{"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__linguistics_semiotics","archetype_slug":"computability_boundary_mapping","domain_slug":"linguistics_semiotics","title":"Classify and govern the solvability boundary for grammar equivalence","opportunity_summary":"Freeze the actual generative formalism and equivalence semantics, then seek independently checked evidence for either a total decider or an impossibility result. Replace forced Boolean answers outside established guarantees with bounded checks, witnesses, restricted fragments, or labeled UNKNOWN. The proposal could materially improve grammar change control, but the actual requirement, target theorem, stakeholder demand, prior art, and useful fallback coverage remain unverified.","adopter_authorizer":"The grammar-tool owner can authorize an isolated classification pilot; production changes require formal-methods review and approval from affected grammar authors and users.","scores":{"meaningful_impact":{"score":4,"rationale":"If the stated unrestricted Boolean requirement exists, incorrect equivalence verdicts can approve inequivalent grammars, reject equivalent ones, and misdirect engineering effort; however, the frequency and breadth of these consequences are unsupported."},"stakeholder_pull":{"score":2,"rationale":"The packet identifies affected actors and a tool owner but provides no interviews, commitments, incident records, usage evidence, or proof that stakeholders demand boundary classification or will accept UNKNOWN."},"incremental_advantage":{"score":4,"rationale":"Relative to retrying timeouts and forcing YES or NO, checked guarantee boundaries and honest fallback states directly address overclaiming and unresolved cases; the gain depends on useful coverage and acceptable UNKNOWN rates."},"distinctiveness_plausibility":{"score":2,"rationale":"The combination is coherent, but prior art is explicitly unsearched and differentiation from existing formal-language analysis, proof workflows, and tool-governance practices is unsupported."},"technical_implementability":{"score":3,"rationale":"Freezing contracts, running bounded analyses, and labeling pilot results are feasible in principle, but semantic fidelity is unresolved and neither a total procedure nor a valid impossibility reduction is established for the actual formalism."},"adoption_authority_feasibility":{"score":3,"rationale":"A named tool owner can authorize an isolated pilot, while production adoption requires formal-methods review and affected-user approval; those additional approvals and willingness to preserve UNKNOWN are untested."},"evidence_readiness":{"score":3,"rationale":"The packet supplies falsifiers, competing proof branches, a bounded pilot, exclusions, and rollback conditions, but lacks a frozen formalism, checked proofs, representative-pair provenance, operational data, and stakeholder evidence."},"safety_net_benefit":{"score":5,"rationale":"UNKNOWN, enforced fragments, bounded claims, witness search, explicit labels, prohibited overgeneralization, independent review, and rollback provide strong safeguards against unsupported universal verdicts."},"scalability":{"score":3,"rationale":"The governance pattern can be reused when formalisms change, but each grammar class and semantic contract may require separate proofs, reductions, tooling, and review, limiting automatic scale."}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"Freeze one formalism and semantics; audit the actual requirement; analyze 20 representative pairs and all pairs within a small bound; attempt constructive and reduction-based classifications; obtain independent proof review.","confidence":"LOW","assumptions":["A versioned workbench and representative grammars are accessible.","One or two formal-methods specialists and tool engineers can complete a time-bounded study.","No new large corpus, custom prover, or production integration is required."]},"initial_deployment_startup":{"band_2026_usd":"250K_TO_1M","scope":"Implement enforceable syntax and guarantee checks, bounded or one-sided modes, UNKNOWN propagation, labels, audit records, regression tests, and review controls in one grammar workbench.","confidence":"LOW","assumptions":["The pilot yields at least one useful classified fragment or fallback mode.","Existing interfaces and storage can be modified without replacing the workbench.","Production approval, user testing, documentation, and compliance work are included."]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Validate production integration, migrate affected workflows, train reviewers and authors, establish monitoring and escalation, and launch controlled equivalence governance for one organization.","confidence":"LOW","assumptions":["UNKNOWN can be represented throughout downstream workflows.","Affected users approve the operating policy.","No safety-critical certification or multi-organization rollout is required."]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Maintain analyzers and proof artifacts, review formalism changes, monitor UNKNOWN and disagreement rates, audit labels, support users, and periodically reclassify guarantees.","confidence":"LOW","assumptions":["One workbench and a limited family of versioned formalisms are maintained.","Reclassification is triggered by material changes rather than performed continuously.","Recurring work does not require a dedicated large research team."]