Governance & Dispute Resolution: the Missing Piece of the Blockchain Ecosystem Puzzle

Xu Yin
10 min readDec 13, 2018

--

Blockchain technology has started a technological revolution, and along the way, it raised new challenges and unanswered questions. One such question is: how do we ensure security and reliability of yet-unproven new technologies, such as smart contracts?

We are used to high security standards, and for the new peer-to-peer ecosystem to succeed, it needs to come with the same high level of security and protection of our rights and assets we see on traditional web platforms.

We can look at “security” as comprised of several layers: transaction reliability, wallet security, code, network security, and arbitration mechanisms.

In most cases, the phrase “blockchain security” refers to the reliability of transactions and security of code from an anti-hacking and cybersecurity perspective.

At Oath, we believe that a fundamental governance layer is crucial for providing end-to-end blockchain security. At this time, however, it is still immature and underdeveloped.

Daniel Larimer is the pioneer in researching and integrating a governance platform for EOS, which has its own Constitution that incorporates a framework for resolving disputes, merging technology with law. The EOSIO Core Arbitration Forum (ECAF) established the EOS Constitution by setting up guidelines, terms, and conditions to establish an arbitration mechanism as described below:

“All disputes arising out of or in connection with this Constitution shall be finally settled under the Rules of Dispute Resolution of the EOS Core Arbitration Forum by one or more arbitrators appointed in accordance with the said Rules.”

By studying ECAF, we can learn more about EOS’s understanding of blockchain governance and their core concept and ideas.

Why Governance & Dispute Resolution Are Important, Suitable Approaches, and Where They Apply

The blockchain ecosystem is similar to the democratic community in that anyone who wants to participate may become a contributing member. From its side, the independent, self-governing community needs to protect the rights of its users, to foster collaboration and emergence of innovative business models. To achieve this, there has to be a set of rules and protection mechanisms for governance and arbitration.

Given the nascent state of the blockchain technology, its vulnerability to hacker attacks, scams, and smart contract loopholes is very real. Black-hat hackers use the blockchain’s privacy features to their advantage, constantly trying to undermine the technology, exploiting leaks of the code, and maliciously attacking through phishing. Such violations are extremely difficult to prevent, and even harder to completely eliminate.

For example, more than 2.09 million EOS tokens, the equivalent of more than USD 10 million, were recently stolen from an EOS account in a hacking incident. Whether such events occur due to the weakness of EOS’s security system or the hacker’s exploiting of the user’s negligence, in safe and reliable systems, users should have recourse against such harmful behaviors.

Fortunately, EOS has an established governance mechanism, ECAF. It is not a perfect solution yet, but it allows users who have been wronged to obtain support in case of technical failures. Presently, most of ECAF’s filed cases allege theft and scam. But blockchain security is not limited to theft and scam; those are extreme forms of violation of users’ rights. The majority of disputes that arise from daily usage will be related to transactional and contractual disputes, just as they do on traditional platforms. If dApps and blockchain companies are going to compete with those traditional platforms, they need to be able to secure their users’ rights.

In decentralized ecosystems, the central authority is replaced through code and consensus algorithms, also removing the decision-making party (distribution of responsibility leads to distribution of decision-making). This leads to a lack of assistance and support in the event of disputes or conflicts. I believe that decentralized platforms should be governed by decentralized communities, in line with the essential values of decentralization and power distribution. What could such a reasonable and reliable high performance mechanism look like?

Governance Mechanism

There are several approaches to manage blockchain governance. At the current stage, it is hard to determine which way is most suitable; it would be like comparing the continental law system and the Anglo-American legal system. It is hard to say which one is better because each has its own advantages and disadvantages

Below, we explore some common use cases to see how different approaches may be more suitable for different scenarios. But before we get into it, let us first examine which factors make a governance system sustainable.

What is Good Governance?

Any credible governance mechanism should have fairness as the fundamental base layer. Transparency and openness are additional core values for structuring a fair system for all parties involved; if the arbitration results are provided by a black-box system, the output is not truly trustable. Extensibility and dynamic adaption are likewise important because the system needs to be able to adjust to new kinds of conflicts and issues arising in an industry as nascent as the blockchain is today.

