This interview uses “distributed storage” as the entry point, and with the help of the actual progress of the Crust project, further explores the similarities and differences of major storage projects, the reasons behind Filecoin’s suspicion, how the industry is implemented, and the face of supervision. topic.
Interview & Writer: Lou Yue
Interviewee: Bova Chen, founder of Crust Network
After the “wealth-making wave” brought by DeFi liquidity mining has slowly cooled down, both on- and off-market funds seem to be looking for the next round of new exits, and the crypto field has never lacked divergent thinking. Some people say that it will break out. Point will be in the NFT field, and some people say it will be cross-chain, DAO, Ethereum 2.0 and other directions. Before the emergence of phenomenal hot spots like DeFi, the industry seemed to embark on a path of diversified parallel development in various fields.
Recently, the news that “Ace Pigeon” Filecoin officially launched its mainnet after a lapse of nearly three years has attracted widespread market attention, and has also driven the topic of ” distributed storage ” into everyone’s field of vision.
In order not to be abandoned by the industry as “classical leeks” again, to learn hard to welcome the next wave of hotspots, and to learn more about the development of the distributed storage track, we have invited our own “Star Public Chain Polkadot” and ” Bova Chen , the founder of the Crust project on the two hot labels of “Distributed Storage,” chatted about recent industry developments.
This interview will take “distributed storage” as the entry point, with the help of the actual progress of the Crust project, to further explore the similarities and differences of major storage projects, the reasons behind Filecoin’s suspicion, how the industry is implemented, and the face of supervision. On this topic, Bova also shared some other wonderful views. The following are the core points of this interview.
Current status of the distributed storage track
Whether it is a positive expectation or a negative complaint, Filecoin has really taken a lot of attention recently. As a “distributed cloud incentive layer protocol” based on the Polkadot parachain structure, some people compare it to Filecoin on the Polkadot network , and it is also on the track of distributed storage, which will inevitably accept everyone’s comparison. In this regard, Bova, the founder of the Crust project, also shared the differentiated characteristics of his own project, as well as the similarities and differences with Filecoin .
Brief understanding of Crust
To put it simply, Crust is a blockchain protocol developed based on the Polkadot Substrate framework. It can provide a distributed storage network under the Web 3 framework, support multiple storage layer protocols including IPFS, and support various storage application layers. provide support.
In fact, many mechanisms are similar to Filecoin, so many people simply summarize that Crust is the Filecoin of the Polkadot ecology. But in addition to supporting storage, the underlying stack of Crust technology can be extended to computing in the future, so what I want to build in the end is a distributed cloud that emphasizes data privacy, data protection, and personal data ownership. It is not only pure data, but can Realize computing and modern cloud functions on the Internet .
Based on the Polkadot ecology, Crust joined the list of Web3 funding projects very early, joined the Substrate Builders Program, Web3.0 Bootcamp, and received technical support from Parity.
Differences in node storage proof mechanism between projects
The concept of distributed storage has been mentioned in the blockchain industry for many years, and there are also various projects in the industry such as Storj, Lambda, Arweave, etc. The core underlying technologies between the projects are mainly divided into several categories.
First of all, some previous distributed storage projects used some super nodes and central guarantee nodes to coordinate and process the relationship between resources and users.
One of the reasons why Filecoin is so popular is because the algorithm is used to truly prove whether there is real data stored on the node. After the algorithm is proved, there is no need to use an intermediary or a guarantor to judge whether the data is stored. In this case, the system can truly be decentralized, and everyone believes that my things will be saved for me by others. This is why Filecoin has attracted so much attention, because it has a certain breakthrough in algorithm.
Crust and Filecoin storage markets are both based on storage orders, in contrast to Arweave, which is another model based on permanent storage.
Crust can actually be classified as similar to Filecoin, which is a completely decentralized storage network, which means that no super nodes are introduced in the storage proof mechanism of nodes. But Crust simplifies the Filecoin algorithm , and also proves that the node really stores things, which is also a big difference from some old- fashioned distributed storage projects.
