A Consensus Mechanism for Efficiency, Fairness, and Security in Blockchain Governance.
Controversies with Delegated Proof of Stake
At the heart of DPoS (a consensus system made famous/infamous by it's inventor, Dan Larimer, the founder of Bitshares, Steemit and EOS) is the core concept of introducing the role of key, fixed delegates. These delegates—members that are voted by the community—work to secure the blockchain, typically for a substantial financial reward. In return, the community receives a fast and secure network (a magnitude of speed higher than most other public blockchains/cryptocurrencies).
To ensure that delegates remain honest, the community is given the power to vote out any delegates (called Block Producers) that are proved to be bad actors.
Despite the clever mechanisms and dynamics that are designed to ensure the system remains as incorruptible as possible, there are a number of criticisms that many esteemed members of the cryptocurrency community have in regards to DPoS.
Without going deeply into the debate (1, 2, 3, 4, 5), it can be summarised that most of the criticisms tend to be towards the human element equation of DPoS. In contrast, some of Dan Larimer's counterarguments rest upon a more optimistic view that the majority of people are good actors, and if given the power to vote effectively, the impact of bad actors will be minimised.
Most of the concerns are related to commonly encountered societal problems such as: collusion, cartels, manipulation, voter apathy, vote buying, voter insignificance, bribes, grooming, vote duplication, centralisation.
It is my view that—although these concerns are valid—there remains a great deal of uncertainty in predicting either the success or failure of such systems. At the current moment, at this early stage of blockchain governance, there does not exist any qualified authority yet predict the outcome of these experiments in blockchain governance with near-certainty.
And this should be expected. Such systems call a wide mixture of disciplines such as psychology, game theory, economics, mathematics, network effects, cryptography, and computer science. It will likely be years—or decades—before this field starts to settle, eventually allowing predictive models to emerge.
Currently, we can only estimate and guess the outcome. This is one of the more exciting aspects of the blockchain revolution; much remains to be tried, tested and determined. The revolution has just started out the gate, and people such as Dan Larimer, Vitalik Buterin and Satoshi Nakamoto are akin to the Edisons and Teslas, or the Jobs, Kays, and Gates of this newly disruptive industry.
Iteration in Governance Models
Perhaps, a wise approach, in this uncertainty, is to develop systems that are flexible enough to adapt these ever-changing circumstances. This is part of the brilliance behind EOS—a platform that has been designed with flexibility in the governance model in mind.
The main genesis block of EOS will include an initial governance model (headed by Thomas Cox of Block.One). Once the main blockchain is launched on the EOS platform, the community and delegates will soon control the fate of the blockchain, and the early governance model will (hopefully) evolve into a system that works for the greater majority.
Even if the critics are right—that the main EOS blockchain does not result in a fair governance model—and many of the predicted calamities prove to be true, alternate governance models can be simultaneously built on the EOS platform.
One early example of such an alternative model is the Evolution blockchain (also built on the core EOS platform) that intends to have a more fair distribution of wealth and also includes a greater number of delegates. Likely additional groups will continue to follow with various other governance philosophies to be tried. Some of these alternatives may succeed, while others will likely fail.
The beauty is that there is now—for the first time in human history—the ability to iterate upon the idea of governance. This is a very powerful development and I suspect the eventual effects of this revolution will utterly transform the human experience in ways we cannot yet predict.
Qualified Demarchy to improve DPoS
One possible alternative, or evolution, that may be tried is to introduce the concept of a Demarchy as a structure of governance on the blockchain.
Demarchy (also known as sortition, or lottocracy is a method of randomly choosing public officials by lot. Utilised in Ancient Greece, sortition both provided a political education for its citizens as well as a practical application of the principles of equality.
I proposed a grassroots alternative to the United Nations under such a model in a previous article.
Demarchy, in particular, may have practical uses for many blockchain organisations due to a number of inherent tendencies towards incorruptibility.
