Sun Ming from Distributed Capital: Interpreting Filecoin products, operations, and economic models from an institutional perspective

Everyone must be familiar with Filecoin. Filecoin is the incentive layer on the IPFS distributed storage network. The project owner Protocol Labs is also the developer of IPFS. By generating reward tokens FIL, it encourages people to contribute storage resources, retrieval services, and bandwidth resources to the IPFS network to build a new data sharing network.

In 2014, Juan Benat, a Mexican, launched the IPFS project and founded Protocol Labs. In 2017, Juan's team conceived Filecoin and conducted research and development on it.

Since the concept of Filecoin was made public, it has immediately attracted attention from all parties: Filecoin is expected to become a groundbreaking product that can be on par with Bitcoin. The main network of Filecoin is expected to be launched in July or August 2020. This project, which has been brewing for more than three years, is about to be launched.

So, how will Filecoin perform?

Can it meet public expectations?

Distributed Capital held a roundtable forum on the FIlecoin project, and will invite big names from different fields to discuss their understanding of Filecoin and their thoughts on Filecoin's future performance.

In the first issue of the Distributed Roundtable, Sun Ming, partner of Distributed Capital, talked about Filecoin and the investment opportunities therein from the perspective of institutional investors.

Sun Ming: Sun Ming, partner of Distributed Capital, senior investor and legal expert in the blockchain industry

Jerry: Distributed Roundtable Moderator

Filecoin Products and Services

Jerry: IPFS appeared a long time ago, but it was not until the emergence of Filecoin that it attracted huge market attention. What did Filecoin do right?

Sun Ming: Filecoin has attracted much attention since its launch, but IPFS has not received such a big response since its launch. The reason is that there was no cryptocurrency community when IPFS was launched, so IPFS was just a peer-to-peer hypermedia protocol for distributed file storage and sharing.

There are multiple competitors for this type of protocol, and there is no special incentive mechanism for using IPFS, and the mainstream HTTP itself is sufficient in most business models. Previously, the main use of IPFS was to build and access some content-sensitive websites, so it failed to become mainstream.

The reason why Protocol Labs has set up an incentive mechanism now is to attract enough users to use IPFS and form a strong IPFS ecosystem. This incentive mechanism can achieve the first phase of Filecoin's mission - that is, to make enough miners willing to provide storage space to meet user storage needs. When a large amount of content exists in the IPFS network, a further ecosystem will be derived, generating the second step of demand, namely the retrieval demand.

In fact, IPFS is a public property and can be developed and used by anyone. But now when people mention IPFS, they only recognize Protocol Labs. Protocol Labs is the inventor of the IPFS protocol. Now, it has taken advantage of the fame of IPFS and cleverly used the incentive mechanism of blockchain to further enhance the influence of the project.

Jerry: The entire Filecoin network, in the final analysis, provides storage services. In the future, after the mainnet is launched, who do you think the user group will be? What pain points does Filecoin's network service solve for these users?

Sun Ming: What kind of user group Filecoin will serve depends on the expectations of Protocol Labs. However, it is foreseeable that Protocol Labs hopes that Filecoin will be commercialized in the mainstream, that is, to allow mainstream institutions to store data on the IPFS network.

Although IPFS has no restrictions on storage objects and content, from the perspective of the Filecoin project itself, mainstream companies will be treated as "verified customers" and their data storage incentives will be greater.

For users, Filecoin's service will solve their worries. If the user's data is stored in a centralized network according to the previous data storage model, it will face the risk of being deleted by the storage institution at any time. Secondly, if the centralized storage institution stops its business, the user's data will also be lost. In the IPFS decentralized network, if the user's data is deleted or lost, the storage miners will be severely punished, so as to urge the storage miners to save the user's data for a long time. Therefore, the IPFS network can at least become a backup place for users to store their data.

Filecoin's services are attractive to users from both economic and non-economic perspectives. From an economic perspective, Filecoin's fees will be cheaper. From a non-economic perspective, users want to find a long-term, stable and less censored storage model, and Filecoin meets their needs.

Jerry: What do you think of Filecoin's market vision, or how big is Filecoin? Ideally, what is the scale of the ecosystem generated around Filecoin? What are the important roles in the ecosystem?

Sun Ming: The most amazing thing about Filecoin is that it has achieved "breaking the circle", that is, jumping out of the cryptocurrency circle. Strictly speaking, BTC has not yet entered the mainstream business world. Ethereum is close to simulating and reproducing a mainstream business world, but it has not yet completely "broken the circle", while Filecoin has completely embraced the real business world.

The concept of Filecoin is simple and easy to understand. Both traditional investors and people in the Crypto field can understand this emerging thing. In addition, Filecoin can properly balance the interests of all parties in the ecosystem, making it easy for all parties to reach a consensus.

