research

Governance in the Age of the DAO: Tooling and Frameworks

research

Governance in the Age of the DAO: Tooling and Frameworks

April 2022

Posted by

Martin Saps

Program Associate

DAOs create opportunities to democratize decision-making, but what does that mean when it comes to day-to-day functioning?

Posted by Martin Saps - April 2022

April 2022

Posted by

Martin Saps

Program Associate

In the previous article of this series we discussed the proliferation of DAOs as a paradigm shift and noted three major categories of DAOs currently active: Protocol DAOs, Investment DAOs, and Social/Worker DAOs. More broadly, digital technology is allowing for a radical shift in opportunities to democratize work; anyone can contribute where they see fit on their own time and terms. But, the day-to-day reality is far less romantic: decentralizing decision-making for large groups is not easy, DAO organization, decision making, and resource management being the three biggest challenges.

Here’s an example: CityDAO gained international recognition with its communal ownership of a piece of land in the most remote corner of America still [barely] accessible by car. CityDAO’s Discord has channels in four non-English languages, sold out of citizenship NFTs, and has  $15+ million sitting in a treasury between ETH, fiat, and other DAO tokens. Things seem to be going really well for CityDAO as a social club, but administratively their resources are dispersed and aligned objectives are just now starting to emerge. The ever-growing community continues to squabble on what to do next, even as the newly-elected council slowly develops the frameworks needed for structuring.

So, what do we make of CityDAO then? On paper, it achieved its goal of forming a decentralized organization and buying a property. On execution, it’s so decentralized that it has problems getting anything done. With such few precedents on governance, it’s a fate doomed to be repeated during this early period of DAOs. At present, running a DAO is like inviting members of a 50,000-person city to a public forum with no moderation or means of adjudication. From a technological point of view, we’re only just now developing the right tools needed for DAOs to take shape.

Enter DAO Tooling

Governance in the Age of the DAO: Tooling and Frameworks Outlier Ventures

Within our DAO series, this article dives into the day-to-day management of DAOs and the technology which makes it possible. Specifically, this article will focus on three basic components of DAO technology: “the management”, “the ballot box”, and “the treasury.” Taken together, these three sets of tools can facilitate the day-to day allocation of resources within a DAO. 

<See this DAOtabase of tools for an increasingly-comprehensive list of resources available to DAOs

The Management

Governance in the Age of the DAO: Tooling and Frameworks Outlier Ventures

Management tools allow DAOs to make decisions and move funds. Operating with the framework of progressive decentralization, they are flexible and can be programmed in a multitude of ways: monarchy, oligarchy, democracy, sociocracy, and plutocracy are all sanctioned ways of interacting with these frameworks.

Aragon

The most well-known administrative tool, Aragon positions itself as a company-building platform tailored to the needs of DAOs. Claiming to have launched over 1,900 DAO’s, its website has countless case studies of “virtual worlds, subreddits, hacker collectives, DeFi protocols, and fashion houses” using Aragon to build their communities. Ironically, for a project named after one of Spain’s 17 Autonomous communities, it doesn’t actually matter what organizations built on Aragon look like – centralized or decentralized, democratic or top-down, all are welcome.

Aragon has rolled out countless plugin tools for different areas of their business including; Aragon Client, Aragon Voice, Aragon Govern, & Aragon Court, designed to cover all aspects of company functioning including dispute resolution, communication, and client relationships. In many ways Aragon utilizes DAOs as the face for an excellent and comprehensive suite of business services.

Colony

With a reputation for consistency and thoroughness, Colony is a newer competitor to Aragon. Colony is far more focused on the quotidian; its heavy focus on governance and decision-making is evidenced by its signature “lazy consensus,” in which all proposals pass automatically unless someone opposes. Much like Aragon, Colony allows colonies (as organizations built on Colony are cleverly called) to organize on varying levels of decentralization. 

Built on Gnosis Chain (formerly xDai), the weight of an individual’s votes within Colony is dictated by their “reputation,” a time-decay rating earned through doing work for the Colony.

DAOHaus

Founded by Ethereum founder Ameen Soleimani, who also founded MolochDAO, DAOHaus is an open-source suite of tools for DAO-makers. The no-code open-source platform runs on Moloch’s stack and has a communal safe which it aptly calls “The Bank”. This bank can store a community’s native token as well as NFTs, and members can request funding from the DAO as compensation for work they have done.

