October 02, 2018

Combatting Vote Manipulation on EOS

Combatting Vote Manipulation on EOS

Aurora EOS is the leading block producer for the informed voter. If you find our work helpful, please vote for our node: auroraeoscom

INTRO

Recently, allegations have surfaced of vote buying and vote trading taking place on EOS. Before discussing this issue further, let’s clearly define these terms:

Vote buying: when block producers pay other parties to cast votes for them, essentially offering kickbacks in return for votes.

Vote trading: when block producers promise to vote for one another in order to mutually cement their positions in the rankings.

Both of these activities are explicitly prohibited by the EOS constitution. Unfortunately, it’s very difficult to prove the occurrence of either, and as a result, enforcement is difficult.

The issue of vote buying/trading is directly tied to another contentious issue: exchanges voting on behalf of their customers. Currently, many exchanges control user funds and thus control their voting power. While some exchanges have allowed their customers to cast votes from within the exchange, others have been unclear about the process. If exchange users can’t or don’t vote, and the exchange instead votes with customer funds arbitrarily, it gives the exchange operators outsized influence in the ecosystem.

This essay will not attempt to determine whether or not the most recent allegations are true. Thus far, no concrete evidence has been presented to support the allegations. However, regardless of the veracity of these particular claims, we believe that this invites the EOS community to engage in a productive conversation around voting and governance. This article will focus on ways the EOS community can combat vote manipulation, regardless of where or how it is happening.

BLOCKCHAIN POLITICS

All blockchain systems are inherently political, and all of them have some system of governance in place. One could make the case that systems of on-chain governance and token voting, like EOS, are even more political than systems with more loosely defined governance, like Bitcoin. However, the EOS governance system also has the advantage of being explicitly defined, auditable, and allows every user to participate and have influence proportional to their ownership stake.

It is difficult to eliminate all forms of vote manipulation in a proof-of-stake system with on-chain governance like EOS. In free markets, parties will act in their own best interest, and some may attempt to game the system to their benefit. Instead of attempting to do the impossible and stamp out this behavior, we must do what we can to make the process less likely and less effective for those that engage in it. We can’t eliminate manipulation, but we can enact a system where incentives make certain types of malicious behavior less attractive and more importantly, less likely to have an effect on the security of the network.

EOS CONSTITUTION  

“No Vote Buying - No Member shall offer nor accept anything of value in exchange for a vote of any type, nor shall any Member unduly influence the vote of another.”

Above is a quote from Article IV of the EOS constitution. The EOS constitution is described as a peer-to-peer terms of service agreement that defines the rules by which block producers, users, and smart contract developers agree to abide. It is an interesting experiment in decentralized community governance, but it has limitations. The EOS constitution cannot be enforced by code; rather, it can only be enforced by community consensus. The constitution is only as powerful as the community that requires that participants adhere to its rules. When it comes to the issue of vote-buying, this is especially clear.

DPOS is unique because it allows token holders to fire validators who do not serve the community well. Block producers can be voted out for objective or subjective transgressions. Accusations of vote buying fall under the latter category. It will be extremely difficult to objectively prove vote buying or trading, so we encourage voters to think carefully before acting on these rumors. If voters act exclusively on allegations and rumors, it could lead to false accusations and witch hunts between competing block producers. That would likely prove harmful to the network in the long run.

WORKING AGAINST VOTE MANIPULATION

Let’s examine the problem at hand. It consists of two major issues:

1. EOS voter participation is low.

  • Low turnout leaves more room for manipulation by whales and other large token holders.  
  • Not all users have the tools and information they need to cast votes.
  • Many token holders and investors have yet to make the connection between the long-term value of their holdings and the integrity of the EOS validator set.
  • On-chain governance and active participation in decentralized networks is a new phenomenon that is still maturing.

2. Many users keep their tokens on exchanges.

  • This gives the exchanges control over a large amount of voting power, which creates outsized influence over governance.
  • For some exchanges, it is unclear how they direct users’ votes, or if they allow users to vote at all.
  • Some exchanges may not allow users to select block producers to vote for, but may still be voting arbitrarily using customer funds.