As for how big Filecoin will grow in the future, it is still unknown. However, it is clear that the valuation model of Crypto is difficult to apply to Filecoin. Crypto's business model is speculative, which means that the above applications can have a high market valuation before they are fully implemented.

Filecoin is more suitable to be compared with traditional mainstream business models, such as Alibaba Cloud, Baidu Cloud, Qiniu Cloud and even Xunlei, etc. The overall industry value of this market is huge, and if Filecoin can get a certain proportion of the pie, it will reach a scale of tens of billions of US dollars.

Operational status

The enthusiasm of miners and the community even reminds you of Ethereum back then. So we are very curious, what are the early signs of prosperity in the community now, and what do you think are the similarities and differences with the early Ethereum community?

Sun Ming: There are some differences between the Filecoin and Ethereum communities. Filecoin’s fans are mostly from the miners, while Ethereum’s fans are from the developers. In the early days of Ethereum, developers were thinking about what interesting things they could do with smart contracts, and issuing coins was only used to assist the application level.

Therefore, Ethereum is more geeky. The early purpose of Filecoin is to attract enough storage devices to join the network, so it will focus on attracting people related to mining (miners and mining equipment manufacturers) to join the community.

Jerry: As far as I know, the government has not done much publicity in China. The enthusiasm of the community and miners is more of a spontaneous response. What is the driving force behind this?

Sun Ming: This kind of spontaneity from the community and miners is closely related to IPFS, because the IPFS brand is highly recognized. For example, around 2018 to 2019, many people began to sell so-called IPFS mining machines, and at that time the technical details of Filecoin had not yet been disclosed.

From another perspective, it is precisely because IPFS is well-known that many people commit fraud in the name of IPFS. Simply put, IPFS is both a protocol and an IP. In the cold start phase of a project, the power of the brand will be far greater than the technology itself.

Jerry: What kind of group are IPFS miners? Are there any overlaps with Bitcoin and Ethereum miners?

Sun Ming: The IPFS miners are very scattered, including geek miners and ordinary miners. Between 2018 and 2019, a small group of the most dedicated and focused mining machine developers emerged in IPFS. They have been studying IPFS and Filecoin since then, and are optimistic about its potential and want to make long-term plans.

These mining machine developers have been planning to build mining machines. After all, mining machines are the most basic in the Filecoin ecosystem (it is not ruled out that these mining machine developers may also participate in mining in the later stage).

These technology pioneers have contributed a lot to the Filecoin ecosystem, and their stories may have some similarities with BTC mining. Back then, BTC mining could be described as "succession": some people who entered the game early may not have made much profit, but as more people entered the game, some winners emerged later.

Filecoin and BTC have completely different requirements for miners. It remains to be seen whether BTC miners will transform and start mining Filecoin. Currently, many of them are still waiting and watching.

Technically, it is difficult for BTC miners to migrate directly to IPFS. Filecoin does not calculate a hash function that has no practical use like BTC, but provides a storage service that is actually needed in the real world.

IPFS mining machines are much more complex in terms of technology, involving many computing devices and storage devices rather than just a single logic chip; in addition, once a Filecoin mine begins operation, it is usually not possible to directly migrate the mining equipment, but it must continue to operate stably in the original location. Therefore, Filecoin mining is more of a mainstream business operation model.

Economic Model

Jerry: What is the essential difference between Filecoin's consensus mechanism and the traditional POW incentive mechanism? Does it feel more like leasing in the shared economy than mining? Can this economic model maintain its efficiency, cost advantages, and miners' enthusiasm in the long run?

Sun Ming: The essential difference is: the goal of POW is to identify the miner with the highest computing power (the one who calculates the fastest) by calculating a hash function that has no practical use, and let it package transactions to generate blocks, but this calculation has no specific use or meaning in the real world; Filecoin's consensus mechanism is much more complicated, combining block generation and storage service provision, the latter of which has significance for real-world commercial activities.

As for mining, it is an incentive for the generation of blocks. It is an accessory to the generation of blocks rather than the essence of it.

For Filecoin, if it wants to maintain a balance between efficiency, cost advantages of storage services and the enthusiasm of miners, Protocol Labs needs to continuously adjust parameters in the later stages. Compared with the later efforts, Protocol Labs needs to start commercial promotion in the initial stage.

Especially when doing To B business, the project party needs to slowly root the advantages of the product into the minds of traditional corporate customers.

In addition to commercial promotion, sales are also a top priority in the early stages of a project. The cleverness of Filecoin lies in its distributed sales. It does not sell the services of the developer's own company, but the services of the entire distributed network outside the developer. Any miner is an agent in the entire network and can represent the network for external sales.

