Moloch DAO & Permissioned DAOs


I am curious what people’s thoughts are on permissioned DAOs, where to be a member of the DAO, you have to be voted on by other members.

This is one feature that intrigues me the most about Moloch DAO because it is breaking the initial DAO paradigm. I initially thought “well this isn’t a DAO because not everyone can join, so it’s not decentralized”.

But is the MVP of DAO basically the organization operates on decentralized technologies (a public blockchain) using smart contracts?

Does the more open and anonymous model bring too much risk and paranoia about incentive alignment?

Curious to hear people’s thoughts on the future of DAOs!


I think we should move away from the term DAO entirely. It is not a useful term for people outside of this space, and even within the space it doesn’t seem to be a particularly useful distinction.

One of the issues is that “decentralized” and “autonomous” are nebulous descriptors. We don’t have a good way to define decentralization (what axis of decentralization are we talking about?, at what degree to we consider something decentralized enough?), and autonomous is similarly vague (autonomous relative to what?).

In practice I think there are two common interpretations for what a DAO is:

  1. An organization which is “decentralized” enough to operate “autonomously” (eg it generally cannot be shutdown by an external process). Aragon organizations fit this definition quite nicely, and by linking the definition of decentralization with autonomy we resolve the lack of clarity in what is meant by “decentralization.”

  2. A coordination or service protocol which is defined by software and coordinates the activities of agents in such a way as to reliably produce some output without requiring a centralized coordination mechanism. Examples of this are things like Proof of Work, Proof of Stake, or Decentralized Oracle mechanisms.

I think both of these things are useful, but because they both are reasonable definitions of a “DAO” but have very different properties the term “DAO” isn’t helpful. The former can be more accurately described as an “On-chain Organization” (though I don’t love this because I don’t think its particularly useful terminology outside of the blockchain ecosystem)-- And I think the latter is more aptly described as a coordination or service protocol.


My 2c:
I have no issues with a curated/invite aspect. For me there are always actual/perceptual barriers to peoples’ participation in projects such as this.
There’s a touch of babies/bath-water here… I’d still say that DAO is a hugely useful term, even only as a rallying call. Perhaps unsurprisingly I’m considering what we’re doing the first iteration of the CoAO concept - we don’t know where this is going yet (& that’s the best bit:).
I also get you’ve been really digging into this hard and your insight is given appropriate weight.
“what axis of decentralization are we talking about?”
A cooperative axis/model of decentralization I’d say - what exactly does that mean? - for me, a fork of coop, platform-coop, & emerging DAO praxis (past,present, future).
I still think of what we are doing as a liquid structured experiment in (people first) governance and organizational autonomy.


I don’t consider it a DAO since it is not permissionless to join but it does seem like an interesting experiment anyways.

That said one could also argue that proof of stake systems are “permissioned” in a way because a holder of the stake must give you permission to buy some before you can participate in the system. Whereas with proof of work, one could theoretically (and nowadays, realistically) build a computer from scratch and start participating, no matter how marginal that participation may be it would still be totally permissionless with no other cooperation by another person participating in the network necessary.


I’m all for it.

Governing the commons:
Design Principal #1 - Clearly defined boundaries.

Life is paradoxical; if we want an open and safe play ground, we have to close it off to perverts… dissonance much? yes, too much, but I like LR for the same reasons. We’ll defiantly see traditional hierarchies emerge as we haven’t yet figured out how to organise scale free like the starlings :wink:

That said, categories break when one changes the examined scale :slight_smile:


Maybe it makes sense to call “On-chain organization”: “Instantiated” as opposed to a widely used term “Incorporated”. It captures both the technical way such an organization is created (using programming command) and the easiness (and permissionless) of creating it.

I believe DAO term to be very powerful and wouldn’t stop using it because so many people recognize it. But, also I don’t mind applying it in a more broad way, which for me includes all the instantiated organizations which could easily be turned into a DAO by changing only the internal governance rules.

MolochDAO and other permissioned instantiated orgs are a welcome step towards more real DAOs in future. But even when we have real DAOs, I don’t think they will amount nearly to the number and importance of organizations that are just instantiated.

This is exactly what we have discussed last week @stellarmagnet.


It makes a lot of sense to me. In a “democracy” DAO, you’d need the existing token-holders to vote to assign tokens to a new address. Make the tokens non-transferable, and you’ll end up with a permissioned DAO.

For a project like Brightid that has no utility token or profit-sharing, but relies on donations and work-for-hire, this structure makes a lot of sense.

MolochDAO is interesting, but I wonder how it would work with members who are aligned in their goals to build a common infrastructure, stand to benefit equally, but have different amounts they can afford to sacrifice. A smaller sacrifice means fewer shares and fewer votes, but also means somewhat of a free ride.