After SBF’s FTX platform thundered in dramatic fashion, the entire crypto market quickly plunged into panic.
Users panic withdrawals
Since Thunderbolt’s FTX platform belongs to CeFi (traditional finance), many users of CeFi platforms have taken stressful withdrawal behavior after the thunderstorm. According to data provided by data service provider Nansen, in the past seven days , users have withdrawn $3.7 billion worth of stablecoins from major platforms (including Binance, OKX, Kucoin, Huobi, Kraken, Coinbase, Bitfinex, etc.) on the Ethereum network alone.
Faced with the panic fleeing of user funds, major CeFi platforms were forced to resort to the old trick of 100% reserves in an attempt to restore user confidence.
So far, CeFi platforms that have announced their reserves are Binance, Crypto.com, OKX, Deribit, Kucoin, Bitfinex, and Huobi.
What these data can prove is only the strength of these platforms. The essence of CeFi means that users need to trust the platform not to do evil. This tests human nature. If the human nature of the platform controller (private key controller) is evil, then the platform is not trustworthy.
DeFi strengthens infrastructure
Following the collapse of the FTX platform, global regulators have reason to tighten their regulation of Crypto, especially for CeFi platforms like FTX.
In addition to supervision, the CeFi platform itself also needs to increase transparency. For example, Binance founder CZ revealed in a recent AMA that Binance will cooperate with Ethereum founder Vitalik Buterin to launch a new method of proof of reserves. For users, if a CeFi platform is unregulated and cannot monitor financial data, its credibility is low.
In the long run, the Crypto market ultimately belongs to DeFi, but the imperfection of the DeFi infrastructure has left the CeFi platform with an opportunity to grow itself. Specifically, CeFi can achieve a better user experience (fast, no gas fees for transactions, full functionality, no mnemonics, etc.), as well as faster product iterations to adapt to changes in market demand. It is for these reasons that over the past few years, CeFi trading platforms such as Binance have been able to achieve faster development than DeFi platforms.
In the next few years, CeFi will still be the choice of the vast majority of existing Crypto users and potential new users during the window period when DeFi infrastructure is still imperfect.
If the first wave of DeFi is due to overdrafting the future ponzi token economic design, the next wave is likely to be based on the improvement of infrastructure.
This includes the expansion of the underlying public chain (L1, L2, etc.), MEV improvement, account abstraction, improvement of cross-chain infrastructure, one-stop integration of DeFi platforms, decentralization of infrastructure such as Infura, Fiat Money import and export Enter and so on.
Ethereum ecosystem case
Below, we will use the most important Ethereum ecosystem as an example to illustrate.
(1) The expansion of the underlying public chain
First of all, what hinders the large-scale development of DeFi is the throughput of the underlying blockchain. On the current Ethereum L1 network, each block can accommodate about 100 – 300 transactions, and the average TPS is about 12.5. high, it will lead to soaring gas costs, which makes users pay expensive handling fees when conducting DEX and other transactions. For ordinary users, this extra cost is obviously unaffordable. To solve this problem, it is necessary to There are two main ways to expand the Ethereum network:
- Bottom layer L1 expansion, such as Proto-danksharding (EIP-4844), danksharding;
- L2 & L3, such as Optimism, Arbitrum, zksync, starknet, scroll, etc.;
For the first method (the underlying L1 expansion), short-term hope can be pinned on EIP-4844, whose development work probably started this spring. According to optimization developer Mofi Taiwo, the complexity of EIP-4844 is roughly between EIP-1559 and The Merge, although many community participants hope that EIP-4844 will be included in the next hard fork of Ethereum. Upgrading Shanghai, but discussions from developers during the 149th Ethereum ACD conference call made me a little concerned about it. In the fastest case, EIP-4844 will be introduced to the Ethereum network with next year’s Shanghai upgrade, and once delayed, this suggests that EIP-4844 may take longer to implement properly.
It is worth mentioning that the expected expansion effect of EIP-4844 on L2 has also been reduced from the initial 100 times to the current 20 times.
As for full danksharding, it may take years to land.