Hardware entry barrier
The characteristics of the Crust project. From a horizontal comparison, Filecoin’s core proof mechanism , including algorithms such as space-time proof, copy proof, etc., is based on a large number of complex cryptographic calculations for proof, although it can indeed effectively achieve the purpose of proof . But it will consume computer resources and network bandwidth. In this regard, Filecoin storage servers, such as mining machines, have a relatively high threshold. Miners need to invest a lot of hardware and equipment overhead.
Crust’s algorithm simplifies this set of processes relatively speaking, and the threshold for hardware is lower. The authentication algorithm is achieved through the “trusted environment space” (based on TEE-based local storage verification) inside the CPU, so Crust will be more economical, more cost-effective, and lower thresholds.
For miners, the cost of entering the network as a resource provider will be smaller, so the Crust economic form will be closer to the storage server form of the current data room centralized cloud , and Filecoin is much higher than this hardware standard.
Staking mechanism
The technology between projects is different. For example, from the perspective of orders and packaged data, Crust will encourage resources for the early network (development), not just orders.
In order to meet the needs of enterprises and individual users and accelerate the implementation of products, Crust proposed the MPoW and GPoS mechanisms, and at the same time designed the GPoS consensus algorithm with reference to the PoS and Polkadot underlying NPoS consensus , in order to achieve higher performance. So relatively speaking, there will be a staking mechanism in the economic model, which is also a difference.
Project ecology
The Filecoin ecosystem is actually a separate ecosystem, and Protocol Labs does not rely on other projects to develop applications.
Crust has been built on the Polkadot platform since the establishment of the project. One advantage is that it can avoid duplication of designing things . Polkadot’s underlying architecture will implement verification consensus. Polkadot’s shared security can also solve some of the underlying things. Crust can focus its energy on algorithm proof, economic distribution, application development and other levels.
Another bigger advantage is that Polkadot’s cross-chain allows projects in the ecology to share resources . Simply put, Crust provides storage. Naturally speaking, all parallel projects of Polkadot can call Crust’s storage resources through a very simple cross-chain protocol. However, if storage resources are developed on Filecoin, a lot of development work must be done in the middle to connect these resources. This is the difference between having an existing ecology and a separate development ecology.
Landing
The last advantage is landing. Take an example to illustrate the difference in landing. In order to resist various file generation attacks, Filecoin and Crust need to encapsulate the stored files before they can be stored correctly.
Filecoin is limited by the algorithm, it takes nearly two hours to encapsulate a 32G file, while Crust is based on TEE’s own re-encryption technology, and it only takes 2s for every 1G to encapsulate.
In addition, Substrate-based on-chain governance and no fork upgrade mechanism . Crust can smoothly upgrade the protocol without forking. This will be very friendly to future Crust-based landing applications.
Why choose to join Polkadot Ecosystem?
From the perspective of the underlying public chain, Filecoin, Arweave, Sia and other related items are distributed storage based on their public chain, why Crust public would choose to join Boca chain, how Crust positioned in Boca ecology of the entire eco-Boca What kind of help is provided to Crust, and will it bring some hidden risks ? In this regard, Bova, the founder of the Crust project, also answered one by one.
Polkadot’s help includes many aspects. To put it simply, when Crust was first established in 18 and developed in 19, it was developed with Ethereum and Cosmos SDK for a long time, because the previous research considered these The architecture is more suitable, but after walking for a long time, I found that the progress is not fast. Then I went to investigate some other platforms, including Polkadot, Filecoin, and Near , which has just launched.
The final feedback is Near and Polkadot. Polkadot officials also quickly sent an email to reflect the needs of the project and gave a complete set of Substrate development tools . Web 3 provided financial support . At the same time, it also participated in the Developer Project Substrate Builders Program established by the Parity technical development team. And Web3.0 Bootcamp, etc., received technical support to help solve some development problems.
Based on this, it can be seen that the Web3 Foundation provides considerable support for the entire ecological project. Not only does it give money, technology, and even PI, and market, it can be seen that a project party has a plan for its own ecological development. This was also decisive at the time. A very important reason for choosing to enter the Polkadot ecology.
