————————————————————
Chain abstraction and intent-centered solutions aim to solve the same core problem: how to achieve automated and asynchronous interoperability between different blockchain networks. In simple terms, it is about enabling different blockchains to communicate and collaborate automatically and asynchronously.
📍 They both introduce the concepts of 'counterparties' and 'cross-chain proofs,' but the implementations are vastly different. This article will detail the comparison of their characteristics:
1) Characteristics of Chain Abstraction 🔻
- Chain-Centric Worldview: This can be understood as having a dedicated blockchain (CA chain) acting as a 'mediator' or 'agent' between the user and various other blockchains.
- User Interaction: Users only need to interact with the CA chain, just as they would operate on a single platform.
- Responsibility Allocation: The CA chain and its associated off-chain components are responsible for helping users achieve their expected results on the target chain, such as transferring coins, executing smart contracts, etc.
- Proof Flow: Cross-chain proofs are always sent from the CA chain to the target chain. The target chain will verify these proofs and only then execute the corresponding operations, such as minting new tokens or using existing tokens.
- User Abstraction: Users only need to submit requests for operations they want to perform on the target chain, such as 'I want to transfer X coins from chain A to chain B.' The CA chain will handle all the remaining details.
- Extensibility: To support various target chains, the CA typically needs a universal proof system that can be verified across all target chains, such as multiparty computation (MPC) schemes.
- Latency: The latency of operations depends on the confirmation rules of the universal proof system, and the latency for each request is fixed, which is not flexible enough.
2) Intent-Centered Solutions 🔻
Imagine you want to operate on different blockchains without dealing with the complexities and differences of each chain. Intent-centered solutions are designed to solve this problem. They do not require users to understand the specific details of each blockchain like traditional methods; instead, they allow users to focus on the goals they want to achieve.
- State-Centric Worldview: This can be understood as an open network made up of many 'solvers.' These solvers act as agents for users on different blockchains, helping them achieve their goals.
- User Interaction: Users only need to interact with one solver chosen by the system. The system selects based on which solver can provide the best results for the user, similar to choosing the most suitable agent.
- Obligation: The chosen solver has the obligation to deliver the expected results to the user. This can be guaranteed through a reputation or commitment system, much like how agents need to be accountable to their clients.
- Proof Process Reversal: Traditional cross-chain proofs flow from the user's starting chain to the target chain. Here, the proof process flows back from the target chain to the user's starting chain. This is enforced by performing proof checks on the starting chain; only if the proof is valid can resources on the starting chain be used. Simply put, it’s like an agent needing to provide proof of task completion to the user.
- User Abstraction: Users only need to focus on the proof obligations of account states on the target chain. In other words, users do not need to understand the complex details of cross-chain operations; they only need to know if the results on the target chain meet their expectations.
- Scalability: To support various target chains, the system needs a programmable proof-checking system that can handle different proof methods. This is likely a 'resource locking' system that ensures only valid proofs can unlock resources.
- Latency: Latency is determined by the confirmation rules perceived by the solver, meaning that latency optimization can be a consideration when choosing a solver. Just like when selecting an agent, their efficiency can also be considered.
👇🏻 Perspective Extension:
Intent-centered approaches provide a better architectural direction. They offer users results-based guarantees by optimizing result states and latencies through solver selection, and are more scalable for customized target chains.
However, achieving this requires reversing the proof process and placing the obligation of proof on the solver network. In other words, while the complexities and heterogeneities of the target chain do not disappear, the burden of integration shifts from the on-chain computation of combinations to the off-chain distributed network of solvers handling proof combinations. This means users no longer need to handle complex cross-chain operations themselves but can delegate these tasks to a specialized solver network.
3) Intent Abstraction 🔻
If chain abstraction serves as the heterogeneous domain execution abstraction for users, then intent abstraction is the heterogeneous proof obligation abstraction for solvers.
Just as developers need to write, orchestrate, and guide computational flows for users in the context of cross-domain execution to achieve chain abstraction, they also need to write, orchestrate, and guide proof flows for solvers in the context of intent abstraction.
The concept of 'chain abstraction' is still evolving, with various approaches covering aspects from 'chain-centric' to 'state-centric.' For simplicity and ease of comparison, I define 'chain abstraction' here as 'chain-centric,' which is more consistent with the architecture of some original chain abstraction supporters.
However, in practice, many newer architectures incorporate elements of both 'pure chain-centric' and 'pure state-centric' models, as seen in frameworks like CAKE.