{"schema_version":1,"research_id":"eoa_inverse_innovation_exp06_external_evaluation_20260803","source_assessment_id":"catalytic_pathway_enablement__religious_studies_theology:P4:v0","cell_id":"catalytic_pathway_enablement__religious_studies_theology","search_queries":["historical calendar conversion uncertainty religious dates digital humanities calendar converter scholarly","EDTF uncertainty approximate date standard Library of Congress","Calendrica calendar conversion software historical calendars","historical chronology calendar conversion intercalation uncertainty research","site:cambridge.org \"Undate\" humanistic dates computation authors GitHub","digital humanities research infrastructure historical dates uncertainty calendars expressed need","DARIAH historical dates uncertainty calendar conversion","Linked Places format dates uncertainty calendar historical","site:cidoc-crm.org official CIDOC CRM 7.1.3 time-span uncertainty begin end pdf","site:neh.gov Digital Humanities Advancement Grants 2026 funding levels","site:bls.gov Occupational Employment Wage Statistics software developers historians May 2025","site:neh.gov/grants/odh/digital-humanities-advancement-grants 2026 Notice Funding Opportunity","site:neh.gov/sites/default/files 2026 DHAG NOFO digital humanities advancement grants"],"sources":[{"source_id":"S1","title":"Undate: humanistic dates for computation","publisher":"Computational Humanities Research / Cambridge University Press","url":"https://www.cambridge.org/core/journals/computational-humanities-research/article/undate-humanistic-dates-for-computation/F6C820D6D40C5F8BB8CBC3570E34378C","source_class":"PRIMARY_RESEARCH","publication_date":"2025-08-05","accessed_at":"2026-08-03","claims_supported":["Historical and cultural dates are frequently incomplete, imprecise, inconsistent, or expressed in non-Gregorian calendars.","A 2022 DHTech hackathon identified uncertain and imprecise dates as a shared problem across projects.","Undate already preserves original-calendar values and precision, converts supported Hebrew and Islamic dates to Gregorian earliest/latest bounds, and supports comparison across calendars.","The work is grounded in the Princeton Geniza Project, Shakespeare and Company Project, and Islamic Scientific Manuscripts Initiative and identifies additional interested researchers.","The software remains incomplete and calendar support is limited, while calendar conversion can retain one-to-two-day ambiguity."]},{"source_id":"S2","title":"dh-tech/undate-python: A Python library for working with fuzzy, partial, or otherwise uncertain dates","publisher":"DHTech Undate project","url":"https://github.com/dh-tech/undate-python","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"2025-10-02","accessed_at":"2026-08-03","claims_supported":["The beta library parses partial ISO 8601 and EDTF forms and converts Hebrew Anno Mundi and Islamic Hijri dates.","The implementation includes tests, examples, documentation, calendar-explicit objects, precision tracking, and earliest/latest calculations.","The project is reusable open-source software under Apache-2.0 but warns that it is not feature-complete."]},{"source_id":"S3","title":"fitnr/convertdate: Utils for converting between date formats and calculating holidays","publisher":"Convertdate project","url":"https://github.com/fitnr/convertdate","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"n.d.","accessed_at":"2026-08-03","claims_supported":["Existing MIT-licensed software converts Armenian, Bahá'í, Coptic, Hebrew, Islamic, Julian, Persian, and other calendars.","The package exposes material assumptions and limitations, including noon-based conversion for calendars whose day begins at sundown, proleptic-calendar conventions, and alternative disputed French Republican leap-year methods.","Calendar arithmetic is technically available, but the package does not itself provide the proposed provenance, scholarly exception, authority, or regeneration workflow."]},{"source_id":"S4","title":"Extended Date/Time Format (EDTF) Specification","publisher":"Library of Congress","url":"https://www.loc.gov/standards/datetime/edtf.html","source_class":"STANDARD","publication_date":"2019-02-04","accessed_at":"2026-08-03","claims_supported":["An official interchange specification represents uncertain, approximate, unspecified, set-valued, and interval dates.","The candidate can reuse standardized uncertainty syntax rather than inventing a new date notation.","EDTF does not determine historical calendar identity or perform source-critical conversion."]},{"source_id":"S5","title":"Calendars and Their History","publisher":"NASA Goddard Space Flight Center","url":"https://eclipse.gsfc.nasa.gov/SEhelp/calendars.html","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"n.d.","accessed_at":"2026-08-03","claims_supported":["Observational calendars have short-term uncertainty, and historical chronology depends on epochs, intercalation, locality, astronomical observations, and changing practice.","Early Hebrew calendrical evidence is fragmentary and sometimes contradictory, with historical disagreement about epochs and intercalation.","Islamic dates can differ according to observational, calculated, civil, and chronological conventions, with disagreement among countries, religious leaders, and scientists.","Conversion must not be represented as deciding living religious observance or proving a historical event."]