Filecoin (FIL)

Filecoin is a decentralized storage network leveraging IPFS

Filecoin

Total supply	40.000.000

Basics

• Based in: San Fransisco
• Started in: 2017
• A DSN (Decentralized Storage Network); data storage network and electronic currency based on Bitcoin. Build over Ethereum by Protocol Labs. If you host files on your computer, you earn Filecoin.
• Mainnet release: 15-10-2020.
• FVM for smart contracts got announced on 4-11-2021. Got live on 14-3-2023.

History

• Filecoin has again (20-4-2020) postponed its mainnet launch, looking ahead to a new window between July 20 to August 21. Final launch date (28-9-2020) is now 15 October.
• "Filecoin Is mailing out (7-5-2020) Hard Drives of Climate Data to Kick-Start its File-Storage Network."

Audits & Exploits

• Bug bounty program can be found here (29-7-2021) up to $100k.

• From Decrypt (20-4-2020):

"According to a new blog post, the mainnet delay is due to the need for additional changes revealed by project’s internal protocol security audit."

Bugs/Exploits

• Had an error (19-12-2020) which led to seven hours of offline time according to Messari (31-3-2021). It was caused by a storage miner actor error. The simultaneous termination of multiple sectors caused nodes to process map entries in different orders and meant miners couldn't reach consensus.
• Had what Binance and Filecoin called, an 'API mistake' (20-3-2021):

"In emails to Decrypt, Filecoin and Binance are chalking it up to an “incorrect use” of APIs, as opposed to a bug. The transaction in question only happened once on the Filecoin blockchain, even though the exchange mistakenly accepted the transaction twice; the result was more like a “double deposit” than a double spend."

Token

Launch

• ICO goal was 40.000.000 but they had the biggest ICO ever and got 257.000.000$ in ETH.

Token allocation

• From Token Economy:

“Protocol Labs will receive all the ETH of the ICO PLUS 50% more coins than investors, so 1.5x. A foundation will receive 50% of the amount of coins “minted” by investors. Total: 66.6% to them, 33.3% to investors. Also, 70% of the tokens that will ever exist will be mined. This means that the investors are only getting access to 10% of the total supply ever.”

Tech

• Whitepaper can be found here (27-8-2020).
• Code can be viewed here.
• Programming language used, from LongHash Ventures (31-5-2022):

1. "Lotus (written in Go)
2. Fuhon (written in C++, implemented by Soramitsu)
3. Forest (written in Rust, implemented by Chainsafe)"

• Filecoin Virtual Machine (Filecoin VM) got announced (4-11-2021). Expected (31-5-2022) in September 2022. Almost ready as of March 2023. Got live on 14-3-2023.

Transaction Details

• Capacity (TPS):
• Latency:

How it works

• From a commissioned Messari report (13-4-2023):

"To keep data safe, Filecoin uses a cryptoeconomic incentive model that regularly verifies the storage with zero-knowledge proofs. To incentivize storage providers to participate in deals, Filecoin rewards them with FIL, the network's native token. Storage providers are also slashed in the event they either fail to provide reliable uptime or act maliciously against the network.

To retrieve data, Filecoin users pay a retrieval provider to fetch the data. Unlike storage deals, which involve transactions on-chain, retrieval deals use payment channels to settle payments off-chain, resulting in faster retrieval."

• From their docs (4-8-2022):

"Filecoin and IPFS are complementary protocols for storing and sharing data in the distributed web. Both systems are free, open-source, and share many building blocks, including data representation formats (IPLD) and network communication protocols (libp2p). While interacting with IPFS does not require using Filecoin, all Filecoin nodes are IPFS nodes under the hood, and (with some manual configuration) can connect to and fetch IPLD-formatted data from other IPFS nodes using libp2p. However, Filecoin nodes don’t join or participate in the public IPFS DHT (Distributed Hash Tables).

IPFS users usually persist the data they want on their own IPFS nodes. This is called pinning. Sometimes the data may be pinned using a third-party pinning service or through groups of individual IPFS users (like IPFS Collaborative Clusters). The data exists in the network as long as one user is storing it and can provide it to others when they request it.

IPFS alone does not include a built-in mechanism to incentivize the storage of data for other people. This is the challenge Filecoin aims to solve. Filecoin is built on IPFS to create a distributed storage marketplace for long-term storage. Nodes with a large storage capacity can rent their storage to users and get paid for it.

