PANews reported on Oct. 28 that columnist Lao Bai sees a better fit for oracle-driven PMM, or Proactive Market Maker, in the RWA market.
His argument is that prices for stocks and other real-world assets are already discovered in external markets such as Nasdaq, so an on-chain DEX is better suited to act as a price taker. If an AMM is still asked to discover stock prices on-chain, he wrote, that only adds noise and arbitrage costs. PMM, by contrast, can reference external prices and concentrate liquidity around the market price, which in his view makes it more suitable for RWA than it was for crypto assets.
Why DODO did not beat Uniswap
Lao Bai revisited DODO’s earlier attempt to challenge Uniswap and said there were two main reasons it fell short during DeFi Summer.
First, prices for major crypto assets were formed on centralized exchanges such as Binance, while oracle updates lagged behind CEX pricing. That gap gave arbitrageurs a window to trade against LPs.
Second, the main battlefield of DeFi Summer was long-tail tokens. Many of those assets had no CEX market and no oracle support, so AMMs had to handle price discovery on their own. PMM was not naturally positioned to cover that segment.
Based on that comparison, he argued that DODO’s design was somewhat mismatched for crypto at the time, but looks more appropriate when applied to RWA.
Why RWA changes the setup
The article said an earlier version of the market thesis involved pairing meme assets with stocks and even producing hot names such as AMC, along with the idea that on-chain trading could feed back into price formation in the real world.
Lao Bai said that idea now looks more like a dream. Even if it worked, he wrote, it would likely affect only a few obscure small-cap names rather than major stocks such as the so-called Magnificent Seven.
His updated view is that assets such as AAPL already have their prices discovered on Nasdaq, and an on-chain DEX should simply take that price. Letting a liquidity pool discover stock prices by itself is, in his words, wrong at the level of first principles. In that framework, the AMM price on-chain becomes little more than noise plus arbitrage cost.
Metric: liquidity moves with the oracle price
Lao Bai said two DEX projects recently caught his attention: Metric and NoteSystems. What they share is the use of external oracle prices instead of the traditional x*y=k AMM model associated with Uniswap.
Metric separates pricing from liquidity. Pricing is handled by oracles such as Chainlink and Pyth, while the pool itself is responsible only for supplying liquidity.
Under that design, LP liquidity is not posted in a fixed price range. It is placed at positions defined by a percentage deviation from the reference price. When the reference price moves, the entire liquidity band shifts with it, so the position does not fall out of range. Lao Bai compared the user experience to Uni V3 or Meteora and said it removes the need for constant manual rebalancing.
He also noted that Metric uses dedicated oracle inputs for stock-specific events such as dividends and stock splits, so prices do not drift after those events. Robinhood stock tokens were cited in that context.
He summed up the distinction this way: DODO concentrated liquidity near the oracle price, while Metric makes liquidity move with the oracle price.
NoteSystems: stale-price handling and market-hours logic
NoteSystems takes a different route. Lao Bai described its S-DAMM swap engine as something close to a full-form version of DODO PMM.
According to the article, S-DAMM keeps three elements already seen in DODO: a reference price, inventory adjustment, and single-sided LP. It then adds a time dimension.
If the price has not updated for about 5 seconds, confidence is cut in half, quotes widen automatically, and depth shrinks. Lao Bai said one weakness in DODO’s earlier setup was that it had no clear way to judge how stale its own price had become.
NoteSystems also includes market-closure logic tailored to U.S. equities. It withdraws regular inventory 60 minutes before the close. During nights and weekends, it uses a dedicated Continuous inventory with wider quotes. If there is no fresh price at all, the system enters Discovery mode. When the market reopens, it revalues against the live price before taking new orders.
On LP risk, the design is layered. The structure includes a Senior USDG vault and a junior Backstop that absorbs losses first. Lao Bai contrasted that with the more direct LP exposure seen in the DODO era.
Current status
The article said Metric is already running on mainnet and has users participating in farming. NoteSystems, by contrast, appears to still be at the testnet stage.
Lao Bai added that Robinhood Chain Infra may be moving into its next phase. The piece was published as the opinion of a PANews contributing columnist and does not represent the outlet’s position or investment advice.