},{"source_id":"S6","title":"Classes & Properties Declarations of CIDOC-CRM version 7.1.3","publisher":"CIDOC Conceptual Reference Model Special Interest Group","url":"https://cidoc-crm.org/html/cidoc_crm_v7.1.3.html","source_class":"STANDARD","publication_date":"2024","accessed_at":"2026-08-03","claims_supported":["CIDOC CRM already models uncertain maximum temporal extents and interval-valued time primitives.","Multiple evidence-derived extents can be retained and intersected for information integration.","The proposed concordance output can map to established cultural-heritage temporal semantics."]},{"source_id":"S7","title":"Digital Humanities Advancement Grants 2025 Notice of Funding Opportunity","publisher":"National Endowment for the Humanities","url":"https://www.neh.gov/sites/default/files/2025-01/DHAG%20NOFO%202025%20-%20Jan%20extension.pdf","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2025-01-09","accessed_at":"2026-08-03","claims_supported":["NEH identified accredited higher-education institutions and eligible nonprofits as authorizable applicants for digital-humanities pilots, prototypes, workflow evaluation, and scaled releases.","The program's Level I, II, and III scopes corresponded to pilots, working prototypes, and mature-project launches.","Published award bands of up to $75,000, $75,001-$150,000, and $150,001-$350,000 provide a public resource-scope benchmark, although this notice is not evidence of current availability or a commitment to this candidate."]},{"source_id":"S8","title":"National employment and wage data by occupation, May 2025","publisher":"U.S. Bureau of Labor Statistics","url":"https://www.bls.gov/news.release/ocwage.t01.htm","source_class":"OFFICIAL_ORGANIZATION_DATA","publication_date":"2026-05-15","accessed_at":"2026-08-03","claims_supported":["National mean annual wages were $148,100 for software developers, $81,240 for historians, $92,040 for postsecondary philosophy and religion teachers, and $70,640 for archivists.","These wage anchors support resource-equivalent staffing estimates but exclude project-specific benefits, indirect costs, specialist scarcity, and procurement."]}],"problem_evidence":{"support":"STRONG","rationale":"Independent research directly reports that humanities dates are often partial, uncertain, inconsistent, and expressed in multiple calendars; the problem was shared across several projects and existing project-specific solutions were difficult to reuse. NASA documentation confirms that religious and historical calendars introduce substantive uncertainty through observation, epochs, intercalation, locality, and competing conventions. What remains unmeasured is the prevalence and labor cost specifically within history-of-religions corpora.","source_ids":["S1","S5"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"DHTech participants and projects including the Princeton Geniza Project, Shakespeare and Company Project, and Islamic Scientific Manuscripts Initiative are identifiable users of closely analogous functionality, and Undate's authors explicitly seek reusable cross-project calendar support. NEH has funded the relevant classes of digital-humanities pilot and software work. No institution or funder was found expressing a commitment to adopt this exact governed concordance service.","source_ids":["S1","S2","S7"]},"prior_art":{"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Undate","similarity":"Directly addresses uncertain and partial humanities dates, preserves the original calendar and precision, converts Hebrew and Islamic dates to Gregorian earliest/latest bounds, and supports cross-calendar comparison. This overlaps the candidate's central transformation and safety-net rationale.","remaining_difference":"The candidate adds a governed intake contract, explicit case-level assumptions and sensitivities, historically scoped and versioned modules, independent scholarly exception review, discrepancy-driven reauthorization or retirement, capacity monitoring, and an explicit prohibition on deciding event truth or living observance. Evidence has not shown that this wrapper improves outcomes.","source_ids":["S1","S2"]},{"name":"Convertdate","similarity":"Provides reusable open-source conversion routines for numerous historical and religious calendars and exposes some alternative methods and day-boundary conventions.","remaining_difference":"It is an arithmetic library rather than a provenance-preserving concordance record and governed scholarly workflow; it lacks uncertainty propagation across disputed inputs, exception stewardship, and regeneration assays.","source_ids":["S3"]},{"name":"EDTF and CIDOC CRM temporal representations","similarity":"Established standards already represent uncertainty, approximation, sets, intervals, and maximum temporal extents for bibliographic and cultural-heritage data.","remaining_difference":"They specify representation and semantics, not source transcription, calendar identification, calendar-rule execution, sensitivity analysis, specialist adjudication, or module recovery.","source_ids":["S4","S6"]}],"distinctive_claim_remaining":"Within two pre-specified calendar-period-locality scopes, adding the candidate's provenance, assumption, sensitivity, exception, and regeneration workflow to existing Undate/Convertdate components will reduce median total analyst labor by at least 25% versus ordinary manual conversion while remaining no more than five percentage points worse on independently judged acceptability, and will reduce omitted material assumptions or unsupported point dates by at least 50% versus a minimally wrapped Undate/Convertdate comparator; after one seeded discrepancy-and-recalibration cycle, held-out acceptability and error rates will return to or improve on their pre-discrepancy levels. This is contrastive and falsifiable but untested.","confidence":"HIGH"},"implementation_evidence":{"support":"MODERATE","rationale":"Open-source