The Filecoin network ensures that data is safely stored. However, the processes of storing, verifying, and unsealing (referred to as sealing, proving, and retrieving, respectively) are computationally expensive and can take time. This is especially relevant for the retrieval of data, which should happen as fast as possible. For this reason, Filecoin enables an additional retrieval market where dedicated nodes can help quickly deliver content from the network for payment by keeping unsealed, cached copies. This delivery mechanism may make use of IPFS but is still in design. See ResNetLab’s Decentralized Data Delivery Market research for more information on delivery mechanisms.

Filecoin aims to add longer-term persistence to safely store large batches of data, while IPFS optimizes for the quick retrieval and distribution of content."

• Uses Proof-of-Replication (PoR); which allows proving that any replica of data is stored in physically independent storage. Introduces a novel useful-work consensus based on sequential proofs-of-replication and storage as a measure of power.
• Proof-of-Spacetime (PoSt); where a verifier can check if a prover is storing her/his outsourced data for a range of time. The intuition is to require the prover to (1) generate sequential Proofs-of-Storage (in our case Proof-of-Rep.
• From LongHash Ventures (31-5-2022):

"The Filecoin network serves as an incentive layer for storing & verifying the data on IPFS. $FIL is the native token of Filecoin. It's used to reward data storage providers for storing IPFS material. Data clients must pay in $FIL to both store and receive data on the network. Most apps using Filecoin utilize multi-tiered storage to speed up access/storage & securely store the data. IPFS pinning (saving files to IPFS) serves as hot storage layer which allows fast data retrieval. Filecoin network is used as cold storage for archival."

Upgrades

• From Messari (28-6-2022):

"The HyperDrive upgrade in July '21 benefited network participants through lower congestion and scaled throughput to 10-25x; however, it hurt the protocol’s revenue."

• Almost immediately after the launch, Filecoin introduced an upgrade to satisfy the miners (23-10-2020):

"The upgrade comes with the FIP-004 proposal – an implementation that increases a miner’s ability to reinvest the network's FIL token by making 25% of mining rewards available with no vesting. Previously, block rewards were doled out over a 180-day period; now, FIP-004 miners are able to receive 25% right away with the other 75% to be vested over the next six months."

• Will implement EIP-1559 on their next testnet (8-8-2020).
• From Decrypt (26-8-2020):

"The capacity of Filecoin’s “incentivized testnet” has exceeded 12 pebibytes (roughly 12.9 petabytes) in less than two days since its launch, according to the project’s website. Miners from around the world are being incentivized to participate as Filecoin (FIL) promised to hand out up to 4 million FIL tokens as rewards. Currently, the capacity of Filecoin’s testnet hovers just over 12 pebibytes that are split between 280 miners, with an average size of 46 tebibytes each."

• From Cointelegraph (10-6-2020):

"Filecoin has completed other trials prior to the testnet unveiling, the statement detailed. The project also transitioned into Proof-of-Replication, or PoRep, consensus, as well as implemented a number of other developments. The testnet trial holds as Filecoin's last experimental stage prior to its mainnet going live, which is expected to arrive in Q3, 2020."

Mining

• From their cryptoeconomics paper (27-8-2020):

"In the first iteration of the protocol, storage miners are the core service providers on Filecoin, with other types of mining in the future. Since Filecoin is a global network that everyone can use, the storage mining demographic is diverse – including individuals, organizations, and companies. Storage miners both compete and collaborate in the Filecoin Network; they compete with each other for storage deals and block rewards, and they collaborate with each other by growing the Filecoin Economy and maintaining the blockchain."

Staking

• From CoinDesk (19-10-2020):

"The miners are required to stake a large amount of FIL tokens as “Initial Pledge Collateral” to start their mining operations."

Scaling

• From Our Network (4-9-2021):

"Filecoin is the largest zk-SNARK network and recently had a 10-25x scalability improvement."

Other Details

EIP-1559

• From their blog (5-1-2021):

"Filecoin has incorporated EIP-1559 into its core protocol, with a small adaptation. Filecoin uses tipsets; its transactions go onto the chain before they are executed, and actual gas usage can’t be determined until that occurs. Filecoin therefore requires users to provide an estimated gas limit on transactions. To properly align incentives, overestimation above a certain margin is penalized via an overestimation burn.

