Analyzing the current development status of decentralized autonomous organizations: What can we do with DAOs?

Written by: Philippe Honigman

Translated by: La

Year after year, DAOs have garnered increasing attention from the crypto community, but we are still far from the golden age. My goal is to update the perspectives on "how we currently view DAOs" and "how to use DAOs."

As in previous years, 2020 has been dubbed "the era of DAOs." However, most people who have heard the term find it difficult to define the concept. I have attempted to provide definitions of this term elsewhere and clearly explain the related narratives. Aragon recently tackled the challenge of "defining what DAOs are" in an excellent manner. In this article, we should remember Vitalik Buterin's original proposition: DAOs are organizations centered around autonomy, with humans as the boundary.

For decades, companies have viewed decentralization as a way to improve the efficiency of production processes and enhance product quality. When we think of automated factories, images of robotic arms functioning like a panacea on the production line come to mind—or, to give a more familiar example: the mobile robots in Amazon warehouses. The natural candidates for decentralization are processes involving routine tasks that require little or no space to perform the task of "making decisions."

On the other hand, tasks related to the design, coordination, or optimization of such processes are considered to require individuals with soft skills, such as understanding complex situations and interpersonal communication. Thus, the core functions of organizations retain the undeniable privilege of humans. This privilege is unlikely to be threatened by the strong AI that is still a long way off.

The typology introduced by Vitalik disrupts this view. He argues that while the margins of an efficient organization may still rely on human input, much of the complex coordination at the center of the organization can operate autonomously and use the organization's internal funds to reward workers. From this perspective, software provides a way to "replace traditional economic and social coordination mechanisms."

So, after years of building and experimenting, what can we say about this concept today? Let us analyze "Decentralized Autonomous Organizations" based on their recent developments.

Autonomy

DAOs are autonomous, meaning they are not controlled by third parties. Because they operate on public blockchain networks, DAOs can manage their own funds and allocate resources according to self-imposed, hard-to-break rules. This is why TheDAO in 2016 was presented as "unstoppable," a term that means no one can change its behavior or terminate the computer it is running on.

While this idea was very appealing at the time, the outcome was that operating a fully autonomous system brought many undesirable potential consequences. When TheDAO was hacked through a vulnerability in its code, millions of dollars were obviously drained from the treasury, but no one could do anything about it. The system operated according to how it was programmed; some argued that it was more likely to be attacked by the system's functionality than by hackers. If "code is law," then the troublesome vulnerabilities in TheDAO's code were also part of the agreement binding the parties involved under TheDAO's smart contracts.

This is the meaning of autonomy. The funds of a DAO are controlled solely by its code.

What Have We Learned?

The use of blockchain-based autonomous code quickly revealed problems, and not just for DAOs. As individuals and organizations, we want to have recourse when issues arise. If we make a mistake in a bank transfer, it means the bank will help recover the funds, and our money will be returned. When two companies disagree on the execution of a contract, they may argue their case in court.

In contrast, consider when someone sends cryptocurrency to the wrong account or when a vulnerability locks up millions in a wallet, such as in the Parity multisig contract incident. If there is no pre-planned method and implementation in the code, there is no way to cancel or reverse those actions. And because it is impossible to guarantee that a complex string of code is entirely free of vulnerabilities and weaknesses, it is possible that even the most carefully protected and edited code cannot safeguard its users.

For those who want to protect their assets and remain independent from corrupt judges or failing states, "immutability" and "resistance to censorship" are valuable. The autonomy of code is a way to establish agreements for individuals and collectives that no third party can interfere with, regardless of how powerful or wealthy that third party may be. However, it is worth emphasizing: the autonomy we truly desire is for humans (as individuals or collectives), and the autonomy of code is a method, not an end.

Autonomy in 2020? Balancing Code vs. Community

The focus has shifted from the absolute autonomy of code admired by cypherpunks to a relative autonomy that combines pure code and human choices.

One way to handle this combination is to provide space for subjectivity in the autonomous process. Kleros and Aragon have pioneered decentralized arbitration services, which can make DAOs more flexible. For example, Aragon Agreements allow organizations to define human-readable agreements, and when disputes arise, a decentralized court jury can enforce these agreements.

Another way to protect people from the blind operation of autonomous rules is to provide them with a way to exit the system before being harmed. This is the idea behind the ragequit mechanism offered by Moloch DAO: any investor in the DAO fund can withdraw their investment when they disagree with a decision voted on by the majority of members. For fairness, this protective measure was already present in TheDAO in 2016, but ironically, it was flawed, and its implementation led to TheDAO being attacked.

Finally, we see increasing concerns about the legal risks for DAO members, which can be addressed through legitimate forms such as wrapping the DAO. Several jurisdictions now offer such legal tools, allowing DAOs and their members to avoid potential legal uncertainties that could lead to infinite personal liabilities. For example, dOrg, as a developer cooperative registered as a Blockchain-Based Limited Liability Company in Vermont, and LAO, as a venture fund organized as a Delaware LLC.

In each case, the ideal of a purely code-driven, purely crypto space is balanced with the legitimate concerns of protecting the people who initially established this space. Decentralized courts make it possible to "protect the community from rogue individuals." Rage quitting prevents the majority from exerting excessive power over minority groups. Providing protection through regulations and applicable laws, using law as a wrapper, prevents threats to DAO members.

Decentralization

Decentralization can be understood as the process of reducing or avoiding the concentration of power within an organization. In a decentralized structure, resources and decision-making rights cannot be monopolized by a few individuals.

Previously, legal arrangements broadly achieved decentralization, such as cooperative statutes or antitrust laws. DAOs and crypto networks inspired by Bitcoin have accelerated decentralization using a combination of economic incentives and programmable rules.

For instance, cryptocurrencies are based on a consensus state of information between parties regarding the holding of funds and transactions between accounts. Using public, permissionless platforms, anyone can participate in negotiations to reach consensus. The absence of gatekeepers means there are no rent-seeking mechanisms, and there are no hierarchical relationships among members in the network; every contributing member receives equal rewards.

What Have We Learned?

It has taken some time to understand and acknowledge: public blockchains are not as decentralized as they appear. For example, Angela Walsh reminds us that decentralization exists on a spectrum, relying on the standards used to measure decentralization; even the largest permissionless cryptocurrencies often exhibit significant centralized power.

The same criticism can be directed at DAOs. Many DAOs have highly uneven token distributions, where a few members can block or pass any proposal. In other cases, the personal influence of founders or key members is more significant than their actual voting weight through tokens.

If we view decentralization as a gradual process, then this is not necessarily a problem. While decentralization is praised as a means to reduce inequality and ensure the accessibility of public goods, this is entirely unverified for launching a new project. As elsewhere, in the crypto space, it seems that nothing can compete with a small team of strong leaders, even protocols designed precisely for decentralized collaboration (such as DAOstack's Holographic Consensus).

A reality check: set aside the dazzling allure of decentralization and look at the real projects in the crypto space. They are all under the control of founders, and most significant projects are organized according to traditional corporate governance systems, with shareholders and accountable managers. Cooperative and elite-led communities, like dOrg, Metagame, or LeapDAO, remain exceptions that still need to prove themselves.

Centralization in 2020? More Options for Effective Decentralization

Decentralization is challenging, but there are numerous ways to achieve it! Here are some interesting methods we've seen recently.

Token-based voting is the primary tool for collective decision-making in DAOs because it is the simplest way to prevent Sybil attacks, which involve voting with multiple identities. Sybil attacks have long been a headache in the crypto space, essentially involving pseudonymous identities. People do not need to verify their identities to participate in public crypto networks like Bitcoin or Ethereum; they only need a pair of cryptographic keys, and anyone can create any number of keys for free. Token-based voting is an effective way to resist Sybil attacks because tokens are not easily created as crypto assets.

However, token-based voting has also faced criticism for its adverse effects on decentralization. As mentioned earlier, most crypto networks are not thoroughly decentralized because a small number of addresses control the majority of tokens. Token-based voting tends to favor oligarchic rule and perpetuates power asymmetries. It also diminishes participation among the majority when they realize their votes carry little weight.

Belief voting can sustain a community's favor over time, rather than organizing votes in a short period and allowing stakeholders to choose which areas they want to influence. This makes Sybil attacks and vote-buying more difficult and helps eliminate voter apathy. 1Hive has implemented a belief voting system via the Aragon App.

Quadratic voting is another method to mitigate the "wealth effect of token-based voting," while still allowing those with more skin in the game to express their preferences. Gitcoin has successfully experimented with funding public goods.

Another approach to decentralization is to rely on decision-making power allocated to those who contribute to the network. When DAO members' voting rights are based on their actual contributions and cannot be transferred to others, the power asymmetry becomes less severe. Various reputation systems have been designed for this purpose, such as DAOstack, Colony, or SourceCred.

Recently, a similar idea has emerged in the DeFi space, referred to as "Simple Agreements for Future Governance" (SAFG). Using Compound and its COMP token as an example, this approach involves granting tokens without economic rights to those participating in decentralized networks or protocols. Later, when the network is widely used, token holders can choose an economic model that adds economic rights and transferability to the tokens. Allocating decision-making power to actual users is intended to prevent rent-seeking and encourage the decentralized development of the network. However, this must be handled cautiously, as it introduces new attack vectors for token holders who may attempt to manipulate the system for their own benefit.

Lawrence Lundy draws from political philosophy to propose that decentralization, as a cornerstone of modern democracy, be applied to crypto networks to prevent power from being seized by specific categories of stakeholders. Token holders are responsible for "legislative power" through voting, while executive power is granted to entities responsible for implementing policies and operations by the legislative body, and judicial power is exercised by decentralized courts, ensuring that executives cannot violate the DAO's intangible principles.

So far, most crypto projects' executive branches are held by companies or foundations that have no fiduciary responsibility to token holders. We have seen a trend of increasing integration between these entities and token holders with voting rights. For example, the Aragon Association is reviewing proposals submitted and voted on by Aragon token holders. Thanks to decentralized courts, there is now a judicial branch. We see DAOs like Pocket Network integrating these three branches, allowing for effective checks and balances to protect decentralized communities.

Another approach to decentralization is to give voice to different stakeholder groups with varying interests. For instance, the governance of the Melon protocol relies on two main components: the Melon Technical Council representing developers and the Melon Exposed Businesses representatives representing users. Multi-stakeholder governance is indeed a very promising method for making collective legitimate decisions among large and diverse groups. John Light shared a detailed tutorial on how to create a multi-stakeholder DAO on Aragon.

Finally, it is worth mentioning that communities can decentralize not only because of institutional or crypto-economic incentives but also because they want to decentralize. Decentralization is a mindset. MetaCartel, MetaGame, 1Hive, or Commons Stack all utilize tokenized tools to compute value and measure voting power, but they also cultivate a participatory culture, which is the most genuine and resilient way to embody decentralization.

Organization

The "O" in DAO refers to what seems to be the simplest term: "organization." This is also the most misleading. When one thinks of an organization, the image of a company comes to mind. However, it is difficult to imagine a company exhibiting decentralization and autonomy. The structure of a company is hierarchical—even cooperatives appoint managers. Companies trade with each other thanks to contractual relationships that heavily rely on legal rules, making them hard to consider autonomous.

Therefore, we still need to question the nature of DAOs as organizational structures.

What Have We Learned?

The founders of TheDAO claimed it was a decentralized fund controlled by a group of investors without any traditional corporate structure: no shareholders, no executives, no management. Companies are sometimes defined as a collection of contracts among different parties (shareholders, directors, employees, customers, suppliers, etc.). TheDAO attempted to eliminate the legal fiction of corporate personhood and maintain only the contractual relationship between parties, expressing and executing as a smart contract running on a public blockchain.

From this perspective, DAOs can replace companies through "disintermediation." Some might argue that this makes them more like "ghosts" of companies rather than a new form of enterprise. Rather than viewing DAOs as autonomous companies, we could say that autonomous and self-managing markets eliminate the need for certain intermediaries. Another way to put it is that DAOs can be seen as organized economies rather than autonomous organizations.

Regardless, we are still in the early days of the DAO era. Over the past few years, we have seen the experimental scope of DAOs range from teams of a few individuals to protocols used by thousands. Some DAOs even include artifacts (like artworks) or natural assets (like forests). Therefore, when considering what types of organizations DAOs can be, we should maintain a very open mindset.