In addition to being fair, a good governance mechanism should also be cost effective and efficient. If the case resolution process is long and expensive, it undermines the disputants’ legitimate rights and interests. A reasonable combination of cost-efficiency and a clear timeline for each case is, therefore, a hallmark of a desirable governance model.

All factors above can be achieved through centralized institutions. Centralized institutions, however, often face conflicts of interest or external influences that lead to malicious acts or false outputs.

Our solution is a decentralized jury, comprised of randomly selected, diverse members of the community from all over the world, to prevent concentration of decisionmaking authority and reduce bias, leading to fair results.

EOS and ECAF

Let us look more closely into how ECAF, EOS’s arbitration management plan, works. ECAF uses a relatively centralized arbitration system of six internal arbitrators. The vast majority of cases submitted to ECAF will be reviewed by a single arbitrator, or panels of three in rare instances. Case decisions are sent back to the Block Producers (“BPs,” 21 total) via ECAF, who have the final say on the ruling because they have the authority to execute the case decision.

It is important to note that ECAF aims to recruit professional arbitrators with certain expertise to help facilitate the decisionmaking process for the BPs. The arbitrators cooperate with the BPs to protect the users’ (claimants’) property and assets. The advantage of this design is that all of ECAF arbitrators are highly professional, experienced, and familiar with the concept of the blockchain technology and related concepts, including the EOS Constitution and arbitration mechanism. At the same time, as an independent arbitration institution, they can also generally avoid conflicts of interest.

On the other hand, limiting the number of arbitrators on the platform and requiring specific expertise and training also leads to a very low case turnover, making it an unscalable and slow solution. In addition, the arbitrators charge fees for their time and expenses associated with resolving cases, leading to high fees and making it a bad fit for low-value cases, e.g. below $1,000.00. Finally, the power of protection of EOS users’ rights is concentrated in the hands of only six arbitrators and 21 BPs, not all of whom are equally dedicated to the stability and success of the platform (e.g., Bitfinex and Huobi).

ECAF Case Flow

Once the case is submitted, it is assigned to an arbitrator according to certain case criteria and amount in dispute. The arbitrator’s identity is public, which leaves him or her vulnerable to external influences, like bribery or threats.

Presently, the ECAF arbitration system is technically under-developed, which is likely because there are no team members with tech background. In addition, there is no official publication platform, and ECAF-related communication channels are inefficient and confusing because there are several working groups trying to resolve the same issues, but without any one official channel for communication. This leads to an overall inefficient process, with uncertainties due to multiple communication channels and parties involved (conflict of interests & opinions), as well as lack of focus on technical development.

On the other hand, ECAF showcases the necessity and importance of blockchain governance and provides a protection mechanism (to a certain degree) for EOS users. It also exposes the complexity and difficulty of integrating such a mechanism on a running system and shows us that the more parties get involved, the more challenging it becomes to implement changes. ECAF is an important example of current challenges in blockchain management and development.

OATH Protocol

Let me now introduce a third-party arbitration service provider that takes a radically different approach to governance. Oath Protocol is modeled on the jury in the Anglo-American legal system, which relies on members of the local community to decide cases. On the Oath Protocol platform, by default, each case is decided by a group of 11 jurors, and the disputants have the option to change the jury size. The disputants also set the jury reward pool, to incentivize the jurors to join the jury and the decide the case. The jurors’ identities are kept anonymous during the process to preserve objectivity and reduce bias and collusion. Once the case is decided, the case information is made public and stored on the Oath Ledger, and either party may request an appeal (up to two in each case) to have the case reviewed by more experienced jurors.

Oath’s built-in reputation mechanism, called the credit level system, will track each juror’s behavior on the platform and assign credit levels based on several factors, such as the number of cases decided, voting accuracy, response time, token holdings, and others. The credit level will not influence the weight of each juror’s vote, but it will affect how many tokens the juror receives from the reward pool.

