DeFi Archaeology: The History of DeFi Development Before Uniswap

Author: 0xKooKoo, Geek Web3 & MoleDAO Technical Advisor, Former Technical Lead at Bybit

Note: This article is the author's current archaeological research on DeFi, which may contain errors or biases, and is for communication purposes only. Feedback is welcome.

Introduction

Most people's exposure to DeFi originated from the DeFi summer of 2020. The sudden popularity of DeFi can be attributed to several reasons:

No reliance on third parties. Like Bitcoin, DeFi does not require any third-party involvement (except for Oracles). Users only need to connect a crypto wallet and sign transactions to conduct transactions on the entire chain. As long as the smart contracts are secure, no one can take away the user's assets. Not your key, not your coin. Those who have experienced the Mt. Gox hack and the new tragedies of players affected by FTX misappropriating user assets can better understand this feeling of distrust.

Increased market demand. Before DeFi Summer, there was a huge demand for liquidity globally. The low interest rates in the traditional financial system and the global liquidity easing policies led to funds seeking higher yield opportunities. DeFi provided a viable alternative, attracting a large influx of funds with higher interest rates and more investment opportunities.

Better protection of personal privacy. DeFi does not require extensive KYC or only requires minimal KYC. DeFi platforms are built on blockchain technology, executing transactions and protocols through smart contracts. Unlike traditional financial institutions, DeFi does not have a centralized management body or intermediaries; instead, it is automatically executed by code and protocols. This decentralized nature makes it impossible for DeFi platforms to directly collect and manage users' personal identity information, thus avoiding the common KYC procedures found in traditional financial institutions. There are indeed many pure on-chain alpha opportunities, and those who can seize these opportunities are professional players who do not wish to expose their strategies and personal information, making DeFi the best choice for this group of players.

Lower barriers to entry, Permissionless. DeFi has indeed addressed some issues and shortcomings of the traditional financial system to a certain extent. For example, anyone can list their token on Uniswap, greatly increasing the breadth of trading. As long as there is a trading demand for a certain token, it can be satisfied on DeFi without waiting for a centralized exchange to go through rigorous reviews to select tokens.

Code auditability. DeFi projects are usually open-source, and their smart contract code can be audited and verified by anyone. This openness and transparency allow people to inspect the code to ensure there are no hidden malicious behaviors or risks. In contrast, traditional financial institutions have relatively closed-source backend programs, making it impossible for people to directly audit their internal operations.

High interoperability. Different protocols and platforms in the DeFi ecosystem can connect and cooperate with each other, forming a seamless financial network. Because of this, the DeFi community tends to maintain principles of openness and interconnectedness to promote more innovation and development.

However, DeFi also has some issues:

Lack of liquidity. Compared to the liquidity of centralized exchanges, DEX still has a lot of room for improvement. According to the latest data from theblock.co on October 16, 2023, the spot trading volume of DEX in the past month was only 13.45% of that of CEX. Additionally, the lack of liquidity leads to significant slippage in transactions. For example, spending 1500 USDT on a CEX can buy 1 tokenA, but in a poorly liquid on-chain liquidity pool, the same 1500 USDT can only buy 0.9 tokenA, resulting in a 10% drop in value for the transaction.

High transaction fees. DeFi transactions are conducted on-chain, so they are subject to the performance and storage capacity of the underlying public chain. For instance, transaction fees on Uniswap may surge due to congestion on the Ethereum mainnet. I have previously experienced a regular transaction that incurred a fee as high as 200 USD, which is quite discouraging.

Fewer features. Compared to the extensive services offered by centralized exchanges, such as grid trading, dollar-cost averaging bots, wealth management products, etc., the current offerings in DeFi are still very basic and fragmented, mainly consisting of simple swap transactions, liquidity mining, staking, and farming.

Poor user experience. The interactive experience of DeFi is much worse than that of mature CEXs. For example, it may take several seconds to complete a transaction, the signature content is not straightforward, terminology is inconsistent, and product flow logic is not smooth. However, this issue is relatively manageable because as standards gradually unify, many frontend codes and product logics can form mature and user-friendly templates, making various platforms quite similar.

