Long Fan is an assistant professor at the University of Toronto and one of the founders of the Conflux project. Before founding Conflux, Long Fan has been doing research on distributed system programming language, software engineering and system security.
Long Fan stated that he had been exposed to blockchain when he was a PhD candidate at MIT, and had a certain understanding and interest in blockchain technology. Bitcoin was bought and sold for $ 1 and $ 5.
On December 27, 2019, the "8 Questions" section of Babbitt conducted an exclusive interview with Long Fan. We talked about the current public chain issues: limited performance, high cost, insufficient decentralization, and insufficient security. We talked about the Conflux vision: to make a good public chain that people will use 20 years later. Long Fan also emphasized that "POW, POS" is not a consensus mechanism, and the speed of the blockchain cannot be solved in the same way … Of course, we also do rumors about "Conflux to be a DEX and alliance chain" Verified.
- Million annual salary, the same platform of 3 major exchanges—Blockchain Cloud Recruitment Video Live Broadcast Replay
The following dialogue is based on the original text of the interview, with cuts. Jia Xiaobei: What does Conflux want to do most?
Long Fan: What Conflux wants to do is the next generation of public chain that is easy to use and really solves the bottleneck of using blockchain. For example, bottlenecks such as performance and the security of smart contracts; the other is to build a good ecological environment for everyone to do blockchain innovation.
Blockchain technology, especially public chains, may become one of the underlying information infrastructures in the future. This does not mean that we already have a very mature system that needs to compete with it, or to completely subvert it. The current system is unavailable, because of the constraints of various aspects of performance, the next-generation underlying information infrastructure has not yet been produced.
In theory, our focus on performance is first and foremost. Because performance determines the cost of using public chain technology and blockchain technology. Just as the communication technology of mobile phones has revolutionized and the bandwidth is high, the data traffic fee can be reduced. Only when the use cost of the blockchain is reduced can we truly empower all walks of life. Many people will say that the blockchain has no applications and the applications are very narrow. In my opinion, a very important reason is that the economic account cannot be calculated at all now, and the cost of interacting with the blockchain once is too high.
What I am facing now is to either interact with a cheap blockchain, but that blockchain sacrifice security, and in the end it seems that the interaction is not a blockchain. On the other hand, platforms that have the characteristics of a decentralized, public chain, such as Ethereum, require a cost close to one US dollar for each interaction, which is a very expensive price.
How much value does an interaction bring to offset this cost. So everyone thinks that blockchain is useless and can only be used for hype and can only be used for things like gambling, because this application has a relatively high single-value content and can afford such expensive expenses.
If the value is interacted with a reliable public chain each time, the price can be reduced from one dollar to one penny, and more applications will appear.
On the other hand, I sometimes feel that it is a national mission. Since the public chain may be a key infrastructure in the future, I would like a group of Chinese to participate in the process of making rules and building the ecology.
Many times when you see the meaning of this thing, it is too late to interpret the application. Like no one now argues that mobile phones are useless, everyone thinks mobile phones are useful. We would say that Android, the operating system for mobile phones, was not developed by us. But let's think if it is technically difficult. I know many Chinese scientists who want to write an operating system. In fact, it is not difficult to copy things like Android. The hard part is how do you rebuild the Android ecosystem. No company in the world has the ability to rewrite the millions of apps in the Android or IOS ecosystem. This is an insurmountable thing. The timing of what we are doing now is right, and we all think that the next-generation layer information infrastructure is bound to iterate. Although the current technology is not available and everyone starts from scratch, some people will come out in the future. Jia Xiaobei: If Conflux can come out, why?
Long Fan: First of all, I have a confidence in my technology. If there is a group of people in China who can make breakthroughs in technology and truly occupy this high ground, it is probably us.
In addition, there are many, many temptations in this field, or some things that may make you go wrong. But our entire team has always been in the right direction, without deviation, according to popular words, it is called never forgetting the original intention.
The purpose of our team, especially the core members, to participate in this matter is very pure. We all do n’t take it as an entrepreneurial project or do things that make a lot of money. More of us think that we are a group of scientists who do system and cryptography, and use what we have learned to change this society and this A rare opportunity for the world. Jia Xiaobei: In your information, I see that Conflux has achieved a high throughput of 3000TPS ~ 6000TPS on the test network, and it uses the POW consensus mechanism. How did you do it?
