A state is sovereign when it has supreme authority to govern its territory without interference from a foreign power. Similarly, a cryptonetwork is sovereign when it runs in a way that resists outside influence. But instead of managing the rules and politics of a geography, cryptonetworks use blockchain protocols to govern the production and exchange of information services over digital space. Achieving sovereignty is necessary to fulfill crypto’s ultimate promise of independent online networks that distribute value more broadly across its participants, instead of concentrating it in a particular company or jurisdiction.
Context
The Internet is somewhat sovereign. Its pioneers etched a rough consensus and running code governance model into its design, a political ideal that supports a permissionless, decentralized, protocol-based architecture. Anyone with the right tools can connect and expand the network, and consensus over the protocol’s rules emerges as a function of which version each individual node chooses to run. So while governments and ISPs have some control over local use (i.e. they can regulate nodes within their jurisdictions), they have limited to no control over the global network because it is practically impossible to corrupt every node.
To achieve this, we had to make the Internet a stupid network. Its only job is to move bits from Server A to Server B. It does not care about what the bits mean: it’s up to the servers, which are “smart”, to process that data and generate intelligence. Pushing functionality to the endpoints removed any network operators’ ability to control or monopolize the services on top. It also made it cheaper to scale by distributing the cost of deploying the infrastructure across a large number of independent-but-interoperable enterprises world-wide.
This tactic was brilliant for breaking up the old telecommunications hegemony, and it’s the Internet’s sovereign-ish qualities that allowed it to reach the level of scale it enjoys today. But the “stupid network, smart node” logic produced an unintended side effect decades later. As open standards and competition took the margins out of the communications business, value moved from pipes and cables to data centers; from the infrastructure layer to the application layer. Transmission at zero marginal cost enabled widely-used online services like Google and Facebook to amass enormous amounts of data very cheaply, which they leveraged to lock users into their proprietary services.
Big Web services benefit greatly from the Internet as a (relatively) open and sovereign network, but rarely pass on those benefits to their users. As a result, end users enjoy little of the Internet’s promise of information independence.
Sovereign Cryptonetworks
In many ways, cryptonetworks do to Big Web what the Internet did to centralized telecom: they introduce open source code and standards and a distributed production model to the information services business, thereby collapsing costs and increasing competition. But unlike the Internet they use a cryptoeconomic consensus and running code model. The rules of each network are tightly specified and rigid, activity is coordinated via cryptoeconomic incentives, and everything is enforced by smart contracts.
This might seem like a regression to the “intelligent networks” the Internet tried so hard to eliminate. But consider that the Internet is a decentralized communications network while cryptonetworks are decentralized information networks. The Internet can afford to be stupid because moving bits is a very narrow, specific task with an objective result. This makes it easier to arrive at general consensus without many strict rules. Cryptonetworks store, organize and interpret the bits. To function, they must both generate the intelligence, and act upon the information – while balancing a decentralized architecture with providing a consistent service. This is difficult to do without a proper system for enforcing the rules.
Sovereignty is a matter of trust, and trust is about scale. In geopolitics, economic activity thrives in jurisdictions where there is a high degree of trust in the rules of the state, and in the state’s sovereign ability to enforce, defend and uphold those rules[1]. The same applies to cryptonetworks: communities thrive when there is great trust in the system, and the way to develop that trust is through continuous decentralization across all elements of cryptonetwork sovereignty.
To understand the nuances, we might compare Facebook’s Libra to Bitcoin, and then again to MakerDAO. Facebook promises a decentralized stablecoin, but designed it to rely on the financial infrastructure of Switzerland (which, ironically, I’m sure they chose out of trust in the country’s sovereignty), and it’s being developed, financed and promoted almost entirely by American firms. This gave the U.S. Congress the power to call hearings and potentially halt the project. That they can do this disqualifies Libra as a sovereign network. By contrast, while an individual state could outlaw local use of Bitcoin or Dai, it would be practically impossible for it to stop the development and operation of those networks world-wide because they are much closer to being self-sovereign than Libra ever will.
Elements of Cryptonetwork Sovereignty
To be sovereign, a cryptonetwork must be able to operate and create, execute and modify its internal rules according to its own governance process. To get there, it has to be sufficiently decentralized across four dimensions: supply, demand, governance and capital market infrastructure.
To illustrate, we’ll use the cryptoeconomic circle model I like to use to describe how tokens, capital and labor flows between the different participants in the cryptonetwork:
1. Decentralized Supply. If you’ve been into crypto for a while, you understand that decentralizing the miner side is critical for trust, reliability, scale and security. I like to point out that Bitcoin is the only online service that has enjoyed 100% uptime for over 10 years. That is because there’s over 1 million independent machines supporting the network, so there isn’t a single point of technical failure.
