V God “hand tears” the Ethereum star expansion party Matter Labs, can his currency be invested?

V God “hands” the Ethereum star expansion party Matter Labs, can his currency be invested?

In the tear project, V God has always been unambiguous.

Yesterday afternoon, the media reported that in response to Matter Labs’ disclosure of the zkPorter expansion plan, V God said that there was an error, and he would rather put the coin in the Optimistic Rollup instead of the off-chain data availability system (referring to its solution is not working).

However, the media did not clarify the related issues behind V God tearing Matter Labs. For example, what exactly is the zkPorter expansion plan that makes V God so disapproved? How did V God fight back scientifically in the article? Also, what impact does it have on your brothers’ investment?

If you want to understand these issues, let the fat brother set the time indicator back to 2 days ago.

Important breakthrough

On April 13, Matter Labs published the article “zkPorter: A Breakthrough in L2 Expansion” on Medium, and the subtitle also emphasized, “Ending the debate between zk and Optimistic rollups.” From the wording, it feels that the project team is very proud and has made a remarkable progress. The original Chinese translation of Fatty is posted here, and those who are interested can read it again.

We know that in order to solve the Ethereum congestion problem, major project parties are moving to Layer 2 one after another to achieve low-cost and controllable on-chain costs. According to Matter Labs, the optimistic rollups and virtual machine-compatible zkRollups currently on the market have a lot of “weak chickens”, which provide linearly increased throughput and cannot meet the rapidly growing demand for on-chain transactions.

Matter Labs said that its proposed zkPorter can solve this problem very well. It has a TPS of 20,000+, and is safer and cheaper than Optimistic Rollups. It will enter the mainnet through zkSync2.0 in the next 6 months.

The project party said in the article that in zkSync 2.0, the Layer2 status will be divided into two parts: zkRollup with on-chain data availability and zkPorter with off-chain data availability. Then, these two parts will be composable and interoperable, that is, the contract account on zkRollup can seamlessly interact with the account on zkporter. The whole process is roughly as follows:

To put it simply, first, users use their zkPorter account to trade on uniswap, which can be done thousands of times, and then transfer the transaction data to the Ethereum chain through zkRollup at one time, and the transaction can be done once for $0.03. Why is it so cheap?

READ  Nexo Finance Accused of Being Behind Zeus Capital and Chainlink Short - Cointelegraph

Because zkPorter transactions take place off-chain and do not need to be synchronized to Ethereum in real time, the cost can be extremely cheap.

How does this solution ensure the security of zkPorter account data?

The data availability of zkPorter account will be protected by zkSync token holders (ie “Guardians”). These “guardians” will track the status on zkPorter by signing blocks to determine the data availability of the zkPorter account. Guardians participate in PoS by staking zkSync Token, so any failure of data availability will result in their pledge deposit being fined. This provides encryption economic guarantee for data availability on zkPorter.

In short, Matter Labs mainly talks about its own solutions. Seeing this, it is estimated that some older brothers think that the plan is fine, the currency should be more fragrant, and you can do it. Don’t worry, let us see how God V thinks about this matter.

V God: “Wrong!”

After the publication of this article, V God immediately published the analysis content on reddit, denying that zkPorter is more secure than Optimistic rollups. Brother Fatty puts the link here. Brother who is good at English reads the original text directly. Below is the translated content of Brother Fatty.

In my opinion, this is wrong. Assuming that neither of the two solutions have bug vulnerabilities, in fact, the bug risk of Optimistic rollups is smaller, but it is another problem. I prefer to put my assets in Optimistic rollups, rather than in an off-chain availability system.

The opinion in the Matter Labs article is that:

As of today, the cost of attacking Optimistic Rollup through 51% computing power coordination is less than $70 million (when PoS comes, the cost will be even less). On the other hand, to make the data in zkPorter unavailable, the attacker will need to accumulate 1/3 of the value of all pledged tokens (it will almost certainly exceed $70 million).

This figure (US$70 million) actually comes from StarkWare’s “Optimistic Rollup Dilemma” article. The basic argument is that an attacker can rent enough hash power for a price of US$300,000 per hour to perform a 51% network attack. If they continue to attack for a week, the cost is about 50 million US dollars.

READ  ChainLink (LINK) Continues to Shrug off Calls of a Bubble - Ethereum World News

There are two things wrong with this argument:

First, if you rent hash power on the market, you will find it difficult to do so. Either you cannot rent it, or the price rises rapidly, and at the same time remind others to raise the price. In particular, the computational cost mentioned in this article is only equivalent to 3% of Ethereum’s hash power.

Second, if a 51% cyber attack does occur, the community will not sit back and watch. On the contrary, everyone will work hard to coordinate the emergency fork, forcing the attacker to work in vain, and this can happen within a week. When solving the Shanghai DOS attack, it only took 6 days from start to completion. Solving the 51% attack will be easier because it is a soft fork. In the POS environment, the community self-repairs 51% attacks. I have analyzed it before; under the POW environment, it is more difficult for the community to self-repair 51% attacks, because the attacker can repeat it continuously, which is very likely.

In fact, Optimistic rollups (security) are okay. In my (V-God) article, the security level of off-chain data availability delegation is far lower than that of the basic chain, and there is no tight coupling. If an attacker purchases tokens for the available layer of off-chain data, there is no guarantee that the Ethereum community will help restore the fork. In fact, the recent precedent is that when the construction of an application layer is attacked, the Ethereum community is even opposed to intervening in the chain.

On the other hand, the data availability of the sharding layer will be protected by the entire Ethereum network and will be tightly coupled. In my opinion, it is okay for some applications to use off-chain data availability during this period, but the premise is that it needs to be clearly stated that the availability of off-chain data is a temporary measure, and there will be data availability based on sharding later. I don’t think that even this is necessary. The data availability of the existing Ethereum chain can meet the demand. There is already 4000+ TPS space. I can’t imagine that the entire transaction space will exceed this level in the future this year.

READ  Data Analyst: 3 Key Metrics Show the Start of a New Bitcoin Bull Trend - Cointelegraph

The induction demand does exist, but at this moment, two orders of magnitude of the induction demand are waiting, which is very unlikely.

Basically, V-God approved all the advantages of Matter Labs’ two-layer expansion plan zkPorter. There are problems with security, and 20000+TPS is not very useful. It is worth mentioning that, for V God’s rebuttal, we have not yet seen MatterLabs officially publish an official response article. Domestic currency media have published news reports on the breakthrough progress of Matter Labs.

Fat brother’s point of view

We don’t know the technology. It’s not easy to comment on the specifics. However, Brother Fatty believes that V God’s words are generally more credible. After all, as the founder of the first public chain in the blockchain industry, V God’s professionalism, reputation, and character are there.

In addition, Fatty also consulted an Ethereum expert. His views are consistent with V God, and he also believes that the trustworthiness of off-chain data needs to be verified, and MatterLabs’ expansion plan is worthy of further study.

For all of you, you must be cautious when participating in zkSync tokens, and its security is worth considering; in addition, when using DEFI applications based on the zkSync 2.0 solution, you should also pay attention to potential uncertainties.

Author/ Translator: Jamie Kim
Bio: Jamie Kim is a technology journalist. Raised in Hong Kong and always vocal at heart. She aims to share her expertise with the readers at blockreview.net. Kim is a Bitcoin maximalist who believes with unwavering conviction that Bitcoin is the only cryptocurrency – in fact, currency – worth caring about.