Eli Ben-Sasson, co-founder and president of StarkWare, talked about the project team, expansion features, and Layer 2 development trends.
Original title: “Hanging Optimistic Rollups? StarkWare’s L2 track is revealed | Unitimes AMAes AMA》
Interviewee: Eli Ben-Sasson, co-founder and president of StarkWare Writer: Unitimes
Today, we invited Eli Ben-Sasson, the co-founder and chairman of StarkWare, a blockchain privacy solution provider.
StarkWare, a blockchain privacy solution provider, is headquartered in Netanya, Netanya, Israel. Two of the company’s co-founders Eli Ben-Sasson and Alessandro Chiesa are also the founders of ZCash.
Its main goal is to further promote the ZK STARK breakthrough blockchain privacy solution developed by the Technion-Israel Institute of Technology. It continues the zero-knowledge proof protocol to protect the privacy of information on the blockchain. On the one hand, it can support the compression of massive data into smaller samples. On the other hand, it is more efficient, transparent and safe than quantum computing. One big advantage is that while proving private information, it also ensures the integrity of the calculation without consuming a lot of computing power.
This time, Eli Ben-Sasson will discuss with us the coming of Starkware, and the secrets of the major players in Layer 2!
Eli Ben-Sasson, co-founder and president of StarkWare
Fixed Q&A
Welcome to the Unitimes online AMA Q&A event. First of all, please introduce yourself and your original intention and development history of StarkWare.
Eli Ben-Sasson: Hello everyone! I am Eli Ben-Sasson, the co-founder and president of StarkWare. Before co-founding StarkWare, I was a full professor in the Department of Computer Science at Israel Institute of Technology.
I started researching encrypted proof systems in 2001, when I was doing post-doctoral research at MIT and Harvard University, and found a way to make this proof system more efficient. Since 2008, I have funded and led research on such systems. In 2013, I realized that they can help blockchain in terms of privacy protection and scalability. So in 2015, I co-founded Zcash (a cryptocurrency company), and after further improving the technology, in 2018, we obviously can use it to solve the scalability of the blockchain, which prompted the establishment of StarkWare.
StarkWare has 4 partners, the other three are our CEO Uri Kolodny, he is a senior entrepreneur (we are close friends for 30 years); our chief architect Dr. Michael Riabzev, he is The doctoral student I brought is also the co-inventor of the ZK STARK agreement; another is our chief scientist Alessandro Chiesa, who is also a professor at the University of Berkeley. He co-founded Zcash with me and is our long-term partner.
Today, StarkWare has 45 employees, all in Israel. We have raised more than $100 million in total from institutions such as Paradigm, 3Arrows, Alameda Research, and individual investors such as Vitalik Buterin and Naval Ravikant. Most of our team is engineers, and we are using ZK STARK technology to build scalability and privacy solutions.
In June last year, the StarkWare team launched the L2 expansion engine StarkEx for Ethereum applications, and has been adopted by several well-known dApps, including the decentralized exchange DeversiFi, the margin trading platform dYdX, and the NFT casting and trading platform Immutable X, etc. . Can you briefly introduce how StarkEx works and the advantages it brings?
Eli Ben-Sasson: Currently, the transaction method on Ethereum (such as transactions on Uniswap) is that nodes in the network need to process every transaction sent to them, and due to the Gas limit of the Ethereum blockchain, users You must pay a very high price to make your transaction packaged into the block.
With the help of StarkEx, Ethereum nodes only need to check a ZK STARK certificate, which shows that thousands of transactions have been executed correctly. The size of the certificate and the time required to check it are much less than the time required to (go directly) check all the transactions contained in the certificate. In this way, on dYdX, the cost of a transaction is less than 1500 Gas, instead of 200,000 Gas as in those mainstream DeFi applications!
It is understood that StarkEx can be deployed in two modes: ZK Rollup mode (publishing data on the chain) and Validium mode (putting data off the chain). For example, DeversiFi and Immutable X both use Validium mode, while dYdX uses ZK Rollup mode. Can you tell us more about the differences, advantages and disadvantages of these two models? When Ethereum applications adopt StarkEx, how should they choose between these two modes?
Eli Ben-Sasson: Yes. As you pointed out, with Validium, data is placed off-chain and saved by the data availability provider in the network; while the Rollup mode puts data on-chain. This means that the cost of each transaction under the Validium model is lower (because there is less data to be published to the Ethereum network), and it also has higher privacy (because no one except the exchange operator can Check these personal transaction orders that brought status changes).