The Past: A History of DeFi

It can be said that since the day BTC was born, people have hoped to trade in a decentralized manner, leading to a continuous stream of on-chain financial innovations. Due to the limited programmability of BTC, people did not think too much along this path until Ethereum emerged, opening up new possibilities and leading many projects to raise funds through ICOs.

With the establishment of the ERC20 protocol, the flow of on-chain assets became more abundant, and a series of financial innovation products emerged.

Now, let's dig into the past and see the difficult journey DeFi has gone through, as well as the remarkable innovations made by various products and figures.

The earliest discussions about decentralized finance can be traced back to July 2013, when Mastercoin founder JR. Willett initiated the first ICO on the bitcointalk forum, stating that only those who contributed would enjoy new features such as decentralized trading and distributed betting built on top of Bitcoin. This allowed him to successfully raise 4,740 bitcoins, valued at 500,000 USD at the time.

In 2014, Robert Dermody and others co-founded Counterparty Protocol, a peer-to-peer financial platform and distributed open-source network protocol built on the Bitcoin blockchain.

The problem it solved was: allowing users to create their own tokens on the Bitcoin blockchain. Counterparty has a native currency called XCP, which is produced from Bitcoin through a "proof of burn" mechanism.

Counterparty provides financial tools that Bitcoin cannot offer, such as derivatives. Counterparty was used by Overstock.com to trade legal securities on the blockchain. It also created a decentralized asset exchange where various digital assets could be traded. Users can use the counterpartyd node software and Counterwallet web wallet to interact with Counterparty.

Counterparty also implemented features similar to smart contracts and dapps on Bitcoin. It provided an open-source and decentralized platform for conducting financial activities without relying on any central authority. Several well-known NFT projects, such as Spell of Genesis and Rare Pepe, were built on the Counterparty platform.

Overall, the Counterparty protocol utilized the Bitcoin network and technology to solve financial products and services that Bitcoin itself could not provide, making it a more comprehensive decentralized finance platform. To this day, the Counterparty protocol still exists and is one of the oldest and most famous decentralized finance (DeFi) platforms.

On September 15, 2015, Gnosis founder Martin published thoughts on how MarketMaker and OrderBook could be combined on his forum, which is the earliest post I found regarding decentralized prediction markets.

Gnosis is a decentralized prediction market built on the Ethereum protocol, providing an open platform for predicting the outcomes of any event, greatly simplifying the creation process of customized prediction market applications. At the same time, Gnosis leverages the trust machine of blockchain and the automatic execution of smart contracts, allowing players to enter the prediction market more flexibly and freely, bringing greater imagination to prediction markets.

By the way, this Martin is quite impressive; later projects like Gnosis Chain (formerly xDai Chain), Balancer, SAFE wallet, and CowSwap are all related to him.

On October 27, 2015, Gnosis founder Martin initiated another discussion on his forum, about how to provide a certain amount of funding as initial liquidity for newly created Prediction Topics to ensure the normal operation of the market.

For example, funding could be provided through project sponsorship or by collaborating with other investors or foundations to obtain financial support. This post emphasized the importance of community participation. It can be said that this is the earliest discussion I found about how to attract more liquidity and participation during my archaeological research.

On September 26, 2016, Nick Johnson, Chief Developer of Ethereum and ENS, posted on Reddit proposing a decentralized exchange concept called Euler. The main content included:

Euler would allow users to purchase Euler coins using different types of tokens. Euler holds these tokens, and the number of tokens determines how many Euler coins a user can exchange. The first Euler coin requires 1 token, the second requires e tokens, the third requires e^2 tokens, and so on, with each Euler coin's price calculated exponentially.

When new tokens are added, a solicitation phase needs to be run, where users can submit bids to provide new tokens in exchange for Euler coins. Finally, the initial price of the new token is determined. The total value of Euler coins should equal the total value of all tokens held by Euler. This can somewhat resist the impact of individual token price fluctuations on its value.

