Object Invocation Protocol · protocol specification
Disconfirming Edge Selection Rule
§SELF — oip-disconfirming-edge-selection-rule
What this page is: the rule for publishing edges where GRAIN patterns contradict each other. What it explains: why confirming edges alone are confirmation bias; when an edge must be published. Why read it: five disconfirming edges against more confirming edges is a suspicious ratio unless the selection rule is public.
Selection rule
Publish a disconfirming edge when any of:
- Two named patterns make opposite predictions about the same observable class.
- Two catalogue nodes are notation-duplicates of one mechanism (variational cousins) — the edge documents the collapse.
- A no-go theorem forces a bound between two families.
- A reviewer can name the tension in one sentence — then the edge must exist or the corpus is hiding.
Ratio honesty
If mapping is honest, tension is normal. Prefer more disconfirming edges than flattering ones until the space is saturated. Absence of tension is a red flag, not a victory.
Current edges (patch wave)
- Free Energy vs Least Action
- Variational Family Collapse
- Catalogue disconfirming discussions in No-Go Theorems