We recently offered a high-level view of Livepeer’s current state, and how the work we have undertaken over the past five years has placed us in an excellent position to achieve broad project goals and to seize a rich store of emerging opportunities.
We’d now like to take a closer look at one of those goals – one that we believe is mission-critical to becoming the world’s open video architecture: true trustlessness.
Since its inception, Livepeer has been on an unwavering path to full decentralization. But while it is already one of blockchain’s most decentralized physical infrastructure networks (DePIN) and resource-allocation protocols, it has not yet fully achieved that coveted objective.
Trustlessness – the removal of all centralized control and single points of failure in a network or protocol – is not just important to Livepeer in the abstract. Decentralization is a key component in bolstering Livepeer’s already highly resilient network.
Thanks to the advances we have made in recent years, it now lies within reach.
Economic ownership, Technology, Governance: The Three Pillars of Decentralization
From the start, we have made a concerted effort to achieve decentralization of stake through measures such as the use of the MerkleMine algorithm to distribute tokens. Over the past five years, the Livepeer protocol has distributed more than 180% of the initial token supply openly and permissionlessly to engaged network participants.
On the technology front, decentralization is nearly total. The network is run by a global nexus of 100 node operators, scores of pool members, and thousands of token-holding delegators. There is work to do on the verification challenge to ensure full trustlessness, but more on that later.
Decentralized governance was introduced in 2020, enabling stake-weighted voting on such steps as protocol updates, parameter updates, and governance changes. The continued presence of a security committee charged with executing all ratified proposals and fixing critical protocol bugs has been a source of community debate.
Bridging the Gap to Trustlessness
Given the project’s diverse staking population, widely dispersed node operator network on-chain decentralized governance, and now – with the recently approved Delta upgrade – a community-governed treasury, Livepeer is in a prime position to eliminate any remaining dependencies and single points of failure.
Taking these final steps toward full decentralization will be good for both the network and the community, and will fulfill a goal the project has had at its core since the start.
So what are those final steps and how will we go about achieving them?
First, we must move the execution of protocol upgrades and parameter updates on-chain. While the community does govern such upgrades through the Livepeer Improvement Process (LIP) currently, their actual execution must be carried out by the security committee.
There is a potentially straightforward path to change that, particularly given the recent introduction of the Governor framework for treasury management via the Delta upgrade, and the similarities and capabilities of the governance mechanisms for both the treasury and the LIP process. This framework could easily be adapted to execute binding protocol upgrades on-chain, thus removing the role of the security committee in enacting protocol upgrades.
The actual mechanisms used should be open for community debate – and potentially for modification, should the community want to make changes to governance in the process of evolving the execution process, but the fact remains that the Livepeer protocol already has the groundwork laid related to the undertaking of placing upgrade execution in community hands.
Second, we must introduce direct community governance over the security committee process itself. At the moment, the committee’s power is ratified via LIP-19, and it operates according to a social contract that includes a commitment to modifying its size or processes, if such changes are proposed by the community.
Despite that pledge, the current set-up falls short of true trustlessness. It would be a major advance toward full decentralization if community decisions relating to the committee were binding: including the addition or removal of keyholders, the modification of signing thresholds, or, potentially, the elimination of the body as a whole.
The community might consider following the lead of the Arbitrum ecosystem and introducing a formal election process for the security committee. Elections for the Arbitrum DAO Security Council are held every six months, with each member serving a one-year term.
Third, we will need to make further advances in the fast and full verification process to make the network truly trustless and eliminate all reputation-based inputs.
Livepeer’s fast and full verification process is designed to solve one of the thorniest problems in decentralized computing: ensuring that transcoding work is successfully carried out. Verification of this work remains a high probability, but not yet perfectly economically secure process.
The fast and full verification process encompasses two strands: the fast portion, which allows broadcasters to receive and verify video segments in near real-time with a high degree of confidence the work was performed correctly; and the “Full” portion, which is a slower dispute-resolution procedure that can result in financial penalties for mis-performed work after the fact.
More work needs to be done on the second of these two strands to achieve full trust assumptions in an open network.
By moving upgrade execution on-chain, placing Security Committee governance in the hands of the community, and by designing a more trustless method for full verification, it will be possible for Livepeer to at last pass the milestone on the road to full decentralization.
Anyone is invited to take part in the journey through Livepeer’s LIP process and to get involved in the ongoing discussion via the #governance channel in Discord.