how the already known voting mechanisms may be applied to DAOs.

One more thing: there’s a difference between a DAO and a DO.

The main difference between these two is that while DAO is automated using smart contracts the DO (Decentralized Organization) might use some automation but most of the times it is expected to function as a normal organization with no level (or at least no high level) of automation. I wrote “expected” because this is still in a highly experimental phase and at there are no operational DOs at least no I know of.

Setting the ground

There are several dimensions to voting systems that I think are of relevance here.

How votes are weighted? — Are the votes equally weighted or some members votes have a higher weight.

Example: Pure Democracy vs. Meritocracy

Who can vote? — Are all members allowed to vote or there are voting groups/circles?

Example: Presidential election vs. Local representatives elections (electing mayor).

How the consensus is reached?

Example: Elected using plurality model (who has the most votes win) or majority model (who has over 50% of the votes)or some other model?

How votes are weighted?

Depending on how the votes are weighted we can distinguish two main types:

• Every vote has an equal weight e.g. Democracy
• Different votes have different weight e.g. Meritocracy, Technocracy.

Having a system where everyone has an equal vote is quite straightforward (assuming that there are not barriers for the person to cast his/her vote. If this is the case the problem is quite different and should be examined in a separate post).

The mind buggering part comes when the votes are weighted differently.

Assuming we have a meritocratic system where voting is done on the basis of “merit”. The main challenges related to this system are related to how to exactly weight the different votes based on people merit.

For example, votes of people that have educational background and experience in medicine shall weight more on topics related to healthcare and less on such that are related to finance. If we measure the weight in points, the big question is how many points a doctor shall have on medical-related topics and how many on finance-related topics? Who shall decide on the weight? Would, for example, the doctor has the same vote weight for all non-medical related topics let’s say finance and education or different (1.3 points for financial matters and 1.5 for education-related).

Meritocracy could be really powerful system if consensus is reached on the weight of the votes and the type of voters inclusion/exclusion (for example doctors may not be allowed to vote on finance-related topics).

One big pointer here is that this contradicts with the democratic principles. Let’s imagine now a similar example but transferred to corporate governance which will be relevant for DAOs/DOs. Instead of voting on financial and healthcare-related topics people are separated into teams dealing with for example infrastructure, development, marketing. Giving ourselves 5 minutes to run a quick simulation in our mind gives us a pretty good perspective on how this would look like in a typical company.

Who can vote?

The votes inclusion has very similar characteristics to the previous point about how the votes are weighted however, in this case, one either can vote or cannot with one big IF.

We still have a type of voters exclusion even in democratic societies and these are the local elections e.g. electing a mayor.

In a democracy we vote for governing body of the whole state, everyone (except those charged with a capital crime and exceptions of this kind) can participate. However, when we vote for a city mayor only people registered and living in this particular city can elect it.

So why is that?

There’s a strong logical reason behind that and that is the ones living in the region have bigger incentive to elect person acting in their interest. On the opposite side, they also bear bigger risk if they elect one that has a real chance of doing damage in their region.

Based on this we can distil two main characters which are really important for voters inclusion:

• Level of incentive for picking party acting in goodwill
• Risk bearing/consequence of picking a party that doesn’t act in goodwill.

Example: When voting for a mayor of London people from Manchester have a lower level of an incentive of picking a party that acts in goodwill and also won’t suffer the consequence of picking a mayor that does harm to the city. Same goes with the organizations, just examples are related to decisions/geographies/functions and etc.

Now, how this could transfer in business and DAOs/DOs (in this post I assume that companies and DAOs/DOs have similar characteristics and goals although this is not 100% but is rather for illustrative purposes)?

NOTE: It’s slightly different whether the organization is NGO or for-profit, also how is setup (functional, project or matrix type) but this topic for another post. In this case, we assume that these are kind of the same.

Based on the characteristics that were outlined above we can determine two main types of inclusion:

• Limited — only people from department/division/give geography, a project can vote for a given topic
• Organization-wide — everyone can participate.

These two main characteristics outlined above can be used as a framework for evaluating which parties shall be included and which not in the voting groups.

How the consensus is reached?

There are many types of voting consensus models here but two are the most distinctive and are widely used:

• Plurality model — an electoral system in which each voter is allowed to cast only one vote, and the candidate/voting option that polls the most among their counterparts (a plurality) is elected.
• Majority model — in the majority model, the candidate/voting option that has 50% + 1 vote of the voters is the elected candidate/option.

Another type of electoral models are the proportional systems which we won’t go over here.

Conclusion

Blockchain-based voting systems have a lot of potential and can make a pretty significant change in the way governance is done. Still, this is a really delicate topic and nailing the system is important because as we see from practice mistakes related to governance are really costly both in a monetary and non-monetary way.

Already accumulated knowledge from the traditional voting systems can act as a starting point, however, they shall be taken with a pinch of salt.

Having a controlled testing environment is also crucial for accumulating enough know-how to be able to make these systems more stable and tailored to the specifics we face in DAOs/DOs.

Disclaimer: This post is a chaotic flow of thoughts that I hope is helpful for someone out there trying to figure out how to implement voting and electoral systems in DAOs/DOs.