An immersion into the cryptocurrency area yields many conceptual contradictions. Cryptocurrencies are designed to be difficult to change, but, like all technology, they need to change to address design flaws and advancements. Blockchains are intended to prevent centralization, yet require strong actors to lead changes or maintain the code.
Perhaps the most frustrating experience comes when there are clear deficiencies that most stakeholders agree need to be corrected, yet consensus cannot emerge on the path forwards.
Bitcoin’s block size debate has now been an active issue for more than two years. Daily, transactions totalling over a billion dollars are pending because the network is at peak capacity.
If changing a simple parameter — even in the presence of temporary solutions — cannot be coordinated, then how can enterprises and governments feel comfortable investing billions of dollars into building infrastructure on top of these systems? For that matter, how can any business gamble on the strategic risk of integrating accountability-free protocols that cannot make rational design upgrades?
Looking back into history, the evolution of the internet has followed a similar pattern with even simple changes like the transition from IPv4 to IPv6 taking decades to realize. Yet there is a strong contrast between blockchain technology and the internet in that they follow a very different style of custodianship.
The internet was a military project that grew out of DARPA into academic circles with strong government backing and a well-defined set of initial custodians. The internet grew under non-commercial conditions without the machinations of corporate influence attempting to monopolize the network. In fact, e-commerce violated the NSF AUP until it was repealed in 1992.
By the time businesses had the luxury of commercializing the internet, there was already a strong set of standards, principles and evangelistic adherents. This did not stop companies like AOL and Microsoft from trying to build wall gardens and creating proprietary technology like ActiveX. This foundation has not stopped next generation actors such as Google from pushing their own agendas given their enormous user bases and capitalizations.
With swarms of rent seeking actors from traders to miners, cryptocurrencies are the ultimate commercially motivated ecosystems. Given this foundation, evolution of the custodianship of cryptocurrencies has resulted in optimization around self-interest.
For example, validationless mining is starting to occur more frequently as it improves a miner’s profit margin, yet this completely disregards the entire purpose and utility of mining. Mining centralization has already occurred with just a handful of actors in control of the majority of Bitcoin’s hash power.
Like the internet, cryptocurrencies require consensus to change. But when such rapid centralization of power to a handful of brokers occurs, what happens when change is not convenient to them?
Unlike the internet, the bootstrapping of most cryptocurrencies is not done through altruistically non-commercial or academic means. From inception, some group seeks to make gains and there are power brokers assigned to help ensure those gains.
Founding centralization is a reality that each cryptocurrency must face in its evolution. We cannot fully escape it, but should at least try to design around gradual decentralization.
What factors promote centralization?
For Cardano, we thought carefully about what factors promote centralization and what techniques could be applied to encourage our protocol to gradually become public infrastructure like the web.
We fully admit that total decentralization is both impossible and perhaps even counterproductive. Yet certain factors can be encouraged to produce a more balanced system.
First, while centralized custodianship of crowdsale funds allows for agile and rapid development of the protocol during the early days, eventually funding has to diversify and the speed of development needs to retire to a more systematic and deliberate pace. Following this point, funding needs to avoid cultural, linguistic and geographic bias.
Second, as the community becomes more informed about the underlying nature of the cryptocurrency’s technology, decisions about the roadmap cannot be centralized to a set of core developers or foundation. There needs to be a blockchain based method for proposing, vetting, and enacting changes to the protocol.
Third, the incentives behind maintaining the Cardano SL blockchain have to be directly aligned with the aggregate desires of all users. We cannot permit a cabal of specialized actors to emerge who are independent of the will of the greater community.
For the first principle, we have chosen to integrate a treasury system into Cardano. For the second, we will deploy a formal process to propose Cardano Improvement Proposals through a system coordinated by CSL itself. For the third, we believe Ouroboros provides an elegant solution.
More detail could be provided on the above topics, but they are extensive in their own right and beyond the scope of a survey paper. Mechanism design is one of the most intricate and interdependent academic fields with incomplete theory and no solid canonical model to stand on.
Rather our science driven approach described in section two serves us well here. IOHK’s Veritas team is working in partnership with a group of researchers from Lancaster University under the direction of Professor Bingsheng Zhang to develop Cardano’s reference treasury model. With the aim of integration in 2018, we expect a dedicated peer reviewed publication by the end of 2017.
For formal description and vetting of changes to a cryptocurrency protocol, this topic is the least understood as it requires both ontological notions as well as a mechanism to incentivize broad participation. Perhaps some form of representative democratic process could emerge or use of liquid feedback to provide more rational voting.
We expect research in this direction to consume most of IOHK’s formal involvement in the development of Cardano. As a starting point, we will deploy alongside the reference treasury model several mechanisms to capture consent. Further study is required for a definitive solution.
Finally, work to improve incentives for Ouroboros is being supervised by Professor Elias Koutsoupias of the University of Oxford. After the cryptographic foundations of Ouroboros are solidified alongside all required scalability work, a broader study of bonds, penalties and exotic incentives will be added to the reference protocol.