{"closest_prior_art":[{"name":"W3C Library Linked Data digital-preservation use case","overlap":"Tests a collection subset before migration, checks whether previously identified significant properties survive transformation, records preservation events, and creates linked object versions.","remaining_difference":"It does not specify poetry features, recurring sentinel inspections, longitudinal persistence or acceleration measures, bibliographic adjudication, or threshold-linked reprocessing and rollback.","source_ids":["SRC2"]},{"name":"KITAB poetry-line corpus quality control","overlap":"Uses corpus statistics to flag anomalous poetry lines introduced partly through automated format conversion, records flags in version metadata, and sends suspected problems for human judgment and correction.","remaining_difference":"It checks a narrow line-length/tag anomaly rather than comparing multiple form-bearing features against designated references across releases and transformation stages with condition bands and escalation authority.","source_ids":["SRC3"]},{"name":"TEI Guidelines for verse and textual features","overlap":"Provides established representations for verse lines, stanzas, indentation, typographic rendition, punctuation, corrections, responsibility, and documented editorial choices—the principal features the proposal would inspect.","remaining_difference":"It is an encoding standard, not a recurring condition-monitoring, trend-detection, or repair-escalation process.","source_ids":["SRC1"]}],"contrastive_claim_falsifier":"The contrast is falsified if an existing corpus control already performs recurring reference-based, multi-feature comparison across transformation stages with persistence thresholds and authority-linked reprocessing or rollback, or if the proposed pilot's flags fail to precede independently adjudicated loss, show poor inter-reviewer agreement, or cannot map confirmed crossings to feasible actions.","contrastive_claim_remaining":"For transforming poetry corpora, recurring comparison of fixed sentinels and stratified samples against designated references, with feature-specific histories and persistence thresholds, will detect analytically consequential loss earlier and activate scoped repair or rollback more reliably than schema validation, generic migration testing, or anomaly-only corpus quality control.","experiment_id":"eoa_inverse_innovation_exp13_second_slot_policy60_20260806","gates":{"adequate_source_search":{"rationale":"Four search lanes covered the proposal directly, verse/poetry and significant-properties terminology, encoding and preservation practices, and combinations of migration testing, corpus flags, human review, and provenance. Exactly four sources from four publishers were retained and opened, including an official encoding standard, W3C guidance, first-party corpus practice, and primary research.","source_ids":["SRC1","SRC2","SRC3","SRC4"],"status":"PASS"},"bounded_next_test":{"rationale":"The proposed read-only test is bounded to 30 sentinels and 70 stratified items, two recent processing stages, two independent bibliographers, and measurable outcomes including agreement, false alarms, persistence, effort, provenance sufficiency, and authority mapping. Subset testing against significant properties and human adjudication of corpus flags are established feasible components.","source_ids":["SRC2","SRC3"],"status":"PASS"},"distinct_testable_claim":{"rationale":"Adjacent work separately supplies feature-rich verse encoding, significant-property migration tests, version/event metadata, and poetry-line anomaly flags. The testable distinction is their integration into recurring, reference-based, multi-feature stage histories whose persistence thresholds trigger owned repair or rollback.","source_ids":["SRC1","SRC2","SRC3"],"status":"PASS"},"no_obvious_safety_or_authority_stop":{"rationale":"The first step is read-only, leaves public derivatives unchanged, limits access to authorized samples, and assigns repair, derivative replacement, rollback, and suspension decisions to named roles. Explicit halts cover inadequate references, restricted-material risk, and inconsistent labeling; no obvious safety or authority stop blocks the assessment.","source_ids":["SRC2"],"status":"PASS"},"supported_problem":{"rationale":"TEI treats the verse line as fundamental and supports explicit preservation of stanzas, indentation, rendition, punctuation, and editorial decisions. KITAB reports actual poetry-line problems associated partly with automated conversion and requiring human review. Primary research demonstrates that OCR errors and insufficient metadata can introduce missed results and false positives in corpus analysis. Evidence is indirect for gradual release-to-release deterioration in poetry specifically, but the problem is visible and researchable.","source_ids":["SRC1","SRC3","SRC4"],"status":"PASS"}},"prior_art_disposition":"ADJACENT_PRIOR_ART","problem_evidence":{"finding":"The