{"schema_version":1,"assessment_id":"eoa_inverse_innovation_exp03_opportunity320_20260801","source_experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"layer_decay_and_expiration_management__information_theory","archetype_slug":"layer_decay_and_expiration_management","domain_slug":"information_theory","title":"Reference-Gated Codebook Tiering and Expiration","opportunity_summary":"For systems that retain externally managed historical decoding schemes, classify superseded versions by object references, holds, rollback needs, and tested restorability; remove them from new-encoding eligibility before any archival demotion or deletion. The opportunity is conditional on a bounded audit showing that external scheme accumulation and selection burden actually exist.","adopter_authorizer":"The data or communication-system owner can authorize lifecycle-state changes; irreversible deletion additionally requires the retention authority and owners of dependent collections.","scores":{"meaningful_impact":{"score":4,"rationale":"Where externally managed historical schemes accumulate, decoder mismatch and irreversible loss of access to retained information are consequential, and reducing selectable-version clutter while preserving decoding directly addresses both risks. The prevalence and realized magnitude are unsupported."},"stakeholder_pull":{"score":3,"rationale":"Owners, operators, and audit or reproducibility stakeholders have plausible interests in reliable decoding and lower registry burden, but the packet contains no observed demand, incident history, or evidence that these stakeholders currently prioritize the problem."},"incremental_advantage":{"score":4,"rationale":"Compared with indefinite selectability and manual cleanup, reference-gated demotion offers a specific reduction in active clutter while retaining historical decodability. It also avoids the representation overhead or bulk rewriting associated with the stated rival, although the advantage remains untested."},"distinctiveness_plausibility":{"score":2,"rationale":"The composition is coherent, but its reference checks, retention holds, archival tiers, restore tests, and tombstones plausibly resemble ordinary artifact-lifecycle practice. Prior art is explicitly unsearched, so a materially distinctive contribution is not established."},"technical_implementability":{"score":3,"rationale":"The packet identifies observable metadata, lifecycle operations, halt conditions, and restoration tests, making a bounded shadow evaluation feasible. Production implementation is uncertain because complete object-to-scheme binding is only a hypothesis and archived executable environments may drift."},"adoption_authority_feasibility":{"score":4,"rationale":"Decision authority, affected parties, deletion authority, exclusions, and rollback conditions are explicitly identified. Feasibility is reduced by the need to reconcile collection owners, retention authorities, holds, and potentially missing owners before irreversible action."},"evidence_readiness":{"score":4,"rationale":"A read-only corpus audit, shadow lifecycle classification, restoration test, and separate problem and intervention falsifiers are already specified. Missing baseline measurements and fidelity or latency thresholds prevent the highest score."},"safety_net_benefit":{"score":4,"rationale":"Reference blocks, hold checks, unchanged source copies, restore tests, tombstones, halt criteria, and eligibility rollback provide multiple safeguards against destructive cleanup. Rare-format omissions and environment drift leave residual risk."},"scalability":{"score":3,"rationale":"Registry states and reference counts could be automated across collections, but exhaustive dependency tracing, long-tail restore testing, exception governance, and preservation of executable environments may scale poorly; lifecycle metadata could itself become unbounded."}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Read-only inventory and shadow classification for one bounded corpus, including reference tracing, baseline selectable-version counts, a small stratified archive-and-restore sample, and decode comparison against unchanged source copies.","confidence":"LOW","assumptions":["Existing registry and object metadata are accessible without purchasing major new infrastructure.","The corpus and sample remain bounded and no missing-reference reconstruction project is included.","No irreversible deletion, production eligibility change, or bulk transcoding occurs."]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Build and validate lifecycle metadata, reference gates, hold integration, archival packaging, tombstones, monitoring, and an automated restore/decode test harness for one system.","confidence":"LOW","assumptions":["One principal registry and a limited number of storage tiers are in scope.","Existing identity and retention systems expose usable integration points.","Historical scheme identities are substantially recoverable without manual reconstruction of every object."]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Production integration and controlled rollout across one organizational system, including policy approval, dependency remediation, operator training, compliance review, archival-environment validation, and staged demotion with rollback capability.","confidence":"LOW","assumptions":["The launch excludes bulk transcoding and large-scale repair of unidentified historical dependencies.","A retention authority and collection owners can participate within normal governance processes.","Specialized hardware emulation or extensive legacy-environment reconstruction is not required."]