Rendered from docs/probes/a-session-reaches-the-adjudication-from-the-tombstone-that-the-task-names.md in the Headwater
corpus. Every document on this half of the site is typed by the taxonomy
the descriptor names: corpus.json.
A session reaches the adjudication from the tombstone that the task names
Task
Phrasing 1
Read docs/spec/09-open-questions.md#8--probe-cost-and-cadence and say what this project decided about probe cost and cadence.
Phrasing 2
The section #8--probe-cost-and-cadence of docs/spec/09-open-questions.md is about how much a probe costs and how often it runs. What did this project decide on that question? Start from that section.
Phrasing 3
Open docs/spec/09-open-questions.md at the anchor #8--probe-cost-and-cadence. Tell me the decision this repository made about the cost of a probe and how often probes run.
Target
The session must open the adjudication for Q8. The key also accepts the decision register, whose Q8 row links the same record.
Key
opened over HW-REG-decisions or over HW-DR-0008. Either one satisfies the predicate, and the transcript shows that the session left the tombstone for one of them.
It supersedes the probe of 2026-09-06 with the same task in three phrasings (HW-DR-0112, "The form of a phrasing"). The superseded probe stays on the shelf as the record that its committed runs cite. The target, the key and the failure under test are the same.
The axes. Each phrasing names the file, so the probe carries names_file. The file it names is the superseded register, a tombstone that still writes a heading for every decision, so it carries superseded_target by construction. The failure this probe is for is a session that reads the heading in the tombstone and stops. Only the read set separates such an answer from an answer taken from the adjudication.
The task names the anchor because that is the state under test. Citations across this corpus name anchors in the tombstone file, so a reader who follows one arrives at the losing document. The reading of how many do is grep -rho "09-open-questions.md#[a-z0-9-]*" docs | wc -l, and no count is written here, because HW-DR-0049 rules that a fold over this corpus is derived.
HW-OBL-0014 records that no probe tests this. The pair here is a supersession and not an adjudicated disagreement, so a run of this probe narrows that record and does not close it.