From Arbitrum's downtime, looking at the different paths of Bitcoin L2 and Ethereum L2

Author: Web3CN

Yesterday, the Arbitrum network stopped operating for about 90 minutes from 10:29 to 11:57. Why did Ethereum L2 go down?

The official response from Arbitrum is:

Due to a surge in user traffic brought by the inscription protocol, Arbitrum's sequencer stopped working, ultimately leading to the network outage.

Why can a surge in user traffic cause Arbitrum to go down? The Bitcoin chain has seen hundreds of thousands of inscriptions without going down. This is because Arbitrum's sequencer is centralized, with only one official node running the network. Therefore, when this node (the sequencer) encounters issues, the network will definitely go down.

You can simply understand that the operation of Arbitrum's POS network ledger relies on the official node (the sequencer). But why do users still dare to use it? Because Arbitrum's ledger will be rolled up and packaged to the Ethereum network, allowing Ethereum network nodes to verify the ledger, thus ensuring ledger security. This is the basic idea behind Ethereum L2. Whether it's OP-Rollup or ZK-Rollup, they both package the ledger to the Ethereum mainnet for the mainnet nodes to verify the Layer 2 ledger. The core goal of this approach is to make the Layer 2 ledger trustworthy.

To put it in a less elegant analogy: the son has no money, and his credit is worthless. Therefore, the son takes a check from the father to spend, using the father's credit to guarantee the son. The Ethereum Layer 2 ledger itself has no credit (the sequencer is centralized, where does the credit come from?). The credit of the Layer 2 ledger is dependent on Ethereum Layer 1, which is the current mainstream design of Ethereum L2.

Of course, this design is currently the optimal solution, but it presents several issues:

1. Layer 2 has a single point of risk due to the centralization of the sequencer, making it prone to outages.
2. Layer 2 assets are not censorship-resistant and can be forcibly frozen.

These are problems faced by almost all Ethereum Layer 2 solutions!

Does Ethereum Layer 2 have such problems? Can Bitcoin Layer 2 solve these issues? What are the similarities and differences in the design of Bitcoin Layer 2 and Ethereum Layer 2?

Before discussing this issue, we need to clarify a few questions:

1. What is Layer 2? What is the essence of Layer 2?
2. What are the design principles of Layer 2? What are the similarities and differences in the design of Bitcoin Layer 2 and Ethereum Layer 2?
3. The correct path for Bitcoin Layer 2.

1. What is Layer 2? What is the essence of Layer 2?

The concept of Layer 2 is well-known due to the Ethereum ecosystem, but the concept of Layer 2 is not an original creation of the Ethereum ecosystem; it originated from Bitcoin.

In the code of Bitcoin version 0.1, there is a preserved original version of the code left by Satoshi Nakamoto. This code allows users to update transactions before they are confirmed by miners. If one user's balance increases, another user's balance will decrease accordingly. Once a user completes a transaction, they can transmit only the transaction result to the main chain network and then close their payment channel. The "payment channel" later gave birth to the Lightning Network, which is Bitcoin's earliest Layer 2 and the first viable Layer 2 in the crypto world.

Therefore, when we talk about what Layer 2 is, we cannot solely look to Ethereum Layer 2 as the benchmark, nor can we use Ethereum Layer 2's design as the only standard for measurement (after all, Ethereum Layer 2 has only confirmed the feasibility of the rollup design direction after two years of development). Instead, we should see through the phenomenon to understand the essence, which is necessary to design a practical Layer 2.

Whether it's Bitcoin Layer 2 or Ethereum Layer 2, the background of their emergence is that when the Layer 1 mainnet cannot achieve more complex and higher-performance application scenarios, it is necessary to move Layer 1 assets to Layer 2 for implementation. Ethereum needs Layer 2 to expand its performance, while Bitcoin needs Layer 2 even more. For example, BTC can achieve fast and efficient payment scenarios in the Lightning Network; ETH can cross to Arbitrum for faster speeds, lower gas fees, and more complex smart contract scenarios.

Therefore, whether it's Bitcoin Layer 2 or Ethereum Layer 2, their essence is the same: both aim to move Layer 1 mainnet assets to Layer 2 to achieve more complex and higher-performance application scenarios. Thus, the essence of Layer 2 is a decentralized cross-chain solution + a high-performance and trustless Layer 2 network.

So, whether it's Bitcoin Layer 2 or Ethereum Layer 2, some basic principles must be followed in their design:

1. It must achieve the trustless transfer of Layer 1 assets to Layer 2, which is the most important first step.
2. The ledger of the Layer 2 network must be secure and trustless.

