I used to make jewellery for a living. Really, really expensive jewellery.
We made some of the most extraordinary jewellery in the world. Work which, if I may say so myself, put the best of the world’s top luxury brands to shame.
One week I’d be in Hong Kong sourcing rare natural blue diamonds. The next, (accompanied by a machine gun toting bodyguard) I’d be hosting a discreet soirée for a dozen Russian oligarchs. Or being interviewed for glossy magazines in New York. Or racing across Europe to hand deliver an engagement ring to a billionaire.
We were critically acclaimed — darlings of the luxury press. The dirty little secret that nobody could have known? We were actually just two dudes listening to Ugly Duckling and making dick jokes in a gloomy Victorian workshop in East London.
We were able to punch so far above our weight because of the way we worked.I had built a large distributed team of the world’s greatest craftspeople. Our workforce was spread across Europe and South East Asia. Our sales team comprised strategically selected influencers in London, LA, New York, Moscow, Paris, and Beijing. Our PR and marketing handled by big name London agencies.
Being a distributed team was our greatest strength. We had low fixed costs, high capacity and could move fast. At the same time, it also sucked terribly.
While we were at the cutting edge in technology, manufacturing, and design, our organisational methodology was stuck in the dark ages. It relied upon a ton of manual coordination and tedious admin.
Far worse, the incentives of some of our suppliers were sorely misaligned with our own.
We were building long term value; they were maximising short term profit.We valued agility; they wanted retainers. We aimed to be lean; they convinced us to spend. I shit you not — our PR agency once said “You know, a private jet to Moscow would only be about £35k…”
By the time we realised the problem we had already spent too much money and travelled too far down the road with these suppliers to easily replace them. The relationships became frustrating and lacking in trust.
Yet, it wasn’t always like this.
Some suppliers acted more like stakeholders. They were personally invested and excited about the work we were doing together. Our success was their success. Needless to say, those relationships were greatly more fulfilling and fruitful.
One day, at an interminable gala dinner in Moscow, a client asked if I could help her launch a company similar to my own. My reaction? Visceral horror expressed in the traditional British manner: a polite smile and nod.
My reaction made me realise the problem. If I was going to work with this client, I was going to be smarter about it this time.
Around the same time, I was getting really into Bitcoin. Specifically, its underlying technology.
Through transparency and economic incentives, the protocol solved some of the same problems of coordination and cooperation of a distributed workforce I was wrestling with, albeit doing one specific job: mining.
I was inspired and wondered if I could abstract the concept of proof of workand design an analogous system suitable for arbitrary work, rather than just mining.
I wanted software to:
- reduce management by making it easier for suppliers to cooperate/collaborate
- reduce admin generated by coordination and international remittance
- reduce the need for trust
- align the incentives of the workforce
If I could find a general solution for these problems, I wondered, could Bitcoin’s blockchain also be used to keep track of company stock?
Then, Vitalik wrote his paper. Everything clicked. I was thinking too small.
Decentralised Autonomous Organisations (DAOs)
Owned and operated by its workforce, Bitcoin can be seen as the prototypical example of a new organisational paradigm: the ‘Decentralised Autonomous Organisation’. It aligns the incentives of its workforce by continuously distributing ‘equity’ in itself in direct proportion to the value each of its miners has contributed.
Vitalik has already written an excellent field guide to the terminological disaster zone that is decentralised application nomenclature, so if you want to dive deep, read that.
What I would call a DAO however, I suspect Vitalik might call a Distributed Organisation (DO). A distributed computing platform reliant on a network of users running clients on their computers for a fee might be considered a DAO. A DO on the other hand uses smart contracts to provide a set of rules by which a group of people may interact.
I would argue however that the concept of ‘autonomy’ hangs on whether or not there is a central point of control or decision making within a decentralised organisation. If there is, then it’s probably not a DAO. To keep things simple therefore, I’m just going to use the term DAO here. If that upsets you, please see here.
While theoretically the Bitcoin network is made up of independent miners with no central control, the actual development and governance of Bitcoin comes from the “top”.
Except there is no top, and that’s kinda the problem. Rather than clear leadership, there’s a core group of fractious developers who control the code. The ongoing civil war within its community has proven Bitcoin’s governance woefully unfit for purpose.
