After Scaling, Why Do Public Chain Still Fail to Land? How Would QuarkChain Change the Current Landscape When It Realizes 1 Million TPS?

20 min readDec 5, 2019


Huoxing Finance APP reports the 55th chat with editor-in-chief with the following theme: after expansion, why do public chain still fail to land? How would QuarkChain change the current landscape when it realizes 1 million TPS? The chat was between the editor-in-chief Meng Xiao she and CEO of QuarkChain, Dr. Qi Zhou.

QuarkChain was founded in 2017 with the headquarter in the Silicon Valley in the US. Its core team are seasoned veterans in developing large-scale distributed systems as infrastructure engineers from Google and Facebook. They have deep understanding in various solutions for expanding capacity within centralized system. CEO of QuarkChain, Dr. Qi Zhou, has been working for a long time in the Silicon Valley as a software engineer at Google and Facebook before with a focus in high-performance systems.

The vision of QuarkChain, is to unite all existing blockchain to form a big alliance to realize the effect of “reusing computing power while mining with consensus”, during which each participant enjoys better privacy protection and more benefits of decentralization. Recently, the QuarkChain team announced the launch and open source of the Go version of the QuarkChain mainnet.

This dialogue revolves around three topics: the prospects of public chain industry, technical discussion about public chain, and the development of ecosystem of QuarkChain. We will cover topics that touch on the pain points of the blockchain industry, technical characteristics of QuarkChain and its direction of future development. It will be a dialogue with plenty of information valuable for the industry.

Below is the compiled notes of the dialogue:

About prospects of public chain industry

Editor: QuarkChain launched the Go version of the mainnet recently. From my basic understanding of programming, Go runs much faster than Python, which would intuitively lead to the speeding up of QuarkChain mainnet. Why didn’t QuarkChain choose to develop in Go from the beginning? Besides, after the launch of the Go version, how far is QuarkChain from reaching the goal of 1 million TPS?

Zhou: This involves a question regarding programming complexity. From early on, we already had a rough plan for developing what languages. In the early stage, with intensified competition in the industry, we decided to develop in Python in order to launch the mainnet as early as possible. For that reason, we did not develop Go at that time, which would require more time for development. We regard the Go version as a version that we can gradually refine. Our reasoning matches with the general logic of software development in the industry.

The advantage of Python is that the speed of development is quick with the tradeoff of lower efficiency in execution. What is impressive though is that, even with a lower efficiency in the execution, our community was able to attain a 55000+ TPS result during our first TPS competition last year.

As for Go, it is a mainstream language for blockchain applications using distributed systems with much higher execution efficiency than that of Python. Our internal testing speed is 5 to 10 times faster than that of Python. This is achieved with no little optimization and we think there is still a lot of room for optimization.

In the last TPS competition, hackers from all over the world provided us lots of inspiration. After the launch of Go, we plan to host a second TPS competition and collaborate more with the global development community. We would like to invite especially hackers from the Chinese developer community. Through the competition, we would like to work with the community to test out the maximum speed of the current Go version and help others to know our functionalities better. At the end of day, we can only find out whether the mainnet is fast or not through testing.

Editor: Since 2017, almost all projects have been emphasizing expansion, where the sharding technology from QuarkChain is also one method to tackle the expansion issue. But I remember that before you said that the biggest pain point of the public chain industry does not lay in TPS. Does that mean you think that QuarkChain’s direction is not a correct one? Also, what is the pain point of the entire blockchain in your opinion?

Zhou: As the basis of system function, TPS is definitely one of the pain points. However, with so many projects focusing on tackling this problem, it is getting solved gradually. Currently, whether the claim is true or not, many projects say it has thousands and millions of TPS.