EIP-1559 is implemented and live on the Filecoin mainnet. At present, around 100k to 150k FIL are being consumed per day as the result of network transactions. Incorporating the standard has delivered at least two key wins:

1. Fast Lane for High-Value Transactions. At times, storage onboarding messages have threatened to price out the more time-sensitive WindowPost messages. As expected, however, integrating EIP-1559 has provided Filecoin participants a straightforward mechanism for navigating this congestion without having to pay exorbitant prices all of the time.
2. Network Capacity Management. So far, the base fee mechanism of EIP-1559 has done very well at keeping network capacity at one hundred percent of its target.

As time has gone on, transaction creators have also gotten much better at estimating the amount of gas their transactions will require, leading to long-term reductions in overestimation burn.

There’s also been some room for improvement: base fee variance has been high, and its change rate has been spiky. This is due to a large amount of congestion as new storage gets onboarded. In the past, this has led to critical WindowPost messages being priced out as well.

There are a number of different approaches the Filecoin community could investigate to address these pricing issues:

1. Gas Control Plane. One possible means of addressing network congestion is creating a dedicated gas lane for control plane transactions, setting aside a fraction of every block to specific messages critical for the functioning of the blockchain.
2. Message Type Fee Structures. Another means of addressing this problem may be reducing the cost of WindowPost, or else increasing the cost of other messages, particularly Pre/ProveCommit messages.
3. Base Fee Change Rate. Spikiness in the base fee over time implies that the rate is not perfectly chosen, and could be smoother."

Oracle Method

• From CoinDesk (24-3-2021):

"Filecoin has integrated with oracle service Chainlink. The integration will enable connection between Filecoin and Ethereum and other smart contract-enabled blockchains."

Roadmap

• Can be found [Insert link here].
• From a commissioned Messari report (13-4-2023):

"The Q2’23 roadmap includes several network improvements as seen in the following FIPs:

1. FIP-0052: Increase max sector commitment to 3.5 years.
2. FIP-0057: Update gas charging schedule and system limits for FEVM.
3. FIP-0060: Set market deal maintenance interval to 30 days.
4. FIP-0061: WindowPoST Grindability fix.
5. FIP-0062: Fallback method handler for multisig and payment channel actors.

Down the road, longer-term improvements include L2 capabilities, hierarchical consensus, and Sealing-as-a-Service. Through FVM, Filecoin aims to build new and strengthen existing partnerships to better foster product growth within the data infrastructure community."

Revenue

• Increased to $4M during Q2 2024:

"While revenue from fees increased 145% in Q2'24 to $4 million, the 19% QoQ decline in average raw byte storage capacity is reflected by the rising penalty fees and batch fees. In this sense, penalty fees grew in Q2’24 to $3.9 million from $1.3 million in Q1’24.

Simultaneously, base fees from storage deals accounted for $9,000 in Q2’24, which is in line with the overall decline in demand-side fees charged to the users across the decentralized cloud storage space."

• From a commissioned Messari report (4-2024):

"While demand for Filecoin storage grew, revenue from fees increased 21% in Q1'24 to $1.6 million. The reduction in storage capacity is reflected by the rising penalty fees and batch feesNotably, base fees from storage deals accounted for only over $8,400, which is in line with the overall decline in demand-side fees charged to the users across the decentralized cloud storage space."

• Protocol revenue from storage fees decreased 68% in FIL terms in Q3'23 (down 73% in USD terms), in line with the overall decline in demand-side revenue across the decentralized cloud storage space. It was $3.8M in Q3. Further went down to $1.3M in Q4. Event though usage went up. Most of the fees came from penalties towards storage provider failures.
• From a commissioned Messari report (13-4-2023):

"As per Messari's revenue analysis, Filecoin's protocol revenue represents the sum of:

1. Base fees – determined by message congestion and required by any storage proof.
2. Batch fees – Used for bundling storage proofs.
3. Overestimation fees – Required to optimize gas usage.
4. Penalty fees – Collected for storage provider failures.

Protocol revenue from FIL fees increased 5% in Q1'23 to 1.3 million FIL (up 21% in USD terms to $6.9 million). While FIL base fees increased 20% QoQ, penalty fees decreased 17% QoQ. Even so, penalty fees were still ~2x higher than in Q2’22 and Q3’22. This may be explained by the early termination of sectors, potentially due to the FIL/USD price dislocations affecting the profitability of storage providers.