SO, HOW DO WE ATTEMPT TO SOLVE THESE TWO PROBLEMS?

1. Increase Voter Participation

Perhaps the single most important way of combating the outsided influence of any group of voters within EOS is to increase total voter participation. Currently ~39% of EOS are staked, but even fewer are actually voting with their staked tokens. Among those that are voting, a minority are actually casting votes for 21 or more total block producers. There is a significant amount of voting power that is directed at a single BP, or a small number of BPs, in order to boost their rankings relative to other block producers.

With such low voter turnout, manipulation of the voting pool by whales, exchanges, and other large parties becomes far easier. A whale with 5% of the voting power can have much more influence when only 25% of tokens are voting than it can when 75% are voting. For this reason, it is extremely important to make it as easy as possible for EOS holders to learn about BPs and to vote.

One of our primary goals at Aurora is to work towards increased voter participation in EOS, and that’s why we’ve made education our primary focus. We believe that proper education and the right tools can drive up the number of voters significantly.

Many EOS holders view EOS as an investment, but few (including professional investors), have fully grasped that active participation in governance directly affects the success of their investment. Developers and other enterprises building on EOS have made significant investments into the token to access the resources necessary, and they, too, should be encouraged to be more actively involved in network oversight. For many users, this is as easy as casting their votes to a proxy who is more involved in the community. One of the best features of EOS is that validators can be held accountable by token holders— but it is up to token holders to exercise this power.

It is important to remember that the EOS community consists of a large number of different groups of participants, all with different wants, needs, and goals. The EOS governance model allows these different groups to actively influence the direction of the network. Through proxies, they can form voting blocs to collectively enact decisions in their best interest. The free-market decision-making competition among these groups is one of the most important features of the EOS governance system and helps single parties from forming a controlling interest in the network. Users should be educated and encouraged to fully take advantage of this powerful setup.

Recently, Block.one released a blog post that strongly implied that it could soon begin voting with its 100M EOS stake. Many in the community see this as a welcome development, since Block.one could use their large stake to vote against bad actors. However, Block.one is just one of many parties that has a vested interest in the long-term success of the network. All of these parties should be encouraged to participate more actively. With the right tools and educational resources, this can be achieved.

2. Adopt the Resource Exchange (REX)

We believe that the REX model proposed by Dan Larimer is the best path forward for EOS. The REX can positively impact usability, reduce developer costs, increase voter participation, and make it much harder for users to manipulate the voting system. We’ll be releasing a blog post soon articulating our views on the many features of the REX, but here we’ll specifically talk about its impact on voting.

The REX model creates an in-protocol liquidity pool where users can rent out their unused capacity (in the form of tokens) to other users who want to access those resources. Proceeds from the EOS RAM market and name auctions are also added to the books of the REX and distributed to participants. This turns the EOS token into a productive asset.

Any user who is holding EOS in excess of their actual resource consumption can earn a return by lending out those tokens, which would otherwise sit idle, to the REX. This more evenly distributes resources and creates a powerful financial incentive for users to participate in the REX. One of the best features of this model, however, is that users must vote for at least 21 BPs (or delegate to a proxy who does) in order to lend their tokens on the REX.

This is an improvement upon the current system for a number of reasons. Financial incentives are perhaps the most powerful way to change behaviors, and users who are currently holding EOS at a 0% return will naturally want to earn a positive return instead. This is especially true for large holders. Many users who currently don’t participate in voting will be compelled to finally do so for this reason. Further, it prevents users from voting for a single block producer in order to manipulate the rankings. If they do so, they’ll lose out on the returns from participating in the REX. This system will likely result in higher voter participation and more votes being cast across a wider set of BPs, which helps combat attempts to manipulate the voting pool.

3. Drive Usage Over Speculation  

The EOS token model is designed in such a way that users of the network are required to hold EOS tokens in order to access the network’s resources. This means that users who want to use dApps built on top of EOS will have to stake EOS (or have it staked to them by the application). Doing so requires users (or dApp developers) to actually keep their EOS in whichever wallet they are using to interact with the application. This is impossible if their tokens are sitting on an exchange.