However, in our opinion, we think that TPS is merely a representation. As a branch of the computing industry, the processing ability of the blockchain system is subjected to Moore’s Law. So in the long run, TPS will not be a problem for sure. For example, during Double 11 (an event that is similar to Black Friday in the US), in 2014 and 2015 the processing power was insufficient to process huge amount of transactions and led to halts in the middle of the celebration but now the transaction process is more or less stable. This example is a proof of how TPS can be overcome gradually.

That goes back to our view that TPS is not the worst pain point. Even though Bitcoin and Ethereum can be congested sometimes, overall 95% of the time we can still use the two networks easily.

On a deeper level, behind the TPS issue, there exists a much more serious problem, which is the fragmentation of the blockchain industry. The sprout of new projects and forking of old projects creates an internal division within the industry. In the long run, this is the biggest pain point which will also bring about the problem of vicious cycle.

Since the bull market at the end of 2017, a lot of projects showed up with injection of manpower, capital and attention. What followed soon after is a stagnant period of nearly two years. It is stagnant on all fronts, be it manpower, capital, or attention. With the inflow of resources slowing down or stopping, there is a lack of new resources, which means that whenever there is a new chain launching, it is consuming part of the existing resources to sustain the project and is diluting the resources shared by the other projects. This kind of dilution of resources will slow down the development of the industry and prevent the entrance of more sources to come in and thus leads to a vicious cycle.

Taking the mining community for example, ASIC and GPU communities are rather independent. Right now the PoW and PoS community are further dividing up the community. Within the ASIC community, for example the Bitcoin mining community, we also see further division due to the presence of BCHABC and BSV. This kind of division reduces the scale of the community and the project: in the past, the community could carry out a large scale project, nowadays it lacks the efforts to do that. On the other hand, each project devotes so much efforts to reinvent the wheels, which does little to invent something groundbreaking for the industry. The dilution of resources also signifies the dilution of computing power, which exposes some projects to the risk of 51% attack. Even though that kind of attack happens very infrequently, but once it does happen, all the money of the community will be stolen. The gravity of such situation cannot be underestimated.

Editor: You just mentioned the division and dilution of resources within the community is a pain point in the industry. Let’s take a look at the facts on the table: since the hard fork of BCH last November, we have not seen any division or dilution happening to any famous public chains. What do you think of that?

Zhou: First of all, a lot of famous projects have noticed this problem. Current situation is that, division of these famous projects results in a situation of “1÷2<0.5”, meaning that division would hurt both parties after division. So right now, projects are more willing to resolve problems within the existing framework. Obviously this only delays the problem from happening but not fundamentally resolving the problem of forking. Once the industry is getting popular again, forking will continue.

Secondly, division these days presents through the launch of new projects and new projects would dilute the resources in the community. Some projects have very specific objective to earn some big bucks and then leave. Of course this is a free market which has little systems involved and every one can be a part of it until the entire industry stops making any money. Other than that, these new projects has no advantages over mainstream projects in terms of community or computing resources and are prone to attacks, which are some of the inherent risks that we see.

Thirdly, the division and dilution of community still exists, however, the forked chain has little threat to the original chain. For example, Ycash is forked from Zcash and ATH from ETH.

Editor: Based on my observation of the public chain, firstly, from an application standpoint, actually traditional internet already matured in application development and has modest demand for public chain. Additionally, we recently see that EOS, as a representative public chain, has encountered major problems in developing application, which is alarming for the whole industry. The commonly held idea for public chain to first launch mainnet and then develop numerous dApp does not seem viable anymore. Do you think the ecosystem of public chain should continue to follow this idea or to venture on a new path?

Zhou: From the performance of current market, the model of launching mainnet with dApps does not seem to work anymore. However, we think dApp is still quite important because application represents the future. Looking at Internet in the era of web1.0, we just put text on the web, and then later on the Internet bubble happened. At that point, many people said the internet was of little use, but with the development of technology, they all changed their viewpoints later on. But before this model can realize in the blockchain industry, we would need to prepare some things beforehand.

