Cross-garden built-environment health score
Fuse HKG's continuous biomarker ground truth with BKG's material and building data into a single, citable "does this building help its occupants" score. A connection no single garden can see: the value lives in the edge between health and building data. Earliest stage — a raw signal — pending evidence that the two datasets actually correlate.
This is RDKG doing its actual job (LD-10): proposing a value in the edge between two gardens that neither could surface alone. It deliberately sits at the signal stage with 0.3 confidence — the honest read of a connection nobody has tested yet.
The de-risking move is small and concrete: don't build the score, first check whether built-environment features and continuous biomarker outcomes correlate at all in data we can already touch. Foresight that doesn't change a decision is entertainment (P8).
Treat as a pure cross-garden hypothesis; the cheapest first experiment is to check whether existing HKG and BKG data even line up.
domains: health · building · embodied-data · cross-garden