{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"computability_boundary_mapping__library_information_science","trajectory_id":"R","attempt_index":0,"archetype_slug":"computability_boundary_mapping","domain_slug":"library_information_science","decision":"CANDIDATE","problem_id":"universal_digital_preservation_fidelity_decision","causal_lever_id":"formalize_and_route_preservation_decidability_boundary","proposal":{"problem":"A digital repository seeks an exact, always-terminating validator that decides whether any migration or emulation preserves every declared significant property of any deposited object, including executable and interactive works. The unrestricted claim may require deciding nontrivial behavior of arbitrary programs, while current practice can mistake timeouts, samples, or format validity for preservation fidelity.","actors_substrate":["repository curators and preservation engineers","depositors and creators","researchers and future users","repository governance and auditors","migration, emulation, and validation systems"],"observable_state":"The validator accepts arbitrary objects and transformations, emits Boolean preserved/not-preserved results, and lacks version-linked definitions of object class, significant properties, computation model, unknown states, and restricted guarantees.","consequence":"The repository may fund impossible universal automation or issue authoritative preservation claims unsupported outside tested formats and bounds.","affected_objective":"Truthful, auditable long-term preservation with useful automation and no loss hidden by over-broad guarantees.","structural_mapping":[{"archetype_element":"unrestricted exact terminating decision demand","domain_realization":"Decide preservation fidelity for every object and transformation.","claim_kind":"HYPOTHESIS"},{"archetype_element":"implicit representation and computation model","domain_realization":"Executable behavior, environment interaction, metadata, and significant-property encodings are not fixed.","claim_kind":"HYPOTHESIS"},{"archetype_element":"impossibility boundary plus weaker fallback","domain_realization":"Separate behavior-rich objects from enforceable decidable fragments, then route to exact, bounded, sound-incomplete, or expert-review modes.","claim_kind":"INFERENCE"},{"archetype_element":"honest status lattice","domain_realization":"Keep preserved, not-preserved, unknown, timeout, out-of-scope, and validator failure distinct.","claim_kind":"INFERENCE"}],"component_map":[{"component":"Problem-Class Specification","status":"adapted","domain_realization":"Define covered objects, transformations, environments, and fidelity predicates."},{"component":"Instance Representation Contract","status":"adapted","domain_realization":"Version object, metadata, environment, transformation, and property encodings."},{"component":"Computation Model Contract","status":"adapted","domain_realization":"Declare whether validation may execute objects, query curators, or use external services."},{"component":"Solvability Guarantee Profile","status":"adapted","domain_realization":"State exactness, soundness direction, completeness, and termination per mode."},{"component":"Quantifier and Scope Map","status":"adapted","domain_realization":"Separate all objects from formats, bounded traces, and named collections."},{"component":"Computability Status Lattice","status":"direct","domain_realization":"Classify modes as decidable, recognizable, partial, relative, or unresolved."},{"component":"Constructive Procedure Witness","status":"adapted","domain_realization":"Provide a terminating validator and proof for each claimed exact fragment."},{"component":"Reduction Preservation Contract","status":"adapted","domain_realization":"Show the encoding preserves the relevant executable-behavior predicate."},{"component":"Computability Impossibility Certificate","status":"adapted","domain_realization":"Certify only any unrestricted class actually reached by a valid reduction."},{"component":"Assumption Register","status":"direct","domain_realization":"Record object semantics, environment closure, property observability, and oracle assumptions."},{"component":"Decidable Subclass Map","status":"adapted","domain_realization":"Map static formats and bounded behavioral models with enforceable membership."},{"component":"One-Sided Recognition Contract","status":"adapted","domain_realization":"Where justified, confirm fidelity violations or witnesses without claiming total decision."},{"component":"Unknown and Nontermination Policy","status":"direct","domain_realization":"Expose unknown, timeout, out-of-scope, and system failure separately."},{"component":"Fallback Solution Contract","status":"adapted","domain_realization":"Specify exact, bounded, abstract, and curator-review guarantees."},{"component":"Computability Guarantee Record","status":"direct","domain_realization":"Version-link each public preservation claim to evidence and scope."