Based on the current situation, the user experience for dApp is terrible. For example, for EOS, the experience of opening an account is much worse compared to that for the Internet. Even for ETH, many dApps need to explain to users what gas is, why one would need to pay gas fee, how to buy ETH and so on, all of which incurs a lot of cost to educate users. In this regard, it is not as comparable to the Internet. But for blockchain and for dApp, what the two can accomplish in low cost is the transfer of assets and open credibility, which Internet cannot offer. Therefore dApp is not directly competing against applications of the Internet, which would not be economical; rather, dApp should do things that Internet is short of achieving.

However, if dApp wants to tackle the problem by launching a chain, it would then again divide up the community. Therefore, we are of the view that in order to enhance the usability of dApp, one would need to do something that the Internet did 10 to 20 years ago, which is to allow applications to run on a high-performance, flexible, and interconnected platform, which can bring about the network effect. A public chain should have high interoperability, for example, to support most functions of existing public chains like privacy, contracts, UTXO, PoS and so on. On the other hand, the public chain should give dApp more downstream control, for instance, dApp can use native tokens to pay for transaction feeds. This design can let all stakeholders, including public chain developers, mining communities, staking communities, dApp developers, private token seekers, normal users, can all enjoy the benefits of using one network ecosystem. This is actually something we have been working on lately as well.

Overall, QuarkChain depends on two-layered structure of root chain and sharding chains to allow customization for each shard regarding consensus mechanism, ledger model, virtual machines, and token economy. This design allows one-button press to issue multi-shard chain, to support dApp to calculate gas fee using mainnet token, as well as to form a major alliance between all the smaller ecosystem using cross-chain format. We can say that this mode of operation would create a very different dynamic in the overall ecosystem compared to what we currently see in the industry.

Editor: I have interacted with many public chains this year. Some public chains described to me some questionable business models while some said outright that they do not have any business models. For the former, the key is to profit from selling tokens, yet this is not a valid business logic. A public chain without business model has a logic that is similar to that of the Android system that hopes to reap benefits from other areas. But as an infrastructure system, the data structure of the Android system is very different from that of blockchain, therefore the business model of the Android system may or may not be reproducible. If so, what do you think should be a viable business model? How does QuarkChain think about and design its own business model?

Zhou: First of all, I think that business model is a systemic question which involves many moving parts such as relationship with applications (application scenario), relationship with users (community and investor), relationship with developers, and the overall ecosystem.

The most central and most direct relationship is the one with token economy of the public chain itself, including functionality of the tokens (for paying transaction feeds and/or for staking), rate of production, how the tokens are circulated in the project and so on. All these details should be designed with care.

Regarding the business model of public chain, it is still at a nascent stage, which ties back to the scale, application scenario, and technology development. For the current dynamics, the business model is still too simple, which can be attributed to the lack of application scenarios.

To create appropriate scenarios then to widely circulate these scenarios, and finally circulate to the aspects of the scenarios that maximize the benefits of public chain is the model we perceive as ideal.

Following this train of thoughts, we have proposed two mechanisms in token economy:

  • Production Mechanism: this relates to our PoSW mechanism and root chain to shards structure. PoSW combines both PoS and PoW to determine the production of tokens and forms the basis of token value.
  • Circulation: the core of our circulation mechanism lays within our flexible structure and multi-native token functionalities. On the basis of app-friendly, high flexibility, and high compatibility, we let all sorts of applications to land on QuarkChain and guarantee interconnectivity. Different kinds of applications or different shards can issue native tokens by burning QKC. The functionalities of these native tokens are identical to that of QKC. With good applications, its tokens would have a higher value, thus guaranteeing the developers the benefits from designing good products.

Our core technology, the Boson consensus, is the key to connect the two cycles. On one hand, the newly developed shards would face the danger of being attacked by dominant computing power due to lower computing power. On the other hand, miners that are not on the shard that sits excellent application cannot participate in the distribution of value from that shard and thus dampens incentives. Using the Boson consensus solves these issues through economic incentives.