libraries already provide calendar-aware partial-date objects, conversion routines, tests, examples, and permissive licenses; EDTF and CIDOC CRM provide output semantics. A two-module prototype is therefore technically plausible. The difficult parts are scholarly rather than computational: historically valid module scoping, provenance completeness, specialist availability, sensitivity semantics, blinded evaluation, governance, and recovery after discrepancies. No external evidence demonstrates the complete workflow, its labor advantage, or multi-cycle regeneration.","source_ids":["S1","S2","S3","S4","S5","S6"]},"scores":{"meaningful_impact":{"score":3,"rationale":"The service could reduce duplicated chronology work and prevent false precision, but domain-specific prevalence, baseline labor, and realized consequences remain unmeasured.","source_ids":["S1","S5"]},"stakeholder_pull":{"score":4,"rationale":"Several named humanities projects already developed or used analogous partial-date and calendar functionality, showing real pull, though not commitment to this exact service.","source_ids":["S1","S2"]},"incremental_advantage":{"score":2,"rationale":"Undate already implements much of the claimed transformation; the remaining advantage is mainly workflow governance and quality control and has not been tested.","source_ids":["S1","S2","S3"]},"distinctiveness_plausibility":{"score":2,"rationale":"The stewardship, sensitivity, and recovery package is distinguishable, but it may be ordinary responsible implementation around substantially overlapping prior art rather than a distinct innovation.","source_ids":["S1","S2","S4","S6"]},"technical_implementability":{"score":4,"rationale":"Core conversion and uncertainty primitives exist as tested open-source components and established representations; adding a bounded service layer is technically credible.","source_ids":["S2","S3","S4","S6"]},"adoption_authority_feasibility":{"score":4,"rationale":"A principal investigator or research-data curator can authorize a historical-research pilot using accessible sources, while interpretive and living-practice authority can remain outside the service. Eligible institutions are a recognized funding class.","source_ids":["S1","S5","S7"]},"evidence_readiness":{"score":3,"rationale":"The intervention has observable outputs and credible comparators, but no corpus audit, workload measurements, head-to-head benchmark, or recovery-cycle data currently exist.","source_ids":["S1","S2","S3"]},"safety_net_benefit":{"score":4,"rationale":"Preserving original expressions, intervals, assumptions, and disputed alternatives directly addresses false precision and misuse of observational or historically variable calendar rules.","source_ids":["S4","S5","S6"]},"scalability":{"score":3,"rationale":"Software execution and standardized outputs can scale, but every calendar-period-locality module requires expert validation and maintenance, and specialist exceptions can become the bottleneck.","source_ids":["S1","S2","S3","S5"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Preregister and run the 24-expression, two-module, three-route comparison; compensate independent calendar specialists and reviewers; instrument labor and errors; run one discrepancy/recovery cycle.","confidence":"MODERATE","assumptions":["Approximately 0.05-0.10 software-developer FTE, 0.10-0.20 historian or religion-scholar FTE, and 80-160 hours of specialist/reviewer effort.","Accessible, nonrestricted expressions and existing open-source components are used.","Salary-equivalent labor is loaded for benefits, stipends, and modest indirect costs; no production interface is built."],"source_ids":["S7","S8"]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Harden two modules and the uncertainty/provenance schema; implement intake validation, audit logs, tests, documentation, exception workflow, versioning, and rollback for one institutional research project.","confidence":"MODERATE","assumptions":["Approximately 0.5-1.0 developer FTE, 0.25-0.5 curator/historian FTE, specialist stipends, accessibility review, and institutional indirect costs.","Built by extending Undate/Convertdate under their open-source licenses rather than reimplementing calendar arithmetic.","Scope remains two modules and research use only."],"source_ids":["S2","S3","S7","S8"]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Launch a supported service with four to eight validated modules, API or web intake, security and access controls, monitoring, documentation, training, independent QA, preservation, and partner onboarding.","confidence":"LOW","assumptions":["Roughly two to four resource-equivalent FTE-years across software, data curation, project management, chronology expertise, and QA over 12-24 months.","Specialist review and institutional indirect costs are material.","The estimate excludes acquiring or digitizing restricted source corpora."],"source_ids":["S7","S8"]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Maintain a small service: software updates, module regression tests and reauthorization, discrepancy response, documentation, preservation, hosting, curator stewardship, and specialist exception stipends.","confidence":"LOW","assumptions":["Approximately 0.5 developer FTE plus 0.5 curator/historian FTE, part-time specialist coverage, and institutional overhead.","Four to eight modules with moderate case volume; large-scale digitization or continuous observance scheduling is excluded.","NEH's cited program disallowed ordinary recurring maintenance, so a separate institutional sustainability source would be