At the same time, a mechanism should be established to quickly stop buying a damaged token to prevent the abuse of that token to redeem other tokens. Overall, this system design is simple and decentralized, but there are some economic impacts that need further investigation.

The Dawn of AMM

On October 3, 2016, Vitalik posted on Reddit, inspired by Nick Johnson, referencing some emerging DEXs at the time, proposing a new method for using decentralized exchanges:

Using a "chain-based automated market maker" mechanism similar to prediction markets to operate decentralized exchanges, without the need for order placement and cancellation like traditional exchanges.

Users can "invest" in this market maker to increase depth and earn profit shares, which also reduces the risks for market makers. Compared to traditional exchanges, this method can significantly reduce spreads, requiring on-chain transactions only during actual trades, without the need for order placement and cancellation. It also raised the issue of how to add new tokens and stop buying when price fluctuations are too large. Subsequent discussions included how to support multiple assets and considerations for fees when investors enter and exit.

This post can be said to have laid the foundation for AMM-type DEXs, opening up a multi-billion dollar market.

In June 2017, EtherDelta officially launched and became the first Ethereum decentralized exchange recognized by regulatory authorities, as it had completed the registration procedures with the U.S. SEC even before its launch.

However, as early as June 23, 2016, EtherDelta founder Zachary Coburn (referred to as Zack) had already submitted the first commit on GitHub. EtherDelta was the first decentralized exchange to register with the U.S. CFTC (Commodity Futures Trading Commission).

Overall, the reason EtherDelta became the first Ethereum DEX in 2017 was mainly due to its realization of a purer decentralization, low barriers to entry, strong anonymity, low costs, and stable performance. The technical principles of EtherDelta are as follows:

It uses smart contracts to implement an order book trading system. Users publish, cancel, and match buy and sell orders through trading contracts. Order book information and transaction records are stored on the Ethereum blockchain, achieving decentralized trading that can be accessed via the web or mobile without the need to download a dedicated application.

The Delta website interacts with EtherDelta's smart contracts through JavaScript, reads order book information, and trades with counterpart users. When users publish or cancel orders, they need to broadcast transactions to the Ethereum network and pay gas fees. When a counterparty clicks on an order, the trading contract automatically deducts the buyer's assets and sends them to the seller, realized on-chain.

Smart contracts record each transaction, including the involved account addresses, types, and quantities of tokens traded. User assets are always kept in their own wallets and are not controlled by EtherDelta's services. EtherDelta charges a 0.3% transaction fee, fully borne by the buyer. The entire trading process ensures decentralization and transparency but relies on the performance of the Ethereum network.

EtherDelta also had some drawbacks at the time:

Manual operation was required during order matching. Traders needed to search for orders on the website themselves to see if they met their needs. Once they found a suitable order, they still needed to manually match it with the counterpart's order. This meant that both parties needed to reach a consensus on the price at the same time. In summary, the entire process required manual operation and could not be automated.

Slow order matching speed. After a user placed an order, it might take a long time to execute because Ethereum's processing speed was not fast, and liquidity was also weak.

Wasted gas fees. Due to the high latency of EtherDelta's order book, some takers might overlook each other's orders. This could lead to multiple takers competing to execute the same maker order, resulting in order failures and delays, with all takers except the winning one wasting gas fees.

Later, EtherDelta faced some criticisms, such as the former CTO being accused of insider trading. For specifics, see the lawsuit issued by the U.S. SEC on November 8, 2018. The report concluded that certain digital assets (such as ERC-20 tokens) are considered securities and thus subject to SEC regulation. The SEC stated that all platforms trading such assets need to register as securities exchanges, but EtherDelta did not do so.

Although Coburn did not formally confirm or deny the SEC's allegations, he agreed to settle with the regulators, paying $300,000 in forfeited funds, a $75,000 fine, and $13,000 in pre-judgment interest. To establish Zachary Coburn's personal responsibility, the SEC demonstrated that:

EtherDelta violated securities laws, and Coburn caused EtherDelta to violate the Securities Exchange Act, knowing or should have known that his actions would lead to EtherDelta's violation of securities laws.