The root chain-to-sharding chains structure of QuarkChain relies on the root chain to verify transactions and provide protection for computing power. As for the sharding chains, it distributes tokens from the local chain proportionally to miners of root chain through a special form of taxation. This arrangement resolves the question of computing power and profit, which is a kind of whether-egg-or-chicken-came-first question.

In the future, QuarkChain can leverage its flexibility and easily add new sharding chain to support new token economies and business models, which provides us with more possibilities in the future.

Editor: The Chinese government has been extremely positive towards blockchain of late, however we can also see that the level of regulation is also increasing at the same time. Historically, the Chinese government usually holds a stance whereby being supportive to technology while being extremely cautious to digital assets. In fact, not only China, the US government expresses a strict attitude towards Facebook’s launch of Libra but does not blockade its path for development. Do you know why the Chinese government reacted so positively these days? How would this reaction impact the entire industry? Would the development of QuarkChain be adjusted because of this change in policies?

Zhou: Blockchain technology provides a highly effective, transparent, cross-border system for the transfer of assets. I think this is a very important breakthrough under the western economic system, which is highly regulated. With the strengthening of national power in China, we can see that the power of China far outshines its influence in finance.

Therefore, starting with blockchain technology, China can build its core blockchain technology and unleash its influence. I think this is a very positive direction. For the entire industry, we can foresee more and more countries and even international federations to join the blockchain industry, which I find extremely motivating.

For us, this will be a special opportunity. Our QuarkChain technology, is developed with self-reliance from ground zero, which matches the Chinese government’s strategic direction for blockchain technology. We hope to service the government’s direction better through QuarkChain’s technology.

About public chain technology

Editor: Previously when ETC fell victim to a 51 percent attack, I remember you mentioned during one of our sharing sessions that the cost of attacking a public chain is quite low. Can you explain to us the reason behind that? How does QuarkChain overcome this issue regarding security? Previously, I thought that PoW consensus mechanism is a safer protocol but in reality, PoW public chains have had plenty of security breaches, which makes PoS and DpoS public chains seem safer. How would you explain the reason that led to this situation? What kind of innovation does QuarkChain have in terms of consensus mechanism?

Zhou: Several things. Firstly, since ETC employs the same mining algorithm as Ethereum, but is a smaller network, it was relatively easy to temporarily transfer hashing power from mining ETH to mining ETC and thus gain a majority hashrate.

Another aspect is that the high mobility of computing power between chains makes public chains easy to attack. In fact, you can rent computing power from online marketplace! You can simply purchase computing power and attack ETC. Someone has done the math that to rent one hour of computing power would amount to about 500 USD but the profit might be millions.

To counter these attacks, we propose our proprietary PoSW consensus, which essentially is proof of work with stake. Specifically, it is an amalgam of advantages of PoW and that of PoS.

The higher the ratio of one miner’s computer power to overall mainnet, the more tokens the miner would be required to stake in order to guarantee one’s efficiency in mining. If not, other miners would overtake that level of efficiency. In other words, with strong abilities comes more responsibilities. If an attack appears, then the token one stakes would be confiscated.

As of now, pure dPoS and PoS have not been put under test thoroughly. For example, PoS of Cosmos has been running for less than a year. Some complain dPOS that is about to retreating back to centralization, as we have seen in EOS.

Other than that, we maintain great compatibility for different consensus mechanisms: with the root chain to sharding chains structure, both root chain and shards can deploy their own consensus mechanism according to specific business needs and modify on the go.

Editor: With the numerous public chain in the market, each has its own point of innovation. For example, new virtual machine, new consensus. From a technical standpoint, how important are these innovations? Are they necessary? What is the point of innovation at QuarkChain?