Then came ‘Governance DAOs’: blockchain-based entities issuing a cryptographic token which may be either purchased or earned. Tokens are publicly tradable, and may entitle their holder to fees or a share of any revenue the DAO generates.
The token entitles the holder to vote on board level decisions about how the DAO runs. Examples might include decisions around setting fees or deciding how long a voting period should last.
“The DAO” does not attempt to tackle the problems of coordination, cognition, or collaboration solved by companies. Rather, it pushes those processes to centralised nodes (AKA ‘Contractors’) at the edge of the network.
A ‘Contractor’ may submit a proposal for funding, and the DAO will vote (weighted by stake) on whether to finance the ‘Contractor’ to deliver that proposal. In return, the holders of the DAO tokens may receive a revenue stream as a result of the delivery of the proposal.
Governance DAOs are designed for infrequent, large-scale stakeholder decision making, not the fast-paced environment of day to day organisational operation.
It is not something a Silicon Valley tech company could use to run and grow their startup. It’s not appropriate for a non-profit that needs to manage day-to-day activities. It’s not suitable for a large, decentralized team to compete in the SpaceX Hyperloop challenge.
So, what would be suitable?
A True DAO.
The True DAO
If “The DAO” is a decentralised investment fund, a ‘True DAO’ is a decentralised company (or non-profit). It solves the problems of effective coordination, cognition, and collaboration which has traditionally been solved by ‘the firm’. The difference is that it does so without requiring any central authority. That means thousands of people, all over the world, can build companies together over the internet without trusting (or even knowing) one another.
A True DAO must assume every participant is rationally self interested; focussed on profit maximisation at any cost. With that focus on game theoretic resilience as the bedrock of the system, a True DAO enables efficient division of labour and effective decentralised decision making.
Management decisions are not typically taken by a company’s investors and neither should they be in a True DAO. While you can (and in some cases should) weigh votes by ownership, that’s not appropriate for day to day operations. Wealth is no proxy for expertise (or even rationality), so rather than weighing votes by stake, a True DAO must have a robust reputation mechanism for accurately and trustlessly identifying competence and trustworthiness by systematic and objective assessment of behaviour.
Indeed, company wide voting should not be necessary for every decision — it must be possible for day to day operation and expenditure to happen with as little friction as possible, without compromising security.
A True DAO should allow its contributors to be paid in either cash (e.g. ETH, BTC, DAI) or ownership of the DAO proportionate to the value of each member’s contribution. It should recognise some roles within organisations are difficult to quantify and it should therefore be able to support salaried roles and performance review.
Where a True DAO is revenue generative, it should be able to determine what it does with it. It may choose to disburse all revenue among its members, retain all revenue to peg the token value, or somewhere in between: distributing some of the revenue as a kind of salary and retaining part as working capital.
To quote His Cryptographicness:
“[A DAO] essentially replicates the legal trappings of a traditional company but using only cryptographic blockchain technology for enforcement.”
DAOs, almost by definition, cross geographical boundaries. The blockchain does not respect jurisdiction and cannot differentiate between who or whatcontrols public keys. All you can know about a party with whom you are doing business is that they control a public key. As Richard Gendal Brown said:
“On the blockchain, nobody knows you’re a fridge.”
DAOs also cannot currently have legal personality. Relying on the traditional legal system is therefore not only messy (and almost certainly expensive), it’s a completely unknown area of law. A True DAO must not therefore require fallback to the meatspace legal system in the event of dispute. Disputes are inevitable and a True DAO should contain robust mechanisms for their resolution.
While we have been fairly quiet about it for the last eighteen months, things have been happening at a furious pace behind the scenes. That vision of the True DAO is what we at Colony have been working on.
Yet, a True DAO is a very pure thing and will not be suitable in every case. While it may be perfect for an open source project, it may not be suitable for a design collective, or a multinational bank. (And yes, both are interested in Colony.)
Colony will make it easy to deploy a True DAO, but it will also allow many other kinds of Open Organisations which require greater levels of trust, or which simply operate differently. A True DAO is not a magic bullet; there is no one right answer.
We see the Colony Network as infrastructure for Open Organisations and part of the remit of the foundation will be supporting the development of applications on top of the protocol.
Stay tuned — this is only the beginning.