Long Fan: First of all, there is a misunderstanding. POW and POS are not consensus mechanisms. POW and POS are just what we call "anti-witch attack mechanism". The consensus mechanism solves the problem of how people vote and decide the next block. The anti-witch attack mechanism is a matter of determining how much influence each person has.
With the "anti-witch attack mechanism", if I wear 15 or 20 vests, I can't cast more. Without this mechanism, all of us would vote online, just like a quarrel in a forum, and I could wear 10 vests to make a lot of people.
The anti-witch attack mechanism solves this problem, it is not a consensus mechanism. We used to call it the consensus mechanism, including the fact that POW consumes a lot of energy, so it is slow. In fact, these are all misunderstandings.
The core of our algorithm is our original tree graph structure. The best point is that it can make the entire blockchain produce faster, without affecting the security caused by the fork, and finally let the entire system Able to process these blocks and transactions more efficiently and in parallel.
Why is the core of our algorithm fast?
The traditional understanding is that the block chain is limited by the block size and the block production speed. If you want to bring it up, you need to make the block larger or make it faster. But if you use the original common algorithm, when you make the output faster, the network will fork, and the fork will become more and more serious, which will have an impact on security and the entire throughput A waste.
If you walk completely along the original structure, if you go to a place, you will get stuck and it will not work. At this time, you can only sacrifice security.
We have made an innovation in the structure of the blockchain. This innovation is not a slap in the head. We are very serious about consensus algorithms. I, Ms. Yao (Yao Qizhi), and a Ph.D. from Ms. Yao have been proving the correctness of this consensus algorithm. The preliminary proof is about eight or ninety pages in English to prove the security mathematically. Jia Xiaobei: According to the current progress, can Conflux's mainnet be online on time in March 2020?
Long Fan: What I can definitely answer is that, according to the current progress, all technical matters can be Ready.
But in addition to technology, there are also strategic and risk considerations. Therefore, if there is a delay, it will not be due to technical reasons. Jia Xiaobei: What are some plans after going online? Long Fan: Focus on 3 points. First, there are high-level and meaningful demonstration applications, and we hope to bring them to the public chain.
In many scenarios, the alliance chain is used because the performance is limited and the cost is too high. It can only be sacrificed by choosing a less secure architecture. But Conflux can solve this problem at low cost and high performance.
When our public chain comes out, we will solve some existing alliance chains, combine the alliance chain and the public chain, and then do some demonstration projects.
Second, we also worked hard to develop with our partners some blockchain products that ordinary people can use. This product does not necessarily require ordinary people to open a wallet and a key, which is very difficult. Of course, the most important of these is the logical design of the product.
Third, there are good projects on mature public chains, and we are cooperating one after another. Especially for some overseas projects, they see the ecological problems of Ethereum on the one hand, and see that China has different opportunities on the other. For example, we are in close communication with some of the infrastructure on the public chain, such as Chain link. Our attitude is that we are willing to cooperate with all, we hope to transfer as many things that can be borrowed and migrated to our chain. Jia Xiaobei: Vitalik and CZ "argued" on Twitter some time ago, Vitalik said that Binance Exchange is not decentralized, and CZ said that there is a certain centralization of core development teams. What do you think of their views? Long Fan: Everyone expresses their opinions from their own standpoint, so whoever is right and who is wrong is the point from which you look at it, and I can't look at it from a fair point of view, because I also have my stand. From my perspective, maybe what I do is more similar to what Vitalik does, and I may be more receptive to or closer to his point of view.
Jia Xiaobei: I heard that you are also going to be a decentralized exchange?
Long Fan: To be precise, we have been very supportive of people in our ecology to do this. We are willing to strongly support it. Exchange-related operations are not our strengths and we cannot do it. But we do have small partners in our ecology.
Jia Xiaobei: How can we achieve a decentralized exchange?
Long Fan: Actually, we have quite a lot of things to do. The decentralized exchange first involves the issue of cross-chain assets. You always hope that more and different assets can be crossed. Then we will have a cross-chain solution. This is a technical need to do.
