Rendered from docs/obligations/0003-the-rebuild-and-cache-design-is-measured-at-spike-scale-alone.md in the Headwater corpus. Every document on this half of the site is typed by the taxonomy the descriptor names: corpus.json.

The rebuild-and-cache design is measured at spike scale alone

Context

Q6 rules that every run rebuilds the graph and that no derived artifact is canonical. The rebuild-and-cache design is measured at spike scale, on generated documents.

Obligation

A real corpus and a real harvesting tier are the instruments, and this corpus owes a reading from each.

Discharge

Neither instrument exists yet. No adopter has brought a corpus of the size the claim is about, and no harvesting tier has been built.

One more reading is on record, and it is still generated. On 2026-09-29, sh tools/measure/large-governs.sh measured one document that governs a glob over 5,000 files. A warm headwater check --strict cost 24 ms of CPU time, and at 50,000 files it cost 187 ms (#1260, spec 5). The tree is not a real corpus, so this reading does not discharge the record.