Independent review · falsification

Contradiction register

Observations that conflict with a stated claim. Nothing here is resolved by restating the claim more carefully.


4 entriesSeverity recordedResolution is a test, not a rewrite
Contradiction C-013
severity mediumOpen
Claim
Common programme profiles reduce implementation work.
Observation
Customer B required a jurisdiction-specific interpretation not represented by the existing profile.
Implication
Could indicate adapter insufficiency, or an overly broad core claim.
Resolution
Do not mutate the core yet. Test against Customer C.
Contradiction C-014
severity mediumOpen
Claim
Portfolio evidence is representative of the external market.
Observation
The first two candidate integrations share an upstream data vendor.
Implication
Heterogeneity may be overstated in the early sample.
Resolution
Sequence a non-portfolio integration before claiming G2.
Contradiction C-021
severity highAt risk
Claim
Integrity results are read as integrity results.
Observation
Two reviewers described a passing verification as the evidence being 'verified'.
Implication
The status vocabulary is not surviving contact with non-specialist readers.
Resolution
Test alternate statement wording before wider release; do not weaken the boundary.
Contradiction C-026
severity lowOpen
Claim
Free SDK lowers acquisition cost.
Observation
Every current integration conversation began with a founder introduction.
Implication
Too early to distinguish; sample is pre-distribution.
Resolution
Re-examine after public SDK release with attribution tracking.
The correct response to a contradiction is usually another observation — not a change to the core.
Handling rule

How an entry is closed

A contradiction closes when a subsequent observation either reproduces it — in which case the affected claim moves to at risk or falsified — or fails to reproduce it under comparable conditions. It does not close because the narrative was adjusted around it.


Independent review

Everything in this room

Contradiction register

Evidence moving against the thesis should remain visible.

A contradiction record should name the claim, the observation, severity, implication, current interpretation and the next test that can distinguish a local implementation problem from a structural failure. The point is to keep negative evidence from disappearing into generic “learning.”

Reading through Overall

The native reading. Claim status, evidence burden, proof gates, falsifiers, contradictions, capital-to-proof, governance, downside, and explicit reasons to stop or narrow the thesis.