An unaddressed conflict: stake vs. expertise

At DAOstack, we often refer to the conflict between scale and resilience as one of decentralized governance’s central challenges. I’d like to point out what I think is another important challenge that’s been largely unaddressed in the DAO space: that of stake vs. expertise.

Lots has been written supporting the idea that organizations should be governed by their stakeholders. The evolution of firms from owner-controlled to shareholder-controlled also supports this. In the DAO space, this looks like: reputation holders in the dxDAO should mostly be people who have earned Magnolia tokens by trading on the DutchX; rep holders in the PolkaDAO are Dot holders; and so on.

The reasoning is that stakeholders ought to have the best incentives to make their organization better, because they’re the ones who depend on its quality. However: having the best incentives is different from having the capability. Stakeholders may know the problems their organization needs to solve in order to improve, but there is no systemic reason to believe they will be good at finding solutions to those problems. Why would traders on the DutchX be able to tell the difference between good and bad solidity code, or proper legal language? What about Dot holders or GEN holders?

There are people who are experts in solidity code or blockchain and law, but these people aren’t stakeholders: they lack the proper incentives toward improving the organization. This is the conflict between stake and expertise. The bigger the DAO, the bigger this problem is: most people are highly skilled in just a few areas, so in managing an organization making choices in many areas (code, marketing, event planning, research, etc.), a majority of the organization is likely to be uninformed in every area! That probably won’t lead to good decision-making about specific projects.

Scale and attention also rear their ugly heads again here: even in a DAO that uses holographic consensus to limit the number of proposals people need to look at, expecting most members to regularly read and understand even a handful of detailed, specific proposals is a stretch.

I think protocols that address this conflict are possible (DAOs can pay other orgs to evaluate this kind of thing if they trust them, or use things like this), but before discussing that, I wanted to ask: do others agree that this is an issue? Have existing DAOs already been experiencing it? Should we start exploring solutions?

5 Likes
  1. I agree this is an issue
  2. Yes, we have for example already experienced this in dxDAO. Our current set of stakeholders are (for the most part) not those with any “expertise”. We have struggled with this very thing as people with “expertise” have struggled with our current reputation distribution. There have been active discussions on how we balance these two things - and some signaling around moving towards a more meritocratic based distribution.
  3. YES

Also, I really enjoyed your article linked above! Awesome stuff :slight_smile:

1 Like

The problem with a meritocratic system in regards to Reputation is that the rep being distributed will always favor the technocratic expertise of only a select few individuals with the ability to perform specialized work.

Considering how nascent DAO’s are, the majority of proposals are centered around code development, and therefore most rep will be earned and held by developers. A meritocratic system in this context has limited diversity in the functions that can be prioritized within a DAO.

It seems to me that this a common bias among DAO developers within DAO’s and one that does not lead to legitimate representation long-term, especially if the DAO wishes to expand beyond specialized functions.

1 Like

See Genesis DAO rep distribution (very far from being the case)

2 Likes

The underlying issue stems from trying to implement several control parameters with a single metric of Rep, Having a personal opinion about some initiative does not imply an ability to assess it’s feasibility or impact, nor represents a commitment to take a substantial and lasting role in either creating or consuming its outcomes. So the vote means a different thing to different participants. Every participant should to be listened to and perhaps even rewarded for showing up, but those dedicating their career and livelihood to the effort are simply not in the same boat, and so need additional ways to express their preferences.
There are unavoidable reasons why entities, for profit or public benefit ones, that start with a few founders do evolve to have the leadership and executives, a steering committee or a board, regular participants, employees, vendors, users, consumers, customers, investors, shareholders, donors public at large, It would be so simple if this tendency was only about power hungry hierarchies intent on hoarding economic returns. Indeed there are very different perspectives, timescales and the type and level of participation. If there are no close votes at different levels of an organization, disagreements flaring into heated arguments, backroom politics, hard compromises, then only a single perspective had usurped the power and the channels of communication, and the effort will hit a wall.

In drawing from the startup community experience, it takes 2-3 key contributors who cover a range of skill sets - vision, domain expertise, product ownership, several sets of implementation skills, operations, finance, promotion, sales. One person may be able to cover 1.5-2 of these well, or not, and any area left unaddressed will slow down the effort or blow it up in the end. People experienced with startups have some insight into their own abilities and value those with complimentary ones enough to share control and rewards with them. This often fails, so I’m surprised at the focus in the DAO community on how to scale to 5000 or a million participants, where scaling to 2 or 3 is so nontrivial, particularly longer term.
The reason that the techies may appear to have undue influence or oversize value in DAOs today is that the effort is so early that there are effectively only the evangelist and the coder roles, often played by the same individuals. And the bootstrapping projects, with participants contributing spare time, have not hit the interpersonal, economic and marketplace realities yet, where filling other roles become unavoidable. Can these projects scale to full time participation by a strong team of diverse key contributors, while already having amassed and dragging along hundreds and thousands of decision-making but largely absentee participants, I do not know.
I intend to grow a DAO one participant at a time, and if it starts as a successful collaborative organization of two or three, I’m fine with that.

