{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp07_retrospective_selector60_20260803","cell_code":"E7C023","selector_replication":3,"assessments":[{"blind_id":"CANDIDATE_A","problem_reality_importance":86,"causal_archetype_fit":92,"distinctiveness_prior_art_resilience":69,"operational_specificity":94,"falsifiability_test_quality":91,"adopter_partner_path":83,"deployability_complexity":74,"authority_safety_reversibility":94,"strict_potential":86,"empirical_partner_potential":89,"scrutiny_priority":88,"biggest_visible_risk":"Calendar identification, textual criticism, and historical interpretation may constitute the substantive work, leaving the reusable service to save only arithmetic while concealing contested assumptions.","rationale":"This is exceptionally bounded and auditable, with explicit module scopes, uncertainty-preserving outputs, matched comparisons, independent review, recalibration tests, and strong rollback rules. The archetype is causally central because repeated rule reconstruction is precisely the claimed barrier. It warrants first scrutiny because a meaningful claim could survive, although familiar conversion tools and the possible inseparability of conversion from scholarship create prior-art and mechanism vulnerability."},{"blind_id":"CANDIDATE_B","problem_reality_importance":69,"causal_archetype_fit":78,"distinctiveness_prior_art_resilience":58,"operational_specificity":84,"falsifiability_test_quality":85,"adopter_partner_path":75,"deployability_complexity":59,"authority_safety_reversibility":88,"strict_potential":64,"empirical_partner_potential":77,"scrutiny_priority":69,"biggest_visible_risk":"The proposition graph may reify one curator's contested doctrinal system, so apparently reusable dependency paths could generate more distortion and review work than useful consequences.","rationale":"The voluntary single-module pilot is bounded, reversible, and capable of detecting false flags, missed paths, anachronism, hidden labor, and authority leakage. However, the claimed transformation is unusually dependent on argument, terminology, period, and school; building and maintaining the graph may absorb the supposed savings. Its strongest lane is an empirical partnership with a willing workshop or journal, not a strict contrastive claim."},{"blind_id":"CANDIDATE_C","problem_reality_importance":82,"causal_archetype_fit":76,"distinctiveness_prior_art_resilience":55,"operational_specificity":87,"falsifiability_test_quality":86,"adopter_partner_path":91,"deployability_complexity":65,"authority_safety_reversibility":92,"strict_potential":69,"empirical_partner_potential":90,"scrutiny_priority":79,"biggest_visible_risk":"The gateway may merely centralize additional librarian, specialist, and community-liaison labor while becoming an informal access gatekeeper, rather than lowering a reusable causal barrier.","rationale":"The preparation and context problem is consequential, and a department or archive could authorize the proposed matched pilot with clear protections and decision-relevant measures. Its partner path is especially credible because burden, completeness, correction, and misrepresentation can be observed directly. Strict potential is weaker: the service resembles a governed referral and consultation workflow, and context work may remain irreducibly relational and case-specific."},{"blind_id":"CANDIDATE_D","problem_reality_importance":91,"causal_archetype_fit":89,"distinctiveness_prior_art_resilience":63,"operational_specificity":95,"falsifiability_test_quality":93,"adopter_partner_path":90,"deployability_complexity":76,"authority_safety_reversibility":94,"strict_potential":85,"empirical_partner_potential":94,"scrutiny_priority":87,"biggest_visible_risk":"A reusable annotation grammar may impose artificial ritual boundaries and manufacture comparability even while formally separating participant terms, observations, and analyst codes.","rationale":"The recurring segmentation, alignment, and reconciliation burden is concrete, consequential, and tightly matched to a reusable bench. The proposal specifies eligible records, layered outputs, bounded automation, matched pairs, blinded review, recovery assays, authority controls, and strong stop conditions. It is the strongest empirical-partner opportunity, but ranks just behind A because annotation benches and codebook workflows look more vulnerable to being standard infrastructure with a refined governance layer."}],"rank_order":["CANDIDATE_A","CANDIDATE_D","CANDIDATE_C","CANDIDATE_B"],"top_choice":"CANDIDATE_A","portfolio_observation":"A and D form the leading pair: both have strong causal mappings, bounded pilots, measurable failure modes, and credible rollback. A has the sharper strict contrastive claim, while D has the strongest willing-partner experiment. C is a valuable partner-led service test but risks reducing to centralized staffing; B is the most interpretation-dependent and curator-sensitive.","confidence":"HIGH"}