Zhou: What we see in most of the single-chain consensus, be it PoW, PoS, DAG, or virtual machine, are more or less the same thing as those we see in the mature public chains. So it is indeed questionable about the necessity of these innovations. For more cutting-edge multi-chain or sharding projects, like ETH2.0 and Polkadot, they are quite similar. To further substantiate my discussion, I encourage you to take a look at our technical whitepaper about our proprietary Boson consensus, which describes most of the multi-chain and sharding projects including the two I just mentioned.

Boson Consensus: A Scalable Blockchain Consensus Algorithm

Besides that, QuarkChain has already realized an innovation in diversity, which adds high flexibility in customizing shards, including its consensus mechanism, ledger, transaction, and token economy. As I said before, both root chain and shards can deploy their own consensus mechanism according to specific business needs and modify on the go. Also, QuarkChain natively support high-performance cross-shard transactions. Normal users do not need to understand the difference between cross-shard and in-shard transaction and different shard can be equipped with different functionalities. We have proven mathematically that all cross-shard and in-shard transaction can benefit from the overall security of the entire mainnet.

In this way, we have two unique characteristics:

  • Innovation: Through adding new shards, QuarkChain will support consensus for application optimization (e.g. dPoS, BFT), Ewasm virtual machines, privacy technologies. All QuarkChain users will immediately enjoy the benefit of the new shards.
  • Enabling new functions for existing cryptocurrencies: Through forking existing coins, we can let current coin-bearing users to enjoy the benefits of the QuarkChain network. For example, after forking Bitcoin, all Bitcoin users can deploy smart contract on QuarkChain and play dApp and participate in DeFi.

About ecosystem development of QuarkChain

Editor: Public chains all share the same story regarding ecosystem development: 1) expansion or better solution for the impossible triangle among security,decentralization, and low cost 2) build up a strong developer community 3) develop as many dApp as possible. Polkadot and Cosmos began with a story of launching a chain with just one click. QuarkChain now talks about launching shards with one click, which feels quite familiar to the stories from the other two projects. Cosmos has now become the developer tools for the public chain for some exchanges. From my understanding, Cosmos is a developer tool but QuarkChain is not quite the same since it is a public chain. So in terms of developing the ecosystem, would being a public chain become a limiting factor? What is the current development of QuarkChain’s ecosystem? For those who are thinking of uniting all chains, comparing to Cosmos and Polkadot, what kind of advantages do QuarkChain possess?

Zhou: Cosmos’s idea is to let each chain be responsible for its own security which we think security problem would still take place. Polkadot is the same as us and share security across network and its substrate is quite remarkable. However, Polkadot is overly reliant on the PoS consensus mechanism, including parachain where consensus cannot be configured easily based on the need of the current code. It means that Polkadot lacks that flexibility that QuarkChain enjoys.

Practically, even though multi-chain, sharding, cross-chain technology may become the future of blockchain, these projects have progressed slowly: Polkadot has not launched its mainnet; Cosmos is still running on Hub. None of these projects have reached production level for multi-chain, multi-shard operations that can carry out cross-chain, cross-shard transactions.

Additionally, all of these projects use PoS as the consensus mechanism of the network. Compared to PoW of Bitcoin, the maturity of PoS consensus, including security and degree of decentralization, remains in question. In comparison, our strong points lay in high flexibility and high efficiency. Our united root layer is like a broad walk that is clear of obstacles for future expansion.

In terms of ecosystem, this year we have new partners like China Unicom Venture Capital, Chainlink and have announced partnerships with TokenRoll, VERA, Crypto APIs, and UDAP. With the shift in policies in China, we have been busy communicating with business partners there as well. Please pay attention to our public announcements for the latest updates.

