EPNs — Part 4: The PRESTO Framework
In Part 1 of this series, we introduced the concept of the Economic Protocol Network (EPN), networks that are built on blockchain technology. In Part 2, we looked at how EPNs encourage participation by distributing network value. In Part 3, we explored an example of how EPNs operate in the wild. In this article, we’ll share a framework that can help developers design EPNs effectively and sustainably. And finally, in Part 5, we’ll explore how they might be deployed in games.
The benefits of Economic Protocol Networks are obvious, but the best means to obtain and sustain those benefits often are not. The illustrative example of Compound in our previous article showcases a high-quality EPN, whose design is simple yet well suited for generating positive network effects. To achieve similar results, we recommend the use of the qualitative PRESTO framework, first outlined by Stefanos Leonardos, Daniël Reijsbergen and Georgios Piliouras of Singapore University.
PRESTO is an acronym that describes a five-axis design space for EPNs, using the following dimensions:
- Optimality: Does the protocol solve the problem in a maximized way?
- STability: Does everyone follow the protocol?
- Efficiency: Does a protocol utilize its resources efficiently?
- Robustness: Does a protocol survive attacks and unexpected behavior?
- Persistence: Does a protocol recover after being forced out of equilibrium?
Optimality: Does the Protocol Solve the Problem in a Maximized Way?
Optimality is the most foundational dimension for protocol design. It asks if two questions of an EPN at the highest level:
- Does the protocol fundamentally solve the problem it is designed to address?
- Does the protocol maximize outcomes for participants participating in it?
The Bitcoin and Ethereum networks are good examples of protocols that solve their stated problem and thus pass the fitness test. However, at this point in time, they do not solve it in an optimal way. The Augur v1 network, an early, highly touted Ethereum-based prediction-market platform that allowed trading of tokens representing complex bets or derivatives on virtually any proposition, financial or real-world, is an example of a protocol that did not pass the fitness test, due to the way the protocol was initially designed.
Stability: Does everyone follow the protocol?
This dimension deals with the question of whether network participants, especially purely self-interested actors, will consistently adhere to the protocol, and whether there are safeguards built into the system to encourage or enforce such adherence.
One such safeguard is simply the scale of the network. Stability in cryptoeconomic protocols increases when there’s a lot of activity (e.g. transactions) and when this activity is distributed across a large number of participants.
On networks where there isn’t a lot of activity and participation is concentrated among a small number of users, the network’s stability decreases and it becomes susceptible to disruptions like “51% attacks” — assaults on a network that are the result of a single entity or consortium amassing the majority of validation power in a network, thus essentially taking control of it. Such attacks have successfully taken place on Bitcoin Cash and Ethereum Classic, networks that are forks of Bitcoin and Ethereum.
Efficiency: Does a protocol utilize its resources efficiently?
The Efficiency dimension asks the question of how well a protocol utilizes its resources.
The current iterations of the Bitcoin and Ethereum networks are both Proof of Work systems that tend to utilize more energy resources as network participation increases. Additionally, their relatively long block/transaction confirmation times can cause network users to not be able to transact with each other in a seamless and frictionless fashion.
A number of relatively new EPNs are trying to solve for the Efficiency criteria by using a different consensus protocol than Proof of Work. The most discussed solution is the switch from Proof of Work to Proof of Stake consensus. Whereas in PoW, miners contribute processing power (and thus, electricity) to validate transactions, in PoS, validators commit capital to have the privilege to attest (e.g. validate) transaction blocks, for which they are rewarded with more capital. The Cosmos blockchain and Ethereum 2.0 are based on Proof of Stake consensus, which is less computationally intense, and therefore more resource-efficient and faster.
Another way to increase efficiency is to reduce the level of decentralization in the network and introduce “trusted parties” to fulfill major tasks of the protocol. Examples are Ripple, which uses Proof of Authority, in which approved accounts can validate transactions; EOS, which uses delegated Proof of Stake, in which holders of EOS “elect” Supernodes that have the ability to validate transactions; and corporate, or “permissioned,” blockchain networks such as R3/Corda, which utilize Proof of Work but only via centralized, Corda-controlled validation servers. These EPNs allow for much higher activity than fully decentralized protocols, increasing efficiency at the expense of absolute network democracy and decentralization.
The decision whether or not to introduce trusted parties in order to increase efficiency ultimately depends on the problem that the EPN is trying to solve. The Bitcoin protocol addresses the need for permissionless, uncensorable money. To solve for this, Bitcoin should be fully decentralized. Therefore, for Bitcoin’s users to benefit, the network should not depend on trusted parties.
There are however, plenty of problems in which trusted parties and centralization are desirable, or even requested by users. For example, in the game development space, most aspects of a game and its ecosystem are governed by the developers themselves. Even in gaming networks with decentralized aspects or open worlds like Minecraft, the network participants — gamers — expect the developer (the trusted party) to introduce patches, release new content, set the rules of the open world environment and act as the trusted authority in the event of disputes.
Robustness: Can a protocol survive attacks and unexpected behavior?
This dimension looks at a network from two angles. Firstly, its resilience to malignant attacks, and secondly, its malleability in the face of behavior that is against the initial network design. In the table below, we look at the three networks from this perspective:
Persistence: Can a protocol recover?
If a protocol fails in one or more dimensions, it is more than likely that participants will leave, causing a multi-sided network to become asymmetrical — dominated by one type of participant — or, in a worst case, being abandoned entirely.
The persistence consideration therefore deals with the question of whether a network has built-in mechanisms to help it recover after being forced out of equilibrium.
There are a number of recent examples of traditional networks that have found a way to recover after risking imbalance or stagnation. For example, Facebook’s introduction of its mobile app and its acquisition of Instagram are examples of ways that operators were able to increase user participation in the network and total network value after a period of flattening growth.
Games are natural candidates for the implementation of EPNs. We’ll explore why in our next and next article about EPNs as Game Networks.
Interested in contributing to our Community Economics series? We’d love to hear from you. Comment below or email us here at firstname.lastname@example.org.
Follow us on Twitter @FortePlatform