Opinion: Why support dYdX leaving ETH to build its own chain
The failure of Luna price does not mean the failure of Terra chain. Those who use Terra chain regularly must have experienced the smooth interaction experience brought by Terra. We can even say that among the public chains with TVL over $10 billion, Terra is the only one that has not experienced any downtime or delay. This is not because the technology of Terra chain is superior, but because the ecosystem is closed, just like IOS is smoother than Android. Compared to other leading chains with hundreds of APPs, Terra has only three native APPs until this year: Anchor, Mirror and Terraswap, which have built the entire ecosystem. And the whole ecosystem is at the service of UST.
Regardless of the failure of Luna and UST, we can be sure that an ecologically independent public chain does provide an incredibly smooth interaction experience. No one cares more about fluency than futures traders, and I think that’s the biggest reason to rebuild dYdX as a standalone blockchain (games are also looking for fluency, so some game developers are considering creating a new blockchain for their games as well).
Currently, dYdX’s trading experience is unmatched in DEXs, but there are still many gaps compared to the leading CEXs, all due to network scalability. It is true that technically speaking, Rollup can meet the needs of dYdX in the future, but how long do we have to wait for this future? Moreover, even if Rollup can work, does it make the transaction experience worse to enjoy an exclusive blockchain alone than to compete with hundreds of APPs for a block space?
There is no doubt that the security of dYdX will be reduced after leaving Starkware. DYdX has not encountered theft of funds yet, but the probability of such situation will increase significantly after independence. However, since the entire chain serves dYdX, a blockchain rollback (soft fork) is enough to fix the bug.
For future vision, you can refer to Thorchain (RUNE), where the whole chain has only one APP, Thorchain DEX. When users add liquidity to DEX, they must add RUNE of equal value as a transaction pair, and the node needs to stake RUNE of twice the value. In other words, $100 million of liquid assets will lock up $300 million of RUNE. If dYdX did the same, it would certainly increase the value the token DYDX significantly.
Choice of blockchain
As for why Cosmos was chosen over others, there is no clear official explanation, only that Cosmos is the best choice at the moment. In fact, it is not hard to understand.
Since we choose to develop a separate chain for an APP, the chain must have cross-chain functions and interoperability. The “cross-chain” here is not a third-party cross-chain bridge, but a blockchain like Polkadot that allows all parallel chains to be compatible with each other. In this area, the only ones that have developed a strong consensus are Cosmos and Avalanche, in addition to Polkadot.
Polkadot is similar to Cosmos in that one is developed using the Substrate SDK and the other is developed using the Cosmos SDK. Avalanche’s P-chain is similar in that you can freely create sub-chains on it, but so far Avalanche has not yet implemented interoperability of sub-chains either. The difference between Polkadot and Cosmos is that Polkadot can achieve global security, but lacks flexibility, so the development progress is slow. The security of each chain of Cosmos network can only be independently responsible for their own, but the advantage is more flexible. Therefore, for dYdX, which is planned to be rebuilt by the end of the year, Avalanche cannot be counted on, and Polkadot is too demanding for security and cannot be realized in a short time, so we finally have to choose Cosmos.
Personally, I think this is a good start (actually Luna was a good start, but unfortunately it was destroyed for other reasons), and more and more APPs will choose to develop a public chain independently in the future. They can choose Rollup to connect to Ethereum, Substrate SDK to connect to Polkadot, or Cosmos SDK to connect to Cosmos like dYdX. Cosmos SDK is the least difficult to develop, Substrate SDK can balance the difficulty of development and network security, Rollup is secure but the most difficult (I haven’t heard of any APP that intends to develop a Rollup alone).
In the past, Ethereum, which had a thriving ecosystem, was able to develop Rollup slowly and without worry. But once the patience of well-funded APP teams runs out (and Antonio’s frustration with Rollup is evident), Ethereum teams don’t have much time left.