In terms of security, Validium users need to trust the data availability committee, that is, they need to trust at least one committee member to be honest. In the Rollup model, such a trust assumption is not needed, but the downside is that its transaction costs are higher (because the data appears on the chain), and for the same reason, financial privacy is also less.
In view of the two modes of StarkEx, it seems to be similar to the L2 solution zkSync (data availability on chain) and their upcoming zkPorter (off chain data availability) also launched by Matter Labs based on ZK Rollup. What do you think about this?
Eli Ben-Sasson: Indeed, zkSync is using many of our innovations-such as zkPorter, which is actually Validium, and other things. Compared with Matter Labs, our main advantages are:
We invented and built our own ZK STARK system, and their technology stack was built by others (PLONK built by Aztec). This also means that we have a stronger grasp of technology and the ability to improve technology.
We already have multiple systems running in a production environment. These systems use a Turing complete programming language called Cairo, which is readily available. Matter Labs has only a simple payment system in the production environment, and no Turing complete language is available.
Finally, our core technology-ZK STARK is faster, more secure (in the sense of cryptography), transparent (no trusted settings required) and post-quantum security, while the core technology used by Matter Labs (built by another team) ) Is slow, requires trusted settings, and can be broken by quantum computers.
The StarkWare team has taken a steady and steady route in laying out the Layer 2 track, and the ultimate goal is to launch your general L2 network StarkNet on the mainnet. According to StareWare product manager Tom Brand recently, StarkNet is expected to go live on the testnet in the next few weeks and the mainnet later this year. It can be said that the L2 scalability engine StarkEx that you have launched, the Turing complete programming language Cairo, and the STARK Provers for generating proofs are all used as a prelude to the launch of StarkNet. Can you briefly introduce Cairo and STARK Provers? And how do they promote the launch of the StarkNet network?
Eli Ben-Sasson: Yes! StarkNet is a very exciting development. Think of Ethereum, it allows you to write and deploy any smart contract, and then others can send any transaction to it. This is really amazing! But due to the amazing success of Ethereum, the network is now very congested, and the cost of executing transactions on it is high.
The STARK proof system can exponentially shorten the time required for the correct execution of the verification program, so it can theoretically be used to exponentially expand the transaction rate on Ethereum (and other blockchains), or in other words, exponentially Reduce the gas cost of each transaction.
So far, the main question is how to use this amazing STARK technology. It is very difficult to write programs in a way that enables STARK to prove effective; it is also very difficult to build and run STARK provers. But all of this will change in the next one or two weeks: StarkNet is permissionless, so anyone can write any smart contract and submit it to the StarkNet chain, just like Ethereum!
At the same time, anyone can submit transactions to StarkNet, and these transactions will be executed. Due to the scalability brought by STARK, the gas cost of transactions through StarkNet is exponentially lower than that on Ethereum.
Now, StarkNet smart contracts are written in a new programming language called Cairo (just like smart contracts on Ethereum are written in a special language, Solidity). The reason for this is that we need a language to optimize the use and scalability of STARK proofs, and this is what Cairo can achieve.
By the way, you can download Cairo through this website
You can also follow StarkNet through this website
It is expected that more L2 solutions will be launched this year and next year, including L2 networks based on Optimistic Rollup such as Optimism Ethereum and Arbitrum, and L2 networks based on ZK Rollup such as StarkNet and zkSync 2.0. Of course, there are also Polygon, xDai, etc. that have been launched on the main network. Side chain scheme. The current mainstream dApps seem to have completed the L2 team. What do you think of the competitive landscape facing StarkNet? And how will StarkNet attract more dApps and user adoption? What is the future plan of StarkNet?
Eli Ben-Sasson: I think there is a lot of room for different L2 solutions, because there is a great demand for scalability, and we can imagine several different L2 solutions using similar technologies, and between them Connect through “bridges”.
So I look forward to all these different (L2) teams deploying! I think StarkNet has the most scalable and safest technology in all L2, and its technology also has instant blockchain finality, which means that its capital efficiency is optimal (definitely better than Optimistic Rollups) many). Therefore, in any L2 configuration, StarkNet will occupy a prominent position.