At the same time, this cross-chain involves the custody of related assets on the chain. In order to ensure the security of assets, what is the overall security architecture of the custody on the chain? That is something else.
After that, decentralized exchanges still have to match, logical collaboration on and off the chain, and what needs to be well supported by the public chain. How can our interface be prepared?
There are some mature things done on Ethereum. Conflux is basically compatible with Ethereum Serenity, but we still hope to do something that can make the user experience better and reflect the performance advantages of our Conflux.
At these points, we do our best to provide technical assistance. In fact, the technical difficulties of the public chain itself are much larger than these, and for us these issues are still relatively easy to solve. Jia Xiaobei: At present, more than 90% of encrypted digital currencies may still be traded through centralized exchanges. Why are so few people using decentralized exchanges now? In fact, everyone will think that decentralized exchanges are certainly better.
Long Fan: There are two reasons.
The first point is technically, because the decentralized exchange experience cannot be consistent with the centralized exchange, which is much worse. We hope to solve it as much as possible. It is impossible to say that it can be completely consistent, but at least it can do much better than Ethereum.
The second point is probably because of the participants. Many decentralized exchange teams are more technical, or technically oriented. And if it is not for technology-oriented people, most of them may choose to be a centralized exchange. The cycle is shorter and the division is faster. Therefore, the decentralized exchange is also one layer behind in operation.
This matter should be a relatively lagging thing. The significance of decentralized exchanges is valued. It must be waited for more centralized exchanges to burst out. Like the entire world is concerned about the privacy of data, it is also Facebook's election thunderstorm ignited attention. Jia Xiaobei: I also heard that you want to make an alliance chain?
Long Fan: After General Secretary Xi's speech, many partners came to our door. We consider that they are technically connected, and many of the technologies of the public chain are more difficult. In fact, the technology of the alliance chain is very mature. Bricks are fine, and we are technically able to make alliance chains without any problems.
On the other hand, we also see that things are developing in stages. At present, the public chain technology is not fully mature. When there are some cost considerations, if you want to use them first, then you must use the alliance first. You can have a communication with customers, and you can change to better technology in the future, which is also a natural thing. So we are really doing this. For us, technically there is no extra work, so why not do it? Jia Xiaobei: If you want to build an alliance chain, what advantages do you have over other alliance chains such as Libra?
Long Fan: Our advantage is in the future in cross-chain and open with the public chain. Looking into the future, the world will always be more and more closely connected. Now that the alliances are linked together, there will always be greater demand in the future. When we expand further, there will be some advantages. Our own technology is not only that it can only be an alliance chain, we also have the genes of the public chain here.
Jia Xiaobei: Many people think that Libra will not succeed. What do you think?
Long Fan: Don't say too much about everything in this era. According to the words of General Secretary Xi, this is a big change that hasn't happened in a century. Many things will develop quickly and change will be rapid. In particular, the blockchain technology is more related to finance, and the world's financial environment is now in an abnormal state. The US Federal Reserve has been lowering interest rates for more than a decade, and QE has been suspended for two or three years. It seems that it can no longer be closed now, so the entire world's financial system is in a period of change.
It is really difficult to say what kind of changes will happen to the blockchain technology and how it will go. Maybe some changes in the world economy or finance and some new trends will push new technologies such as blockchain to a more important position. This is my opinion.
So it's not just Libra, including blockchains, even public chains, and bitcoin. I'm more willing to look at the problem from a more macro perspective. Libra itself actually represents the use of blockchain technology, bypassing the banking supervision system or foreign exchange system of small countries, directly facing users in this country, and breaking through the existing structure of banks in this country.
Using blockchain technology, a basket of strong currencies can do this, which is just what technology can bring. In such a period of change, it may conclude that it will change repeatedly. Now the US Congress may say that you cannot do this thing, and maybe there is something right now that we will strongly support it. Jia Xiaobei: What is the culture of your company?
Long Fan: Generally speaking, I don't forget my original intention. Everyone in this team is very convinced about the significance of blockchain technology. I hope to promote the blockchain technology so that more people can use it and make a useful public chain. Many people ask me you: What is the goal? My goal is very simple, just to make a public chain for people after 20 years. It does not mean that the market value of the day after tomorrow, or the market value of the day after tomorrow, that is not our goal.