Conservatism in Precedence Resolution

Abstract

This RFC proposes a conservatism tie-break: where more than one condition is satisfied, the most conservative outcome governs, while precedence continues to determine which condition is recorded. It also reprints the §10.1 matrix in precedence order, which is editorial and carries no change in meaning. Both provisions are non-breaking: neither can produce an outcome less conservative than the current text on any input. §9.2 fixes six ordered precedence conditions and states that the first matching condition governs. On the baseline 0–5 scale more than one condition is routinely satisfied at once, and the order does not always place the most conservative outcome first. The most consequential instance: a weighted mean below 50 necessarily implies at least one layer below 3, so condition 4 always matches before condition 5, and the §10.1 row for a Maturity Score below 50 is unreachable — with the condition that does match yielding AL-2 rather than the AL-0 that row specifies.

Evidence base

The unreachability of the sub-50 row is demonstrable arithmetically from the baseline aggregation in §8.2 and the layer-minimum condition in §9.2, without reference to any implementation. The printed-order discrepancy is demonstrable by comparing §10.1 with §9.2 and Annex B.2.
Both were surfaced during the build of the eighteen-module WEKID Training Programme. Module 10 currently teaches the interaction factually — stating both rows, noting that the Standard does not comment on the collision, and directing candidates to record which condition governed.
Supporting material available to the working group:
• The findings register, items S1 and P1.
• The arithmetic demonstration: on the baseline scale a weighted mean below 2.5 entails at least one layer below 3, and therefore always satisfies condition 4 before condition 5.
• Worked resolutions of five verified cases through the full precedence order, including the two that expose the defect.

Where this proposal stands

• Proposed · 2026-09-20 23:01:22
  Submitted by Jim Judge
• Triaged · 2026-09-20 23:33:05
• Working-group review · 2026-09-20 23:36:36
• Public comment · 2026-09-20 23:37:28
• Council decision
• Versioned release