{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"layer_decay_and_expiration_management__mathematics","trajectory_id":"R","attempt_index":0,"archetype_slug":"layer_decay_and_expiration_management","domain_slug":"mathematics","decision":"CANDIDATE","problem_id":"superseded_formal_artifact_accumulation","causal_lever_id":"lifecycle_gated_formal_artifact_retirement","proposal":{"problem":"In a long-lived formal mathematics library, superseded definitions, lemmas, proofs, and released snapshots can remain active and discoverable beside their replacements. They create ambiguous imports and proof-search noise, yet maintainers cannot safely remove them when downstream dependencies and reproducibility obligations are uncertain.","actors_substrate":["formal-library maintainers","theorem authors and reviewers","downstream proof developers","researchers reproducing cited results","definitions, lemmas, proofs, namespaces, and release snapshots"],"observable_state":"Multiple versions of a mathematical artifact remain searchable or importable without consistent current, superseded, archived, or preserved status; maintainers lack a complete dependency and reconstruction view.","consequence":"Users may select obsolete statements or interfaces, maintenance and proof search become harder, and cleanup can break dependent proofs or the ability to reconstruct published results.","affected_objective":"Keep the active formal corpus unambiguous and maintainable while preserving dependency integrity and reproducibility.","structural_mapping":[{"archetype_element":"sequential deposits","domain_realization":"Successive versions of definitions, lemmas, proofs, and library releases accumulate over time.","claim_kind":"INFERENCE"},{"archetype_element":"changed active usefulness","domain_realization":"An artifact can remain mathematically valid while becoming obsolete as the preferred interface or proof dependency.","claim_kind":"INFERENCE"},{"archetype_element":"stale layers mistaken for current truth","domain_realization":"Superseded artifacts remain discoverable or importable without an authoritative status marker.","claim_kind":"HYPOTHESIS"},{"archetype_element":"deletion dependency risk","domain_realization":"Removing an artifact can invalidate downstream proof terms, imports, citations, or release reconstruction.","claim_kind":"INFERENCE"},{"archetype_element":"preservation exceptions","domain_realization":"Published-result reconstruction, historically significant proofs, and frozen releases can require continued preservation.","claim_kind":"INFERENCE"},{"archetype_element":"managed disposition","domain_realization":"Artifacts are refreshed, marked superseded, archived, compacted through refactoring, or retired after dependency checks.","claim_kind":"HYPOTHESIS"}],"component_map":[{"component":"layer_inventory_and_identity_map","status":"direct","domain_realization":"Canonical registry of formal artifacts and version identities."},{"component":"deposition_order_and_age_index","status":"direct","domain_realization":"Creation, release, and supersession chronology."},{"component":"retention_policy_matrix","status":"adapted","domain_realization":"Rules by artifact class and reproducibility role."},{"component":"decay_function_or_aging_rule","status":"adapted","domain_realization":"Time since supersession informs review priority but never mathematical validity."},{"component":"expiration_trigger_definition","status":"adapted","domain_realization":"Supersession plus elapsed deprecation period triggers review."},{"component":"layer_value_and_risk_score","status":"adapted","domain_realization":"Review ranking from use, uniqueness, dependency reach, and reconstruction value."},{"component":"dependency_and_reconstruction_check","status":"direct","domain_realization":"Trace proof/import dependents and test cited-release reconstruction."},{"component":"pruning_or_archival_pathway","status":"direct","domain_realization":"Retire from active namespaces or move to a versioned archive."},{"component":"preservation_exception_register","status":"adapted","domain_realization":"Named exemptions for cited, historical, or uniquely reconstructive artifacts."},{"component":"deletion_audit_and_rollback_window","status":"direct","domain_realization":"Recorded retirement with a reversible quarantine interval."},{"component":"resource_budget_and_storage_tier","status":"adapted","domain_realization":"Budgets for active search/index complexity and archival storage."},{"component":"review_cadence_and_revalidation_loop","status":"direct","domain_realization":"Review each release or maintenance cycle."},{"component":"legal_hold_or_compliance_override","status":"adapted","domain_realization":"Publication, citation, or institutional reproducibility freeze."},{"component":"hot_warm_cold_access_state","status":"adapted","domain_realization":"Active, deprecated, archived, and deep-preserved corpus states."},{"component":"orphan_layer_detection","status":"direct","domain_realization":"Detection of artifacts with no live proof or reconstruction dependents."},{"component":"supersession_marker","status":"direct","domain_realization":"Machine-readable replacement or retirement annotation."},{"component":"layer_compaction_candidate_flag","status":"adapted","domain_realization":"Flag redundant theorem families or proof variants for refactoring."},{"component":"rehydration_or_restore_procedure","status":"direct","domain_realization":"Restore an archived release and replay its proof check."},{"component":"deletion_impact_estimate","status":"direct","domain_realization":"Estimate affected proofs, citations, namespaces, and recovery effort."},{"component":"stale_context_detector","status":"adapted","domain_realization":"Detect references to replaced definitions, interfaces, or assumptions."}],"mechanism_dispositions":[{"slug":"age_weighted_value_score","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Age is not evidence that a theorem has lost validity; composite scoring risks false precision. Use supersession and dependency evidence instead.","counterfactual_removal":"No causal step is lost because review can be prioritized without an age-discounted scalar."