Editor: I would like to share a historical story with you. During the 1000s, Tangghut leader Lu Jipeng pledged allegiance to the Song dynasty and offered his territories in the West to the Song emperor. His actions brought reactionary effects upon his little brother Li Jiqian who turned down the choice to live a debauchery life in the heartland of China under the Song rule and refused to stay loyal under the Song. His leadership of the Tangghut tribe ultimately cemented the foundation of the Western Xia dynasty that rivalled with the Song dynasty. The interesting point of the story is that, ambitious people prefer achieving things with their own hands and wish to unite people with their own vision. That goes back to many blockchain developers: they all have their goals and ambitions. Even though QuarkChain may be able to provide low costs and plenty of benefits, it does not necessarily mean that all public chain developers would choose to work under QuarkChain’s ecosystem. So regarding that, do you think that QuarkChain can add values for other public chains?

Zhou: This historical story is enlightening. I think the story reflects a matter of choice. Whether to conquer the Chinese heartland or to venture into the wild west, these are two different choices. The core principle is of course to maximize one’s own benefits, which echoes what is mentioned in Three-Body Problem, “the most critical mission for a civilization is survival”. Blockchain had been drifting outside of mainstream projects but now is reentering the arena. We will need to wait and see whether this means it will conquer the technology world or will disappear from the limelight soon after.

Of all things, we should look far ahead. There are “mainland” and other “wild West” within the industry. What is important is that we find a comfortable foot hold that suits our positioning and ability. On one hand, we can collaborate with large enterprises and support alliance chains and private chains. Our Go version has already been used by some entreprises but we cannot announce publicly due to confidentiality agreement. On the other hand, we are also friendly to SME entrepreneurs who build their businesses from scratch.

If we reach the success that BTC or ETH enjoys (or even just one tenth of what they have), then firstly, all the sharding chains can enjoy security similar to that of BTC and ETH. Secondly, users and ecosystem of the chain can immediately become the users and ecosystem of the new sharding chain, which lowers the cost of innovation and entrepreneurship. Since all users of the entire public can join the chain easily, I think this is a massive benefit. The two problems that exist since the cold start of blockchain will be solved quickly. So I think our solution is highly attractive to both new sharding chains and to the entire blockchain industry.

Editor: Since this year, the entire blockchain industry has been underperforming with poor market value for the tokens. QKC Price is suboptimal, even though QuarkChain repurchased its token QKC with 100 million USD. After the short term effects faded from the repurchase, it is not able to raise the price of the token substantially. Why do you think such is the case? How would the future change?

Zhou: We think that the happening of the bear market is in fact an opportunity. We saw a lot of projects that previously attracted plenty of capital but with little discipline in financial operations have already exited the market. Those ambitious projects with clear vision, patience, and strong technical capabilities are slowly standing out. Even though QuarkChain has not been in existence for long, but through one year of refinement and testing, I think we are now competitive to mainstream projects. We are still in the sowing phase of development, but we have the conviction that with the continuous growth of the blockchain industry, one day we will harvest what we have sown. At this point, I think the present stage of blockchain development is quite similar to the Internet bubble in its early days. The question is who will become the next Google and Amazon?


Online Question: Is it true that QuarkChain is collaborating with China Unicom? To what extent are the two teams collaborating? What are you planning to develop more businesses in the next step?

Zhou: We will form a joint laboratory with them in the area of 5G and blockchain technology. About Unicom VC, the China Unicom Venture Capital was established in April 2014 and is a subsidiary of the China Unicom with 200 million RMB capital. The venture company uses a business model of incubation and investment platform to operate and invest in projects related to internet, value-adding telecommunication services and will choose to invest in projects pertaining to internet, mobile internet, cloud computing, logistic network, e-Commerce, culture, and new media and are highly related to main businesses of China Unicom. These prospective projects would be in early stage or growing stage and Unicom VC would incubate them and provide them with capital.

Online Question: How much does it cost to develop a public chain? How much does your team spend on R&D each year?

Zhou: Our team consists of about 30 persons, with a dozen or so in the Silicon Valley. I believe you can have a rough estimate yourselves based on these numbers.

If you have any questions about this competition, please leave your questions on the developer forum.

QuarkChain developer forum: