By Doug Petkanics
March 9, 2020
The Livepeer Project’s mission is to build the world’s open video infrastructure. Through the power of open source software, combined with the harnessing of underutilized resources like compute and bandwidth, combined with the use of cryptoeconomic incentives for bootstrapping and participation, there is an opportunity to deliver an infrastructure that can power video streaming applications at a highly efficient price, and infinite scale. Along with the proliferation of high quality cameras and ubiquitous bandwidth, the belief is that this will enable video applications to be created that were never before possible under traditional, centralized cost structures, which unlock communications, economic opportunity, and entertainment for society going forward.
These three enabling factors (open source, an open market for resources, and cryptoeconomic incentives) inherently imply decentralization, as they indicate that the development, the infrastructure, and the economic policy are each contributed to or adhered to by many independent parties, all around the world. It is this network of stakeholders - developers, infrastructure operators, and network participants - who give Livepeer the opportunity to push down the stack from single application, to base layer protocol which can power every video streaming experience on internet connected devices. Such a base layer, can inherently be thought of as a public good. And as such, it should benefit the public and be governed by the public, rather than benefiting a single entity whilst being governed by that same entity.
However, for all the benefits that a worldwide network of incentivized stakeholders bring to a project like Livepeer, they also create challenges, particularly in the realm of decision making and resource allocation - in short, governance. Each party may have different priorities, different incentives, or different beliefs about what may be the best way to build, deploy, and operate the worldwide video infrastructure. Video streaming itself is a very complex, cutting edge technology, where few possess the knowledge to develop on the latest innovations, and productize in a way that delivers a reliable, usable platform for many applications. Knowledge is a key prerequisite for informed participation in governance, and a network can only be as decentralized as the number of people who are qualified to make informed decisions. Combine these differences in opinions, along with the technical complexity of making sound decisions, and the challenges of finding alignment amongst a loose collective of global stakeholders, and it’s evident that governance over Livepeer is no simple matter.
As there are no shortage of possible governance structures available to Livepeer, it should be stated from the beginning what governance structures are not likely to be successful long term - those of centralized, corporate or individual authority. While a corporation, or a benevolent dictator can provide a highly efficient form of governance, where decisions are made quickly, and it is clear what direction the project is going in, it comes with too many tradeoffs that wouldn’t allow Livepeer to achieve its ultimate potential.
- Open source contributors would likely yield to the roadmap, gatekeeping, and authority of the entity, rather than seeking competitive improvements out of their own self interest.
- Infrastructure operators would feel as if they were on an uneven playing field with the entity itself, knowing that the entity could make changes or take control of the system, de-platform them, and affect their economics in a negative way - hesitating to get involved in the first place.
- Cryptoeconomic network participants would be skeptical of the incentives, be unsure what the future holds, and be disempowered to participate in a known process for affecting updates as needed in the future.
Although Livepeer started off as the idea of a small group of founders and early core developers, and although the early incarnations of the protocol enabled this group to iterate rapidly, fix bugs, and protect user value, this group also has recognized the importance of decentralizing the governance and control over time, for all the reasons mentioned above.
The transition process will not be trivial, but it will be worth the effort. The answers for the perfect governance system are not known in advance, but a few key founding principles are known, that can guide the creation of the proper governance system and decentralization. The remainder of this document is an attempt to list out and elaborate on these principals, so that they can serve as guide posts for those participating in Livepeer and its governance going forward.
Prioritize practicality and utility
It is easy to get lost in the theory of the elegant design of perfect systems. But the likelihood of designing something perfect on the first go that will be thrown up against the messiness of human psychology, economics, irrational markets, complex technology, existing industry, and regulation is near impossible. Instead of attempting perfect design and outcomes, focus on practical utility. Livepeer should work. It should be usable. It should make an impact. It should begin to deliver on its value propositions. Make decisions that enable it to do these things, and iterate towards perfection after it is clear that you have something worth decentralizing and governing.
Governance should be extensible
Governance systems can get very complex. Related to the above points on practicality, it could be an exercise in futility to attempt to design the perfect mechanisms in absence of real world impact and the effects that come from the influx of stakeholders that leverage this. Recognizing that not just the protocol will need to be upgraded in the future, but the governance processes themselves will need to be as well, start out with a governance process that is clear, extensible, and amendable itself.
Remember the Mission
"Build The World’s Open Video Infrastructure”.
Make technical first decisions in service of the mission
For Livepeer to fulfill its mission, there are many different types of stakeholders who will have a part to play - some technical, and some not. However, as Livepeer is base layer infrastructure and protocol, the stakeholders who contribute directly to enabling its functionality are largely technical - open source developers, network infrastructure operators, and the users who will integrate the infrastructure with their own applications. It’s critical that the technical voice be a guiding hand in the governance process. This can be accomplished through education and information dissemination shepherded by the technical participant set, such that the majority of non-technical users can make informed decisions that support technical progress, and reduce the chances of technical harm.
Processes should be transparent, documented, and well understood
Every community stakeholder should have a very clear view of how governance decisions get proposed, reviewed, and executed. This doesn’t mean that every participant is committing to every communication being made in the open, but it does mean that the technical means for which governance actions are executed adhere to a well communicated and documented process, in which the mechanisms for participation are clear and understood to everybody.
Community should have a clear path to come to consensus on governance decisions
If the above requirements around transparency, documentation, and process hold, then it should be clear to everyone in the community what the steps are to reach a consensus based action on the network.
Stakeholders should have the opportunity to opt-out of the network
If someone disagrees with a governance decision, then they need to be able to exercise their right to exit the network. Possible mechanisms include forking the protocol code, forking the underlying network itself, liquidating their LPT position, or transitioning to another video infrastructure solution.
Incentives should encourage participation
Remember that as a global network and video infrastructure backbone, Livepeer can only succeed through the active participation of the key stakeholders mentioned above. The incentives within the protocol were initially designed to incent this participation and route ownership towards those who were doing active work. Governance should adhere to this principal, at the expense of other short term motivations.
Those who do work on the network should accrue ownership and benefit from the economics Work comes in many forms - infrastructure operations, development, security and QA. Decisions should be made and resources should be allocated to incentivize these types of useful participation.
The more a network grows, and the more diverse the interests of the stakeholder base become, the more likely there are to be contentious governance issues that arise in the future. Although potentially painful at the time, they are a sign that the network is useful and is achieving its promise to the point where these parties care enough to voice conflicting opinions about the future direction. Hopefully in those debates, the parties can look back on these founding principles, and use them as a guiding hand to represent the intentions of the project in the early days.
A technical first, practically usable network for video developers, with the goal of serving as the backbone for the world’s video infrastructure, run by thousands of worldwide participants who are actively engaged in its success and financial upside as owners, is an incredibly powerful thing.