Only by meeting both of these conditions can a Layer 2 be practically usable and completely decentralized.

2. What are the similarities and differences in the design of Bitcoin Layer 2 and Ethereum Layer 2?

Now that we understand the essence of Layer 2 and the basic principles of its design, let's look at the similarities and differences in the actual design of Bitcoin Layer 2 and Ethereum Layer 2.

1. It must achieve the trustless transfer of Layer 1 assets to Layer 2.

In addressing this issue, Ethereum's approach is: the Layer 2 official deploys a smart contract for custodial assets on the Ethereum mainnet first. When users transfer ETH from the Ethereum mainnet to Layer 2, their ETH is locked in that smart contract, and new ETH is generated 1:1 on the Layer 2 network. When users issue a command to transfer back to the mainnet, the Layer 2 ETH is destroyed, triggering the smart contract on Layer 1 to unlock the ETH for the user. This is the cross-chain implementation method between Ethereum Layer 1 and Layer 2. It is achieved through Ethereum's smart contracts and communication between Layer 1 and Layer 2 networks, allowing for trustlessness.

So, how does Bitcoin's Layer 2 achieve trustless BTC cross-chain?

Before the Bitcoin Taproot upgrade in 2021, it was impossible to achieve a completely decentralized BTC cross-chain. However, the Taproot upgrade introduced Schnorr signatures and MAST contracts, making fully decentralized Bitcoin cross-chain a reality.

Schnorr signatures are a signature algorithm more suitable for Bitcoin than elliptic curve signatures (this is not just my opinion; when Satoshi Nakamoto created Bitcoin, he actually intended to use Schnorr signatures, but at that time, Schnorr signatures had not yet been open-sourced. After being open-sourced in 2009 and undergoing 12 years of scrutiny and validation, Schnorr signatures were officially introduced to Bitcoin through the Taproot upgrade by Bitcoin Core in 2021. Ethereum has also wanted to support Schnorr signatures, but due to the complexity of upgrading the signature algorithm involving Ethereum's account system, it has not yet upgraded to Schnorr signatures).

The main feature of Schnorr signatures is aggregation, allowing 1,000 Bitcoin addresses to sign and manage the same asset. This not only achieves signature privacy but also allows the data from 1,000 signatures to be combined into one, completely solving the data accumulation problem caused by multi-signatures. Therefore, Schnorr signatures can break the original limit of 15 signatures in Bitcoin, achieving completely decentralized signature management.

MAST contracts, short for Merkle Abstract Syntax Tree, use Merkle trees to encrypt complex locking scripts. Its leaves are a series of non-overlapping scripts, and when spending, only the relevant scripts and the path from that script to the Merkle tree root need to be disclosed.

Simply put, MAST contracts are equivalent to VM functionality (similar to smart contract functionality) and can execute predetermined operations through instructions. For example, the combination of MAST contracts and Schnorr signatures can trigger MAST contracts to allow 1,000 nodes participating in decentralized asset management to sign, thus intelligently executing Bitcoin's inflow and outflow according to the rules set by the contract, with no human intervention, relying entirely on contract execution, achieving decentralized management of Bitcoin.

The organic combination of Schnorr signatures and MAST contracts can achieve a completely decentralized BTC Layer 2. To facilitate understanding, we can take the BTC Layer 2 project BEVM as an example (BEVM uses Schnorr signatures and MAST contracts to achieve this) to see how completely decentralized BTC Layer 2 is realized.

When a user transfers BTC from the Bitcoin mainnet to BEVM, the user's BTC enters a contract address managed by 1,000 nodes, and simultaneously, new BTC is generated 1:1 on the BEVM, which is the BTC Layer 2 network. When the user issues a command to transfer BTC back to the mainnet, the BEVM network nodes will trigger the MAST contract, and the 1,000 nodes holding the assets will automatically sign according to the established rules, returning the BTC to the user's address. The entire process achieves complete decentralization and trustlessness.

From the above content, it can be seen that by using the combination of MAST contracts and Schnorr signatures brought by Taproot, Bitcoin can achieve completely trustless cross-chain just like Ethereum Layer 2. This is the most important first step in realizing a fully decentralized BTC Layer 2.

2. The ledger of the Layer 2 network must be secure and trustless.

The ledger of Ethereum Layer 2 is managed by the sequencer. When processing transactions, it rolls up the Layer 2 ledger and packages it to the Ethereum mainnet at a certain ratio, generally 10:1, and then it is verified by Ethereum nodes. However, the sequencer of Ethereum Layer 2 (which is generally only one official node) is completely centralized, operated and controlled by the Layer 2 official.