Judging from the current results, this choice is not wrong, because the blockchain has developed to this day, and its underlying infrastructure is relatively complete. Although it cannot be said to be particularly perfect, compared to the previous years, it is already sufficient to develop applications, including The current DeFi and some other applications have already left the bottom layer . The construction of financial facilities is the logic of thinking on the business layer.
The underlying framework of Polkadot Substrate allows the team to build a lower-level infrastructure without having to repeat the wheels. It has built an infrastructure , and then can build business models and achieve algorithm optimization on this platform, so let the development progress of Crust It has improved a lot than before. This is a place that has gained a lot.
Secondly, just as Polkadot has received many expectations from the market, the market attention of its ecological projects has also increased. When developing on the Polkadot platform, everyone thinks that you (already) have a good tool, so the development of the project Will accelerate, including this month’s Web3 conference, Polkadot will also regularly hold such conferences around the world. For some relatively high-quality projects in the ecology, we will give some opportunities for everyone to show , and then provide resources to support as much as possible, which is also an advantage obtained from Polkadot.
Technically speaking, in addition to tools that are more useful, if the entire Polkadot ecological cross-chain agreement is formed, it will form a state of blossoming, because it is known as the ” king of cross-chain “. Any blockchain project as long as it meets the protocol standards, It can be accessed. When more and more valuable projects appear on its platform, the synergy effect of everyone will be very good. When the synergy between the projects increases, the ecology is not just a project. It will be the effect of “one plus one is greater than two”.
As a distributed storage project, Crust is not good at making payments and may use native tokens to pay, but in fact this is not the optimal solution. If there is a mature payment system, everyone can directly use a close system.
Take Taobao as an example . Before Alipay, users would use PayPal and UnionPay payment systems, but after Ali’s own development of Alipay, a very mature settlement system, was released, Taobao would have no settlement burden and no need to build a settlement system by itself. You can continue to focus on the project itself.
The same is true in the cross-chain system. Some are focused on payment, some are focused on underlying transmission and storage, and there may be projects focused on identity authentication. When everyone collaborates, such as when registering an account system There is no need for authentication, because there is already a corresponding identity authentication system that can be connected.
Just like many Internet companies nowadays, when users register, WeChat directly authorizes login, and there is no need to develop a separate account system. It is easy to implement these current Internet 2.0 models under the cross-chain mechanism .
Polkadot’s cross-chain capability can effectively expand Crust’s potential market size and will be more friendly to cold start. For example, Crust has reached cooperation with projects such as Maskbook and Darwinia, and will provide services such as decentralized data distribution and NFT data storage for these projects after the mainnet goes online.
Are there potential risks in joining Polkadot?
Crust’s choice of the Polkadot public chain did bring a lot of gains. Since the construction is based on the Polkadot public chain , such as a problem with a certain underlying protocol tool, will there be a situation that affects the upper-level project from the bottom ?
In response to this issue, Bova said that everyone is developing based on a set of standardized tools. If the tool itself has loopholes, all projects based on this tool should be affected. For example, some exchanges now write based on a certain set of codes. If the code has loopholes, all exchanges may be affected. This does exist.
However, the Polkadot Substrate framework has gone through two or three generations of updates and iterations from the initial 0:00 version to the current version 1.0 and 2.0, and the underlying foundation is being improved step by step.
In addition, since Polkadot founder Gavin Wood is from Ethereum, the entire Polkadot architecture is very similar to Ethereum. Many people understand Polkadot as the Ethereum 2.0 version that came out in advance. Ethereum has been running for so many years, and it has basically verified that its underlying core is no problem. This is why everyone believes that Polkadot will not bring risks in some underlying layers and protocols.
Of course, some additional things are not ruled out, such as proposing a new concept, these concepts may have loopholes, but in the development process, as more and more developers enter the ecology, it will promote the ecology of Polkadot. more perfect.
