Evidence Pack and Verifiability Boundaries
Explains how sources, facts, attribution, counterfactuals, and uncertainty enter a reviewable record.
Summarizes the sources, facts, and uncertainties an assessment should record so readers can check where its conclusions come from.
A publicly reviewable contribution record must distinguish at least whether a person or outcome exists, where its core facts come from, when the result was realized, who actually benefited, how an individual's increment can be identified within a team result, and what material counterevidence or uncertainties remain. The strength of a conclusion must match the quality of its sources: institutional promotion, second-hand retelling, and unverified judgment cannot replace primary evidence.
When evidence is still incomplete, the current research edition may retain a reasoned single-point allocation while marking it as provisional, stating what is missing and what evidence would trigger revision. Sparse documentation does not automatically mean zero, and displayed decimals cannot manufacture evidential precision. New sources may trigger reassessment; a revision must retain its date, reason, and before-and-after difference.