{"schema_version":1,"assessment_id":"eoa_inverse_innovation_exp03_opportunity320_20260801","source_experiment_id":"eoa_inverse_innovation_exp03_full320_20260801","cell_id":"computability_boundary_mapping__film_media_production","archetype_slug":"computability_boundary_mapping","domain_slug":"film_media_production","title":"Scoped render-completion preflight with explicit uncertainty routing","opportunity_summary":"Define an enforceable restricted project fragment whose completion can be checked exactly, route other render and compositing projects to bounded or sound-incomplete analysis, and preserve UNKNOWN, timeout, out-of-scope, and failure as distinct outcomes. The proposal could reduce overconfident delivery and compute-allocation decisions, but the sealed record does not establish that unrestricted completion claims or truth-valued timeout misuse occur in production, that the restricted fragment covers useful workflows, or that the integration is distinctive from existing practice.","adopter_authorizer":"A post-production engineering owner can authorize a synthetic offline pilot; any change to production admission or delivery policy requires both the production executive and pipeline reliability owner.","scores":{"meaningful_impact":{"score":3,"rationale":"False clearances, false rejections, uncontrolled render consumption, and missed delivery commitments would matter materially if the described binary preflight behavior occurs. The sealed candidate supplies no evidence about occurrence, frequency, or the share of losses addressable by completion-boundary routing."},"stakeholder_pull":{"score":2,"rationale":"Artists, production managers, render operators, vendors, and clients have plausible interests in reliable scheduling and compute allocation, but no adopter interviews, requests, incident records, willingness to change workflows, or other demonstrated pull are present."},"incremental_advantage":{"score":4,"rationale":"Relative to binary heuristic preflight or an unguaranteed predictor, mechanically enforced fragment admission, exact checking within that fragment, and explicit UNKNOWN routing provide a clear safety and correctness advantage. The practical size of that advantage depends on useful-fragment coverage and whether current preflight is actually interpreted as definitive."},"distinctiveness_plausibility":{"score":3,"rationale":"The composition of a checked computability boundary, exact restricted analysis, labeled fallback modes, and reclassification triggers is coherently differentiated from the stated baseline and nearest rival. Prior art is explicitly unsearched, so distinctiveness beyond the sealed comparison cannot be credited."},"technical_implementability":{"score":3,"rationale":"A synthetic restricted grammar, sandboxed cases, mechanical admission, and bounded comparison are plausible to build. Implementability remains uncertain because successful completion lacks a fixed operational predicate, plug-in expressiveness is unverified, and enforceable totality contracts may exclude important workflows."},"adoption_authority_feasibility":{"score":4,"rationale":"The candidate identifies an owner able to approve an offline pilot, separate authorities for production policy, excluded live-decision uses, and explicit halt and rollback conditions. Production adoption would still require coordinated approval and resistance to coercing UNKNOWN into NO."},"evidence_readiness":{"score":4,"rationale":"The candidate supplies a synthetic non-production test setting, looping and terminating cases, a baseline comparison, fixed compute limits, oracle-based correctness checks, and explicit falsifiers. Corpus construction, ground-truth adjudication, completion semantics, and outcome measurement still require preregistration."},"safety_net_benefit":{"score":4,"rationale":"Separating UNKNOWN, timeout, unsupported input, and failure directly limits overclaiming, while sandboxing, advisory-only use, exact-verdict checks, and rollback protect the first step. Benefits could be weakened if operators reinterpret UNKNOWN as rejection or if the completion predicate omits crashes, corrupt assets, nondeterminism, or quality failure."},"scalability":{"score":2,"rationale":"Each expansion of the project language, plug-in set, execution model, or external dependency may require renewed contracts, proofs, and checking. The restricted fragment may exclude common creative workflows, and no evidence establishes reusable coverage across projects, vendors, or facilities."}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"Audit one pipeline's claimed completion semantics and timeout interpretation; define one restricted manifest grammar and observable completion predicate; build a sandboxed synthetic corpus and exact checker; compare tri-state routing with the existing preflight under fixed limits.","confidence":"LOW","assumptions":["The work is confined to one pipeline and a synthetic non-production corpus.","A pipeline engineer and formal-methods or static-analysis specialist are available.","No proprietary production assets or live delivery decisions are required.","Existing sandbox and test infrastructure can be adapted rather than created from scratch."]},"initial_deployment_startup":{"band_2026_usd":"250K_TO_1M","scope":"Integrate mechanical fragment admission, exact and fallback routing, labeled results, telemetry, versioned guarantees, sandbox controls, and operator-facing workflow into one controlled production environment without granting automatic delivery authority.","confidence":"LOW","assumptions":["The pilot demonstrates correct exact verdicts and decision-relevant improvement.","Only one facility or production pipeline is integrated initially.","Existing render orchestration, identity, logging, and rollback systems expose usable interfaces.","Security and production-policy review are required but do not demand major infrastructure replacement."]},"operational_launch":{"band_2026_usd":"1M_TO_5M","scope":"Harden and validate the router and checker for production use across representative project types, train operators, establish governance and incident procedures, certify supported operators, and connect results to admission or scheduling policy under human authority.","confidence":"LOW","assumptions":["The restricted fragment covers enough real work to justify launch.","Multiple creative tools, plug-ins, and vendor workflows require validation.","Production authorities approve policy changes after an advisory phase.","The launch includes reliability, security, evaluation, documentation, and partner-coordination labor."]