The only portion of FIL fees that isn't burned by the protocol is the “tip” collected by block miners. This mechanism is used to speed-up transactions on the supply-side of the network. Therefore, this “tip” counts as supply-side revenue. Block rewards account for more than 99.9% of supply-side revenue in Q1'23, while “tips” accounted for only a small portion. Supply-side revenue decreased 5% in Q1’23 to 18.9 million FIL (up 12% in USD terms). The decrease was driven by an overall reduction of FIL reward issuance due to exponential decay as well as the baseline minting model."

Usage

• TVL went up to a new high of $273M, but a look into the numbers shows it is purely due to LSTs (30-7-2024).
• From a commissioned Messari report (13-4-2023):

"As of March, 2023, 484 known projects were being developed on Filecoin, IPFS, and the Protocol Labs Network ecosystem, down from the peak of 632 projects in October 2022.

While the Filecoin ecosystem committed a total of $17 million in grant funding in 2022, the grants program focused on a few core initiatives around the adoption of the Filecoin Virtual Machine in Q1’23. As a result, the number of eligible applications has come down, i.e., 62 projects in initial stages March 2023, down from 258 at the peak in October 2022. Simultaneously, March 2023 saw 211 projects originating from accelerators, up from 194 in December 2022. The grants program will remain an integral part of driving ecosystem growth.

Most applications and protocols that leverage Filecoin offer data services:

1. Ocean Protocol: Developer tools and platform for data marketplaces.
2. Lighthouse: Perpetual data storage service with a one-time payment pricing model.
3. Slate: Search engine for handling and sharing personal data.
4. Berty: Secure messaging and social media application.
5. Dether: Cash on- / off-ramp and diverse financial transactions.

Media and entertainment-focused protocols include:

1. MoNA: 3D art gallery in the metaverse.
2. NFTwitch: NFT minting platform for Twitch content.
3. Huddle01: Decentralized video conferencing.
4. Curio: NFT marketplace for entertainment brands profiting off intellectual property.
5. OPGames: NFT minting from games.

Several use cases aim to leverage Filecoin’s infrastructure to power highly specific data needs:

1. Koios: No-code data DAO platform.
2. ZKsig NFTs: Access control to marketplace.
3. DataMarket: Data purchase and checkout functionalities."

• From LongHash Ventures (31-5-2022):

"Total network storage capacity - 17.087 EiB, Utilized storage capacity - 97.6181 PiB, Total Storage Deals - 4,079,535, Active Miners - 3959"

• From Our Network (4-9-2021):

"Since its launch, Filecoin has reached 10EiB in capacity, more than 3000 storage providers & 10,000 developers building applications on Filecoin. The active usage of the protocol has generated more than $1.4B in protocol revenue."

• From Decrypt (20-11-2020):

"The decentralized storage network hit 1 exabyte of capacity. That's much higher than estimates of the size of the Netflix library. The next step is getting applications to join the network."

Projects that use or built on it

• Livepeer. From Our Network #51 (18-12-2020):

"On the web3 side, the file.video integration between Filecoin and Livepeer serves as a template for a decentralized Youtube, and allows users to upload videos to be ingested and transcoded by Livepeer, and stored on Filecoin for long term access and playback."

• From a commissioned Messari report (4-2024):

"Teams from Solana are working to archive Solana’s block history on Filecoin. Once completed, Solana data will become more accessible to infrastructure providers, explorers, and data indexers. Solana will benefit from higher redundancy and enhanced security. Additionally, the price feeds of Solana’s leading oracle, Pyth, are now available on Filecoin."

Old Faithful, the project to store the Solana ledger on Filecoin was announced to have gone live (21-9-2024).

Competition

• It has a detailed comparison in its docs (4-8-2022) with Amazon and Google Cloud.

Filecoin Forks

• From Decrypt (18-9-2020):

"MIX Group, a Filecoin mining company, plans to fork, Weiping Han, MIX’s chairman said during a press conference in Xiamen earlier today. Han added that he had the support of many Filecoin miners and influential VCs such as Jun Du, co-founder of Huobi and founder of Node Capital.

Han is not the first one to propose a Filecoin fork. An anonymous Filecoin fork project entitled Filecoin Vision was released on Github earlier in China today. Its proposed “new rules” are basically the reversal of Protocol Labs’ research paper.