},{"component":"Recheck Trigger","status":"adapted","domain_realization":"Reassess after format, emulator, property, environment, or language changes."},{"component":"Termination Condition","status":"direct","domain_realization":"Set explicit bounds for non-total modes."},{"component":"Scope Boundary","status":"direct","domain_realization":"Mechanically reject or quarantine inputs outside enforceable fragments."},{"component":"Decision Record","status":"direct","domain_realization":"Record the chosen boundary, fallback, authority, and rationale."},{"component":"Uncertainty Residue","status":"direct","domain_realization":"List unproved fidelity obligations and model-to-collection gaps."},{"component":"Independent Proof Review","status":"adapted","domain_realization":"Have an independent reviewer check reductions and fragment proofs."},{"component":"Complexity Follow-On Gate","status":"direct","domain_realization":"Assess cost only after a class is shown decidable."}],"mechanism_dispositions":[{"slug":"abstract_interpretation_or_model_checking","disposition":"selected_supporting","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Use finite over-approximations for declared behavioral properties, labeling one-sided guarantees.","counterfactual_removal":"Behavior-rich objects lose a sound finite fallback."},{"slug":"bounded_domain_exhaustive_search","disposition":"selected_supporting","contribution_type":"TEST_DESIGN","adaptation_or_rejection":"Exhaust controlled finite objects, states, and traces in the pilot.","counterfactual_removal":"The bounded test loses its complete-within-bound claim."},{"slug":"computability_boundary_decision_record","disposition":"selected_supporting","contribution_type":"OPERATIONAL","adaptation_or_rejection":"Version the boundary, evidence, guarantee, and recheck triggers.","counterfactual_removal":"Guarantees can drift without provenance."},{"slug":"computational_complexity_analysis","disposition":"selected_supporting","contribution_type":"OPERATIONAL","adaptation_or_rejection":"Gate feasibility assessment after decidability.","counterfactual_removal":"Decidable fragments may still be operationally unusable."},{"slug":"constructive_algorithm_and_correctness_proof","disposition":"selected_supporting","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Require totality and correctness proofs for exact fragment validators.","counterfactual_removal":"Exact-fragment claims lack positive witnesses."},{"slug":"diagonalization_impossibility_proof","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"A direct self-reference proof is unnecessary if a valid reduction suffices.","counterfactual_removal":"No proposed causal step changes."},{"slug":"enumeration_and_dovetailing","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"No complete enumerable witness system for fidelity is established.","counterfactual_removal":"No justified recognizer is lost."},{"slug":"fallback_mode_router","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Route each case to exact, abstract, bounded, or curator-review mode with labels.","counterfactual_removal":"Weaker methods would be exposed as one misleading Boolean validator."},{"slug":"halting_problem_reduction","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Test whether unrestricted behavioral fidelity can encode halting while preserving answers.","counterfactual_removal":"The proposed impossibility boundary would lack decisive evidence."},{"slug":"language_fragment_restriction","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Restrict exact validation to mechanically recognizable object/property languages.","counterfactual_removal":"There is no enforceable exact-answer region."},{"slug":"many_one_reduction_proof","disposition":"selected_supporting","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Supply totality, computability, and answer-preservation obligations for the reduction.","counterfactual_removal":"The hardness transfer could rest on analogy alone."},{"slug":"promise_problem_restriction","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Unenforced promises permit authoritative answers on violating deposits; syntactic fragments are safer.","counterfactual_removal":"No selected protection is lost."},{"slug":"proof_by_counterexample","disposition":"selected_supporting","contribution_type":"TEST_DESIGN","adaptation_or_rejection":"Use in-scope objects to refute over-broad validator claims, not prove undecidability.","counterfactual_removal":"Cheap falsification of universality is weakened."},{"slug":"proof_checking","disposition":"selected_supporting","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Independently check formal reduction and fragment proofs.","counterfactual_removal":"Boundary claims depend more heavily on author authority."},{"slug":"reduction_direction_checklist","disposition":"selected_supporting","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Gate any source-to-target hardness claim and expose assumptions.","counterfactual_removal":"A reversed reduction may be accepted."