},"annual_recurring":{"band_2026_usd":"250K_TO_1M","scope":"Maintain fragment contracts and checker code, review new plug-ins and execution assumptions, rerun regression and oracle tests, monitor UNKNOWN handling and errors, support users, and periodically revalidate guarantees.","confidence":"LOW","assumptions":["Deployment remains limited to a modest number of related pipelines.","Language and plug-in changes trigger recurring review but not continuous redevelopment.","Existing compute and observability platforms absorb most routine execution and storage.","Independent review is repeated for material model or guarantee changes."]}},"research_burden":"HIGH","earliest_credible_horizon":"3_TO_12_MONTHS","pipeline_gates":{"recognizable_externally_supportable_problem":{"status":"YES","reason":"The candidate specifies observable binary handling of pass, failure, timeout, and sampled success; identifiable scheduling and compute harms; and an audit-based falsifier. Whether this behavior exists in an actual pipeline remains an evidence question rather than a failure of problem recognizability."},"identifiable_adopter_or_authorizer":{"status":"YES","reason":"The post-production engineering owner is identified for an offline pilot, and the production executive plus pipeline reliability owner are identified for production admission or delivery-policy changes."},"distinct_testable_incremental_claim":{"status":"YES","reason":"The proposal claims that mechanically admitted projects can receive exact checking while other projects retain explicit uncertainty, reducing false clearance, false rejection, or uncontrolled compute relative to binary baseline preflight under fixed limits."},"bounded_next_evidence_step":{"status":"YES","reason":"The sealed candidate authorizes a synthetic, non-production comparison using one restricted grammar, looping and terminating cases, a bounded oracle set, fixed compute limits, and explicit correctness and outcome falsifiers."},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The first step excludes live delivery decisions, requires sandboxing of untrusted plug-ins, preserves UNKNOWN, assigns pilot authority, and stops on containment failure, unenforceable membership, or any wrong exact verdict."},"implementation_cost_scope_and_range":{"status":"YES","reason":"The proposed first step and subsequent integration, launch, and maintenance scopes are separable enough to assign broad resource-equivalent bands, although confidence is low because the pipeline architecture and fragment coverage are unknown."}},"blocking_evidence":["Audit whether any actual preflight makes or is interpreted as making a class-wide completion guarantee, and whether timeouts or sample success are treated as truth-valued verdicts.","Establish whether the admitted production language has unbounded computational expressiveness or instead has a finite state space and known total completion procedure.","Define a versioned, observable meaning of successful completion that addresses crashes, corrupt assets, nondeterministic services, external state, and the exclusion or inclusion of output-quality criteria.","Demonstrate that fragment membership and operator-totality contracts can be mechanically enforced without silently admitting unsupported behavior.","Measure useful-workflow coverage and compare error and compute outcomes against baseline using independently adjudicated ground truth.","Determine through external prior-art research whether the proposed integration is distinct from established render-farm verification, workflow restriction, and static-analysis practice."],"next_evidence_step":"Run a preregistered offline study for one pipeline: first audit its language and actual interpretation of preflight outputs; then define one versioned completion predicate and enforceable restricted grammar, independently review the model and reduction, and compare the current binary preflight with tri-state boundary routing on a sandboxed synthetic corpus containing adjudicated terminating and looping plug-in cases under identical compute limits. Reject the intervention if the audit finds only finite total workflows with no truth-valued timeout misuse, if membership cannot be enforced, if any exact-fragment verdict differs from the oracle, or if routing does not reduce false clearance, false rejection, or uncontrolled compute.","research_questions":["Does the target pipeline support recursion, unbounded state, or plug-ins capable of encoding unbounded computation under the relevant execution model?","What precise observable predicate constitutes successful completion, and which crashes, asset failures, external services, nondeterminism, and quality conditions are included?","Are timeouts and sample renders currently interpreted as definitive failure or universal safety in decisions that affect delivery or resource allocation?","What proportion of representative production workflows can satisfy a mechanically enforceable total restricted-fragment contract?","How will ground truth, false clearance, false rejection, uncontrolled compute, and operator treatment of UNKNOWN be measured consistently?","Can operators preserve UNKNOWN as uncertainty rather than coercing it into rejection or approval?","Which elements are established practice, and which—if any—constitute a distinct technical or governance contribution?","How often do plug-in, dependency, or execution-model changes force reclassification and renewed proof or validation?"] ,"recommendation":"VALIDATE_PROBLEM_FIRST","uncertainty_constraints":["The assessment is closed-book and cannot establish prior art, novelty, prevalence, market size, realized impact, or exact cost.","The candidate describes a pipeline that may make an unrestricted completion promise; it does not demonstrate that a real adopter currently does so.","The undecidability argument depends on the actual project language and execution model supporting an appropriate reduction.","The operational completion predicate is not yet stable or validated.","Useful-fragment coverage, comparative benefit, and workflow burden are hypotheses without empirical results.","Cost bands are resource-equivalent planning ranges based only on the stated scope, not on inspected systems or vendor quotations."],"closed_book_prior_art_boundary":"Prior art is explicitly unsearched. This assessment credits only internal differentiation from the candidate's stated heuristic binary baseline and unguaranteed learned predictor; it makes no claim that restricted workflow languages, termination checking, tri-state analysis, render-farm verification, or their proposed composition are novel or uncommon."}