EtherDelta was somewhat unfortunate; it registered with the CFTC in the U.S. but did not register with another important U.S. regulatory agency, the SEC. EtherDelta registered with the CFTC mainly because it traded primarily in digital currencies rather than financial securities. However, the SEC later issued guidance classifying many tokens as securities, so theoretically, EtherDelta should also have registered with the SEC. At that time, the SEC's regulatory stance on blockchain innovations was not clear, and EtherDelta did not proactively register with the SEC.

There were some dramatic stories of team conflicts within EtherDelta, such as a fork resulting in ForkDelta, and even due to centralized equity disputes, it became the first decentralized exchange to exit the market.

The approximate timeline is as follows:

In early 2018, the EtherDelta founding team sold the platform to Chinese businessman Chen Jun. According to a leaked document dated December 15, 2017, EtherDelta underwent a share transfer and prepared to start raising ETH (Ethereum) in the market.

On February 9, 2018, the team issued a statement saying that EtherDelta was undergoing a technical upgrade. On February 18, the exchange was reported by the media to have suspended trading. On February 19, the founding technical team, after selling the EtherDelta platform, forked the EtherDelta project and launched a new platform called "ForkDelta."

On February 21, 2018, EtherDelta suspended trading again, and the actual controller, Chen Jun, was reported to have fled.

The AMM Era Officially Begins

Bancor Protocol launched on June 12, 2017, raising $153 million through ICO. The most important innovation of Bancor was the introduction of the AMM mechanism into the decentralized exchange space, addressing a series of challenges in decentralized trading, which effectively laid the foundation for AMM applications within the Ethereum ecosystem. Unlike the traditional order book matching method for buying and selling orders, Bancor used liquidity pools to solve the pricing and matching issues of decentralized exchanges, allowing users to trade without waiting for counterparties.

On September 29, 2017, IDEX, co-founded by brothers Alex Wearn and Philip Wearn, officially launched its beta version, although its project source code was first uploaded to GitHub in January 2017.

2017 was the peak of the ICO bubble, with various ICO projects emerging, but most projects had varying quality and chaos. As the ICO market cooled, holders of various tokens began seeking ways to trade them. However, the mainstream exchanges at that time were not decentralized and posed risks controlled by third-party institutions, which provided an opportunity for IDEX.

It mimicked the previously established Counterparty protocol on Bitcoin, implementing the first generation of decentralized trading functionality on Ethereum. Users can trade various Ethereum and ERC20 standard tokens through IDEX, avoiding the need to trust third-party organizations and institutions.

IDEX focuses on:

Speed. IDEX uses offline order book matching, making it faster than EtherDelta, with a user experience more akin to that of a centralized exchange.

High security. Its core is smart contracts, and user assets are not controlled by intermediary institutions, reducing risks.

Comprehensive functionality. It supports instant cancellation of unfilled orders (which is free since it is canceled off-chain), market trading, and is easy to operate.

Support for multiple tokens. At launch in 2017, it already supported over 200 ERC20 tokens, offering good selection.

Low transaction fees. The transaction fee is 0.3%, which is relatively lower compared to other decentralized exchanges.

High anonymity. IDEX did not require real-name authentication at launch, making it suitable for users seeking privacy.

However, at that time, the entire DEX space was just starting, with low trading volumes. In 2017, the total trading volume was only around $50 million. Although IDEX was quite popular at the time, the trading volume was still very low. This proved that the decentralized exchange products and ecosystem were still immature and needed continuous enhancement of products and user experience. On November 8, 2018, an article summarized that IDEX was then firmly in the top position among DEXs.

MakerDAO (launched in December 2017)

The main innovations of MakerDAO include:

Low volatility: By introducing the stablecoin Dai, MakerDAO provides a cryptocurrency pegged to the U.S. dollar, allowing users to trade and store value in the cryptocurrency market while reducing the risk of price volatility.

Weak centralization risk: Traditional centralized stablecoins are issued and supported by centralized institutions, leading to trust and risk concentration issues. MakerDAO's decentralized model avoids the risks of a single centralized institution through smart contracts and collateralized assets, allowing users to participate in and control the system directly.

