CAI Founding Papers
Applicability, Coverage, and Missing Evidence in Code Assurance Scoring
Absence, partial coverage, and survey completeness.
Version 1.0 · signed 17 September 2026 by Jimmy Borch · Canine Development · 27 pages
Abstract
What the paper argues.
The Code Assurance Index (CAI) calculates a headline score from a producer-supplied evidence bundle. Measurements at dimension level are combined into categories; categories and direct meta-dimension inputs are combined into lenses; and the measured lens results enter an ordered weighted average (OWA) that produces the headline score and its reporting band. Evidence that is absent, partly covered or incomplete at different stages of this process has different arithmetic and interpretive consequences.
Check that you have the signed document.
The PDF is the signed artefact. Same bytes, same hash, whoever you got it from. Run this against the file you downloaded, and compare the result with the digest below it.
sha256sum applicability-coverage-and-missing-evidence.pdf502fb7ecfc99a2b392cf313bb409b0d4d672759d81e4b0aa291f23a5afaf4ca6