• Althought forks did not gain traction, protests did continue, with some big miners 'striking', and other simply not running due to them not having enough FIL tokens to start mining (19-10-2020):

"The miners are required to stake a large amount of FIL tokens as “Initial Pledge Collateral” to start their mining operations. “All the miners have been off since the mainnet went live, this is not some sort of protest but we have to shut them down because we really don’t have the tokens as collateral to mine,” ST Cloud CEO Chuhang Lai said in the report. 

In response to miner’s concerns, Filecoin has decided to release 25% token rewards in advance once a miner builds a block on the blockchain. “The revision could enable 80% of our mining capacity,” Xiaoming Zhan, CEO at IPSFMain said."

• Almost immediately after the launch, Filecoin introduced an upgrade to satisfy the miners (23-10-2020):

"The upgrade comes with the FIP-004 proposal – an implementation that increases a miner’s ability to reinvest the network's FIL token by making 25% of mining rewards available with no vesting. Previously, block rewards were doled out over a 180-day period; now, FIP-004 miners are able to receive 25% right away with the other 75% to be vested over the next six months."

Pros and Cons

Cons

• From Decrypt (18-9-2020):

"Filecoin miners, a majority of whom are based in China, have been grousing that Filecoin’s mainnet launch has been delayed. (The blockchain was originally slated for a March start.) The miners have further complained that Filecoin has been unresponsive to community concerns, and is seemingly centralized in its governance.

More specifically, Han listed a number of miners’ technical grievances ranging from hardware restrictions to cryptography and transactional bandwidth. But the most important factor, perhaps, was what he labeled Filecoin’s “harsh” reward system to miners.

According to Han, a research paper published by Protocol Labs on August 27, Filecoin’s economic model could wash out as many as 80% miners after the mainnet launch. That’s because miners must now provide an “Initial Pledge Collateral,” equivalent to 20 days worth of block reward and 30% FIL circulating supply target. These are capital and opportunity costs that many miners cannot afford, which irks the miners.

Second, the protocol will only allocate 30% of its storage mining allocation in “simple minting,” and the remainder in “baseline minting,” meaning miners can only harvest the initial 30% of their reward while waiting for the rest to be released linearly in the next 180 days. For miners who borrow to mine Filecoin, such a slow return mechanism could bankrupt them, Han said.

Lastly, Filecoin will implement a severe penalty for miners who commit technical errors or fail to follow processes such as submitting the certification of their storage commitment."

• From Glassnode (22-10-2020):

"On top of the system's design struggles, the emerging scandal surrounding the early release of additional tokens certainly hasn't helped Filecoin's public image."

Team, Funding, Partners

Team

• Full team can be found [here].
• Created by Protocol Labs
• It is often mentioned that Filecoin has an experienced core team, but it is difficult to find a lot of information except direct github commits.
• Juan Benet; founder of both IPFS and Filecoin (under the same company). IPFS needs an incentive system build in for it to be truly usable and useful, filecoin is building that incentive system.
• Has two foundations (12-2020) working on growth, The Filecoin Foundation and the Filecoin Foundation for the Decentralized Web.

Funding

• From this post:

“Filecoin counts, among its many accredited investors, a few big names in venture capital who have never actually been publicly named by Protocol Labs, but have been reported to include Winklevoss Capital and other investors in Protocol Labs itself. These early investors, dubbed as advisors, were able to purchase Filecoins at a price of 75 cents, whereas outside investors were given a starting price of $1. The average price for Filecoin was $2.65 in the first hour of the token sale to outside investors.”

• Has investment from Consensus Capital ("we’re a partner who supports their growth.")

• From CoinDesk (11-12-2020):

"BlueYard Capital and Union Square Ventures led [a] round, kick-starting half a decade of development."

• From Paradigm (10-12-2020): "Huobi Group and Protocol labs have partnered to launch the Huobi Filecoin Incubation Center with a $10M USD fund."
• The Fenbushi Ecosystem Fund for Filecoin Development has been launched (11-1-2021).

Partners 

• Ethereum Foundation partners with Filecoin to build ASICs that support the upcoming Ethereum 2.0.
• Has a partnership with Decrypt 18-3-2021).

(:

Knowledge empowers all and will help us get closer to the decentralized world we all want to live in!

Making these free wiki pages is fun but takes a lot of effort and time.

If you have enjoyed reading, tips are appreciated :) This will help us to keep expanding this archive of information.

ETH tip address: 0x83460bE5F218b1520B69D702cE60A1DE37dD8E31