A Qualified Demarchy differs from other Demarchy models in the sense that qualified candidates—with practical experience and abilities—are chosen beforehand to be placed in the selection pool. Once this group has been filtered, the random sortition mechanics can choose an appropriate delegate.
In addition, delegates can be recalled via a voting mechanism that is triggered by a petition of a certain size. Two types of petitions may exist in this model: an internal petition (called by one's peers within the governance system) or an external petition (called by the public). Each type of petition may have different thresholds to reach a recall vote for any individual position.
A Qualified Demarchy—combined with the ability to recall delegates—has many advantages over democratic voting such as:
- Qualified Demarchy eliminates the need for expensive campaigns with charismatic, crowd-charming, career politicians.
- The relative ease of selecting candidates through sortition can allow for non-disruptive, staggered and efficient transitions of a single delegate to the next.
- A strong likelihood of an accurate representation of actual demographics, including age and gender.
- Lobbying, grooming, bribing, vote-buying and influencing becomes more difficult due to the randomness of the final selection process.
- a Qualified Demarchy reduces the conditions and avenues of power that allow political parties, special interests, and oligarchies to form.
The flow for this process, in the selection and terms of the Block Producers of a system such as EOS, are as follows:
[A] Prove basic qualifications towards adequate Block Producer (BP) capacity.
[B] Assess potential delegate's capacity as being sufficient (true) or insufficient (false).
[C] If "true", BP candidate goes to selection pool (with other candidates) for the duration of their rein/term-limit. If " false" BP candidate needs to try again to reach qualifying criteria OR they decide to withdraw their candidacy.
[D] BP candidates in the pool are randomly sorted. Winners become BPs and non-winners remain on standby (with the same, or similar, proposed rewards that EOS is sharing for active BPs and stand-by BPs).
[E] If petition numbers are triggered and reached during any winner's reign, the delegate can be recalled. Once the mandatory voting is triggered by a successful petition, the community chooses to keep/kick the BP in question.
[F] If the BP remains, and no other petitions are triggered during their reign, once the reign ends the BP is put back into the standby BP selection pool. (Go to section "C").
[G] If a Bp is voted out, the BP has to go to the very start of the process (section "A").
Note: BPs are selected in a way that staggers their reign throughout the year as to prevent any major disruption to the network
Some advantages are that it would be unnecessary to vote a delegate into the BP position, therefore vote buying would be unnecessary, pre-grooming would be realised as being highly inefficient, lengthly campaigning would not be required and a greater diversity of Block Producers could add a wider, and fairer, representation and sharing of the inflation rewards.
Under such a system, when abuse happens there is a clear path of accountability and retribution. Even if a petition triggered DOESN'T end up voting a BP out, such a petition is a clear and strong incentive for the BP to improve.
Qualified Demarchy is a proposed solution to some of the criticisms of standard DPoS with expected improvements in efficiency as well. The philosophy extends and supports Daniel Larimer's assertion that "The vast majority of people have good intentions" and efficiently flips the voting procedure to be used to remove bad actors, instead of spending time in selecting the most promising and charismatic actors (which, like modern politics, may prove to be the opposite of what they convey). If most people have good intentions, then the selection becomes less important than the systems of accountability and retribution.
Digital technologies have transformed nearly every industry they have come across. Part of the reason behind this is the ability to iterate rapidly and inexpensively. DPoS allows for human systems of governance to be tried, tested and iterated upon. Economic incentives ensure that large networks of people collaborate, cooperate and compete within a defined ecosystem. Over time, new models, mechanisms and dynamics can be found that have particular effects, and improved efficiencies and desirabilities amongst the participants.
Governance may very well be the true 'killer app' of the blockchain, with currency simply being relegated as the incentive component within the greater governance umbrella.
With Qualified Demarchy as a proposed addition to the experiment that is Delegated Proof of Stake, we have an additional model in which to iterate upon on our road towards improved systems of human governance in our rapidly changing modern world—spurred on by the blockchain revolution.