Notably, Oath Protocol’s arbitration system is designed to prevent conflicts of interests and juror collusion. Distributing the decisionmaking power among the general public (blockchain users), instead of restricting it to a small group of individuals with specific expertise, will also facilitate long-term scalability because it will provide a virtually indefinite jury pool to fill in any number of juries. Cost reduction is yet another advantage because there is no cost associated with arbitrator training or professional fees.

Our approach is based on decentralized, community-driven decisionmaking, meaning that whatever the general public believes is right or wrong should be generally accepted. Because the jurors cannot interact with each other during the voting process, each juror will have to make his or her own independent decision based on the facts of the case. At the end of the set voting period, the majority’s vote will determine the winner.

Oath’s mission is to redesign the arbitration process to improve security for any blockchain or dApp, with a focus on core attributes like fairness, transparency, scalability, low cost, efficiency, and decentralization.

As a third-party service provider, Oath Protocol’s main challenge will be gaining widespread recognition and acceptance in the blockchain community, while providing a highly secure, reliable, and automated technology for integration. We believe that our commitment to providing a scalable, secure, and fair blockchain governance and dispute resolution platform will lead us to become the standard governance and dispute resolution protocol for the blockchain ecosystem.

Governance and Dispute Resolution Use Cases

So far, we have explored why arbitration and governance are of crucial importance and where those concepts apply. They are not, however, limited to blockchain-powered platforms. In fact, decentralized arbitration is particularly relevant in the traditional context because it presents an opportunity to replace central decision-making authority with mathematically secure algorithms, smart contracts, and consensus algorithms.

Next, I would like to examine several scenarios in traditional industries and the Internet. These use cases include e-commerce, leasing/short-term rentals, social networks, trading, insurance, and travel, among others.

In 2016, Alibaba launched a customer-to-customer e-commerce platform called XianYu (means “Idle Fish”). Today, it has over 500 million registered users. Even though it is one of the largest e-commerce platforms in the world, XianYu does not provide traditional customer service. Instead, the platform offers an arbitration mechanism, managed by XianYu users, to resolve thousands of disputes occurring on XianYu daily.

For instance, a buyer purchases an item and pays the price through an external service provider (mostly AliPay). The payment is held in escrow until the goods arrive to the buyer. If the purchased item is not as described and the buyer wants his money back, he can initiate a dispute and publish it on the platform so that other users can vote on whether or not he should get his money back. Seventeen users are randomly selected to vote on the case (to prevent collusion and bias), and if the majority votes for the buyer, the platform will automatically initiate a reversal of the escrowed payment.

This is a simple case study from the traditional internet using a “decentralized jury” to resolve user disputes. The main purpose of XianYu, notably, is to reduce costs associated with traditional customer service, which inherently lacks such important features as transparency, shared communities, juror reward mechanisms, cryptographic security, and much more.

My team and I have examined several other examples of real-life community-based governance models, and you can review them in my previous post, Real-life Governance Use Cases.

Those examples demonstrate the evolution of modern-day arbitration models, not only for the blockchain space but also in traditional industries.

For example, the same community-based model may be adopted and used by familiar home-sharing platforms like AirBnB, car-sharing companies like Uber or DiDi, and travel platforms like CTrip because their existing arbitration systems lack transparency and efficiency.

Allowing a community of ordinary users to resolve disputes based on their own experience and reasoning would introduce fairness and credibility to the process that is often plagued by bias, conflict of interest, and even simple mistakes.

Combining the encryption and immutability features of blockchain technology, consensus algorithms, and juror incentives, Oath Protocol’s pioneers a new approach to governance and dispute resolution for the blockchain and traditional web platforms alike!

Please tap or click “👏below or on the left side of the screen to let Yin and others know that you appreciated this post. The number of claps indicates how much you liked the post and support its content, so put those hands together as many times as you like.

To learn more about Oath Protocol, sign up for email updates on our website. To stay up to date on Oath Protocol news and announcements, follow us on Medium,Telegram, Facebook, Twitter, LinkedIn, Steemit, Reddit, and Instagram, and join the conversation in our Telegram and SynQ communities.

--

--

Xu Yin

Co-founder and CEO of Oath Protocol, the first dispute resolution protocol based on blockchain technologies