Therefore, the ecosystem created by Filecoin is to fully mobilize the enthusiasm of miners and enable them to promote To B business in the early stage.

Jerry: In the incentive model, a core issue is the verification of effective data storage. There are two controversial views on the market regarding the contradiction between the validity of the verification data and the decentralized system. One is biased towards decentralization, and the other is biased towards centralization to strengthen effective verification.

Which strategy do you think is more beneficial in the project start-up phase? What is the long-term solution?

Sun Ming: Centralized or decentralized operation depends on the founding team's positioning of the project. In fact, Protocol Labs prefers to treat Filecoin as a commercial project rather than a common Crypto community project.

Therefore, it is natural to complete some things off-chain, such as "verification of validly stored data", rather than relying solely on the blockchain to complete all operations.

Although off-chain operations go against the geek spirit, the completely open, transparent and code-controlled on-chain processing method cannot achieve the optimal commercial effect, because the functions of blockchain are limited after all.

As a quasi-corporate blockchain project, it will be a relatively long process for Protocol Labs to complete some of its business off-chain. Moreover, these off-chain operations themselves can be decentralized to a certain extent, rather than being necessarily centralized. Therefore, in the future there will be distributed customer verification to determine valid data.

In this regard, Vitalik does not think that blockchain projects are best governed entirely on-chain, but rather that on-chain and off-chain governance should be combined.

Jerry: As the main network is improved and its efficiency is increasing, the space for technical optimization of third-party companies that manufacture mining machines and mine will be compressed. How do you view the situation where the technical advantages of mining machines/miners are gradually eliminated?

Sun Ming: When the Filecoin mainnet is first released, there will definitely be a technical gap between mining participants, but as time goes by, this technical gap will gradually decrease. The same is true for Bitcoin mining. As each mining machine is upgraded, the technical gap between mining machines is getting smaller and smaller.

Therefore, compared with soft resources such as technology, mining participants (whether mining machine manufacturers or miners) should pay more attention to timeless hard resources, that is, those resources that gradually accumulate but do not decrease in effect over time. The hardest resources are the ability to obtain various types of hardware (supply chain resources) and business relationship resources.

For example, business relationships with B2B will not be easily snatched away. In the future, entrepreneurs in the Filecoin field will definitely make layouts based on fields with moats, rather than simply focusing on technical development capabilities.

Team Background

Jerry: Early-stage projects require a lot of judgment of people. Why Juan’s team? What are the strengths of Juan’s team?

Sun Ming: The IPFS protocol itself has proven the team's capabilities. Filecoin has further commercialized IPFS. If you are optimistic about the development of IPFS, you will definitely choose to trust Juan's team.

Of course, there are other incentive layer projects developed based on IPFS on the market. However, the capabilities of these project teams may not have been proven yet, and investing in other blockchain projects based on IPFS at this time is risky, and it is difficult for the crypto community to reach a consensus on these teams (but there are exceptions).

Competitive Analysis

Jerry: How can you compete with centralized cloud storage projects in terms of price and performance? Similar to Amazon Cloud Glacier?

Sun Ming: Compared with traditional cloud storage projects, Filecoin's strategy is to compete in a different way. Filecoin can provide something that other companies don't have. For example, the data stored by Filecoin for users will not be lost due to the closure of the operator, and it is less subject to censorship and supervision.

Jerry: And existing decentralized storage projects, such as Sia, StorJ, etc.?

Sun Ming: Compared with other distributed storage projects, Filecoin's strategy is to leverage the brand advantage of IPFS. The operation methods of Sia and StorJ are not much different from Filecoin - they are mainly focused on developing To B business, supplemented by personal business.

However, Sia and StorJ do not have the strong brand appeal and influence of IPFS, so there is less participation from community miners and they will face greater resistance at the sales level.

Jerry: JD Cloud has already used the IPFS protocol when storing data; how do you view this competitive threat?

Sun Ming: Even though JD Cloud uses the IPFS network, it is still a centralized service provider in essence, and is not comparable to Filecoin and has no direct relationship with it. However, it is not ruled out that companies within the JD system will participate in Filecoin mining as miners in the Filecoin ecosystem in the future. At this time, the company will have certain advantages based on JD's business resources.

Timing of investment opportunities

Jerry: Looking back at the early days of the market, Amazon Cloud and Alibaba Cloud had a huge motivation to use idle computing power and storage due to the peaks and valleys of storage and computing power, and entered the cloud storage market. At this moment, what changes have taken place in the market? What changes have occurred in the supply and demand levels of the market?