Speaking of different L2 solutions, what do you think will be the future development pattern of L2 in the long run? Is it a single big family, or a hundred families?
Eli Ben-Sasson: I think we will see a lot of different L2, I don’t think there will be a certain L2 monopoly. In addition, a decentralized L2 is not a real monopoly, just like Bitcoin and Ethereum are not a monopoly, because they are decentralized. We will also see bridges between different L2s. In fact, we have been discussing with various other L2 teams how to build such bridges!
Since the birth of Bitcoin for more than 10 years, we have witnessed many new encryption technologies and applications. In your opinion, where will the next battlefield after L2 be? Can you make us a wave of predictions?
Eli Ben-Sasson: I think we will start to see the adoption of NFTs, gaming applications, and virtual reality (and the social networking and interaction that it brings). Then eventually, traditional enterprises and financial services will enter this field, but they will be the last (not the first) to adopt.
In addition, we will see the public demand that the transparency and integrity standards of the blockchain and ZK STARK certification systems be applied to other areas, and that traditional companies will apply similar standards even in a licensed environment. We have already begun to see this in the field of CBDCs (central bank digital currencies).
Since last year, DeFi has attracted a lot of traditional capital. Some people think that DeFi is the killer application on Ethereum that we have been looking for. Some people think that the threshold of DeFi is still too high, and its user experience is very unfriendly, the logic behind it is very complicated, and the risks are also high. . What do you think of these views? What do you think will be the breakthrough point for the large-scale application of blockchain in the future?
Eli Ben-Sasson: I think the killer application of Ethereum will actually come from the virtual world of games, entertainment and virtual reality. The reason is that in these worlds, people have given a lot of value to digital and scarce things, so putting these things on the blockchain is a natural evolution.
So, although DeFi is really cool and great, I think games, NFTs, social networks, and virtual reality will become killer apps. The reason these applications have not taken off is due to the limited scalability of Ethereum. But StarkNet will solve this problem, so we will see this aspect really take off on StarkNet.
Free Q&A
Hello, I would like to ask if Starkware will have its own native token? Can you introduce its token economics?
Eli Ben-Sasson: About the Token of StarkNet-I will not mention it for now. What I want to say is that we have thought a lot about the cryptoeconomic incentives of various elements of StarkNet. The best way to ensure that you can participate is to start building on StarkNet and learn the language needed to write smart contracts-Cairo.
If Starkware launches its own independent mainnet, does that mean Starkware will be separated from Ethereum instead of Ethereum’s L2, and will also require dapps on Ethereum to migrate to StarkNet in the future?
Eli Ben-Sasson: No, StarkNet will be an L2 connected to Ethereum. Its security depends on Ethereum. You can seamlessly migrate things into or out of StarkNet, so StarkNet is not separate from Ethereum.
Hello, I see that some blockchains want to use TEE SGX as a solution for enterprise trusted computing. Do you think it is a reliable solution in a decentralized scenario? Or is ZK STARK better?
Eli Ben-Sasson: No, I think TEEs (Trusted Execution Environment) are a very bad and insecure solution. the reason is simple. Imagine that there are 100 billion US dollars on the TEE network, which means that the security type relies on the assumption that no one can extract the key from the SGX chip, and the key is actually on the SGX chip. To me, this is an absurd assumption. In fact, before SGX, there were TPMs in the world, and those TPMs were compromised, and their keys were extracted at a price of $5,000 or less. SGX will follow the same route. ZK STARK cannot be destroyed, that’s it. This is mathematically proven (remember, ZK STARK has no trusted settings and no private key!)
Why does Starkware choose ZK Rollup over other L2 technologies?
Eli Ben-Sasson: Our core technology is ZK STARK, which expands the blockchain by performing calculations off-chain and proving its correctness. Therefore, our technology requires the implementation of L2 solutions like Rollup and/or Validium, where data is maintained off-chain. We will make StarkNet provide these two options, namely, on-chain data (Rollup) and off-chain data (Validium).
At what stage is the current market, and has the bull market ended? What impact will the upgrade of Ethereum to 2.0 have on the future of L2 projects?
Eli Ben-Sasson: I don’t know if the bull market is over. But I’m sure that as a technology (regardless of the current price), encryption technology is still in its early stages and there is still much room for growth. As for Eth 2.0, I think we need it, and we also need an expansion solution like StarkNet.