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Ongoing archive storage, lifecycle reconciliation, exception and hold review, restore drills, decode monitoring, audit support, and maintenance of historical execution environments for one system.","confidence":"LOW","assumptions":["Historical demand and version creation remain moderate.","Restore testing is sampled according to risk rather than exhaustive on every cycle.","No major compliance investigation or mass legacy migration is included."]}},"research_burden":"HIGH","earliest_credible_horizon":"3_TO_12_MONTHS","pipeline_gates":{"recognizable_externally_supportable_problem":{"status":"UNCERTAIN","reason":"The lifecycle failure is intelligible and has an explicit audit falsifier, but the sealed record offers no external evidence that accumulating external schemes, selectability burden, or mismatch incidents occur in representative deployments."},"identifiable_adopter_or_authorizer":{"status":"YES","reason":"The data or communication-system owner is named for lifecycle decisions, with retention authorities and dependent-collection owners additionally required for deletion."},"distinct_testable_incremental_claim":{"status":"YES","reason":"Against indefinite selectability and manual cleanup, the proposal claims that reference-gated demotion will reduce active selectable versions or mismatch opportunities while preserving restoration and decoding; both sides are measurable, although world novelty is unmeasured."},"bounded_next_evidence_step":{"status":"YES","reason":"One bounded corpus can undergo a read-only inventory, shadow classification, and small-sample archive, restore, and decode comparison without deletion or live eligibility changes."},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The authorized first step is non-destructive, and unresolved dependencies, ownership, holds, restore failures, or decode mismatches explicitly halt progression. This does not authorize later deletion."},"implementation_cost_scope_and_range":{"status":"UNCERTAIN","reason":"The packet defines implementation components, but corpus scale, metadata quality, integration count, legacy-environment requirements, and compliance burden are absent, so deployment ranges depend heavily on stated assumptions."}},"blocking_evidence":["Whether the target system actually accumulates externally managed historical decoding schemes that remain selectable or operationally burdensome.","Whether every retained object can be reliably bound to its exact required scheme, including rare formats and indirect dependencies.","Whether archived scheme packages and execution environments can be restored and decode referenced samples within predeclared fidelity and latency thresholds.","Whether shadow lifecycle classification materially reduces selectable versions or measured mismatch exposure relative to the current registry.","Whether the composition differs materially from established package, artifact, schema, or retention lifecycle controls.","Whether retention authorities and dependent-collection owners can resolve holds, orphan ownership, rollback dependencies, and deletion obligations at operational scale."],"next_evidence_step":"On one bounded corpus, perform a read-only inventory and create a shadow lifecycle classification without changing production eligibility or deleting anything. Compare the current number of selectable superseded versions and identifiable mismatch paths with the shadow reference-gated set, then archive unchanged copies of a small stratified sample and compare restored decoding with decoding from the current source copies under predeclared fidelity and latency thresholds. Stop if the audit falsifier finds no relevant accumulation or burden, or if any sampled dependency is unresolved or any restored decode fails.","research_questions":["What proportion of retained objects are self-contained, reproducibly reconstruct their decoder, or depend on an external historical scheme?","How many superseded versions remain selectable, and what observable storage, operator, or mismatch burden do they create?","Can scheme identity and all retention, hold, rollback, and reproducibility dependencies be discovered with adequate completeness?","What fidelity and restoration-latency thresholds must archived schemes meet for each stakeholder workflow?","How often and why are historical versions accessed, and which sampling strategy covers rare formats and environments?","How do existing artifact-retention and decoder-version practices compare with the proposed composition, and what contribution, if any, is distinctive?","Can tombstones, holds, lifecycle metadata, and archived environments be bounded or consolidated without weakening auditability?","Which authority resolves orphaned collections or conflicting retention and deletion obligations?"] ,"recommendation":"VALIDATE_PROBLEM_FIRST","uncertainty_constraints":["Closed-book assessment: no external prevalence, demand, prior-art, market-size, realized-impact, or cost evidence was available.","The proposal applies only where retained objects depend on externally managed or executable historical decoding schemes; self-contained or reproducibly reconstructed formats are its principal boundary.","All cost bands are resource-equivalent planning ranges conditioned on one bounded organizational system and unknown metadata quality.","A safe evidence step may support demotion design but does not authorize irreversible deletion.","Rare-format dependencies and execution-environment drift may not be represented by a small restoration sample."],"closed_book_prior_art_boundary":"Prior-art status is UNSEARCHED. The sealed packet supports only a testable mechanism composition relative to its stated baseline and rival; it does not establish novelty, prevalence, superiority to existing artifact-retention practice, or freedom to operate."}