Canton Network Blog

The fine print on cross-chain capabilities: separating fact from fiction

Written by Bernhard Elsner | Apr 18, 2024 4:00:00 AM

In the blockchain ecosystem, where innovation and competition are continuously pushing boundaries, it's crucial for market participants to critically evaluate the claims made by platforms regarding their technological capabilities. As a particular example, let’s look at one property of Canton’s that I’m particularly proud of: The ability to make atomic smart contract calls across sovereign subnets.

Reading recent statements from Polygon about their (future) AggLayer today, I could easily conclude that they will offer the same.

Polygon claims to facilitate "near-instant, atomic cross-chain transactions" and asserts that "contracts can safely call contracts on other chains at super low latency." These statements are compelling, painting a picture of a capability similar to what Canton Network offers with atomic smart contract calls between subnets. However, a closer look reveals a crucial distinction: in Polygon's framework, even in the future, atomicity and smart contract calls across chains are not simultaneous capabilities.

Dissecting the Claims

Polygon's description initially suggests a seamless integration akin to Canton's, but delving into the details provided in their fine print—which likely few will read—reveals the nuances.

Atomic Cross-Chain Transaction
Firstly, they describe the process of submitting a “bundle” or set of transactions across multiple chains. The atomic bundle is designed so that transactions are executed on each relevant chain if and only if all transactions are successful. If one transaction fails, the bundle is not included on any chain.

For example, if AcmeBank were to issue a Acme Coin on both Chain A and Chain B, it could theoretically operate a bridge by atomically burning in Transaction A and minting in Transaction B. However, a different scenario, such as DEX on Chain A wanting to move AcmeCoin on Chain B via a smart contract call as part of a trade settlement, would not be feasible under this setup.

Asynchronous Cross-Chain Calls
Secondly, they tout super low-latency cross-chain composability through asynchronous cross-chain calls. While this is a powerful feature, it prioritizes liveness over safety from the perspective of a single chain. This means that independent of whether the transaction on Chain A goes through, the cross-contract call to Chain B can fail.

In essence, if the DEX on Chain A in the above example were to make cross-chain calls to transfer Acme Coin on Chain B and some other asset on Chain C, the transfers could succeed or fail independent of each other. There is no guarantee of atomic settlement across chains like exists for constructing "transaction bundles" off-ledger.

Conclusion

While Polygon continues to innovate and expand its capabilities, the claim of “feeling like a single Blockchain” would in my opinion require atomic smart contract calls between subnets. This capability is implied by a combination of claims that are made close to each other, but it’s not actually claimed, which makes me believe that they are not and will not be able to provide this essential capability. They are able to tick every box individually, but crucially, not all at once, which is a limitation for those in capital markets seeking robust transaction solutions across sovereign applications.

As discussions around blockchain capabilities continue, it is essential to engage deeply and critically, understanding the specific functionalities and limitations of each platform. For those in doubt, we encourage a thorough discussion to unpack these complex technical details further.