needed."],"source_ids":["S7","S8"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"Primary research and official calendar documentation independently establish multi-calendar uncertainty, inconsistent historical-date representations, and repeated project-specific handling.","source_ids":["S1","S5"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"Named digital-humanities projects and their institutional researchers already use closely analogous functionality, and eligible universities or nonprofits can authorize and seek funding for bounded pilots. This is adopter credibility, not a commitment to this candidate.","source_ids":["S1","S2","S7"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The remaining claim is explicitly relative to manual conversion and minimally wrapped Undate/Convertdate, with thresholds for labor, independent acceptability, assumption omissions, unsupported precision, and post-recalibration recovery.","source_ids":["S1","S2","S3"]},"bounded_next_evidence_step":{"status":"YES","reason":"A 24-expression, two-module, three-route study with independent blinded assessment, fixed outcomes, and one seeded regeneration cycle is bounded and falsifiable.","source_ids":["S1","S2","S3"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"A research pilot can be restricted to already accessible historical expressions, retain original notation, avoid claims about event truth or living observance, and quarantine failed modules. Wider deployment would require source-specific access and authority review.","source_ids":["S4","S5","S6"]},"credible_cost_scope_and_range":{"status":"YES","reason":"The four stages are tied to explicit staffing assumptions, current federal occupational wages, existing open-source components, and public digital-humanities project scopes. Operational and recurring estimates remain low-confidence until time logs and institutional overhead are observed.","source_ids":["S2","S3","S7","S8"]}},"next_evidence_step":"With one participating historical-research project, freeze two calendar-period-locality modules and select 24 already accessible expressions spanning routine and ambiguous cases. Have independent analysts process every expression through three randomized routes: ordinary manual scholarship, minimally wrapped Undate/Convertdate, and the proposed governed service. Blind two qualified reviewers to route and precommit primary criteria: candidate median total operating labor at least 25% below manual; acceptable-concordance rate no more than five percentage points below manual; material-assumption omission or unsupported-point-date rate at least 50% below the minimally wrapped software route; no wrong-variant execution left unquarantined; and no downstream treatment as an authoritative event or observance date. Seed one documented rule discrepancy, recalibrate the affected module, then rerun a fixed held-out suite. Falsify the incremental claim if any primary threshold fails, if gains disappear after specialist and correction labor is counted, if reviewers cannot distinguish acceptable concordances from false precision, or if post-recalibration performance does not recover to the pre-discrepancy level. Report module-development cost separately and show amortization at 25, 100, and 500 cases rather than hiding it in operating time.","blocking_evidence":["No corpus-level estimate shows how frequently religious-studies chronology work encounters reusable conversion rather than irreducible textual or historical interpretation.","No head-to-head evidence compares the proposed governance wrapper with Undate/Convertdate or ordinary manual scholarship.","No measured turnover, specialist saturation, discrepancy rate, module-degradation rate, or successful regeneration cycle exists.","No named institution has committed staff, data, or authority to the proposed pilot.","Cost bands lack project-specific time logs, benefit rates, indirect-cost rates, and specialist-market prices.","World novelty, patentability, freedom to operate beyond the observed open-source licenses, market size, and realized impact were not measured."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"Bounded web research found substantial collision with Undate, Convertdate, EDTF, and CIDOC CRM but cannot establish exhaustive world novelty. Patentability, freedom to operate beyond the identified Apache-2.0 and MIT components, market size, and realized impact remain explicitly unmeasured.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Secure one institutional research partner, a named chronological-data curator, and qualified specialists for both module scopes.","Preregister the three-route 24-expression protocol, allocation, blinded review, thresholds, exclusions, and incident rules.","Measure total labor, acceptability, assumption completeness, false precision, wrong-variant execution, exceptions, corrections, and downstream review for all routes.","Demonstrate one discrepancy-driven recalibration and held-out recovery cycle without silent rule changes.","Produce observed startup and recurring-resource estimates from time logs and document whether the remaining advantage over Undate/Convertdate is large enough to justify a separate service."],"reason":"Web evidence establishes the problem, adopter class, implementability, and substantial prior-art overlap. The only remaining decision-relevant claim is an incremental workflow and quality advantage over existing software and manual scholarship; resolving it requires partner fieldwork with real expressions, specialist labor, blinded review, and live regeneration testing rather than additional bounded web search."},"proposal_index":4}