Welcome to the second in our four part series on strategies to navigate the present and future landscapes of composable organizations.
Understanding the foundational, conceptual history of DAOs is critical to understanding their potential. Marshall McLuhan famously coined the phrase "The medium is the message" and this certainly holds true for the organization of communities. For some types of organizations, it will be better to be registered with paper documents and with bylaws, operating procedures, etc. memorialized by paper. For other types of organizations, it will be more preferable to spin up an organization digitally and making use of blockchain technology.
This is all to say that each container that is used to represent the various functions of an organization, such as membership, roles, permissions, decision-making, etc. has different strengths and weaknesses. DAOs and their ilk are no exception to this rule. In fact, the primary motivation of researching this subject is to better understand the costs, safeguards, market conditions, and norms under which different architectures are preferable to others. To that end, the first part of this post will be about why there is such an interest in the concept of composability for Web3 products; the second part of this post focuses more upon the specific features of DAOs that could be composed in different ways.
Why The Interest in Composability?
Building upon the previous discussion in Part I, a prominent theme in this discussion is composability. In this part, the discussion will be geared toward exploring why composability is important and analyzing DAOs (and their ilk) to gain a better understanding about how cryptographic projects can be composed.
There are lots of opinions about why composability is important in crypto. Before getting to those, though, I want to look at an example about what happens when systems are not composable and interoperable.
In the early days of the railway industry, people all over the world developed their own types of tracks. As such, the length between the tracks (that is, the rail gauge) were different widths. Some of the popular rail gauges were as wide as 7 feet/2.134 meters and others were as narrow as 4 feet 8 and ½ inches/1.435 meters. This lack of standardization makes it much more difficult to engineer a system to get from one point of the country to any other part of the country. As a result, a whole industry developed to fix these break of gauge problems. On top of that, there was even The British Gauge War wherein the operators of one railway would intentionally build their tracks to different gauges than the nearby tracks in the hope of controlling new territories and capturing certain bands of commerce on their individual tracks.
Eventually, however, a standardization occurred across British railways and the system in place is more reliable and efficient. If you are looking for a much more global and comprehensive economic analysis about the standardization of rail gauge, I recommend checking out: Path Dependence in Spatial Networks: the Standardization of Railway Gauge. If you are interested in some additional examples where standardization of complex ideas resulted in a relative consensus, I would recommend checking out the Typing Wars and the VoIP Wars.
All of this is a prelude to understanding the complex state of affairs that led to the modern architecture of the internet; this is also why the dominant model of commerce on the internet today is built on open. A byproduct of openness, collaborative development, and the broad marketplace for ideas that is the internet is that it is much simpler to compose digital networks, markets, businesses, services, interactions, documents, and information items in a way that achieves a particular purpose. The graph below from CommerceNet's eCo Framework Project in 1999 is an early example of how some pioneering folks thought the internet could be composed for commerce.
Looking back through the history of railway gauge, typing keyboards, VoIP, and even early models of the internet demonstrates something critical – composability is innovation. This is a critical building block for blockchain generally, and the Web3 community in particular.
"Smart contracts are public on Ethereum and can be thought of as open APIs. You don't need to write your own smart contract to become a dapp developer, you just need to know how to interact with them." - "Smart Contract Composability" from Ethereum.org
"[S]mart contracts can act as building blocks that are able to be assembled into larger systems. Just like a software library, smart contracts for different protocols and applications can easily plug into each other like Lego pieces." - Lindy Xie "Composability is Innovation" a16z
"Within crypto, composability is the ability of decentralized applications (dApps) and DAOs to effectively clone and integrate one another (syntactic composability), and for software components such as tokens and messages to be interoperable between them (morphological composability)." - "What is Composability?" Aragon
These different flavors and variations on composability have become the competitive edge of decentralized projects over their equivalent legacy industries. This discussion provides a conceptual answer to the question of why composability, the next section examines more thoroughly what can be composed for these new cryptographic organizations.
What Are The Core Features of DAOs?
As a practical matter, identifying a core set of features for DAOs is limited to a specific point in time. New innovations are always happening and new opportunities always emerging. As such, the following analysis focuses partially on the legal status of DAOs, the governance of DAOs, the behavior of DAOs, and the technical mechanics of DAOs.
Legal Status :: Registered v. Unregistered
This category comes first for a few different reasons. Firstly, this typology is relatively simple - DAOs are either registered with a central authority or they are not. In the first draft of their Model Law for DAOs, the COALA research group points out:
"Decentralized Autonomous Organizations (DAOs) can be classified into two distinctive categories: registered DAOs, i.e. DAOs that are organized according to the laws of a State and that are registered in a corporate registry, and unregistered DAOs, i.e., DAOs that are created outside of the legal frameworks defined by national laws and are not registered in a corporate registry. The vast majority of existing DAOs are unregistered DAOs and their legal status is currently uncertain: they are alegal."
Registration of a DAO carries with it myriad benefits, the primary of which can be associated with forming any type of business entity. In cases similar to an C-Corp, there is a limitation of liability. In cases similar to an LLC, there is an ability to spin these up rather quickly. In all cases, there is a degree of rigor required to register that is rewarded with a series of benefits, and subsequently increased predictability around the behaviors that a registered DAO can undertake.
