The Design Principles Behind Nervos
This is the first in a series of blog posts exploring the key ideas behind the development of the Nervos Network. Each article will alternate between a core tenet of Nervos’s governing philosophy, and a use case that illustrates how these principles are intended to work in practice. Today, we focus on the principles that make up the foundation of Nervos.
Layered network
We have always believed the best way to harness the potential of blockchain technology is through a layered approach to the network design. Structurally, this is the main concept driving Nervos’s architecture. A multi-layered solution solves, or avoids, most of the key dilemmas currently facing the blockchain community.
When we think about the following problems:
- decentralization vs. scalability
- neutrality vs. compliance
- privacy vs. openness
- store of value vs. transaction cost
- cryptographic soundness vs. user experience
We can identify that all of these conflicts stem from attempts to “square the circle” using a single blockchain. From a basic standpoint, this doesn’t work, for the simple reason that fundamental qualities of a blockchain cannot exist in contradiction.
We believe blockchains should be layered, with Layer 1 used for the most high-value purpose: store of value. This model allows other functions that require fast transaction speeds to be offloaded to Layer 2, where they can scale far better than if we were to try to transact at the base Layer 1. Layer 1 blockchains should be designed to complement, not compete with, Layer 2 solutions.
Layer 2 for Scaling
With Layer 1 for store of value, we can reserve Layer 2 for other functions requiring high throughput. Indeed, Layer 2 offers the best scaling options, providing nearly unlimited transactional capabilities, minimal transaction costs, and an improved user experience. This should also lead to a greater array of functionality for dApps because Layer 2 networks can be designed to suit their use.
With Layer 1 focused on being a secure, neutral, decentralized and open public infrastructure, Layer 2 offers endless opportunities for customization.
In the Nervos Network, the Common Knowledge Base (CKB) serves as Layer 1 and is the value preservation layer of the entire network. It is an open, public and proof of work-based blockchain, inspired by the best attributes of Bitcoin. Its design is meant to be maximally secure and censorship-resistant, to serve as a decentralized custodian of value and crypto-assets. Layer 2 protocols are then free to provide unlimited scalability and minimal transaction fees. They also allow for application-specific trade-offs over trust models, privacy and finality.
Proof of Work
Nervos utilizes Proof of Work (PoW) rather than Proof of Stake (PoS). This is because only PoW meets all the demands for a truly decentralized system.
The most important reason for this is that PoS systems permission the consensus process, allowing the largest validators to consolidate their status over time and entrench themselves in positions of power on the network. PoW, on the other hand, offers far more open access to competition, and no one is likely to maintain a leading edge in technology and operation forever.
Blockchains that are highly decentralized through PoW can also ask higher use charges as a “service premium”. When making a transaction, users are not only paying for IT infrastructure, but they also secure the right of private ownership over their own assets. The latter is exactly the “service premium” offered by this type of blockchain.
Ownership Model
Nervos utilizes an innovative ownership model to avoid the problem of “state bloat,” in which the amount of occupied storage capacity grows to the point that it cripples the blockchain.
In existing blockchains, where ownership is based on a one-time fee, this is a real risk. To avoid this, CKB is taking a completely different approach to the design of a blockchain network.
To best allocate resources and avoid the problem of state bloat, state storage must have a clear and fine-grained ownership model. To deliver consistent long-term rewards to miners (regardless of transaction demand), state occupation must have an ongoing cost.
Nervos imposes a cost for storing data on CKB, similar to ‘state rent’ proposed in the Ethereum community, however this cost is imposed seamlessly through CKByte inflation. This properly incentivizes actors on the network, dissuading them from storing data longer than required by use case, and mitigates state bloat.
Next Up
As part of this series, we will be exploring use cases that demonstrate how our principles apply to real-world application. Our next blog will cover one of the most anticipated use cases—decentralized finance.