Hit the Filecoin market controversy
When talking about distributed storage, many people may still think of Filecoin, because the market has indeed accumulated a lot of expectations, but from the appearance of fake mining machines and “Songjiang Declaration” before its launch, many people questioned the testnet after the mainnet was finally launched. A series of negative events such as a series of negative events such as a series of negative events such as a series of negative events such as a series of negative events such as the linking of the main network with the currency of miners, the lack of currency to pledge, the economic model being complained about. In response to these “unexpected” market reactions, Bova also shared his views,
First of all, this matter has indeed been ups and downs recently. The so-called “where there are people there are rivers and lakes.” Some projects that involve great interests may have some contradictions and will have a relatively large impact.
We think Filecoin is a relatively good project . A very important factor in these problems like its existence is that some of the models in the middle have finally changed several times , including the economic framework, including the test currency, and finally turned into a truly valuable currency. In the beginning, it was not designed in advance, and it did not give everyone an expectation.
According to the market reaction, everyone will complain, but digging will still dig. Participants are actively participating, because regardless of whether the test currency is converted to official currency, the profit of the miners is still very high , because at least they can make money , So in the short term , people will not easily escape the market and escape this project.
After all, in the early stage, the entire market value of Filecoin mining machines should be at least tens of billions . Everyone has so many hardware resources, so many manufacturers, players or investors have already entered, and they cannot withdraw at any time, so everyone wants to make them based on their own interests. To maximize benefits, more complaints are also based on benefits .
The negative news is mainly because some of the previous parameters and some rules were not fixed in advance for everyone. The changes in the middle make everyone a little undecidable about the future expectations, so that someone will fight for their own interests. If you write these things down, for example, Bitcoin is halved every four years, and there will be no miners to make trouble during the halving, but if you never write about it, and then suddenly the income is halved at a certain point in time, then There will certainly be many miners complaining about their interests.
Filecoin is more that it may not be able to predict too many things in advance, or because the project complexity is too high , so the speed of the intermediate transformation is relatively high , which leads to some negative information, such as mining due to the complexity of the proof algorithm The high threshold eventually made some machines unable to mine.
Incentive, application and supervision of distributed storage
The token economic model of a project is an important part of supporting its network security and sustainable development. Based on the official website information, we have simply sorted out the Crust token CRU system, overall distribution form and economic model.
Crust token CRU system
According to Crust’s official information , the initial circulation of the mainnet is 20 million , and the main uses are early private equity, ecological expansion, team incentives, BD and foundation reservations.
- Token Sale: 5,000,000 CRU (25%)
- Business and marketing: 2,000,000 CRU (10%)
- Community development: 5,000,000 CRU (25%)
- Foundation reserve: 4,000,000 CRU (20%)
- Team: 4,000,000 CRU (20%)
The ecological expansion part is mainly used for community delivery. Through the following four channels: user placement, miner placement, placement to exchanges, and bidding for the Polkadot parachain.
The tokens sold in the private sale round and the team incentive department are locked when the mainnet is launched, and released linearly. The linearly released part will be mined by the network together with the mining part.
Incentives
After a brief understanding of its token economic model, more people will be concerned about how to determine the incentive tokens to be issued to, how much, what are the standards, whether it is fair, how to prove the workload and other issues for a project. Ask Bova to talk about his thoughts on this topic.
As a storage network, or we think of a distributed cloud, it will also include computing in the future. There is nothing more than one resource provider , one resource user , the other is the merchant, the other is the user , this is the traditional cloud .
Decentralization is basically this form. In a decentralized network, the solution is to use protocols to call these resources and join the network. In order to encourage them (resource providers) to come in, Crust will allocate corresponding benefits. In other words , it is called mining , which allows resources to enter the network first and returns.
From the user side, more users are users who have pure data needs and are willing to spend money to buy resources or lease space. What Crust does is to mobilize the matching of these two (resources) through economic models . Use algorithms to verify that this node really helps you save things.
Filecoin proves this through an algorithmic mechanism. If you really save it, I can give you something, but as to whether the data you save is valid and whether it is the data sent by the user, this may have to be determined through the market or through other means. .
Crust’s consideration here is that no matter who the data comes from, since it is sent through the market, even if an individual sends data that is not necessary at all, it should be stored, because I have already formed a contract with you , such as on Baidu Cloud Some of the documents may not be needed many years ago, but it cannot be judged as invalid data, because it is sent by users, so our market basically comes from the data stored by users.
