I’m seeing a lot of questions about DAOs on Reddit and elsewhere. So I’ve decided to offer my explanation — hopefully, it will help someone to better understand this complex topic.
DAOs (Decentralised Autonomous Organisations) are mechanisms by which we can align economic incentives (distribute risks/rewards) over the Internet using software. Using DAOs, humans can coordinate themselves to work towards a common goal at the global scale without relying on trust or third parties. This mechanism has been enabled by the invention of the blockchain, pioneered by Bitcoin.
The ‘alignment of incentives via risk/rewards distribution’ is a fundamental purpose of a DAO. Some participants in the DAO will seek to take more risk today for the expected higher reward in the future, other participants will seek to minimise risk today by sacrificing future rewards. If incentives are compatible, then each participant will contribute to the DAOs purpose, just by working to maximise their own interest.
This is nothing new. We’ve been doing this for centuries using inventions such as joint-stock companies, insurance schemes or even nation states. Groups of people sharing a common goal pool resources together, agree on risk/reward distribution and enjoy the benefits (or not) of the shared activity in the future. This way humans can cooperate on a large scale and align incentives between individuals who never met each other face to face.
Nation states, joint-stock companies, corporations, insurance companies are just a few examples of ‘inventions’ that allowed us to achieve large-scale cooperation. Essentially, they are just abstractions that we use to organise ourselves to collaborate. But their function is no different from the function of DAOs but technologies used to implement them are different (paper, print, enforcement by the judicial systems).
While we’ve mostly replaced paper & print for transferring information, we still use print & paper for aligning and communicating economic incentives (the entire legal system).
Bitcoin proved we can align incentives using just software, without third parties, to achieve a common goal of creating a digital currency with gold-like properties. We might call Bitcoin a proto-DAO. It proved that a certain design pattern is possible and works in the real world.
Ethereum takes this concept to the next level. Ethereum provides a platform to coordinate ourselves to achieve economic goals other than just currency creation.
Ethereum is to DAOs what Blogger was to publishing. It allows everyone to code economic incentives without having to build the underlying delivery infrastructure from scratch.
The Internet allowed you to exchange information with anyone in the world at no cost. DAOs will allow to your exchange economic value with anyone in the world. That means to invest, raise money, speculate, trade, get insurance, lend, borrow, get paid, form joint-ventures in ways that were impossible before. Generally, make a living in a digital world of the future.
It’s early, it’s risky, it’s a Wild West. But this is the way, the global networked economy will create value in the future.
In order to scale, DAOs will have to continuously reinvent themselves.
The concept of DAOs (Decentralised Autonomous Organisations) has been explored in the blockchain space for some time but the recent rise of Ethereum and upcoming DAO crowdsales generate a lot of new interest in the community.
I’d like to share some thoughts on the nature of DAOs and their importance in the emerging programmable economies enabled by the blockchain.
What does the Wikipedia say about DAOs?
“A decentralized autonomous organization or distributed autonomous organization (DAO) can be thought of as a self-governing organization under the control of an incorruptible set of business rules. These rules are typically implemented as publicly auditable open-source software distributed across the computers of their stakeholders. A human becomes a stakeholder by buying digital tokens, similar to shares in a company, or by being paid in those tokens to provide services for the company. These tokens may entitle its owner to a share of the profits of the DAO, participation in its growth, and/or a say in how it is run.”
Personally, I’m not a big fan of this definition. It explains the features and implementation details but doesn’t explain the benefits.
I’d like to offer another definition:
DAOs are the software-based mechanisms for aligning economic incentives over the internet by distributing risks & rewards among people that share a common economic goal but don’t know each other.
So DAOs are not really autonomous and are not organisations. At least not by the common definition of the word ‘organisation’. DAOs can be understood as risk/reward schemes, programmable sets of incentives that can be distributed over the internet that aim to affect change in the world without a formal, legal organisation.
I intentionally didn’t include the concepts such as blockchains, smart contracts, decentralisation, tokens, voting, crowdsales etc. These concepts are obviously fundamental building blocks of DAOs in the current state of the technology. But the real value of DAOs is not in ‘being recorded on the blockchain’ or ‘distributing tokens’ .
The real value of DAOs lies in providing a mechanism for evaluating, funding and distributing risks & rewards of new ventures that is accessible to everyone in the world.
In this regard, DAOs are similar to joint-stock companies of renaissance Italy.
The early joint-stock companies allowed shareholders to share risks & profits in sea expeditions that were notoriously dangerous — high profit but also high risk. Some of the voyages generated huge returns while unsuccessful ones brought zero. (Read more: http://szabo.best.vwh.net/jointstock.html) So it made sense to pool resources together to fund these expeditions and distribute risk to ensure more consistent returns. The invention of the joint-stock company as a mechanism for risk/reward distribution has generated a great wealth for Italian cities at the time.
In the XXI century, sea expeditions were replaced by startups as the wealth-generating engines with founders being new captains and startup teams being the new ship crews.
But the opportunities for funding and participation in new ventures haven’t changed much since the early sea expeditions. These opportunities are still geographically concentrated (the centre of wealth accumulation is now Silicon Valley), limited to wealthy individuals and still rely on national, legal and judicial systems for enforcement. The technologies of print & paper and hand signatures are still used to finalise agreements.
We praise the technology for being the great equalizing force in the world. It surely equalizes access to information but the missing component is the equal access to investment opportunities. Everyone can access information using Google, but only a selected few had access to investing in Google back in the 90s.
What if the ability to invest in the next Google was as widely available as the ability to access Google is today? On the flipside, the ability to lose money on the next Pets.com (one of the great failures of the dot-com era) should be also equally distributed.
I believe, that due to Ethereum and programmable blockchains, we’re on the cusp of the major transition in the way new ventures are evaluated & funded and DAOs will be central to that transition.
But even though we’re using the XXI century tech, old rules of human economic activity still apply.
To generate returns for investors DAOs will have to solve the same challenges that traditional organisations aim to solve:
Attract and allocate various forms of capital (talent, technology, money) to create new solutions and deploy them in the real world.
Allocate the risk of failure — define who absorbs the risk if things don’t work out as intended. Economic activity by definition has uncertain outcomes and this uncertainty has to be taken into account.
Distribute rewards — define who reaps the benefits of success? how these benefits are distributed? What does success really mean?
Design the evolutionary/scaling mechanism — organisation has to grow and adapt to the changing external conditions. It has to be able to attract more resources over time, by acquiring more talent, money and develop organisational structure that allows it to scale.
Bitcoin (we can call it a proto-DAO) proved that 1–3 are possible to achieve without the legacy organisational structures.
To use an example of Bitcoin, the Bitcoin’s software managed to attract talent, deploy thousands of specialised chips in the real world, evolve its code base, push the Moore’s law to its limits and bootstrap an entire industry just by providing certain incentives in code. Without any legal entity at the top, Bitcoin’s code made people behave in a certain way, work on certain things by exploiting their curiosity, ambition, greed, fear and appetite for risk.
But most importantly it distributed rewards of its success to people based on objective measures — everyone could participate. People made money, and people lost money but the risks were visible to those who wanted to inspect them.
The challenge we have is the point #4 — how DAOs can grow and adapt over time?
The topic that is rarely discussed in is the evolutionary/scaling mechanism for DAOs. How DAOs can adapt and evolve over time, attract new talent & capital and become more competitive.
Most of the blockchain and DAO designs focus on incentivising early adopters. But not many discussions are focused on how to incentivise the late adopters to join the original DAO rather than to fork.
What we learn from Bitcoin and it’s recent scaling discussions is that DAOs won’t be immune to universal market forces that affect traditional organisations.
The combination of talent, capital, community and products that contributed to organisation’s early success, very often is not optimal to take the organisation to the next level.
Different skillsets will be required at various stages of organisation’s evolution and organisational design should account for that.
In the traditional organisations, founders cash out and are replaced by professional CEOs, early investors who no longer contribute are bought out and replaced be the new ones. New talent and new investors have to be vested in the success of the organisation to help it evolve and compete externally.
How to achieve this feature in the context of decentralised organisations where the participants don’t know each other, the organisation itself is embedded in code and there are many conflicting visions for the future?
How to accommodate interests of early adopters who came in early and want to sit on their investments and watch them increase in value with interests of newcomers who could contribute to the DAO but the equity has been allocated already.
Without the equity, they’re incentivised to fork the DAO and allocate new equity to themselves effectively launching a competitor.
