Pay People Not To Vote: Dan Larimer's Radical Solution to EOS Vote Buying
Share this article
EOS has been wracked by issues of governance and centralization during its first year of existence. However, the most pressing issue is a simple one: vote buying. The community is searching for solutions, and now, Dan Larimer, the CTO of Block.One, has proposed a radical fix―which ironically involves encouraging people not to vote.
The Problem With EOS Voting
EOS relies on a system called delegated proof-of-stake, which allows holders of the EOS token to vote for block producers. EOS holders can also vote in referendums, but block producer elections are the blockchain’s bread and butter. Basically, elected block producers govern EOS and represent the interests of EOS token holders.
However, there are a few problems that come along with EOS’s democratic voting model: First of all, many investors consider EOS a speculative investment, so they hold EOS tokens, but do not vote. Secondly, many users store their EOS on an exchange, which gives those exchanges plenty of power when it comes to voting.
These problems have resulted in allegations of vote buying scandals. Huobi and several other EOS block producers allegedly engaged in a mutual vote buying scheme in October 2018. In a separate incident, a block producer called StartEOS attempted to explicitly buy votes from EOS users.
Finding a Solution
The simplest way to solve vote buying merely involves encouraging users to vote so that their voices are heard. However, this is easier said than done: block producers and exchanges have plenty of voting power, and they are more motivated to vote than basic users. As such, workable solutions require complex changes and additions to EOS.
One part of the solution involves EOS REX, a new resource exchange that allows EOS users to earn revenue. REX requires users to vote for 21 block producers in order to use the system. This gives users a reason to vote―but it is only a partial solution, since not everyone cares about REX.
Brendan Blumer, CEO of Block.one, has also proposed a more complex solution. He suggests a “one token, one vote” system, in which a single vote carries more weight than several votes. This voting model should eliminate mutual voting cartels, and it should also increase voter turnout, among other things.
Larimer’s Radical Proposal
This brings us to EOS creator Dan Larimer’s latest proposal.
He suggests paying users not to vote.
This idea may sound counter-intuitive, but it is easy to see how it might prevent vote buying. As Larimer explains, his idea essentially involves outbidding exchanges and block producers that want to buy votes. He writes:
Larimer’s proposal is the direct opposite of almost every other solution that has been proposed, since it doesn’t attempt to encourage voter participation. It would also require REX to change its incentive model―that is, users would need to abstain from voting in order to use the resource exchange.
The proposal is also in conflict with Blumer’s “one token, one vote” proposal. Larimer suggests that his own proposal will move toward a de facto “one user, one vote” model. That is, if EOS holders have a reason not to vote, even the largest exchanges and block producers will cast very few votes.
The Wider Implications
Larimer’s proposal may or may not succeed, but if it does, EOS’s critics will probably see the change as evidence of EOS’s overwhelming centralization.
Larimer acknowledges this: “Paying people to give up control will centralize things, but fairly…the only way to decentralize power from exchanges is to pay exchanges not to vote.”
EOS has always been controversial due to its centralized nature. It does, after all, centralize power in the hands of just 21 block producers. However, this must be considered in light of the fact that other blockchains experience centralization too: Bitmain, for example, has at times heavily dominated Bitcoin mining.
This means that if you are feeling favorable toward Larimer, EOS isn’t merely embracing centralization. Instead, it is bracing itself for the inevitable reality of centralization and adapting itself as problems arise. Of course, adaptations can have unwanted side effects, so EOS’s solutions to centralization may be a bandage at best.
If that is true, modifications to EOS voting can only change so much. Voting doesn’t necessarily increase or decrease EOS’s overall degree of centralization. Instead, users ultimately vote for centralized representation.
Most changes to EOS’s voting system, no matter how radical, will not change that basic fact.