Sun Ming: What we need to see clearly is that the market demand has always existed, but it is relatively scattered. We need a platform to bring together the scattered demands. It is difficult for the traditional centralized business model to achieve this because the cost-effectiveness of serving such a long-tail market is not high; but the distributed network can do it because it solves the cost problem of serving scattered customers through distributed supply and sales, making the data storage of the long-tail market more efficient.

Jerry: What are the driving forces behind these changes? What is the current pattern of industry development? Is it the Spring and Autumn Period and the Warring States Period, the Romance of the Three Kingdoms, or a unified state?

Sun Ming: The market for distributed storage has not yet been developed and is still in its infancy. Many people do not realize the potential of the Filecoin market, let alone imagine what this market will look like. Currently, this market is still in its primitive and wild state. Since the market has not yet been effectively formed, there is no market structure to speak of.

Jerry: What do you think will be the future outcome of the distributed storage track?

Sun Ming: In the future, when the development of distributed storage is mature enough, the following phenomena will occur: First, Filecoin, like a public chain project, has a network effect, and there will be a trend of one dominant player in the future. The more data stored on this network, the stronger the ecosystem, and the lower the chance of being overtaken by others.

Second, the trend of regionalization of miners is more obvious. Miners will follow the principle of proximity: for example, Chinese miners will look for Chinese customers, and American miners will look for American users. The benefits of localization are as follows: first, miners are more familiar with the local market and have stronger business relationships; second, the efficiency of localized transmission is relatively high.

Third, as the project develops, miners like "local tyrants" will appear in the market and may occupy a larger share of the local market, but will also be more easily regulated by the government.

Fourth, the types of miners will be diversified, such as industry-specific miners that specialize in serving customers in a certain field. As for the possibility of a single miner becoming dominant, it is very small, because the hardware required for Filecoin mining is in large supply and most of it is general-purpose hardware, with an open market supply, and there will be no monopoly. Even if there are FPGAs or even ASICs in the future, their impact on the overall mining efficiency of Filecoin is far less than that of Bitcoin mining machines, and they are also easier to design and manufacture.

Jerry: What risks and challenges does Filecoin face before it achieves its vision?

Sun Ming: Technical risks always exist. For example, the storage data is stored in a single location at the order-taking miners, rather than being dispersed across the entire network. In terms of business model, improper incentive distribution will be the biggest potential risk of the project. If speculators find loopholes in the incentive layer, it will lead to the phenomenon of wool-pulling, so that honest miners will not get enough incentives.

The Protocol Labs project puts verification offline to prevent the phenomenon of wool parties. Another potential risk is that Filecoin's overall commercial development in the early stage is unsuccessful and no mainstream corporate customers are found, which will lead to the lagging development of the project.

The layout of distributed capital

Jerry: How does our distributed capital participate currently? What investment layouts have been made in various links of the ecosystem? What else are you optimistic about?

Sun Ming: We have invested a lot of FIL coins in 2017, and invested in a mining machine manufacturer this year, and are preparing to invest in more. Distributed Capital's strategy is to invest in mining machine manufacturers first, and then observe the development of the industry and make judgments, gradually infiltrating into the Filecoin mining farm and mining pool fields, and even IDC computer rooms, one of the "new infrastructure", may be one of the investment directions.

Because the industry will definitely be divided and tend to be specialized as it develops, for example, a mining farm will only be a mining farm, and a mining pool will only be a mining pool. Since both distributed storage and IDC computer rooms are in line with national policy orientation, traditional capital has a lot of investment and cooperation intentions in this regard, which is very different from our previous blockchain project investment.

Jerry: As an investment institution, we often find that a specific project supports the entire portfolio, such as Alibaba to Masayoshi Son. How do we view this opportunity? Does the Filecoin-related track have such great potential?

Sun Ming: As a capital provider, we will bet on major projects with epoch-making significance. Previously, the company made a big bet on Ethereum. At present, Filecoin is the one with the greatest chance of bringing revolutionary impact among the known new projects, so it is worth investing the most resources in its ecosystem.

Jerry: Valuation issue: including the futures trading price on gate, it is close to RMB 60 (late May 2020). Even so, its valuation has entered the top 10 (calculated based on the circulating market value one year after the mainnet launch). Many people in the market believe that its valuation is too high. What do you think of this statement?

Sun Ming: Currently, the price of Filecoin is around $8, which is a relatively low price. It is easier to value Filecoin than pure blockchain projects because it is an extension of traditional business activities and can be estimated using traditional valuation methods.

In addition, Filecoin mining covers many physical resources (such as the rental costs of IDC computer rooms, the purchase costs of mining equipment, and equipment maintenance costs, etc.). This part of the cost is relatively easy to calculate and is also one of the references for considering valuation.

Jerry: Are there any other important issues that I haven’t mentioned that you think need to be added?

Sun Ming: Not for now.

——End——