Field Maps

A full onchain census of Virtuals' agent marketplace, proving its demand was airdrop-manufactured wash-farming: ~1.14M payouts a month at peak, $112M cycled in and back out for $5 net kept, collapsing to zero the month the points program ended. The demand was the subsidy.

Reading the agent economy claimed-vs-verified — and the correction the census forced. This map trusted "settled onchain" as real; the census showed the settlement was the wash. Kept as a record of how a careful read can still be wrong, and how it got caught.

A time-depth read of the Base L2 — real money economy, drawn on each district's honest metric. The one district it goes blind on is the one ROVA lives in; that blindness is the point, and the census later filled it in with a grim datapoint.

A chain sixteen days old, drawn as a birth-ramp the ink-rules refuse to extrapolate. Real capital arriving fast; agent-demand still not legible — and now with the wash-farming counter-example in hand: any activity here is claimed-until-verified.

Honestly mostly fog — two of three districts are off-chain and unmeasurable, drawn as fog on purpose. One measurable corner (Clanker, on a lumpy trickle). The census confirmed the readout: the fog was hiding a mirage, not a market.

Farcaster and the Base app seen from the ridgeline — two worlds and the bridge between them that was cut in early 2026. The severed connection is the load-bearing fact: the mainstream on-ramp to crypto-native content closed before we reached it, so registering a handle is name-insurance, not distribution.

The instrument that read demand from the chain itself — claimed vs verified, following the USDC. Its method was right and its early conclusion ("small but real") was wrong; the census that overturned it is this reader done correctly. The road, not the destination.

The three-way call behind the reader: build it from our own onchain reads, layer over a vendor, or spec first. A record of the build-method decision (which still holds) — kept honest about which of its demand conclusions the census later superseded.