In the traditional company, the early investors would be diluted to allocate new equity to talent coming in at the later stage and create a win-win for everyone. Early adopters would have a smaller piece of the larger pie so the absolute value of their investment grows. Newcomers are compensated with equity and have ‘skin-in-the-game’ to join forces rather than to go on their own.
So far, achieving this has been impossible in the cryptocurrency world. The evolutionary mechanism that emerged in the cryptocurrency world forces innovators to fork the original chain and launch a competitor.
The success of Bitcoin inspired the founders of Ethereum and the recent rise of Ethereum, immediately resulted in the Lisk ‘competitor’ and its succesful crowdsale.
“Why join Ethereum and be the 10000th investor, when I can join Lisk and be the 1st. After all, it’s like Ethereum but better because XYZ” (I’m not talking about the technological merits here, just the reasoning of the average investor).
Will the same process repeat itself with the successful DAOs on the Ethereum blockchain? Probably yes.
If a DAO achieves even a moderate level of success, why late-comers should contribute to the original DAO instead of launching a fork with a new allocation of shares?
My bet is that Ethereum based DAOs will be forked even more often that Bitcoin was in its early days. You don’t have to bootstrap the entire network from scratch and the DAO code has to be open source anyway.
But what if we designed a DAO that leverages forking as its decision making mechanism? If forks cannot be prevented, we might very well leverage them as a feature not a bug.
This way we could decrease one of the major risks facing the DAO investors — the risk of being forked.
I’ll explore this topic in more detail and propose an ‘evolutionary DAO’ model in the next article in this series.
Ideas presented in this article were developed in collaboration with Kuba Kucharski. We spent many hours discussing Ethereum, DAOs and ways to scale attention marketplaces described in the Userfeeds article. We’ve realised that the model we design for Userfeeds can be applied more generally to other types of DAOs. Most of the ideas presented here are as much his as mine.
A couple of days ago, I’ve read Bedeho Mender’s article — “Why Your Ethereum Project will Most Likely Fail”. It’s an insightful and important piece and if you haven’t read it before you should do it now.
Today, we’re here — the ‘newspaper on the iPad’ stage
I look at Ethereum and the dApp ecosystem as a paradigm shift of similar magnitude as the transition from desktop to mobile.
With every technological transition of this kind, the first stage of experimentation often results in plugging old models into the new platform. If you remember the early days of mobile, the ‘Adobe Flash on the iPhone’ controversy, or some terrible experiences of desktop apps ported to mobile you know what I’m talking about.
When the new platform emerges, the first step is often to repurpose the old stuff for the new tech. It rarely works but this is a necessary step where lots of failures happen that allow everyone else to learn. It doesn’t matter whether that’s a transition from desktop to mobile, from TV to online video, or from horse carts to automobiles (technological tiller).
With Ethereum’s ecosystem, we’re now at this stage. I agree with Bedeho that a lot of current projects follow the pattern of ‘the same as before but decentralised’ . But I don’t necessarily think it’s a bad thing. To me, it’s just a signal we’re just at the first step of the transition.
So what’s next?
From Information Based Interactions to Value Based Apps
Let’s take a look at the fundamental nature of the transition from Web 2.0 to Web 3.0/Ethereum apps.
All interactions with the Ethereum computer cost a transaction fee.
So from the user perspective every time I’m clicking, submitting, registering anything in the Ethereum application I’m actually spending small amounts of money. Compared to the Web 2.0 apps this is a drastic change and some might say it’s simply a bug.
After all, why should I pay money to click buttons in the application? And then wait 15 seconds for the application to respond?
We got used to the client-server model of the Web where the only value-based interactions are the buy/sell transactions on the checkout page. (whether that’s the purchase of physical goods/services or payments for access to Software-as-Service apps).
Every other interaction is information-based, doesn’t involve sending or receiving any value.
Ethereum expands the range of possible value-based interactions beyond the simple ‘pay for goods’ or ‘pay for access’ UX metaphors that most people are familiar with.
Before the Internet most people were familiar with few, well-defined information-based interactions:
face to face or phone conversations (one to one)
reading a newspaper/watching TV/listening to radio/ attending a seminar or a lecture (one to many)
It’s not an exhaustive list but you get the point.
In all those cases, the roles were defined and mostly fixed and rules of interaction were established either by a tradition or social convention.