},{"slug":"archive_restore_test","disposition":"selected_supporting","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Periodically restore sampled archived releases and rerun proof checking.","counterfactual_removal":"Archival could silently cease to support reproducibility."},{"slug":"cache_eviction_rule","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Access recency is unsuitable for deciding mathematical authority or preservation.","counterfactual_removal":"The lifecycle chain is unchanged; ordinary index caching can remain separate."},{"slug":"dependency_safe_delete_check","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Gate retirement on live proof, import, citation, and reconstruction dependencies.","counterfactual_removal":"Cleanup could break formal proofs or published-result reconstruction."},{"slug":"lifecycle_storage_tiering_policy","disposition":"selected_supporting","contribution_type":"OPERATIONAL","adaptation_or_rejection":"Map tiers to active, deprecated, archived, and deep-preserved corpora.","counterfactual_removal":"Authority can still be clarified, but affordable long-term preservation weakens."},{"slug":"log_rotation_and_cleanup_job","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Formal artifacts are dependency-bearing objects, not a position-disposable append-only stream.","counterfactual_removal":"No required relation changes."},{"slug":"retention_schedule","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Replace record-law classes with artifact and reproducibility classes plus named freezes.","counterfactual_removal":"Disposition becomes ad hoc and preservation exceptions lack authority."},{"slug":"soft_delete_quarantine_window","disposition":"selected_load_bearing","contribution_type":"SAFETY_GUARDRAIL","adaptation_or_rejection":"Remove an artifact from active resolution before irreversible erasure and retain a tested restore path.","counterfactual_removal":"A classification or dependency error could become immediately irreversible."},{"slug":"stale_layer_detection_dashboard","disposition":"selected_load_bearing","contribution_type":"CORE_CAUSAL","adaptation_or_rejection":"Inventory artifacts and surface superseded-but-active definitions, lemmas, and interfaces.","counterfactual_removal":"The target population and stale-authority problem remain largely invisible."},{"slug":"time_to_live_ttl_policy","disposition":"considered_rejected","contribution_type":"NONE","adaptation_or_rejection":"Fixed age-based self-expiry is incompatible with enduring validity and unpredictable future citation value.","counterfactual_removal":"No viable intervention step is removed."},{"slug":"tombstone_or_deletion_marker","disposition":"selected_supporting","contribution_type":"OPERATIONAL","adaptation_or_rejection":"Leave a durable identity marker naming retirement reason and successor.","counterfactual_removal":"Old references become ambiguous and superseded artifacts may be mistaken for never-existing ones."}],"causal_chain":["Inventory identity-resolved artifacts and detect superseded items still active.","Assign explicit lifecycle and preservation classes.","Gate retirement on dependency, citation, and reconstruction checks.","Demote or archive safe candidates, quarantine removals, and publish successor markers.","Revalidate archives and exceptions so the active corpus stays bounded and reconstructible."],"baseline":"Maintainers leave obsolete artifacts active, add informal deprecation notes, and perform occasional manual cleanup when conflicts or maintenance costs become visible.","nearest_rival":"A deprecation-only policy that warns users but neither assigns archival states nor gates eventual retirement through dependency and reconstruction checks.","authority_safety":{"affected_parties":["maintainers","artifact authors","downstream proof users","authors relying on reproducible releases"],"decision_authority":"Designated library maintainers approve state changes; artifact owners and cited-release custodians review contested or irreversible cases.","authorized_first_step":"Run a read-only inventory and shadow classification on one namespace for one release cycle; compare proposed states with maintainer judgments and execute no retirement.","excluded_actions":["automatic hard deletion","expiry based solely on age or low access","altering theorem statements or proofs","retiring artifacts with unresolved dependencies or preservation freezes"],"halt_rollback":"Halt on any missed live dependency, disputed preservation obligation, or false-current/false-stale rate above the preregistered threshold; restore quarantined artifacts and revert active-resolution markers."}},"negative_tests":{"strongest_counterevidence":"The packet supplies no observation that formal libraries actually contain enough superseded-but-active artifacts to impair search, imports, or maintenance; mature libraries may already control this through namespaces, versioning, and deprecation practice.","analogy_break":"Mathematical truth does not normally decay with age. The mapping applies only to an artifact's active interface, contextual currency, and preservation role—not to the truth of its theorem.","failure_condition":"The approach fails if artifact identity or dynamic dependencies cannot be resolved reliably, or if most old artifacts remain uniquely necessary and therefore cannot leave the active corpus.","problem_falsifier":"A representative audit finds negligible superseded artifacts in active resolution and no material search ambiguity, wrong-version use, maintenance burden, or uncertainty about safe retirement.","intervention_falsifier":"In the bounded shadow test, lifecycle classification fails to identify stale-active artifacts more accurately than deprecation-only review, or dependency checks miss any known live dependent.","risks":["False staleness labels could undermine valid results.","Hidden dependencies could make retirement break proofs.","Preservation exceptions could become permanent clutter.","Archive formats or proof tooling could become unrestorable.","Lifecycle metadata could cost more to maintain than the problem warrants."]},"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":0.79,"generator_notes":"Closed-book structural transfer. The candidate concerns lifecycle authority of formal artifacts, not decay of mathematical truth; prevalence and effect size remain untested."}