Currently, the two main types of DAOs are protocols and collectives. Protocol DAOs enable stakeholders to make collective decisions on key parameters of crypto networks based on open protocols: such as blockchain protocols like Tezos or Decred, and financial protocols like Compound, DeversiFi, or Maker.

Collective DAOs are closer to traditional organizations: compared to protocol DAOs, they have fewer active members and broader activities. Compared to typical companies, 1Hive, dOrg, or MetaGame are collective examples that use DAOs to be more transparent, open, flexible, and fair to members. Notably, DXDao is a hybrid DAO that collects an open assembly to manage multiple DeFi protocols.

DAOs in 2020? Not a Thing, Just an Evolving Process

It is clear that there are no pure "Decentralized Autonomous Organizations." DAOs are not fully decentralized, exhibit moderate autonomy, and are difficult to compare with traditional organizations.

In 2020, it is time to acknowledge this fact: most DAOs are actually associated with companies or foundations. In most cases, this is not a problem but a necessity. DAOs are not a gradual, ever-evolving process. It is possible to "DAO-ify" organizations, companies, markets, and protocols, meaning to use blockchain technology to transform these entities into transparent, rent-free networks owned by their members.

There is now ample evidence that decentralization can be introduced gradually. Melon started as a company, and once their protocol entered production, it merged into a DAO. DeversiFi was initially a byproduct of a centralized exchange and is now making significant strides toward "fully DAO-managed DEX." The venture-backed company "Compound" launched its governance token in 2020, and its protocol has managed hundreds of millions of dollars.

It is worth noting that this path can go both ways. The story of MakerDAO is one such exception: as one of the first DAOs, it has a strong spirit of decentralization and has gone through many twists and turns, from the founder reclaiming tokens last year to arranging a new plan to "empower the community with full authority over the protocol." Another compelling case is Justin Sun's Tron acquiring Steemit, notably storing STEEM tokens in centralized exchanges, ultimately leading the community to establish a new blockchain called Hive.

DAOs are like living entities, and they can even die. At the beginning of 2020, DigixDAO announced its dissolution. Its members faced an arbitrage between the project's value creation potential and 366,000 ETH, which was raised in the first token sale in 2016. DAO members decided to burn their tokens to receive a share of the treasury, ultimately leading to the DAO's demise.

Conclusion

So, what can we do with DAOs in 2020? It is not yet the golden time. I do not believe we are about to see widespread adoption of DAOs. DAOs are still a strange name, an ambiguous concept known only to a small group of people, even compared to the number of those interested in cryptocurrencies and blockchain technology.

This seems contradictory. Today, many people are interested in self-organization, collectives, and cooperative enterprises. Many appreciate the shortcomings of free markets and political institutions in addressing many social and environmental challenges of our time. One might think that a technology aimed at helping communities govern themselves, make decisions, produce, and share value with each other would attract more public interest.

My personal belief is that once "crypto" itself becomes mainstream, DAOs will be ready to succeed. DAOs generally suffer the same fate as Dapps: for those unfamiliar with crypto, the user experience can be unbearable. I appreciate the efforts of Aragon, Colony, Abridged, and others to make this more appealing, faster, and simpler. We just have not yet reached parity with Web2 applications. As long as the situation remains as it is now, only a few will be willing to use DAOs.

That’s okay. Now is a fortunate time for experimentation, testing, and learning. It is time to create and innovate. If you are ready to join the crypto team (I hope most readers of this article are), then today's DAO technology is already good enough for you.

If you are starting a project, consider raising funds through connected curves, facilitating initiatives through public proposals, and using token and reputation-weighted voting rights to make decisions, collectively and transparently introducing new members.

If you are creating a software product or financial service, consider building an open network composed of users, developers, and investors, and designing a token system that serves them: giving them a voice and sharing the value created by their contributions.

Considering externalities can also add value. The environment and society as a whole are intangible contributors from which we all benefit and rely on. DAOs help create inclusive economies through transparency and collaborative incentives.

When the era of DAOs finally arrives, it will be because they serve the public good, not the limited interests of the crypto community.

Thanks to Luke Duncan, Jordan Ellis, Griff Green, Hammad, Jack Laing, Auryn MacMillan, Kirstin Maulding, Ori Shimony.