Much like Colony differentiates between funds and reputation, DAOHaus differentiates between Loot and Shares. While the former corresponds to the DAO resources a member is entitled to, the latter gives that member ability to take part in DAO activities such as voting. DAO members can lead insurrections against others by filing a GuildKick, a proposal which, if passed, converts a member’s shares into loot, thus stripping them of their voting power. Once that member is “jailed”, they can easily be “Ragekicked”, and their loot returned to the bank. On the flipside, there is the infamous “Ragequit” function, which allows any member of the DAO to leave and withdraw their share of loot at any time. DAOHaus is an interesting open-source protocol with far more functionality for community governance. It remains to be seen how that functionality will play out as their user base grows.

One to Watch: Orca Protocol

Still in its private beta phase, Orca is a DAO-building protocol organized around small working groups, called Pods. On Orca, each Pod within the DAO has its own Gnosis Safe. Access to that safe is token-gated through a membership NFT, which acts as an access “keycard.” Orca also offers a Software Developmment Kit for developers to program their Pods in unique ways.

The Ballot Box

Governance in the Age of the DAO: Tooling and Frameworks Outlier Ventures

We’ve seen how these DAO tools work, so how are decisions resolved under the hood? In Colony, all decisions unanimously pass unless someone opposes. But what if someone does oppose?

Snapshot

Collective decisionmaking is where tools like Snapshot, a Web3 chain voting protocol developed by Aragon and Balancer Labs, come in. Snapshot allows DAOs, or “spaces,” as it calls them, to vote off-chain while supporting on-chain integration. Voting can be configured in countless ways: first-past-the-post voting, ranked choice voting, and quadratic voting among others. As a platform Snapshot stays in its lane: it is very good at what it does and promises nothing else. It is utilized by prominent DAOs like OlympusDAO, Uniswap, and BanklessDAO and plugs in with Colony and Aragon through an API. The results of that vote are then executed directly by the DAO treasury (Gnosis Safe, etc.).

Discord:

At the moment, most DAO’s do their voting informally through Discord. Not necessarily because it is any better than Snapshot, but rather because many DAOs have their beginnings as a Discord server and would rather not migrate their users to a new system. Until recently, there was a fragmentation between Discord polls and on-chain decision mechanisms; the two had to be connected via SafeSnap, an application which turned Snapshot votes into financial decisions via Gnosis. Thanks to Zodiac’s Reality module, Gnosis Safe now allows for Discord polling as on-chain decision making.

Tally:

A platform that offers similar tools to Snapshot although less widely adopted.

The Treasury

DAOs also need a place to store their funds such that they are not in the hands of one person. And they need a mechanism by which to dispense these funds for community activities. Unless they run on DAOHaus, which has its own safe, DAOs will likely use Multisignature wallets such as Gnosis to keep their funds.

Gnosis

Gnosis is perhaps one of the most exciting protocols in the Web3 space today. The organization, which recently absorbed xDai chain and renamed it eponymously, sees itself as an on-chain treasury of organizations. The project runs  “three interoperable product lines allow you to securely create, trade, and hold digital assets on Ethereum.” Most relevant to DAOs, Gnosis Safe is communal safe (colloquially called a multisig) holding assets to which multiple people can have the key. In this sense Gnosis is more on the level of essential tools like MetaMask and TrustWallet. Whereas the latter allow individuals to store assets, the former allows communities to store assets collectively.

The tool is essential for building organizations in a trustless, decentralized way. Giving all or some members of the organization access to a safe ensures that the money can remain in the hands of the community. What’s more is that safe can be permissioned in a multitude of ways, giving different wallets access to different assets. Adding in SafeSnap, there is a trustless protocol for ensuring that the results of a Snapshot vote are actually put into effect.

Some Missing Pieces

DAO tools have thus far worked in limited use-cases around capital management; in these cases “the management” run the day-to-day functioning of the organization, “the ballot box” allows members to decide on those proposals, and “the treasury” dispenses the money to the “management,” repeating the cycle. Delegation is also emerging as a key principle, e.g. as a default for governance participation in many DAOs.

Outside of this limited scope there are countless missing pieces to the DAO puzzle. One example of this is organization and social governance. If we see Discord as an “Agora,” how can DAO members turn discussions into actionable insights? Some de-facto leaders in CityDAO are trying to solve these difficult questions. As great as their community manager is, though, it’s nearly impossible for a team of less than ten people to organize a large community using Discord alone. Protocols like Sobol and DeWork are trying to crack the Admin/HR dilemma, and others like Collabland are trying to make community management more efficient; but social governance might be unsolvable purely because contributing to the Discord community is the objective that communities are trying to manage.

It’s clear that DAOs will have to address challenges like these if they want to operate as productively as Web2 companies. But that’s the thing – maybe they don’t. And maybe that’s okay too.