component problem is visible: poetry contains structurally and typographically significant features that encoding policy must represent; automated corpus conversion can produce poetry-line anomalies; and OCR defects can distort scholarly corpus results while remaining compatible with searchable text. The retained sources do not directly demonstrate persistent or accelerating loss across successive releases of one poetry corpus, so support is partial.","source_ids":["SRC1","SRC3","SRC4"],"status":"PARTLY_SUPPORTED"},"research_id":"eoa_inverse_innovation_exp13_light_screen_20260806","schema_version":1,"screen_id":"E13P150","screen_survival":true,"search_lanes":{"component_combination":{"no_result_note":null,"queries":["\"migration\" \"significant properties\" compare source target digital preservation validation","\"poetry\" \"quality control\" digitization corpus line breaks OCR","\"sentinel\" passages digital corpus quality monitoring OCR"],"source_ids":["SRC2","SRC3"]},"direct_problem_and_intervention":{"no_result_note":"No direct match was found for a named recurring Poetic-Form Fidelity Watch combining reference passages, stratified sampling, longitudinal feature-loss bands, and rollback thresholds.","queries":["poetry corpus OCR lineation stanza indentation digitization quality errors scholarly analysis","\"poetry\" \"quality control\" digitization corpus line breaks OCR","\"sentinel\" passages digital corpus quality monitoring OCR"],"source_ids":["SRC3","SRC4"]},"products_practices_and_standards":{"no_result_note":null,"queries":["TEI verse encoding line stanza indentation page breaks typography poetry guidelines","digital preservation quality assurance migration significant properties validation recurring monitoring","site:loc.gov digitization quality review sampling OCR layout metadata guidelines"],"source_ids":["SRC1","SRC2","SRC3"]},"synonyms_and_historical_terms":{"no_result_note":null,"queries":["\"verse\" \"significant properties\" digital preservation migration","digital preservation migration quality assurance compare source target significant properties tool","OCR errors impact digital humanities corpus analysis literature digitized newspapers primary research"],"source_ids":["SRC1","SRC2","SRC4"]}},"sources":[{"claims_supported":["Verse lines and line groups such as stanzas have established structural encodings.","Indentation and other typographic rendition can be recorded when considered essential.","Punctuation, corrections, responsibility, gaps, and editorial choices can be explicitly represented and documented."],"publisher":"Text Encoding Initiative Consortium","source_id":"SRC1","source_type":"OFFICIAL_STANDARD","title":"TEI Guidelines: Elements Available in All TEI Documents—Core Tags for Verse","url":"https://guidelines.tei-c.de/en/html/CO.html"},{"claims_supported":["Digital preservation extends beyond backup and includes migration plus lifecycle metadata.","A migration may be tested on a collection subset for loss of previously identified significant properties.","Preservation events, versions, formats, tools, and agents can be recorded and linked."],"publisher":"World Wide Web Consortium Library Linked Data Incubator Group","source_id":"SRC2","source_type":"OFFICIAL_GUIDANCE","title":"Use Case Digital Preservation—Library Linked Data","url":"https://www.w3.org/2005/Incubator/lld/wiki/Use_Case_Digital_Preservation.html"},{"claims_supported":["Large poetry-containing corpora present practical quality-control challenges.","Statistical outliers in annotated poetry-line lengths can identify problems concentrated in particular files.","Some tags arise from automatic format conversion; suspected cases require human judgment and can be flagged in version metadata."],"publisher":"KITAB Project","source_id":"SRC3","source_type":"FIRST_PARTY_PRODUCT","title":"Using the Many to Spot the Few","url":"https://kitab-project.org/spot-the-few"},{"claims_supported":["OCR conversion produced substantial errors in a scholarly text corpus.","Comparison of raw and corrected datasets showed that errors caused missed results and false positives.","Metadata quality materially affected analysis, and interdisciplinary interpretation was needed to evaluate outputs."],"publisher":"Digital Humanities Quarterly","source_id":"SRC4","source_type":"PRIMARY_RESEARCH","title":"Mining for the Meanings of a Murder: The Impact of OCR Quality on the Use of Digitized Historical Newspapers","url":"https://dhq.digitalhumanities.org/vol/8/1/000168/000168.html"}],"world_novelty_boundary":"This bounded public-web screen found adjacent standards and practices but no opened source implementing the full combination. It cannot establish world novelty, patentability, market size, expert acceptance, realized value, or absence of unpublished, proprietary, differently worded, or patent prior art."}