How can such a centralized design gain user trust? Mainly by rolling up the Layer 2 ledger and packaging it to the Ethereum mainnet for miner nodes to verify. If users do not trust that ledger, they can initiate off-chain complaints to verify it. Therefore, Op-Rollup is also known as optimistic proof, based on the optimistic assumption that the official will not act maliciously. If they do, it can be proven through complaints. These combined designs can basically ensure that the Layer 2 ledger is trustworthy.

However, this also leads to a single point risk for the Ethereum Layer 2 sequencer, which also means that assets like ETH on Layer 2 are not censorship-resistant and can be forcibly frozen by external forces. This is because the ETH Layer 2 sequencer is just one official node, which can be centrally controlled. This will also lead to an upper limit on the asset scale of ETH Layer 2, as many large funds will hesitate to enter due to the lack of censorship resistance. Imagine if you have 100,000 ETH, would you dare to transfer these assets to a non-censorship-resistant Ethereum Layer 2? The recent Arbitrum network outage also exposed the issue of single point risk in the sequencer.

At the same time, this gives rise to two user-unfriendly issues:

a. Due to the 7-day complaint mechanism of Op-Rollup, when users transfer ETH from Layer 2 back to the Ethereum mainnet, they must wait at least 7 days for the complaint period to expire.

b. Since the sequencer of ETH Layer 2 is completely controlled by the project’s official single node, the cross-chain and transaction fees of ETH Layer 2 are entirely enjoyed by the project’s official (it is reported that ETH Layer 2 like Base, ZKsync, etc., have sequencer revenues exceeding 5 million USD per month, peaking at over 10 million USD), while Layer 2 users cannot share these network growth dividends.

So, how does Bitcoin Layer 2 ensure ledger trustworthiness?

We still take BEVM as an example. As mentioned earlier, BEVM achieves decentralized cross-chain for Bitcoin through the combination of MAST contracts and Schnorr signatures. To achieve real-time communication between Layer 2 and Layer 1, BEVM's network consists entirely of Bitcoin light nodes, making BEVM a trusted network composed of 1,000 Bitcoin light nodes.

To ensure the absolute security of the Layer 2 ledger and guarantee that network nodes do not act maliciously, BEVM draws on the economic game mechanism of the Bitcoin network. BEVM combines the nodes that hold Bitcoin with the nodes that run the Layer 2 network, meaning that the nodes running the Layer 2 network are also the nodes holding BTC assets through staked assets. Additionally, BEVM has designed a fully automated dynamic staking mechanism based on economics, ensuring that the total value of BTC/mainnet tokens staked by BEVM's Layer 2 nodes is always greater than the value of the assets they hold, using economic game mechanisms to ensure that Layer 2 network nodes have no incentive to act maliciously, thereby ensuring that the Layer 2 ledger is absolutely secure and trustworthy.

In addition, BEVM's design brings two benefits that Ethereum Layer 2 does not possess:

a. The network nodes of BEVM are completely decentralized and not controlled by any project party. Therefore, BTC on BEVM Layer 2 is censorship-resistant and cannot be frozen by any external force, allowing for seamless cross-chain interactions with the Bitcoin mainnet. This can solve the trust issues for large funds.

b. Since the BEVM network is operated by decentralized nodes, the cross-chain and network fees generated are shared with nodes and users, rather than being solely enjoyed by the project party.

3. The correct path for Bitcoin Layer 2

Through the above comparison, we can clearly see the similarities and differences between Bitcoin Layer 2 and Ethereum Layer 2. Due to the inherent differences between Bitcoin and Ethereum, the design of Bitcoin Layer 2 cannot simply copy the Ethereum Layer 2 model. Instead, it should see through the essence of Layer 2 and combine it with Bitcoin's characteristics to find the correct path for Bitcoin Layer 2.

The correct design direction for Bitcoin Layer 2:

1. Bitcoin Layer 1 is inherently not Turing complete. The minimalist UTXO design and limited block space of Bitcoin cannot verify complex data and programs. Therefore, attempting to improve the solution through client verification or within Bitcoin's limited UTXO and block space is not feasible. This direction not only makes the implementation extremely complex but also limits application scenarios, at most supporting asset issuance. Expanding towards higher-performance Layer 2 is not viable. The only correct direction is to move BTC to Layer 2 in a decentralized manner to achieve more complex and higher-performance scenario expansions.