Registration is also important because the default type of organization that is implied in the US, where none is explicitly stated, is a General Partnership; General Partnerships are likely not the most desirable for a big project with lots of participants, because each participant can be liable for the acts and omissions of other general partners.
Some legal scholars will be wringing their hands over the fact that it is possible to obtain a limitation of liability (one of the primary benefits for registering an organization with a legal status in the first place) without actually registering with a central authority. In A Legal Framework for Decentralized Autonomous Organizations, David Kerr and Miles Jennings point out that a more lightweight way to launch a DAO with legal protections will be via an Unincorporated Nonprofit Association. It is important to note, however, that Unincorporated Nonprofit Associations are typically products of a single contract among members and can become quite unwieldy as more complexity arises and may lose their fitness for purpose with the natural progression of time.
Decision-making :: Modes of Governance
One layer beneath the Legal Status of DAOs that can actually be legally mandated is the modes of governance for DAOs. Typically, for registered organizations, this will come in the form of bylaws. For unregistered organizations, there might be standardized processes for governance and there might not be.The life work of Elinor Ostrom actually focused on how informal networks of governance for common pool resources can be better than through formal networks, challenging legacy assumptions and earning her the Nobel Prize in Economics. It follows that the choice of governance is best left to those who are deciding. However, as the focus of this article is composability in the context of governance, it is worth highlighting that governance may indeed be composed.
CommunityRule, a project from the Media Enterprise Design Lab at the University of Colorado Boulder has a book that serves as a framework for composing governance in a number of distinct flavors – Benevolent Dictator, Consensus, Circles, Do-ocracy, Elected Board, Jury, Petition, and Self-Appointed Board. They also have a forum where anyone can view templates of these forms of governance or upload their own template to a library of templates for governance. Already, several DAOs, including MetaGov, are uploading and sharing their own rules such that they might be adopted, iterated, or otherwise analyzed.
Behavior :: Coordination DAOs v. Venture DAOs v. Hybrid DAOs
Additionally, there are substantive considerations that will shape the landscape of DAOs. Some of these considerations relate to the behavior of DAOs. The behaviors of DAOs are important because they will determine how they will be treated from a legal standpoint. To illustrate this point, in the United States, investment contracts are subject to the Howey Test and, thus, would be considered a security. The criteria articulated in Howey for an investment contract is as follows: a) an investment of money, b) in a common enterprise, c) with the expectation of profit, d) to be derived from the efforts of others. If an organization engages in behavior that meets these criteria, then the transaction is subject to the disclosure and registration requirements under the Securities Act of 1933 and the Securities Exchange Act of 1934.
The two easiest paths to avoid this are to either function without attempting to achieve a profit or to participate in the creation of a DAO. The latter example is similar to an employee earning stock in a company.
- Coordination DAOs :: function without profit as a driving motive are coordination DAOs – they serve to coordinate activity for a group of people.
- Venture DAOs :: function with profit as a driving motive and that are subject to the heightened requirements of the SEC are Venture DAOs.
- Hybrid DAOs :: function with some combination of the above DAO tooling (e.g., a mechanism for coordinating and a mechanism for investing); these mechanisms could be blended into one organization that meets the heightened requirements of the SEC or the mechanisms could be stratified into two organizations that closely collaborate together.
In order to avoid the onerous requirements of the SEC, many DAOs will seek to at least preliminarily start out as a Coordination DAO, only later to a) blend venture activities into the operation of the DAO or b) establish an affiliated Venture DAO under specified conditions.
From a practical perspective, however, perhaps the biggest issue that arises when categorizing DAOs is the degree to which they are autonomous. For example, if the operations of a DAO (especially voting) are enacted entirely onchain, then using the moniker "DAO" is likely not an issue. However, if the operations of a distributed organization are not entirely onchain, this invites an interrogation of whether such organization is actually a DAO or whether it might be something else?
In his CryptoLaw Newsletter, Gabe Shapiro went about Defining Real and Fake DAOs in a more thorough manner:
"A popular misunderstanding of smart contract technology is that smart contracts work 'automatically' without human intervention—that they are some sort of 'non-human agents' which engage in 'algorithmic governance' of organizations. On the contrary, smart contracts are just passive object code stored on the blockchain. This code is called by miners/validators when users request it to be called by offering to pay the miners/validators for making such calls and writing their results to a new block. In other words, smart contracts never do anything unless one of their functions is specifically and manually called by a human (or the agent of a human—e.g., a 'bot')."
One popular example of a cybOrg would be an organization with decentralized governance that requires some number of signatures to a shared wallet, such as a multisig wallet. From a technical standpoint, it is also worth pointing out that a project may be decentralized from a technical perspective (e.g., the architecture of the organization), a governance perspective (e.g., the decision-making of the organization), both together, or neither. For example, the use of a multisig wallet in combination with snapshot voting is decentralized from a governance perspective, but less so from a technical perspective. This architecture could be desirable in cases where there would be a high cost of voting for a large number of people with lots of proposals, but less so for ensuring total technical decentralization of the organization with fewer decisions and fewer proposals.
At first, labeling something as a DAO or not might only seem like a cosmetic or marketing decision. However, because each technological architecture of a digitally-enabled organization relates specifically to a different set of safeguards, this naming matters. For example, the safeguards that need to be in place for a multisig + snapshot are going to be different than the safeguards that need to be in place for on-chain token governance.
This gets us to the big question – what can I do to ensure my crypotgraphic organization is composed in the right way?