The categorization of Ethereum Layer2 has been a topic of debate among researchers and developers. On Monday (6th), two researchers engaged in a debate on the topic of Plasma, which drew a response from Vitalik Buterin, the co-founder of Ethereum.
Summary:
Responding to V God’s call to action! A quick look at the L2 Redstone, a chain game that combines OP+Plasma.
Background:
V God throws a new direction for “ZK+Plasma”, will the dominance of Rollup change the Ethereum L2 landscape?
Table of Contents:
V God emphasizes the core function of Plasma as “asset exit mechanism”
V God proposed “ZK + Plasma” at the end of last year
Foundation researchers advocate for Ethereum orthodoxy
How to distinguish between different types of Layer2
The categorization of Ethereum Layer2 has been a topic of debate among researchers and developers. L2BEAT, a well-known Ethereum Layer2 research institution, and researcher Emmanuel Awosika engaged in a debate and exchange, with Vitalik Buterin joining in and mentioning the long-silent Rollup technology Plasma, which has been rarely mentioned by the Ethereum community since it became mainstream in 2021 after Optimism and zero-knowledge proofs.
On Monday, L2beat researcher donnoh.eth engaged in a debate and exchange with researcher Emmanuel Awosika on the X platform. Unexpectedly, this drew a response from Vitalik Buterin (V God). The content of the tweet is as follows:
donnoh.eth: TLDR, Redstone (Layer2): OP Rollup with off-chain data availability (DA), no DA proof but DA challenge. They claim this mechanism is “Plasma,” which means the network is as secure as Rollup but easily exceeds the budget of honest challengers for the sequencer, making it easy to steal funds.
Emmanuel Awosika: If they claim it is Plasma, does that mean it is not as secure as Rollup? I think someone needs to explain why Plasma chains are different from Rollup with on-chain data availability, clearly indicating that “Plasma is not equal to Rollup.”
donnoh.eth: If a project has off-chain DA and provides guarantees for users to withdraw funds, then the project is Plasma.
Emmanuel Awosika: This sounds like what RedStone is doing, and it is not completely different from other implementations of Plasma, is it? You always need to trust that someone has retained the necessary state data copies to generate withdrawal Merkle proofs (or store the data yourself). I don’t think early Plasma implementations had challenges with data availability, so Redstone’s approach can be seen as a security enhancement in this regard. Specifically, how does Redstone ensure data availability using which chain?
donnoh.eth: As long as there are altruistic challengers willing to burn money, you can withdraw. However, if the sequencer spends more than these challengers, they can submit a promise that cannot actually be obtained.
Emmanuel Awosika: How does your Plasma version prevent data withholding attacks?
donnoh.eth: (Quotes V God’s article)
Emmanuel Awosika: However, the Plasma protocol described here does not prevent data withholding attacks. You can prevent invalid state transitions by generating SNARKs to prove that the computed state root published on the chain is correct, but you cannot accurately prove that the data behind the state transition is available.
Excerpt from the article:
Unless I missed something, you still need to deal with unavailable blocks, right? If the sequencer engages in malicious behavior and there is a network failure of off-chain data availability, Redstone will also encounter the same problem. Therefore, they don’t seem to be completely different. If I’m confused here, please let me know!
Vitalik.eth (V God): The main purpose of Plasma is not to prevent invalid or unavailable state changes, but to allow users with valuable assets to exit using previous (valid and available) states in such situations, and to prevent duplicate exits of any assets through a consistent set of rules. Plasma was first proposed in 2017 by Vitalik Buterin and Joseph Poon in their paper “Plasma: Autonomous Smart Contracts,” announcing it as a new scaling solution for Ethereum. However, the technology did not progress, and Plasma’s complexity in terms of fund withdrawals made it difficult to become a truly viable scaling solution. But the failure of Plasma was not in vain, as the emergence of zk-Rollup showed the potential direction of Layer2 scaling, as it can solve Plasma’s data availability and exit complexity issues.
In November 2023, V God delivered a speech titled “Exit games for EVM validiums: the return of Plasma” at L2DAYS in Istanbul, and later updated a blog with the same title. In the article, V God pointed out that while Rollups are more stable in terms of security, Plasma has an advantage in reducing transaction costs. However, with the emergence of validity proofs (ZK-SNARKs), Plasma can effectively address the biggest challenge for payments – client data storage – and create Plasma-like chains that support EVM, greatly enhancing the potential of Plasma.
Furthermore, V God concluded at the end of the article:
In addition, in August last year, Dankrad Feist, a member of the Ethereum Foundation and the proposer of Danksharding, made a controversial statement about “what is Layer2,” stating that modular blockchains not using Ethereum as the DA layer are not Rollups and therefore not Layer2. He also specifically mentioned that the solution Validium (a ZK Rollup that does not release DA data on Ethereum) is also not Layer2 due to its security flaws.
Interestingly, Dankrad also stated that solutions like Plasma and state channels, which do not require on-chain data availability to ensure security, are still considered Layer2.
However, this differentiation between Plasma and Validium, both of which use off-chain DA layers, as Layer2 and non-Layer2, has raised suspicions of a double standard among netizens. In response, Dankrad stated that the criterion for differentiation lies in whether “users can safely withdraw funds to Layer1.”
Following Dankrad’s viewpoint, two weeks later, the Ethereum Layer 2 research institution L2Beat renamed Optimistic Chain to Optimium (an OP Rollup that does not use Ethereum as the DA layer). At the same time, they added a “Rollups only” button on the L2Beat website homepage, allowing visitors to hide information related to Validium and Optimium and directly distinguish between Rollup and those scalability network solutions that do not use Ethereum’s DA layer.
Further reading:
A Brief Discussion on Celestia’s Business Ideas: Attracting Users to Ethereum Layer2
However, there are dozens of solutions in the market claiming to be “Ethereum Layer2.” How can we classify and summarize these projects? We can make quick judgments based on the proof mechanism and the introduction of the DA layer:
Validium: Validity proof + off-chain DA layer
Plasma: Fraud proof + off-chain DA layer
OP-Rollup: Fraud proof + Ethereum DA layer
ZK-Rollup: Validity proof + Ethereum DA layer
At the same time, L2Beat classifies solutions that use “fraud proof + third-party DA layer” as “Optimium” to distinguish them from projects using third-party DA layers.
On the other hand, V God also pointed out in “Different types of layer 2s” that an important security indicator for Layer2 is whether assets can be securely withdrawn from Layer2. Similarly, Dankrad also considers security as an important criterion for judging Layer2.
However, as V God mentioned earlier, the core function of Plasma is to “ensure that assets can be safely withdrawn,” and Dankrad also explicitly stated that Plasma is considered Layer2. So, if a scalability solution that uses a third-party DA layer also allows users to safely withdraw assets, can it be considered as Dankrad’s definition of “Layer2”? This is worth considering.