}},"research_burden":"HIGH","earliest_credible_horizon":"3_TO_12_MONTHS","pipeline_gates":{"recognizable_externally_supportable_problem":{"status":"UNCERTAIN","reason":"The packet describes concrete symptoms, actors, and consequences, but supplies no audit showing that a deployed workbench actually asserts unrestricted total equivalence or forces timeout cases into Boolean verdicts."},"identifiable_adopter_or_authorizer":{"status":"YES","reason":"The grammar-tool owner is explicitly authorized to approve an isolated pilot, with formal-methods reviewers and affected users identified for production approval."},"distinct_testable_incremental_claim":{"status":"YES","reason":"The proposal claims that fixed contracts, checked boundary classification, and labeled fallbacks will prevent timeouts and bounded tests from being treated as universal Boolean evidence, directly contrasting with the stated heuristic baseline."},"bounded_next_evidence_step":{"status":"YES","reason":"The authorized step freezes one formalism, limits empirical analysis to 20 representative pairs plus a declared bounded set, pursues competing classifications, requires review, and keeps results non-authoritative."},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"For the isolated pilot, authority, excluded actions, labeling, halt conditions, and rollback are explicit; production verdict changes remain outside pilot authority."},"implementation_cost_scope_and_range":{"status":"UNCERTAIN","reason":"The candidate specifies functional components but not staffing, workbench architecture, proof complexity, data access, integration dependencies, or approval effort, so implementation ranges remain assumption-driven."}},"blocking_evidence":["Audit evidence that the real workbench makes an unrestricted total-equivalence claim, accepts grammars outside a proved-decidable fragment, or collapses unresolved cases into YES or NO.","A frozen, mechanically enforceable grammar syntax, alphabet, encoding, generated-language semantics, computation model, and quantifier scope faithful to deployed use.","An independently checked correctness-and-termination proof for a total procedure or an independently checked faithful impossibility reduction for the exact target class.","Measured coverage, disagreement, error, witness, runtime, and UNKNOWN rates for restricted and bounded modes against the current checker.","Stakeholder evidence that grammar authors, reviewers, engineers, and downstream users value the change and can operationally preserve UNKNOWN.","A bounded prior-art review of formal-language equivalence analysis and grammar-tool governance.","Work breakdown and integration discovery sufficient to refine staffing, review, migration, and recurring-maintenance costs."],"next_evidence_step":"Run a non-production classification study on one frozen, versioned formalism: audit whether the universal Boolean requirement actually exists; compare the current forced-verdict checker with a contract-aware prototype across 20 representative pairs and every pair within a small declared bound; record termination, disagreements, witnesses, and UNKNOWN; attempt both a checked total-decider proof and a faithful impossibility reduction. Stop or redirect if the audit finds an already enforced proved-decidable fragment with preserved UNKNOWN, or if independent review rejects encoding fidelity or a proof obligation.","research_questions":["What exact syntax, encoding, alphabet, semantics, computation model, and quantifier scope define the deployed equivalence requirement?","Does the workbench actually promise a terminating YES or NO for every accepted pair, and how are timeouts currently converted into decisions?","Can a total algorithm be proved correct and terminating for the frozen class, or can a faithful impossibility reduction be independently checked?","Which enforceable fragments, bounded checks, or one-sided witness procedures provide useful coverage without unsupported generalization?","What error, disagreement, runtime, and UNKNOWN rates result relative to the current heuristic checker?","Will downstream workflows preserve UNKNOWN rather than treating it as NO, and who has authority to enforce that behavior?","Which existing methods or governance patterns constitute the nearest prior art, and what incremental contribution remains?","What staffing, integration, review, and maintenance resources are required for one-workbench production adoption?","Does the formal model preserve the pragmatic or contextual distinctions that affected users consider material?","Which linguistically important constructions would be excluded by any proposed decidable restriction?"],"recommendation":"PARTNERED_RESEARCH","uncertainty_constraints":["Closed-book assessment; no external validation was performed.","Problem prevalence, stakeholder demand, realized impact, market size, and prior art are unmeasured.","The deployed formalism and semantic fidelity are not yet specified.","No target-specific decidability or undecidability conclusion has been established.","Cost bands are resource-equivalent estimates based on broad assumptions, not observed project data.","The usefulness of restricted modes and UNKNOWN depends on coverage and workflow behavior that have not been measured."],"closed_book_prior_art_boundary":"Prior art is explicitly UNSEARCHED. This assessment makes no claim that computability-boundary analysis, grammar-equivalence methods, checked reductions, restricted fragments, UNKNOWN-preserving interfaces, or their combination are novel, rare, or absent from existing tools and scholarship."}