Transparency and autonomy: MakerDAO adopts a decentralized autonomous organization (DAO) model, enabling holders of MKR tokens to participate in decision-making and governance of the platform. This model increases the system's transparency and community participation, enhancing the fairness of decisions and the reliability of the system.

Kyber Network (launched on February 26, 2018)

The main innovations of Kyber Network include:

Instant exchange: Kyber Network allows users to directly exchange tokens instantly without the need for an exchange. Users can trade directly through Kyber Network's smart contracts without having to buy and sell on centralized exchanges.

Decentralized liquidity pools: Kyber Network introduced decentralized liquidity pools, aggregating funds from multiple participants to provide a deeper and more liquid market. These liquidity pools are provided by users holding tokens and managed through smart contracts.

Best price execution: Kyber Network automatically selects the best price and liquidity sources to execute trades through smart contracts. This means users can obtain the most favorable exchange rates without needing to compare and choose between multiple exchanges.

Flexible integration: Kyber Network provides open APIs and smart contract interfaces, allowing other decentralized applications (DApps) and services to seamlessly integrate and utilize Kyber Network's liquidity.

0x Protocol (launched in May 2018, raised $24 million through ICO)

The main innovations and problems solved by the 0x protocol include:

Providing an open-source decentralized trading protocol and API, supporting DApps to develop on top of it, lowering development barriers and integration costs. 0x positions itself as the "settlement layer" for decentralized trading. It is not a facilitator of trade but rather an infrastructure on which any type of marketplace can be built, such as eBay, Amazon, order book DEXs, and even traditional financial giants' familiar order flow granularity and control.

Supporting trading of any ERC20 tokens, not limited to just two tokens. It adopts an economic incentive model based on the governance token ZRX. It provides a unique 0x Mesh network connecting various relayer nodes.

The 0x protocol built the consumer-facing DEX aggregator Matcha, which uses the 0x API and smart order routing to aggregate liquidity and provide optimal trade execution. Subsequently, other DEX aggregators also emerged, benefiting from aggregating on-chain liquidity, akin to wholesalers sourcing from different factories and then selling at a markup.

Compound (launched in September 2018), TVL first broke $100 million in 2019

The main innovations of Compound include:

Introducing digital asset lending to the Ethereum ecosystem for the first time, Compound is the first protocol to achieve cross-asset lending of ETH and ERC20 tokens.

No physical collateral is required; simply depositing digital assets into a smart contract allows for borrowing, greatly lowering the cost barrier to obtaining loans.

Interest rate market-driven mechanism, Compound adjusts the interest rates of different assets in real-time based on supply and demand, bringing the market toward balance.

Supporting lending of various mainstream stablecoins and tokens, such as USDC, DAI, etc., providing users with higher choices.

Borrowed assets can be used directly without delivery, simplifying the lending process, allowing users to repay loans and retrieve collateral at any time.

Providing open and non-custodial APIs, greatly promoting the application of lending services across DApps.

Implemented with simple-to-operate and easily auditable smart contracts, which is also an important reason for DeFi's global sweep.

Overall, Compound utilizes digital assets and blockchain technology to provide global users with convenient and efficient decentralized lending services. It addresses the cost efficiency and localization issues faced by traditional finance, opening a new chapter in DeFi development.

dYdX (launched in October 2018), with a peak TVL exceeding $1 billion. The main innovations and problems solved by the dYdX protocol are as follows:

Building a decentralized perpetual contract trading platform, allowing users to trade perpetual contracts on-chain, avoiding the risks and asset custody issues of centralized exchanges. By using a hybrid on-chain and off-chain order book, the off-chain order book improves trading efficiency, while the on-chain order book ensures transparency. Through the off-chain order book, dYdX can provide lower slippage and deeper liquidity, enabling high-frequency trading and low trading costs.

It allows users to participate in governance and earn miner rewards by collateralizing assets. It provides decentralized margin trading, supporting multiple assets, allowing users to achieve leverage trading of up to 20X. It supports overnight margin trading and isolated margin trading, allowing users to adjust their position margin rates according to their risk preferences.

(To be continued)