Protocols today use intents in ways that are a significant advancement from the HTLCs and Atomic Swaps we discussed in our previous article. This installment of our “History of Intents” series dives into how modern intent systems influence interactions across various blockchains, showcasing technological advancements and the challenges they seek to address.
TL;DR:
- By early 2021, interoperability between programmable blockchains began to gain momentum, with the rapid growth of BSC and Polygon as fast, cheap, scalability solutions for Ethereum Mainnet.
- Around this time, Connext redirected its focus from scalability improvements to Ethereum using state channels to developing an intent-based cross-chain protocol, which functioned almost identically to most of the intent systems we see today.
- The current bridging landscape is dominated by two technologies: M-of-N Bridges, which rely on multisigs or PoS systems to verify transactions between chains, and Intent Bridges, which utilize a series of solvers to quickly execute and settle transactions on the destination chain while waiting for reconciliation/settlement to occur using either an M-of-N bridge or an optimistic bridge.
- Beyond simple asset transfers, intent-based systems now support complex transactions like cross-chain swaps with conditional settlements, showcasing the potential for more intricate blockchain operations.
- Addressing solver liquidity issues and the complexities of rebalancing across chains, newer systems like Connext V2 have introduced mechanisms that abstract these challenges, aiming to enhance solver efficiency and decentralize operations.
In Q1 of 2021, the interoperability of programmable blockchains took off with the rise of Binance Smart Chain (BSC) and Polygon as affordable and fast scalability solutions compared to the increasingly costly Ethereum Mainnet (L1). Users would access BSC via Binance and then try to use bridges to get to Polygon, xDAI, Avalanche, and other chains directly, bypassing L1.
Given this landscape, by Q2 of 2021, Connext redirected its focus from building scalability mechanisms using state channels to creating its first intent-based system, aligning with the broader trend towards more sophisticated blockchain interactions.
Connext’s Evolution to Intent-Based Systems
Connext initially implemented state channels to batch multiple one-to-one user interactions off-chain, with final updates being posted on-chain. This setup served as an early model for the more comprehensive intent-based systems that would follow. By 2021, it became evident that cross-chain transfers dominated Connext’s transaction volume, prompting a strategic refocus towards these systems with the introduction of NXTP (Connext V1). This version adapted Connext’s framework to better support the prevalent user behavior of transferring assets like stablecoins and ETH across chains, primarily for farming yields.
This shift paved the way for Connext V2 (Amarok), which represents the current iteration of the protocol. Amarok has built upon the foundational work of its predecessors to enhance the decentralization and efficiency of managing cross-chain transactions.
Bridging Models and Technologies
The maturation of the blockchain space has seen the rise of two leading bridging technologies: M-of-N bridges and Intent bridges. M-of-N bridges, utilized by protocols like Multichain, Stargate, Axelar, Wormhole, and others, operate based on a multi-signature or Proof of Stake set to verify transactions between chains. Conversely, intent bridges, such as those implemented by Connext, Hop, Across, DLN, Squid, and others, rely on solvers to execute transactions on the target chain, often with reconciliation/settlement occurring optimistically or supported by an underlying M-of-N bridge.
The development of these models indicates a diversification within the blockchain space. They cater to varied user needs and enhance the functionality of cross-chain transactions.
Solver Selection and Intent Pools
As Connext evolved from its initial versions to more advanced intent-based systems, a significant area of development was solver selection — how transactions are matched with solvers and who can fulfill them on the target chain. This process has seen various approaches from different protocols, each with distinct advantages and trade-offs.
A Deeper Exploration into Approaches to Solver Selection
No Selection/Mempool: In the simplest form, transactions are made on-chain in the chain’s mempool, where solvers compete to fill them first.
- Benefits: This model offers high guarantees on latency as solvers vie to submit transactions quickly.
- Drawbacks: This model exposes transactions to potential issues like Miner Extractable Value (MEV) and can lead to wasted resources if solvers do not win the race to fill the transaction.
Request for Quote (RFQ): In this approach, users express their intent off-chain through a user interface (UI), and solvers respond directly with their quotes. Users then select a solver’s quote and proceed with the transaction.
- Benefits: This decentralized model is efficient for pricing, minimizing latency akin to estimating gas costs.