2 Likes

I agree with this last view - trying to deconstruct an entire organisation down to a few simple rules is impossible, however learning from the startup / corporation / public utilities / private / public market governance models that operate at an ever increasing scale is key. There is no silver bullet to solve ‘governance’, no code that will accurately define the nuance and humanity of any scale of organisation.

Thinking about DAOs at the simplest possible group size, then adding team members into a project that already has some sort of defined direction and mission North Star, with outsized weight of reputation assigned to those who are articulating and driving that same vision and delivering the work required to build towards that vision is the only real way these DAOs can work. I do believe that early reputation should be respected but just as founders might have limits to their impact, influence and interest in a company at various scales, then so should DAOs. Creating meaningful additions and clauses to the reputation - potential decay over time for founders needs to be baked in, but equally not in such a way that a 1 / million founding team can’t maintain a solid grip on the overall direction.

I would say that in the excitement of the potential of DAOs to solve all the problems, at all the scales, there is a need to go back to finding teams with projects that are messy and complex and very human, but that collectively have the stamina to build something over the long term.

DAOs are just great new coordination mechanisms, a tool in the arsenal of a motivated team that can help drive efficiencies, openness and alignment with stakeholders in the management of projects - they are not an answer in of themselves.

In summary, I think that the issue isn’t the tech, it’s the lack of projects with already engaged communities and well articulated narratives.

1 Like

This is a pretty common criticism of DAOs, and while it may seem true on the surface, it’s not necessarily applicable in practice. It’s not clear if DAOs should:

(1) aim to decentralize currently centralized organizations
(2) create new structures that are decentalized from the start (e.g. dxDAO)

I’m not frankly sure that (1) is possible – especially if there’s already a management structure distinctly in place. It may lead to greater success / sustainability just to deploy decentralized structures from the get go rather than try to transition centralized structures. That being said you could decentralize some component of a centralized entity, e.g., use a DAO as a curation instrument for articles to be published in a magazine that’s privately owned by a publishing firm (say, by the readers).

Do note that I’m referring to DAOs as emergent orgs embodying a sufficient quantity of persons to facilitate decentralization (~150+) and not the currently frequent misuse of the term in reference to small groups leveraging a multisig.

2 Likes

Thanks for the replies, all. The discussions around how large DAOs distribute reputation are quite interesting – sounds like there’s still some more communications we should work on in that area – but I actually don’t think it’s that relevant to the OP here.

Regardless of who is given reputation, stake vs. expertise is a problem (especially for large DAOs): either

  1. you give voting power to all stakeholders, and those stakeholders aren’t experts in everything the DAO does, leading to naive decisions; or
  2. you give voting power only to experts in everything the DAO does (if such people exist…), and you exclude many stakeholders, leading to decisions that don’t represent the DAO’s real needs.

Something in between just gives you a mix of both problems (a mix of misaligned & naive decisions). It’s not the reputation policy that can address this problem, but the decision-making mechanism. See the article I linked in my first post for some thoughts in this direction.

Another direction might be creating “expertise oracles” that are hired to give proposals a grade corresponding to their expertise. These would be incentivized through pay and their business reputations (i.e. bad oracles don’t get business). This could start simply as individuals offering their services to a DAO–e.g. make a proposal to Genesis to grade all boosted proposals on their technical soundness for the next month in exchange for X monetary reward (no rep).

A third direction is for DAOs to be siloed into departments of expertise, as I think Colony is trying. I’ll admit I still don’t totally understand their structure, though.

2 Likes

The intersection between representation and expertise is a delicate balance that even our best bureaucratic governments struggle to deal with. Its the paradox of living within a representative society, as the will of the populace is not always logical, and often runs counter to the logic provided by the expertise of a technocratic workforce.

All risk contains an element of chance. It has always surprised me that even in a place like San Francisco, where almost everyone is aware of the long-term risks posed by earthquakes…somehow believe they will not be present on the day that it occurs.

When it inevitably does, we all wonder why it was ever allowed to happen in the first place.

The function that expertise plays within an organization or government, especially one that strives to be representative, is to educate! When education fails, it remains the obligation of those with expertise to then mitigate such a compromise that reduces the element of chance for the progression of the organization and its members.

From this perspective, the questions become:

(1) How do we enable those with expertise to effectively communicate the risk of proposals, and to mitigate that risk if members of such an organizational body choose to move forward?

and

(2) How do we reward such expertise, so that it doesn’t undermine representation through the consolidation of voting power?