As a node , as long as it provides resources to access the network and can form mining, it will reward him through the model. If he provides storage for users through the storage market and solves the user’s storage order, then he will also get market settlement Income, this is our economic model.
- Integrate Internet Economic Model
The other piece of design incorporates some considerations of the current Internet economy . Take Didi Dache as an example. Didi is nothing more than dispatching some private cars or making full use of the car usage rate to connect to the platform, and users can deploy through the network. To use. In the early days, in order to encourage users to use the Internet, coupons were issued to let everyone discover that taxis may be more economical than other modes of travel.
This model is more to increase the early market share and user participation, but it is not a long-term development. In the later period, users will still pay the drivers through market settlement. When the market share increases, the drop The Di platform will withdraw its subsidies and user coupons .
In fact, in the blockchain industry, mining and the recent DeFi liquidity rate of return actually have some concepts like this. Through early subsidies, users will increase their willingness to enter and increase market share or user volume. When the real demand is on the platform After being posted, the subsidy will be withdrawn.
Therefore, Crust will also be adopted in the economic model. Simply put, users can earn income by storing data. Through the token design, users can receive the distribution of native tokens while storing data on the platform. The income will increase with the market. Mature, then slowly withdraw.
- How to avoid “being evil”
DeFi, including Didi Dache, has some phenomena of ” mining for mining “. These can be avoided as much as possible through algorithms, but they cannot be completely eliminated. However, as the proportion of users “wholesale” decreases, or participate in this The cost of this method becomes higher, but it can still be restricted.
From the perspective of the Crust model, users do not know where they exist when a storage order is sent. If a node wants to use the storage capacity of its own node, it may not necessarily take the order after the order is sent. However, just like Didi, if you take a mobile phone and place an order next to the car, it will be distributed nearby. You can swipe your car through the principle of optimal algorithm. This situation cannot be completely eliminated, but you can increase it by random dispatch of the algorithm. cost.
Is there a possibility of large-scale application of distributed storage?
It seems that distributed storage in the decentralized field has not been applied or verified on a large scale in the market. Regarding the reliability and usability of its verification, Bova gave us the following answer.
Talking about the concept of distributed storage, there are actually a large area of applications now, but many distributed storage systems are not decentralized, that is , decentralized storage and distributed storage are two concepts .
Just talk about distributed, Alibaba Cloud can be made distributed, just like Guizhou on the cloud, a big data center can be built in Guizhou to respond to all data storage in the country. It is conceivable that user data in Northeast China and Beijing will be stored in Guizhou, and the data will be sent to the main site network and then stored in Guizhou. The delay will be high and the efficiency will not be high. If a data center is built in Beijing, there is no need to run The backbone network has low response delay, which is optimal. Therefore, many centralized manufacturers will also try to build data centers in some relatively scattered places to respond to data requests in various places. This is distributed.
Blockchain is more about decentralized scheduling . This becomes difficult because without the main body of Alibaba Cloud, everyone who accesses the network is not a main body. How to allocate their resources?
This question is more difficult to tell the truth. First of all, you have to prove that these people really saved this thing because they are not controlled by the subject. If he lied that he saved it, it is actually unfair, or the user can’t get the data. , So it is necessary to prove first, and then to distribute benefits and dispatch relations.
Before the emergence of Filecoin, the established projects did not really prove that the node really saved things. It may be a third-party intermediary or guarantor, or create a company to provide a guarantee. When I find that you don’t have a deposit, I may remove it. , Is an afterthought or a guarantee mechanism,
Filecoin tries to use algorithms to prove that by removing the role of the central node as an intermediary, it has truly achieved decentralization. This is one reason why Filecoin will be placed in high hopes. The other is that the IPFS protocol itself is a decentralized scheduling protocol, which can schedule many decentralized node resources without a centralized organization.
The reason why the distributed storage in the decentralized ecosystem has not undergone major commercial applications is not that distributed data cannot be commercialized, but that the added resources are uneven . It is necessary to ensure that each resource reaches a certain service level , There are more things to do.