2. It is essential to solve the problem of decentralized cross-chain from Bitcoin to Layer 2. This is the foundation of everything. Traditional Bitcoin cross-chain methods such as hash time locks, pegging, encapsulation, and multi-signatures are difficult to gain user trust. The technical combination brought by the Taproot upgrade in 2021, namely MAST contracts and Schnorr signatures, can solve the decentralized cross-chain problem for Bitcoin and is a direction worth exploring for Bitcoin Layer 2.

3. In ensuring the security and trustworthiness of the Layer 2 ledger, it is absolutely necessary not to copy Ethereum Layer 2's model, attempting to roll up the BTC Layer 2 ledger and package it to the Bitcoin chain for verification. This is not feasible because the Bitcoin blockchain does not support OP or ZKP verification, and miners will not participate in the verification of Layer 2 ledgers. Storing these ledgers on the Bitcoin chain is merely a proof of storage and has no significance. To ensure the safety of the Layer 2 ledger, we can learn from Bitcoin's economic game mechanism and design a dynamic staking mechanism for nodes based on economics and game theory, ensuring that Layer 2 network nodes have no incentive to act maliciously, thus safeguarding the security of the Layer 2 ledger.

Of course, we also hope that in the future, Bitcoin will undergo BIP-level upgrades to enable the Bitcoin network to verify OP or ZKP, allowing Bitcoin miners to perform ZKP calculations. At that time, ZK-rollup can enter the Bitcoin network, and Bitcoin Layer 2 can achieve a more ultimate solution. However, this may take 5-10 years or even longer to realize.

Based on the above analysis, we can see that the most practical and feasible BTC Layer 2 solution currently is based on the combination of MAST contracts and Schnorr signatures brought by the Taproot upgrade, combined with a dynamic staking network of Bitcoin light nodes to achieve real-time communication and network security between Layer 2 and Layer 1, thus realizing a truly decentralized Bitcoin Layer 2. This is precisely the solution that BEVM has already implemented (for specific details, please refer to the BEVM white paper: https://github.com/btclayer2/BEVM-white-paper).

So, does Bitcoin Layer 2 have the opportunity to surpass the scale of Ethereum Layer 2?

The answer is almost certainly yes:

I believe there are at least the following reasons:

1. There is already a fully decentralized BTC Layer 2 solution available.

Before the emergence of a fully decentralized solution, the largest Bitcoin wrapped asset was WBTC issued by the centralized institution Bitgo, currently around 6.5 billion USD in scale. After the emergence of fully decentralized solutions (like BEVM), it is predicted that this market can grow by 5-10 times or more, reaching a scale of 32.5 billion to 65 billion USD, far exceeding the current total TVL of 20 billion USD for ETH Layer 2 (this data includes cross-chain ETH and other assets on ETH Layer 2; the actual cross-chain ETH has not reached 20 billion USD).

2. Bitcoin, due to its inherent non-Turing completeness, needs Layer 2 more than Ethereum to develop its ecosystem. Therefore, in the future, a large amount of BTC will move to Layer 2 to build various decentralized BTC applications. This is determined by market demand.

3. Bitcoin Layer 2 can be more censorship-resistant than Ethereum Layer 2, making it easier to gain the trust and favor of users, especially large funds.

4. The market capitalization of Bitcoin is three times that of Ethereum. Currently, the total TVL of ETH Layer 2 is about 20 billion USD, accounting for about 10% of Ethereum's market cap. By the same ratio, if 10% of BTC enters Bitcoin Layer 2 in the future, the total TVL will reach 85 billion USD, which is three times the scale of Ethereum Layer 2.

Summary:

Layer 2 solutions originated from the Bitcoin ecosystem and were developed within the Ethereum ecosystem.

The current solutions for Ethereum Layer 2 are not perfect, nor are they the ultimate solution for L2, and they cannot serve as the only reference standard for all L2.

Bitcoin Layer 1 is inherently non-Turing complete, and its minimalist UTXO and limited block space cannot handle complex data and computations. Therefore, Bitcoin must develop Layer 2, and it must be a completely decentralized Bitcoin Layer 2.

Before the Bitcoin Taproot upgrade in 2021, Bitcoin could not achieve a fully decentralized Layer 2 solution. However, the combination of MAST contracts and Schnorr signatures brought by the Taproot upgrade has made fully decentralized Bitcoin cross-chain a reality, allowing fully decentralized Bitcoin Layer 2 to become a reality. The BTC Layer 2 project—BEVM has provided its answer.

Bitcoin Layer 2 cannot completely copy the Ethereum Layer 2 solution; it needs to be designed in conjunction with Bitcoin's own characteristics.

Finally, the scale of Bitcoin Layer 2 will inevitably surpass that of Ethereum Layer 2; this is an inevitable trend!