Vitalik Buterin, founder of Ethereum, has published an essay on his blog entitled “Notes on Blockchain Governance”. In this article I will offer a brief summatary of a closely-argued text in which he analyses the pros and cons of different types of decision-making with regard to the development of a blockchain, and arrives at an interesting conclusion.
Righty tighty lefty loosey
The premise of the essay is the fact that coin holder voting has had a resurgence as a ‘multi-purpose decision mechanism’. In itself, this has the same advantage as democracy does over other forms of government:
“…first-generation blockchains empower de facto, centralized core development teams or miners to formulate design choices.” – Tezos, “Governance”
Buterin begins by explaining that there are two types of coin voting: ‘loosely coupled’ and ‘tightly coupled’.
He says that when we look at governance as a ‘coordination model’, we deal with two layers: the bottom layer comprising reality, and a top layer consisting of ‘coordination institutions’. The bottom layer in the real world is physics, force, and in a blockchain, whatever a protocol allows its users to be. The coordination institutions act as consensus gatherers. When there is enough consensus on the top layer, the bottom layer can be changed.
This relates to blockchain voting in this way: voting can be attached to the bottom layer, or the top layer. Tightly coupled voting, the former, means that a change approved by a vote will happen automatically. If a minority disagrees with the fork, the onus will be upon them to “exert great effort to coordinate a hard fork to preserve a blockchain’s existing properties”.
Loosely coupled voting is a coordination tool, and users will have to actually download and run the fork themselves.
Both types of voting system have advantages and disadvantages, which Buterin goes on to discuss. And he spends time explaining that regardless of model, voting as a rule of law has weaknesses.
For example, different classes of user will inevitably clash as their differing needs affect how they would like the blockchain to develop, and anyway, if a non-technical public must defer to a small group of experts in order to make an informed decision, what actual meaning do their votes have?
Another example of a weakness shared by both systems is that voter participation in blockchain upgrade decisions is always very low. As examples, Buterin cites Ethereum’s DAO Carbonvote with a turnout of 4.5%, and a winning vote in the Bitshares system receiving only 17%, which is noteworthy given that Bitshares (as the name implies) is a social system designed around voting.
Such concentrations of power lead to bribery, and Buterin offers examples of exactly such a case on the Lisk platform.
Tradefora Completes Integration with Serenity EscrowGo to article >>
As relates to this risk, loosely coupled voting has the agility to resist organised bribery more easily, because the community simply needs to agree to ignore malicious sources, while a tightly coupled model would have to write such commands into the underlying code. On the other hand, tightly coupled voting systems have powerful, centralised actors with the motivation to protect their system.
assert total_supply <= 21000000
Buterin argues that codified protocols should also “be viewed skeptically”. His argument is that the enshrinement of core properties (values) should be loosely coupled, as “any meaningful norm is actually quite hard to express in its entirety; this is part of the complexity of value problem”.
He presents an interesting example of how the Bitcoin 21 million coin limit could be circumvented in real terms without having to alter the code by installing a very similar coin as a mandatory stage in all Bitcoin transactions. This would create a currency fungible with Bitcoin, and the number of bitcoins would, de facto, be multiplied.
While Buterin describes the weaknesses of voter-based systems, he also expresses a lack of enthusiasm for the alternative – core developer consensus, or “ivory tower intellectuals” divorced from user experience.
Reading this essay, one is reminded of the old Churchill quote: “It has been said that democracy is the worst form of government except all the others that have been tried.”
Formal on-chain blockchain governance is not a panacea, and it’s not even clear that it’s an improvement.
New article from me: https://t.co/p7FcgBdSUN
— Vitalik Buterin (@VitalikButerin) December 18, 2017
Buterin advocates a system whereby decisions are made by a multitude of systems, which include the initial roadmap of the project, core development consensus, coin holder votes, user votes, and “established norms”.
He calls this system ‘multifactorial consensus’, and likens it to the checks and balances of the multifaceted government of a neoliberal democracy. All facets have their intrinsic weaknesses, but if spliced together in such a way that no one pillar has too much control, we come as close as we can to having them cancel each other out.
To this writer it seems that Buterin is wise enough to recognise that the fact that humanity has yet to see an ideal society shows that the imposition of ideas is best avoided; in reply to a Twitter user begging for leadership, he responded: “Let cultural evolution figure that out; it’s smarter than me or any of us.”