{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp05_complete_proposal_portfolio20_20260803","cell_id":"computability_boundary_mapping__veterinary_medicine","portfolio_valid":true,"proposal_assessments":[{"proposal_index":1,"complete":true,"causally_faithful":true,"materially_distinct":true,"reason":"Operationally specifies the unrestricted policy-verification claim, matched reduction review, enforceable finite-state restriction, conservative model checking, labeled abstention, authority boundaries, falsifiers, rollback, and a bounded evidence step. Its causal path correctly moves from an unsupported universal decider to a model-relative decidable fragment and governed fallback."},{"proposal_index":2,"complete":true,"causally_faithful":true,"materially_distinct":true,"reason":"Fully defines the simulator-matching instance, one-sided witness guarantee, fair dovetailing, explicit UNKNOWN behavior, replay certificates, optional finite-box negatives, safety limits, falsifiers, and tests. It faithfully replaces an unsupported two-sided decision procedure with recognition and bound-qualified fallback rather than treating timeout as NO."},{"proposal_index":3,"complete":true,"causally_faithful":true,"materially_distinct":true,"reason":"Provides precise pipeline, input, environment, termination, and report-equivalence contracts; a scoped reduction; parallel certificate and counterexample evidence; bounded exhaustion; release routing; versioning; falsifiers; and rollback. It correctly preserves failed proof search as UNRESOLVED and confines each equivalence verdict to its evidence."},{"proposal_index":4,"complete":true,"causally_faithful":true,"materially_distinct":true,"reason":"Operationally separates impossible complete termination prediction from decidable checking of supplied termination certificates, with runtime capability enforcement, explicit uncertified states, a separate complexity gate, governance, falsifiers, rollback, and a concrete checker test. The sound-but-incomplete proof-carrying path matches the archetype's causal structure."},{"proposal_index":5,"complete":true,"causally_faithful":true,"materially_distinct":true,"reason":"Precisely distinguishes unrestricted adaptive synthesis from an enforceable finite promise, then supplies exhaustive protocol-tree enumeration, coverage certificates, bound-qualified negative results, UNKNOWN and OUT-OF-MODEL states, complexity review, welfare authority, falsifiers, and synthetic validation. It correctly derives termination and completeness only inside the declared finite fence."}],"pairwise_assessments":[{"proposal_a":1,"proposal_b":2,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 1 addresses universal safety and termination verification of clinical care-policy programs by restricting the language and model-checking an over-approximation; Proposal 2 addresses existential reproduction of an observed trace by open-ended research simulators using fair witness search and explicit UNKNOWN."},{"proposal_a":1,"proposal_b":3,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 1 certifies prohibited-action safety for a single admitted care policy through a decidable fragment and abstraction; Proposal 3 governs relational equivalence between legacy and replacement laboratory pipelines through checked proofs, counterexamples, and finite-envelope evidence."},{"proposal_a":1,"proposal_b":4,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Although both concern executable veterinary workflows, Proposal 1 targets medication-safety verification and regains decidability through language restriction plus model checking, whereas Proposal 4 targets recurring surveillance-job termination and admits expressive programs through producer-supplied, mechanically checked proofs followed by complexity review."},{"proposal_a":1,"proposal_b":5,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 1 prevents unsafe actions by verifying restricted clinical policies over patient-state abstractions; Proposal 5 constructs bounded adaptive assessments that distinguish paired behavior models by exhaustively enumerating decision trees and covered outcomes."},{"proposal_a":2,"proposal_b":3,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 2 asks whether any simulator execution matches a past behavior trace and uses one-sided dovetailed recognition; Proposal 3 asks whether two laboratory programs agree on every admitted record and uses positive proof certificates, decisive counterexamples, and release gating."},{"proposal_a":2,"proposal_b":4,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 2 searches an unbounded candidate space for a replayable existential behavior-match witness; Proposal 4 checks author-supplied universal termination certificates for surveillance workflows and then separately assesses resource feasibility."},{"proposal_a":2,"proposal_b":5,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 2 retrospectively seeks one parameter-state-seed execution reproducing a fixed trace while retaining unrestricted partial simulators; Proposal 5 prospectively synthesizes an adaptive protocol guaranteed to separate two finite promised response models and can certify nonexistence only within explicit bounds."},{"proposal_a":3,"proposal_b":4,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 3 evaluates semantic and termination equivalence between two laboratory pipelines to govern a migration; Proposal 4 evaluates proof-carrying universal termination of one surveillance workflow to govern recurring scheduling, with no program-pair comparison."},{"proposal_a":3,"proposal_b":5,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 3 gates software migration using equivalence certificates or divergent specimen records; Proposal 5 synthesizes action-conditioned behavioral discrimination protocols through bounded exhaustive search and routes any live use to welfare review."},{"proposal_a":4,"proposal_b":5,"same_problem":false,"same_intervention":false,"independent_opportunity":true,"key_difference":"Proposal 4 changes surveillance-workflow admission from universal termination inference to checking submitted proof artifacts; Proposal 5 changes unrestricted behavior-assessment synthesis to complete enumeration over a finite promised model and protocol space."}],"replacement_indices":[],"rationale":"All five proposals are operationally complete and preserve the archetype's core sequence: formalize the claimed class and computation model, obtain appropriately scoped constructive or impossibility evidence, avoid inferring status from timeout, and ship weaker guarantees with explicit unknown, bound, or scope labels. Every pair differs materially in the affected decision problem, intervention, and causal path; shared boundary records, proof review, or fallback labels are supporting archetype infrastructure rather than the defining intervention. No replacement is required."}