For the second method (L2 & L3), more attention needs to be paid to the risk factors of the rollup project itself. In Vitalik’s recently published article “Three Stages of the Rollup Layer 2 Network”, he mentioned that most of the current Rollup projects all use a temporary auxiliary round mechanism, which means that these networks are still exposed to various types of single point of failure risk.
And DeFi projects built on these two-layer networks will naturally inherit the risks of these networks themselves. Although Vitalik did not mention how long it will take for Rollup to implement the final phase, it is a guess that it may take another 2-3 years.
(2) MEV improvement
Due to the transparency of the Ethereum blockchain and the rules for confirming transactions in the protocol, currently on DeFi platforms such as Uniswap, users may encounter poor MEV experiences such as front-running transactions. It may lead to heavy losses for users. According to statistics, when users use Uniswap, only 1/3 of the fees are paid to liquidity providers, 1/3 is spent on gas fees, and the last 1/3 is lost On the MEV issue. Regarding this issue, either a protocol that protects users from malicious MEVs, or an osmosis-like Lisk approach is needed to address this issue.
(3) Account abstraction work
Currently, the vast majority of Ethereum accounts are external accounts (EOAs) controlled by private keys. Although they are created for free, users need to keep the private key or mnemonic and can only use ETH as the gas fee. It’s a disadvantage in terms of outreach to new Crypto users.
The smart contract wallet with an abstract account does not require users to manage mnemonics (using a social recovery mechanism), and users can choose to use ERC-20 tokens or even legal currency to pay for network gas fees, which is a kind of user experience. Very big improvement.
In terms of account abstraction, the current most noteworthy proposal is EIP-4337 ‌, which does not require any changes to the Ethereum consensus layer protocol and introduces the concept of a “paymaster” (payer), which means application development Providers can subsidize users’ network gas fees to better improve user experience.
However, the creation of smart contract wallets requires a certain cost, which makes it unrealistic to achieve large-scale expansion of smart contract wallets under the current limited throughput of the underlying network.
Therefore, the development of EIP-4844 and Rollup infrastructure will be very important for account abstraction.
(4) Improvement of cross-chain infrastructure
Due to the inevitable development trend of Crypto multi-chain in the future, cross-chain has become an integral part of DeFi, but due to its complexity, cross-chain has also become a hardest hit by hackers.
According to incomplete statistics, in the past few years, the cross-chain bridges attacked by hackers include Poly Network, Ronin, Wormhole, Horizon, Nomad, Binance Bridge, Multichain, QBridge, etc., with the highest loss reaching 620 million US dollars.
The reasons for the attack include smart contract vulnerabilities, private key leaks, Merkle tree vulnerabilities, etc. However, what has not yet appeared is the evil of the centralized cross-chain bridge and the economic attacks on weak chains that Vitalik is worried about.
As of now, most users are still using multi-signature cross-chain bridges, which require more trust and are easier to be attacked. The IBC protocol adopted by Cosmos is considered to be the most secure and mature cross-chain bridge. method, but it still has security holes in theory.
Going forward, we expect to see trustless ZK technology and infrastructure such as IBC become more mature and serve as a secure linker for the entire DeFi ecosystem.
(5) One-stop integration of DeFi platforms
Unlike the one-stop services such as spot trading, lending, derivatives trading, and wealth management provided by the CeFi platform, the services provided by the current DeFi platform are relatively simple. For example, the Uniswap platform only provides spot DEX trading services, Compound only provides lending services, and dydx Only leverage and derivatives trading services are provided.
For users, what is more needed is a one-stop service platform, rather than switching back and forth between various applications, which may be the next development trend of DeFi.
Final opinion
Due to the close relationship between Crypto and centralized fiat currencies, DeFi and CeFi will coexist for a long time, but will tend to use decentralized wallets to store their own Crypto assets and rely on relatively decentralized infrastructure to trade them, rather than relying on a hosted CeFi platform. Of course, when there is a demand for deposits and withdrawals, the CeFi platform is indeed more suitable, and supervision can bring some guarantees for the legitimacy of funds.
With the development of infrastructure such as public chain L1&L2, we can expect that DeFi will usher in a new round of explosive growth in the next 2-3 years.