The Web has changed that. It opened up multiple new ways in which many-to-many interactions could occur, with people we don’t know, at the global scale. With blogs, social media, discussion groups and feeds we developed new many-to-many interactions that were unknown and impossible in the previous technological era. New metaphors for many-to-many interactions had to be established — with terms like ‘blogging’ or ‘tweeting’.
I remember the first time I’ve joined the IRC chat in the 1990s — seeing all these messages from people all over the world, passing through my screen, in real-time — it was something magical at the time. And way beyond the usual human interaction patterns that were known to me at the time.
Ethereum does to value-based interactions what the Web did to information-based interactions. It opens up a new world of possibilities for exchanging value, aligning economic interests, at the global scale with people you’ve never met.
It will require new UX metaphors that go beyond the known ones such as ‘buy/sell’, ‘pay-for-access’,‘receive payment for goods’, ‘receive payment for work’.
But at the end of the day, fundamental things will stay the same.
If you’re designing a dApp for Ethereum, you’re architecting a new economic interaction rather building an app.
So there are 3 basic questions you have to answer before writing any code:
what are the categories of users that will interact using your dApp?
what is the main value-based benefit that applies to each group. There are only three you should consider: 1/ increased revenue, 2/reduced costs or 3/insurance. If you can’t point out how your dApp will translate to one of these three benefits then don’t go further.
define core interactions that will help them achieve this benefit.
To explain this even more simply — how people will make more money, save more money or achieve a greater sense of financial security using your dApp compared to the alternatives (centralised or decentralised).
In many cases, disintermediation of the existing industries will realise these benefits. In other cases, it won’t, because the benefits won’t be substantial enough to compensate users for the high switching costs and learning curves (learning new tech, managing private keys, fear of losing money etc).
But the key message is — how can you enable new categories of users to make money in ways that were impossible before due to centralisation, friction, monopolies or high transaction costs.
To the end user of Ethereum, the decentralisation itself is a component as important as packet-routing of TCP/IP was to the early IRC users. It makes things work but it’s hidden in the background.
The value is what matters.
The Generalised Platform For Marketplaces
Ethereum is sometimes defined as a ‘world computer’ but this definition doesn’t really convey its core value proposition.
I prefer to explain Ethereum as a generalised platform for marketplaces where individuals can adopt various roles and freely participate in economic interactions.
And by marketplaces, I don’t only mean Ebay, Amazon, Etsy, AirBnb or even financial platforms.
I also mean Facebook, Google, Instagram, Snapchat which are essentially the multi-sided marketplaces for attention/reputation/money.
These are particularly interesting to me as most people are oblivious to the economics of these platforms. We participate in them daily, we see the actions (posts, photos, search results) but not the incentives (who benefits economically from us seeing these posts, photos or search results).
If you’re interested in hearing more about attention/reputation/money marketplaces , I’ve recorded a podcast with Epicenter Bitcoin where we discuss these topics:
Understanding the Asymmetric Payoff in Building dApps
What’s unique to Ethereum though is that all economic incentives will be open-source and public. If a particular dApp model works, it will be immediately copied, modified and improved upon. If the improvement offers more revenue, decreased costs or better insurance to the users, it will be adopted by them.
Because of full transparency, open source nature and interoperability between dApps, Ethereum is a giant feedback loop that will result in a very fast pace of iteration that we haven’t seen before. It might not be obvious today as the platform has just launched but will become apparent in a year or two.
The costs of failure for dApps are negligible but the potential rewards are huge, so dApps will evolve like memes do today. Everyone will copy and remix promising models. And when switching costs for the users are zero because identities and reputations are in their control, this will create a very powerful combination.
The 99,9% of these projects will fail and that’s fine. It’s the speed and open nature of learning & iteration that matters. Ethereum just needs 0,1% of successful dApps to become a multibillion-dollar platform. Power laws and network effects wil be at play here.
The asymmetry of payoff that’s built into Ethereum projects is so huge that I can’t point to any other platform with similar risk/reward characteristics.
What is the asymmetric payoff I’m talking about?
It’s basically the risk/reward ratio where the upside potential is greater than the downside risk.
Developers risk only their time in building new or modifying existing dApps. Infrastructure, security, identities and payments are all built into the platforms. But if the dApp takes off — then the upside is unlimited… until someone else remixes it and comes up with a better model.
This feedback loop is what makes me very optimistic about Ethereum’s future.