Following Fusaka's upgrade in the past week, I am bearish on third-party data providers (DAs), considering Aave's changes, its breakup with Sky, its new partnership with Maple, the continued Perps trend, and HyENA. 1. One of the more significant data changes after the $ETH Fusaka upgrade is the noticeable increase in blob base fees. A detailed analysis post is here ? What's worth mentioning is the competitive landscape with third-party data providers (DAs). (1) Firstly, in terms of price competition Compared to third-party DAs, especially solutions like Celestia, Ethereum's native blob is definitely more expensive. This is determined by the underlying consensus of the blockchain and economic security. The core issue here is the trade-off between security and cost. Therefore, the future ecosystem will likely see professional L2, enterprise-level L2, or projects with relatively strong funding and ample budgets opting for blobs. Smaller teams and projects, with limited budgets and small initial scale, will certainly prefer third-party data analytics solutions. (2) So we need to make some judgments about the future. What kind of entrepreneurs and projects will primarily use L2 in the future? I believe it will most likely be geared towards professional needs and established project teams. Small teams, projects with low budgets, or even cold-start projects are more likely to choose to build a Protocol first rather than L2. Meanwhile, Ethereum will continue to scale L2 in the future. For professional teams, the trend is that the cost in this area will not be lower than that of third parties, but it will be negligible; in other words, it will reach a point where "there's no need to save money on this." From this perspective, I am not optimistic about the long-term development of third-party data analytics (DA). However, it will exist in the market and solve a small part of the needs, rather than becoming mainstream. 2. Aave proposed removing USDS and DAI as collateral, officially parting ways with Sky. At the same time, Maple's syrupUSDT was listed on Aave. Judging from the situation, it may be a long-term binding relationship similar to Ethena. This event foreshadows a future trend in DeFi: leading protocols will begin to expand their businesses horizontally, and unless there is a strategic long-term interest ties, most will turn from partners into competitors. For example, Aave has started to develop stablecoin business, and MakerDAO has started to develop the lending market. The composability of DeFi will gradually tend towards internal composability. This actually first appeared in the Curve ecosystem. Curve projects are usually highly bound together at the protocol and community levels, which has formed a culture. This phenomenon has both advantages and disadvantages, and it's a result of increasing commercialization. In the future, it may be a few large ecosystems dominating the market, each with its own prominent business, while also covering almost all DeFi tools and directions. This business isolation also isolates risks to some extent. The syrupUSDC/USDT pair on Maple currently offers around 6% yield, so it was quickly filled up as collateral on Aave. It's estimated that Maple will be permanently tied to Aave, becoming the second Aave TVL cash cow after Ethena. 3. The Perps trend continues (1) Ethena deployed its own Perps marketplace on Hyperliquid HIP3. (2) Rumors also suggest that the next version of Fluid dex will be Perps. Hyperliquid earns approximately $1 billion annually, making it one of the most profitable DeFi projects currently. Therefore, everyone wants a piece of the pie. The first wave was driven by new projects, with the market buzzing with activity. This current trend is more about established DeFi projects leveraging their existing brands to expand their revenue. Even assuming Hyperliquid's size is only 10%, the additional revenue would be substantial. (Ethena's revenue over the past year was around $400 million.) However, it is difficult to increase the market share in this area. It requires incentive measures. Without subsidies, it is almost impossible to poach talent from other places. I estimate that Ethena's HyENA must have some incentives. Considering that I might use it to generate some trading volume, the short-term returns should be pretty good, even if it's just buying trading volume with tokens. I think the problem with this sector is that it only has existing on-chain users. For example, someone like me who isn't very familiar with CEXs would prefer Hyperliquid. However, it's difficult to attract users within CEXs, especially those who rarely come to the blockchain. In the long run, penetration rate and user education are key factors in determining the overall ceiling of Perps. 4. Will MSTR sell its coins? MSTR announced the establishment of a $1.44 billion cash reserve to pay dividends. Regarding the issue of selling tokens, I think MSTR definitely doesn't want to sell. In the short term, having already begun building cash reserves, the possibility of selling tokens is very low. However, at the recent BBW event, Saylor mentioned that when there is a negative premium, they might sell BTC derivatives or BTC. I think this is to leave myself a way out. After all, to be serious, there is no such thing as a perpetual motion machine that never sells coins. So, if an extreme situation arises that requires selling coins, it will prevent myself and MSTR from being too passive. 5. My thoughts on Stable token issuance The unique aspect of the token economics design is the use of $USDT as the gas and settlement asset. $STABLE is used for staking and governance. Staking earns USDT rewards. The only advantage of this design is that it benefits USDT users, as they don't need to prepare additional gas fees, which is indeed convenient. My expectations are relatively low, primarily because various questionable practices occurred during the early stages of the project, raising concerns about the project's seriousness. Meanwhile, the overall goal of the project is to gain more USDT market share. However, Plasma has essentially proven that the existing USDT market is very fixed, making it difficult to compete for market share. Therefore, the key becomes the story both stablecoin projects have told: attracting more enterprise-level demand. Whether this can be achieved is a crucial factor, and everyone can assess it for themselves. If they cannot expand into the external market and cannot capture the existing market, they may have to rely on subsidies to survive. Once token rewards stop, there will be no more supply. 6. Other (1) Curve begins deploying in the foreign exchange market Swiss Franc <-> US Dollar; current liquidity is low. Adding to watch for future deployments and data. (2) Aerodrome repurchased 800K tokens To date, over 150 million AERO units have been acquired and locked in. New to watch: some AERO holdings were added during this recent market downturn. (3) Revert now supports Uniswap V4 I've been using this LP data tool for a while now; it's great.Following Fusaka's upgrade in the past week, I am bearish on third-party data providers (DAs), considering Aave's changes, its breakup with Sky, its new partnership with Maple, the continued Perps trend, and HyENA. 