unpopular take: a lot of "decentralization" in web3 is aesthetic, not functional.
A chain with 9 validators, all in AWS us-east-1, is not decentralized.
Geography, client diversity, and stake distribution all matter. very few chains score well on all three.
The hardest thing in web3 research is separating "this is technically impressive" from "this will actually be used."
Lots of technically impressive things go nowhere, a few ugly but useful things win everything.
I try to ask both questions every time.
Underrated web3 insight: the chains winning on UX rn aren't winning on technology.
They're winning on defaults : better wallets, faster onboarding, smarter gas abstractions.
Account abstraction (ERC-4337) might matter more for adoption than any consensus improvement.
For those still looking for good projects, here’s an on-chain lottery giving away a jackpot tomorrow!
You can play for free with a Gmail account: https://t.co/58yxqU7uz2
The next Megalé draw has increased for tomorrow, Wednesday, April 22, as there were 0 winners yesterday.
Come play for free, with a simple Gmail account, for a chance to win $EPI 933,332 🚀
Hot take that is actually lukewarm: most new L2s launching in 2026 do not need to exist.
The ones that survive long term are the ones with a real application reason to be their own chain, not just a fork looking for a community
And now it’s probably even more important to focus on hidden gems that break out without strong backing, without VCs, and with a revenue-first logic that doesn’t rely on a token from the start.
3 things every early stage web3 project gets wrong:
- raising too much money too early (kills scrappiness)
- building for investors, not users
- mistaking discord activity for product validation
The AI × web3 narrative is mostly noise.
But the signal underneath is real: verifiable compute (proving AI inference happened correctly) is an actually hard and actually important problem.
The projects that survive bear markets aren't the ones with the best marketing.
They're the ones where the devs kept shipping when nobody was watching. Always check github activity before you check token price.
Exactly, the focus should be on building for the user. And users don’t want to see that kind of information; they want simplicity. It has to be simple.
For me, the gap between "interested in web3" and "actually using web3" is still huge, and it's almost entirely a UX problem.
-> gas fees, seed phrases, bridge confusion. We know the problems, the solutions are slow...
hot take: most people are tracking the wrong metrics in web3.
TVL is vanity. developer activity is sanity. Look at commit frequency, new wallet addresses interacting with a protocol, and fee revenue.
Everything else is noise.