Many users keep their tokens on exchanges because they are simply speculating on future price increases in the EOS token. They have no need or desire to self-custody their tokens, because they don’t actually want access to the network resources upon which they have a claim. This is partially a consequence of the nascency of the network. If users don’t have applications to use, then the “killer app” of EOS is simply speculation. And speculation doesn’t require users to take their tokens off of exchanges.

Currently, the cryptocurrency market is almost entirely driven by speculation rather than usage. EOS is a notable exception. Because of its architecture and focus on usability, EOS is by far the most widely used blockchain, and usage is growing more quickly than any other chain despite the fact that EOS is only 5 months old.

As EOS matures and more applications get built, we’ll see a shift in these statistics. Users will be compelled to own EOS because it allows them to use applications, and they’ll have to take their EOS off of exchanges in order to do so. So one simple way to combat the outsized influence of custodians is to continue to do what many in the ecosystem have already begun doing— build useful applications and secure tools that allow users to interact with those applications.

4. Increase Usability and Security for Voting

Aside from education, the single biggest factor that limits the number of voters is the actual logistics and usability of voting software. While we can educate users about the importance of involvement and self-custody, they will not actually participate unless it can be done in an easy, secure, and intuitive way.

Currently, users of EOS have a number of options for securing their private keys, but usability is still an issue. Users must be highly technical to take advantage of the most secure offline voting tools, and even the best desktop and mobile wallets are still new and not fully trusted by all users. For large holders like crypto funds, enterprises, and other institutions, secure custody is especially important. EOS token holders need better education around the options available to them, as well as improvements in the ease of use for these tools.

We are seeing good progress on this front. Ledger announced EOS support recently, and several existing wallets have integrated Ledger functionality. We’re seeing improvements in the usability of offline voting tools. More developers are taking advantage of the EOSIO software’s powerful system of account permissions and account recovery. And Block.one is working on an iOS wallet that uses the Apple secure enclave to essentially put a hardware wallet inside of everyone’s phone. As these options mature, users will be able to more easily and confidently participate.

5. Push for Exchange Transparency

Exchanges and other third-party custodians will likely always play a role in this industry. These parties offer extremely valuable services that can help strengthen the crypto ecosystem. However, users should demand transparency and accountability from the exchanges that custody their tokens. Exchanges should be fully transparent, both with their customers and with the community at large, about how they are using customer funds. Users should have the option to vote, as well as the ability to verify that their vote has been cast. Exchanges should not be allowed to arbitrarily vote on behalf of users without their explicit consent.  

The rise of decentralized exchanges like EOSfinex will alter the landscape of crypto exchanges. Users will be able to trade assets on-chain without going through lengthy withdrawals and deposits and trusting their funds to exchanges. DEXs cannot yet offer all of the value-add services that centralized exchanges can offer, but the existence of DEXs serves as competition to centralized exchanges that will force them to bolster their services in order to remain competitive.

FINAL THOUGHTS

EOS is less than five months old, and the network has already achieved a number of great successes. The EOSIO software contains some of the most compelling features that can drive forward usage of blockchain software at scale. But every network faces growing pains and challenges as the software, community, and governance model mature.

For all of the successes of EOS, the network is still in its infancy. Low voter turnout, lack of education, and lack of tooling all leave more room for voter manipulation than would otherwise be present. However, there are clear ways to address these problems, both from a community perspective and from a technology perspective. As EOS grows and supports more use cases, those invested in the long-term success of the network will combat the forces, like vote manipulation, that degrade the long-term security of the network. We encourage all EOS holders to start today.

——————

Aurora EOS  is the leading block producer for the informed voter. If you find our work helpful, please vote for our node: auroraeoscom

If you prefer to proxy your vote, our proxy account is auroraeosprx

Our Voting Guide

Find us on TwitterMediumTelegram, and Steem

Subscribe to our newsletter for weekly updates on the EOS ecosystem