1. One of the more significant data changes after the $ETH Fusaka upgrade is the noticeable increase in blob base fees. A detailed analysis post is here ? What's worth mentioning is the competitive landscape with third-party data providers (DAs). (1) Firstly, in terms of price competition Compared to third-party DAs, especially solutions like Celestia, Ethereum's native blob is definitely more expensive. This is determined by the underlying consensus of the blockchain and economic security. The core issue here is the trade-off between security and cost. Therefore, the future ecosystem will likely see professional L2, enterprise-level L2, or projects with relatively strong funding and ample budgets opting for blobs. Smaller teams and projects, with limited budgets and small initial scale, will certainly prefer third-party data analytics solutions. (2) So we need to make some judgments about the future. What kind of entrepreneurs and projects will primarily use L2 in the future? I believe it will most likely be geared towards professional needs and established project teams. Small teams, projects with low budgets, or even cold-start projects are more likely to choose to build a Protocol first rather than L2. Meanwhile, Ethereum will continue to scale L2 in the future. For professional teams, the trend is that the cost in this area will not be lower than that of third parties, but it will be negligible; in other words, it will reach a point where "there's no need to save money on this." From this perspective, I am not optimistic about the long-term development of third-party data analytics (DA). However, it will exist in the market and solve a small part of the needs, rather than becoming mainstream. 2. Aave proposed removing USDS and DAI as collateral, officially parting ways with Sky. At the same time, Maple's syrupUSDT was listed on Aave. Judging from the situation, it may be a long-term binding relationship similar to Ethena. This event foreshadows a future trend in DeFi: leading protocols will begin to expand their businesses horizontally, and unless there is a strategic long-term interest ties, most will turn from partners into competitors. For example, Aave has started to develop stablecoin business, and MakerDAO has started to develop the lending market. The composability of DeFi will gradually tend towards internal composability. This actually first appeared in the Curve ecosystem. Curve projects are usually highly bound together at the protocol and community levels, which has formed a culture. This phenomenon has both advantages and disadvantages, and it's a result of increasing commercialization. In the future, it may be a few large ecosystems dominating the market, each with its own prominent business, while also covering almost all DeFi tools and directions. This business isolation also isolates risks to some extent. The syrupUSDC/USDT pair on Maple currently offers around 6% yield, so it was quickly filled up as collateral on Aave. It's estimated that Maple will be permanently tied to Aave, becoming the second Aave TVL cash cow after Ethena. 3. The Perps trend continues (1) Ethena deployed its own Perps marketplace on Hyperliquid HIP3. (2) Rumors also suggest that the next version of Fluid dex will be Perps. Hyperliquid earns approximately $1 billion annually, making it one of the most profitable DeFi projects currently. Therefore, everyone wants a piece of the pie. The first wave was driven by new projects, with the market buzzing with activity. This current trend is more about established DeFi projects leveraging their existing brands to expand their revenue. Even assuming Hyperliquid's size is only 10%, the additional revenue would be substantial. (Ethena's revenue over the past year was around $400 million.) However, it is difficult to increase the market share in this area. It requires incentive measures. Without subsidies, it is almost impossible to poach talent from other places. I estimate that Ethena's HyENA must have some incentives. Considering that I might use it to generate some trading volume, the short-term returns should be pretty good, even if it's just buying trading volume with tokens. I think the problem with this sector is that it only has existing on-chain users. For example, someone like me who isn't very familiar with CEXs would prefer Hyperliquid. However, it's difficult to attract users within CEXs, especially those who rarely come to the blockchain. In the long run, penetration rate and user education are key factors in determining the overall ceiling of Perps. 4. Will MSTR sell its coins? MSTR announced the establishment of a $1.44 billion cash reserve to pay dividends. Regarding the issue of selling tokens, I think MSTR definitely doesn't want to sell. In the short term, having already begun building cash reserves, the possibility of selling tokens is very low. However, at the recent BBW event, Saylor mentioned that when there is a negative premium, they might sell BTC derivatives or BTC. I think this is to leave myself a way out. After all, to be serious, there is no such thing as a perpetual motion machine that never sells coins. So, if an extreme situation arises that requires selling coins, it will prevent myself and MSTR from being too passive. 5. My thoughts on Stable token issuance The unique aspect of the token economics design is the use of $USDT as the gas and settlement asset. $STABLE is used for staking and governance. Staking earns USDT rewards. The only advantage of this design is that it benefits USDT users, as they don't need to prepare additional gas fees, which is indeed convenient. My expectations are relatively low, primarily because various questionable practices occurred during the early stages of the project, raising concerns about the project's seriousness. Meanwhile, the overall goal of the project is to gain more USDT market share. However, Plasma has essentially proven that the existing USDT market is very fixed, making it difficult to compete for market share. Therefore, the key becomes the story both stablecoin projects have told: attracting more enterprise-level demand. Whether this can be achieved is a crucial factor, and everyone can assess it for themselves. If they cannot expand into the external market and cannot capture the existing market, they may have to rely on subsidies to survive. Once token rewards stop, there will be no more supply. 6. Other (1) Curve begins deploying in the foreign exchange market Swiss Franc <-> US Dollar; current liquidity is low. Adding to watch for future deployments and data. (2) Aerodrome repurchased 800K tokens To date, over 150 million AERO units have been acquired and locked in. New to watch: some AERO holdings were added during this recent market downturn. (3) Revert now supports Uniswap V4 I've been using this LP data tool for a while now; it's great.