Haru

Unpopular opinion: engagement isn't a bad metric. It's just an incomplete one. Likes and replies tell you something. A post that gets a lot of reactions probably touched a nerve. But on their own they can't tell you why. Was it useful? Was it honest? Or was it just loud? That's the gap I keep noticing around @0xhazels. The scoring doesn't throw engagement away. It asks whether the engagement is relevant, genuine and connected to something meaningful. A smaller reaction to a thoughtful post can count for more than a big one to an empty post. I think that's a fair way to look at it. The number is a signal, not the answer. The question underneath is always whether the work was worth someone's attention. I know some people will disagree, so tell me where I'm wrong. Is engagement the best measure we have, or are we leaning on it because it's easy to count?

Putting a stock onchain solved one problem: getting it to people. It left the other one alone. Nothing is wrapped around the asset. That gap is easy to miss. Tokenized equities are the fastest-growing corner of onchain real world assets, and they arrived with market hours but without market makers. A stock onchain is still just a stock, with no capital structure around it. @2FactorFinance is working on that missing layer. Its public beta is live on Base, working with Coinbase on Kelly-optimized leverage for Coinbase tokenized stocks, with Bitcoin first. The idea is to split an asset's volatility into two perpetual parts: a senior side with deep downside protection, and a junior side that carries the leverage with no liquidation risk and no external hedging counterparty. It only needs an asset with positive long-run drift, so the same logic can travel from Bitcoin to equities. The Points Program is open for Season 1. Marks come only from verified social actions, education and referrals, never from buying, depositing or holding, and they carry no cash value. When the leaderboard freezes at launch, the top 10 split 1 BTC paid in cbBTC. Referral link for anyone who wants to follow along: points.2factor.finance/r/jgg… If you could add one missing layer to a tokenized stock, what would it be?

Would you let an AI agent anywhere near your money? I wouldn't, unless there were rules. That's the part of @meetmotes I find most interesting. Every Mote is one NFT tied to one agent identity and one smart account. The owner funds that account and then sets signed permissions and limits. The agent works inside those limits instead of having open access to everything. Think about how a company handles spending. A new employee doesn't get the whole bank account. They get a card with a limit and a clear set of things they can pay for. The design here follows the same logic, except the owner is the one writing the rules. For me, that's what makes the idea more believable than just "AI plus NFT." The permissions are the point. Without them, an agent is a risk. With them, it's a tool you control. What limits would you want to see before trusting an agent with funds?