- Drawbacks: RFQ systems rely heavily on the stability of browser environments, which can be unreliable due to browsers lacking robust support for P2P messages. This may lead to users losing connection or failing to receive quotes from solvers. This approach also requires increased 1:1 interactivity between users and solvers, creating a liveness dependency. Similar to atomic swaps, if solvers go offline (or run out of funds/gas) after providing the quote but before completing the transaction, the user becomes stuck until a timeout occurs. Ideally, this process should involve multiple solvers (1:n) where any solver can complete the transaction.
Private Intentpools: This model involves intents being broadcast to a private mempool where an operator runs an auction to select a solver.
- Benefits: This approach decouples the user-solver direct interaction into a user-network and network-solver relationship, enhancing reliability.
- Drawbacks: This method introduces a central operator that could exert undue influence or experience downtime.
Public Intentpools: A more decentralized alternative, public intent pools operate similarly to private ones but with the added benefit of public verifiability.
- Benefits: This model uses a shared mempool that acts as a global auction or ordering mechanism for solvers, improving transparency and trust in the process.
- Drawbacks: This model makes it difficult for competing intent protocols to give up their order flow to a public ecosystem.
Each method reflects different priorities in system design — such as speed, decentralization, and reliability — and showcases the diverse strategies employed by different protocols to optimize the intent fulfillment process in modern blockchain networks.
Generalized Intents
The majority of conversation today centers around using intents for trading, even for cross-domain intents. This makes sense, as trading is currently the largest addressable market in the space. However, intent-based bridging does not need to be limited to transfers of fungible assets.
Expanding the Scope of Intents
In 2021, Connext realized that solvers could provide calldata as part of filling an intent, and their settlement could be conditional upon that calldata being correct. This is just another condition upon which solver settlement is predicated. This approach allows for actions like bridging and depositing from one chain to another in a single transaction.
We found it interesting that adding arbitrary conditionality to solver settlement can lead to more intriguing behavior from solvers. For instance, UniswapX enables users to set a minimum amount to receive on the target chain for cross-chain swaps. In this model, solvers can determine the routing of the swap as long as they meet the minimum-amount-received requirement.
Adopting UniswapX’s strategy makes it feasible to expand intents further by allowing settlement conditions to be programmable. Yet, historically, this proved challenging for cross-domain intents because each chain required independent implementation of settlement conditions, posing a high risk of fund loss if not executed correctly.
Settlement & Rebalancing
Challenges in Solver Liquidity and Rebalancing
One key challenge unique to cross-domain intents is optimizing settlements to provide the best solver experience. Most intent systems (like Connext v0 and v1) typically settle funds back to solvers on the source chain.
Feedback from solvers indicates that this model is suboptimal for several reasons:
- Solver liquidity moves away from the chains with high utilization and towards chains where it is least useful. Solvers need to rebalance their liquidity to continue earning on it.
- Rebalancing is expensive. Solvers must bridge or transfer funds from less liquid chains, influencing their pricing quotes.
- Rebalancing is complex. Rebalancing requires solvers to engage with bridges, centralized exchanges (CEXs), OTC desks, and more, effectively building an in-house aggregator.
- Solvers can offset costs by netting A→B flows against B→A. However, they need large amounts of capital on each chain to do this effectively, giving market makers an outsized advantage in solving.
Connext v2, Across, and Hop have all introduced mechanisms to abstract rebalancing from solvers. While this in-protocol settlement/rebalancing improves solver decentralization and pricing, the question remains: is it enough?
Conclusion
The present state of intents in blockchain technology is a dynamic and evolving field. Intent systems today are far more sophisticated than their predecessors, capable of supporting complex, multi-chain operations and catering to the needs of an increasingly diverse set of applications. As these systems continue to develop, they promise to abstract the complexities of blockchain interactions further, making them more accessible and efficient for users across the globe.
As we continue to push the boundaries of what is possible with blockchain technology, the role of intents is likely to grow even more prominent. The next installment of our series will explore the future of intents, delving into emerging trends, anticipated technological advancements, and their potential impacts on various industries. Stay tuned to explore how the future of blockchain may unfold, driven by the innovative use of intents.
About Connext
Connext is a network for fast, trustless communication between chains and rollups. It is the only interoperability system of its kind that does so cheaply and quickly without introducing any new trust assumptions. Connext is aimed at developers looking to build bridges and other natively crosschain applications.
Website / X / Discord / Blog / YouTube / Documentation / Github / Audit
