1 September 2019
For many, the promise of blockchain lies in the idea of a truly open source platform, designed in such a way that it opens up the potential for innovation and healthy competition in an ecosystem free from centralized control. In this model, blockchain is a trust machine, built on transparent, decentralized governance. Proponents of this idea like to contrast blockchain with the centralizing tendencies of incumbent platforms, such as Twitter and Facebook. But a danger lurks in the attitude shown by some leading blockchain networks toward governance. That danger is that certain initiatives, which profess to follow principles of transparency and accountability, in fact collapse into opaque, consortia-based governance models and create, at best, a ‘synthetic trust’ – partially decentralized, but unsupported by any solid principles.
In this age, tech giants are the incumbent trust providers. Platform businesses – scalable networks of users and resources – have a near-monopoly on the interchange of information that connects consumers to producers. This was previously most obvious in retail marketing on social media, where Facebook, Twitter, and others have grown to wield huge influence over consumer spending habits. But it is increasingly clear that a new economic paradigm is emerging, in which the platform model itself facilitates the exchange of digitized goods and services (including both information and physical goods) between producers and consumers. For the distribution of news media, a case in point is the personalized newsfeeds delivered by Facebook. In this economic paradigm, the digital platforms are becoming the keepers of our digital identities and the focal point of trust between market participants.
But recently, they have lost credibility as repositories of trust. Most obviously, a series of high profile hacks and data leaks from the likes of Facebook, Experian, and others have damaged consumer faith in the security of our digital identities, a trend which has been exacerbated by frequent allegations of surveillance and data misuse against many of these same platforms. In some cases this activity is compounded by platforms acquiring potential challengers before they become a threat, preventing the emergence of a healthier market for innovations in users’ data privacy. At the same time, the network effects of platforms have accelerated the centralization of information flows across the internet. This often manifests in limits being placed on what kind of content users can (and do) interact with online, these limits attracting a spectrum of labels from “content curation” to “internet censorship”. The result is power being concentrated in a few dominant platforms, who act as gatekeepers to the digital market. This is a model that requires us to put our trust in Facebook, Twitter and others to provide us with commercially and politically impartial digital products.
Is blockchain a solution? The idea of blockchain as a “trust machine” views the foundations of DLT as being (i) transparency, and (ii) decentralized governance. With these two qualities comes the possibility of an open source innovation platform designed such that everyone can build on it applications and products, and add value through their services. Where this model is successfully implemented it can provide an answer to Hobson’s choice, between entrenched private monopolies and state-controlled platforms: the third way is an ecosystem that is decentralized by design, where users control both their data and their power to choose between producers (of content, news, goods, services, etc.), all without any commitment to a centralized authority.
But is this achievable? Blockchain faces a conundrum: in decentralized markets, trust and growth are at odds with one another. This closely relates to the well-known ‘trilemma’, which suggests that blockchain can maintain only two characteristics out of decentralization, scalability, and security. The success of network business models depends on there being a growing number of users (for which scalability is key), and the implication is that either security or decentralization must be compromised in a blockchain’s attempt to scale toward mass adoption. Either way, the more a network grows, the less trust each user can be said to have invested in the network.
A compromise on decentralization can already be seen in some leading blockchain networks, manifesting as governance models that lack transparency and concentrate power in the hands of select individuals who have no direct obligations to the users of the system. Some of these are accidental: for instance, the fact that the top five largest ETH pools secure around 80% of transactions on the Ethereum ledger is a consequence of market forces, not of code design. In other cases, a lack of transparency is clearly intentional. In Hyperledger’s Open Governance Model, the Governing Board’s meeting minutes are confidential and voting power is allocated according to financial contributions to the initiative. Libra’s governance model is centralized and controlled by invested members who meet certain eligibility criteria, including a market valuation of at least $1 billion USD and a consumer reach greater than 20 million people per year. These are two examples where control over the ledger protocol is restricted to just a few individual entities.
These factors combine to increase the risk of collusion in blockchain networks; and where there is a real risk of collusion, the promise of blockchain as a “trust machine” has not been realised. It is, at best, a synthetic trust, unsupported by the principles of open, transparent, free-to-use and universally accessible blockchains. Until the decentralized model is fully realised, trust in the digital market will continue to face a dilemma: either we bargain away our privacy to centralized, but commercially accountable trust providers (the ‘platforms’ of the current model); or we keep direct control over our data in a trust machine designed and run by anonymous entities who are unaccountable to users of the network.
By Paolo Tasca