Whether it is to quantify the quality of service and quantify the network, to ensure that things are stored, not deleted, or to establish a punishment mechanism, etc., this is easy in centralization, because it is its own node, but it is different if it is decentralized. He can withdraw at any time regardless of your control, so this is a more difficult problem. But now I think it has been solved step by step by many project parties.
So in the future, distributed storage may enter from certain fields and gradually provide real user needs , including some temporary file transfers and cold data file storage that Crust is currently doing, and Yotta Chain and other real commercial storage data In the system. What is foreseeable next is that there will be more and more distributed systems, and there will be commercial applications running on the platform.
Regulatory issues
In response to the question “As more people join the distributed storage ecosystem, there may be confidential and sensitive data, will it cause censorship and regulatory risks?” Bova responded to this. Everyone in the blockchain industry It is inevitable that they are all talking about supervision .
Not only distributed storage, but all decentralized projects have regulatory issues. From the storage field alone, it is theoretically possible to supervise , because it is not more troublesome than other blockchain projects.
There is a principle that ” who stores the data is responsible for his own data. ” Individuals need to deposit through the portal of the platform. Whoever provides this portal should theoretically have this responsibility to control, and the source of the data can be traced. A link may be needed in the middle of whether this person has been authenticated by his real name or whether he can be found. For example, in the Bitcoin network, if there is an anonymous address for money laundering, it will be more difficult to find, but if a money laundering operation occurs at the address of the exchange, it can be directly traced to the exchange.
Whenever a certain responsibility can be defined, whether it is decentralization or centralization, who does this thing, who is responsible for this thing, so basically this logic is also in the decentralized storage network. From the perspective of the project party, an agreement is provided, and whoever uses this agreement can pursue the responsibility of the user.
Crust will provide some UI or front-end operation interface. The data that enters the background from this interface will undergo a content review, but if it comes from your interface, then another content review should be done.
When it is stored on the network, you may not be able to delete it, or you cannot review it through the background method, but you can definitely trace the source of this data. From the perspective of the regulatory framework of decentralized projects, this should be the principle.
Crust project progress
From the release of the first version of the testnet in February and the second version of the Maxwell testnet in August, more than half a year has passed. According to expectations, the entire development schedule should be completed by the end of this year. Publish the mainnet .
There may be several points of concern in the middle. For example, Filecoin has been delayed for two or three years. In fact, Polkadot has been pushed for many years from 2016 to the present. In fact, many problems in blockchain projects have been slow to solve and are unpredictable. Sometimes it may delay our development progress, but from the current development situation, the test feedback is basically relatively optimistic.
For example, Polkadot’s cross-chain mechanism has not been finalized until now, and the testnet ran through it once, because of throughput issues, it may eventually be upgraded. We also hope that at least when the Polkadot cross-chain protocol is launched , the Polkadot slot can be accessed at the same time .
Other things, including Filecoin, are actually a lesson for the past. Crust will change or optimize accordingly. These may take some time. If you are optimistically expected, the official network should be launched at the end of the year or early next year.
The testnet currently has no place that affects the development progress. The team is more pursuing excellence and is constantly building storage models, because Filecoin has not verified that its storage market can actually store things. It is hoped that the storage market can be improved in the same period. It was released so that users can directly store things on it. It took a lot of time. Other developments are basically fairly smooth.
At the end of 11 test network will launch incentives, with Filecoin space race would be more similar, large-scale miners will reward test, with some commercial interests test make the network closer to the real usage scenarios, in November or December There will be several phases of such bonus tests. It is expected that 300,000 tokens will be invested in the first phase to encourage miners to enter. The first stage is mainly to test the software, hardware and network environment of the node, and at the same time, the Candy airdrop activity on the testnet will be carried out simultaneously.
In addition, Boca slots for auction issue, Bova pointed out that the slot auction is a very important milestone in Boca above, the project will enjoy access slot on the ecological resources, it will be a lot of programs are expected to participate in the auction slot. For example, a part of the token economic model may be reserved for users to let users vote for us and rent for us. However, how much this part is reserved and how it is distributed still depends on the end of Polkadot’s auction mechanism. Stereotyped.