Developing a decentralized supply side requires a cryptoeconomic model which properly incentivizes people to provide resources at scale. Ideally, supply scales as demand grows, so it’s alright if in the beginning there’s few miners for few users. But if it is possible for a relatively small number of miners to handle a lot of demand by themselves at scale, supply side concentration might occur. In some cases it might be possible to discourage people from taking too much of the network, for example by requiring miners to put up an amount of stake at risk proportional to the amount of supply they wish to create for the network.
Another important element to consider is the geographic diversity of supply. This matters to reduce or eliminate the impact a single state can have on a network, a kind of jurisdictional risk. For example, we know that a significant portion of Bitcoin miners are located in China, but Bitcoin mining is sufficiently distributed across the globe that if China were to suddenly ban the activity, the service would continue to operate globally non-stop.
Incentivizing geographic diversity at the protocol layer is not always practical, but some projects like Filecoin and FOAM are developing mechanisms for distributing production capacity and rewards based on the geographic location of miner nodes. A complementary strategy is to reduce friction for new miners by nurturing international markets for the token. The easier it is for a miner to buy and sell tokens in their jurisdiction, the more practical it is to invest in the infrastructure to mine.
2. Decentralized Demand. Here, we look at the total number of users, their geographic distribution, and we also consider the number of interfaces to the cryptonetwork’s service. Having a large number of individual users increases the economic value to the supply side and reduces demand-side jurisdictional risk. One case to study is India’s alleged attempts to ban the use of Bitcoin: while this would be unfortunate for Indian citizens, it would be unlikely to have a major impact on the overall Bitcoin economy because Bitcoin demand is not very concentrated in India.
Another is Numerai, a hedge fund with a cryptonetwork to crowdsource market predictions from tens of thousands of users, who are rewarded in the token Numeraire ($NMR). The problem with the original implementation has Numerai the company as the sole “customer” of the network, so all those users depend on Numerai existing. To solve this problem, the Numerai team designed Erasure, a decentralized version of Numerai’s prediction game that allows anyone to buy and sell predictions using NMR, removing the community’s economic reliance on Numerai.
Interfaces are the apps people use to consume a cryptonetwork’s service, such as Bitcoin wallets or Ethereum dapps. Ensuring there is a wide variety of independent interfaces increases the overall reach of the network, especially as individual “interface teams” mount user acquisition campaigns independently. For example, a lot of people in the U.S. came to Bitcoin through Coinbase, and the overall network benefited from their success. But Bitcoin did not grow at the pace of a single company: there are hundreds of businesses, all over the world, continuously working to increase the network’s reach.
Individual apps target different users across languages, and geographies, and serve as the customer service and marketing layer for the network. They are also key to reducing that demand-side jurisdictional risk. Ensuring demand trends to decentralization has a lot to do with the design provided by the network and the diversity of use cases for the service. It’s also important to consider the potential business opportunities for the people providing these services, as this will determine the potential amount of resources available to develop this layer of the network.
3. Decentralized Capital and Market Infrastructure. This is a crucial, but often-overlooked element for cryptonetwork success. Market Infrastructure refers to all the tools and services miners, users and investors need to store, use and trade a cryptonetwork’s token. That means exchanges, wallets, custodians, liquidity providers and sources of financial capital. Like the previous elements, geographic diversity matters a great deal, especially on the exchange and capital side.
Having lots of wallets and custodians increases the pool of potential users and holders of the token. But it’s particularly important to have the token trade against as many pairs, exchanges, and geographies as possible – especially because jurisdictional risk is highest at the capital markets layer today because financial regulations regarding cryptonetwork tokens remain unclear in many places. It’s also valuable to cultivate a diverse group of investors that can capitalize and “liquify” the network with a variety of geographies and currencies.
A cautionary tale is how Bittrex delisted over 70 tokens in its US exchange in June 2019 (presumably over regulatory concerns) to the detriment of networks which relied on Bittrex US for much of their liquidity. As described in the cryptoeconomic circle (which I realize looks like a triangle), a network’s token facilitates the exchange of labour and capital, and it is important for it to be highly liquid in order for the network to function properly; that is, so miners can always sell some of their tokens to cover their production costs, and so users and investors can always buy some tokens to use and capitalize the network. Making sure there are many ways and places to trade, including decentralized exchanges, helps solve this problem.
Developing market infrastructure is, for the most part, off-chain human work. It involves developing relationships with exchanges, investors and market makers, some of which can take time and money. The work can be carried out by the core development team or other community members, depending on the makeup of each network. Decred, for instance, requires some of this work to be managed through its public proposal system, while some core dev teams hire people dedicated to this work. It’s also an area where teams can rely on their institutional investor community (at Placeholder, we’ve been developing our network and expertise in this area).