},{"slug":"semi_decision_with_explicit_unknown","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Adopt only if a sound one-sided recognizer is first established.","counterfactual_removal":"Current routing still preserves unknown."},{"slug":"theorem_prover_guided_search","disposition":"selected_supporting","contribution_type":"TEST_DESIGN","adaptation_or_rejection":"Attempt machine-checked certificates while recording open goals as residue.","counterfactual_removal":"Evidence can remain hand-checked, but assurance decreases."},{"slug":"turing_reduction_analysis","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Relative oracle degrees do not answer the needed enforceable product boundary.","counterfactual_removal":"The main reduction and routing chain is unchanged."}],"causal_chain":["Formalize objects, transformations, fidelity predicates, quantifiers, and computational capabilities.","Seek a constructive total validator for restricted classes and a valid halting reduction for the unrestricted behavioral class.","Independently check both arguments; leave mismatched or incomplete cases unresolved.","Enforce recognizable decidable fragments and route remaining cases to labeled bounded, abstract, or curator-review fallbacks.","Version guarantees and reclassify when assumptions change."],"baseline":"Format validation, checksums, sample migrations, regression tests, and curator judgment are combined without a formal class-wide termination or fidelity guarantee.","nearest_rival":"A stronger semantic-difference test suite that improves coverage but still treats finite passing tests as support for unrestricted fidelity.","authority_safety":{"affected_parties":["creators and depositors","future researchers","communities represented in collections","users needing accessible renderings","repository staff and funders"],"decision_authority":"Repository preservation governance, with collection stewards and independent technical review; the validator team cannot redefine significant properties alone.","authorized_first_step":"Run a bounded, non-production study on synthetic and duplicated objects: formalize one behavioral fidelity predicate, attempt one reduction and one decidable-fragment validator, and measure routing coverage and unknown rates.","excluded_actions":["discard or overwrite originals","declare undecidability from timeout or failed proof search","treat unknown as not-preserved","publish unrestricted guarantees from the bounded study","migrate production holdings automatically"],"halt_rollback":"Halt if the formal predicate fails curator semantic review, a reduction obligation fails, the abstraction is unsound, or any original is endangered. Retain originals, withdraw pilot labels, revert to existing review, and record status as unresolved."}},"negative_tests":{"strongest_counterevidence":"If operational fidelity is always a finite, mechanically observable checklist over bounded non-executable formats, a total validator may already exist in principle; the issue would be complexity, standards, or requirements rather than computability.","analogy_break":"Cultural significance and future user value are not automatically program-semantic predicates. A halting analogy transfers only after a faithful formalization and computable answer-preserving encoding of executable behavior.","failure_condition":"The mapping fails if real repository accountability excludes arbitrary executable behavior or if fragment membership cannot be enforced without discarding essential preservation meaning.","problem_falsifier":"Audit requirements and deposits show no universal exact terminating claim: every fidelity decision is explicitly bounded, format-specific, or acknowledged as expert judgment.","intervention_falsifier":"A checked total algorithm decides the faithfully formalized unrestricted class, or the proposed reduction fails totality, computability, or answer preservation while routing adds no reduction in false certainty.","risks":["Formalizing fidelity may omit socially important properties.","Restrictions may exclude the collections most needing preservation.","False alarms may overwhelm curators.","Official labels may launder weak guarantees.","Expert escalation may shift burden without resources or accountability."]},"null_rationale":null,"classification":{"candidate_kind":"DOMAIN_TRANSFER","prior_art_status":"UNSEARCHED","evidence_maturity":"HYPOTHESIS"},"revision_change_log":{"revision_kind":"ORIGINAL","prior_problem_id":null,"prior_causal_lever_id":null,"problem_changed":false,"causal_lever_changed":false,"conceptual_changes":[],"operational_changes":[],"repairs_addressed":[]},"confidence":0.84,"generator_notes":"Closed-book structural inference. The domain problem and empirical prevalence are hypotheses; no claim of novelty or established undecidability is made."}