4. Decentralized governance. Governance is a catch-all term for all network and community management activities. This includes core development, treasury management, forks and upgrades, and even the operations of the consensus protocol itself. Proof of Work, Proof of Stake, DAOs, Github repos, subreddits, community chats, calls and meetups all fall under the umbrella of governance in some way. We’ll leave aside the debate about whether decentralized, “on-chain” governance works, and focus instead on why it matters by looking at what happens when it’s too concentrated.
What’s happening with Kik and the SEC is an example of the risks of centralized management. Several aspects of the Kin network are decentralized, but Kik maintains control of several key management activities. They write and maintain the code of the network as well as its primary interface, and they also control the funds it raised with the Kin token sale, which is the basis of the SEC’s complaint. The reason the SEC can target Kik easier than other projects is because Kik is a single, established company that can be sued by a regulator. This means the SEC, an American institution, potentially has the power to shut down or halt the development of a global network.
A counterexample is Ethereum. The core team also raised money with a token sale, which the SEC frowned upon but decided not to pursue, presumably because it would’ve been a more difficult target. The Ethereum Foundation drives much of core development and still manages a network treasury, but the Ethereum ecosystem is diverse enough that the EF does not have as large an influence over Ethereum as Kik has over Kin.
Fully decentralized governance is a long, complicated topic for another day. But for now it’s useful to think of it as an evolving gradient. In many cases, “decentralized enough” is a perfectly fine goal, and we have a couple of well-understood tools that can help us get there. A good place to start is encouraging independent development teams to produce alternative implementations of the protocol in a variety of programming languages, as we see in Ethereum. This reduces the network’s reliance on a single core team. Another is to make sure the network can be forked.
Developing Sovereignty
With these high-level concepts in mind, we can begin assessing how sovereign a cryptonetwork is and make a host of design and participation decisions. We can also use this as a reference for communities to plan and build towards sovereignty. The reality is that most, if not all, current cryptonetworks lack this quality. This is acceptable during this experimental phase of the market, especially as it’s an attribute that takes several years to develop. In practice, 100% sovereignty does not exist in the physical world and it is unlikely to exist in an ideal configuration in the digital world. However, I believe the world’s largest and most successful networks will be those that are the most sovereign, as they will have the easiest time convincing miners, users and investors to commit their time, work and money to their growth.
As a practical exercise, we can apply this framework to study a network like Decred, an intellectual fork of Bitcoin. The founders of Decred, concerned with the lack of agency bitcoin holders have over protocol governance in Bitcoin, decided to build a cryptocurrency network with a more robust consensus protocol. Decred augments Bitcoin’s proof-of-work model with a layer of proof-of-stake that allows end-users to participate in the network’s governance and block reward process. In addition, Decred dedicates 10% of each block reward to its development fund, a kind of decentralized trust dedicated to fund Decred development, in which all DCR holders are trustees. By contrast, Bitcoin’s core development decisions are almost exclusively driven by a small number of core developers, increasing the risk of corruption within the network. It also lacks an independent, predictable funding model for core development. As a result, many Bitcoin core developers are employed by outside companies, potentially increasing their influence over the network. Therefore, while Bitcoin today has an advantage in terms of scale across the supply, demand and investor sides of the equation, Decred is better set up for long-term sovereignty thanks to its decentralized governance principles.
But my favorite thing about this principle is that it forces us to look at decentralization across many dimensions. Many cryptonetwork developers and communities focus on decentralization of supply, and perhaps of the token’s distribution, but think little about how their designs and strategies affect the diversity of market infrastructure, capital providers, interfaces and geographies. This becomes a conversation about distributing data, wealth and power across a larger number of people, in a larger number of places – and that’s why thinking in terms of a path to sovereignty is more useful than thinking about a path to “decentralization”.
[1] Mario comments “Max Weber famously defined the state as an entity that lays claim to the legitimate use of physical force in a given territory, or in other words - to the monopoly on physical violence. What you're describing here is very similar to Bourdieu's extension of this definition to include the legitimate use of ‘symbolic violence’, i.e. the ability the define, inculcate, and enforce the meaning and proper use of symbolic structures, from official terms and forms to rules and standards to institutional norms and culture, etc. In other words - legitimate control over information and knowledge. Also interesting to note: Weber said that bureaucratic administration means fundamentally ‘domination through knowledge’.
In a way, cryptonetworks represent an attempt to break the state's monopoly on symbolic violence by creating ‘information management systems’ - bureaucracies, basically - that can legitimately define, handle, exchange, and enforce symbols/meaning (information) in a way that is relatively independent from the state, at least initially.”