{"schema_version":1,"research_id":"eoa_inverse_innovation_exp06_external_evaluation_20260803","source_assessment_id":"catalytic_pathway_enablement__sociology_anthropology:P2:v0","cell_id":"catalytic_pathway_enablement__sociology_anthropology","search_queries":["official qualitative data repository deposit ethnographic data de-identification metadata deposit workflow","ICPSR Guide to Social Science Data Preparation and Archiving deposit confidential qualitative data","UK Data Service deposit qualitative data anonymisation consent repository requirements","Qualitative Data Repository deposit guidance data deposit curation","site:qualitative-research.net qualitative data sharing barriers archiving ethnography researchers study","qualitative researchers data sharing archive barriers empirical study ethnography repository","QDR curation handbook official deposit workflow qualitative data repository PDF","CARE Principles Indigenous Data Governance official cultural authority archives","qualitative data de-identification software repository deposit automated tool ICPSR official","QualiAnon software qualitative data anonymization tool official study","Dataverse curation workflow automation checksums metadata deposit official","Data Curation Network CURATE checklist repository workflow official","NIH Data Management and Sharing Policy official qualitative data 2026","NSF data management sharing policy official social science research data repository","CoreTrustSeal requirements repository appraisal preservation metadata official 2026","Local Contexts Traditional Knowledge Labels archive authority official","\"Barriers and facilitators to qualitative data sharing in the United States\" DOI","\"Are we ready to share qualitative research data?\" DOI","\"Enabling qualitative research data sharing using a natural language processing pipeline for deidentification\""],"sources":[{"source_id":"S1","title":"Barriers and facilitators to qualitative data sharing in the United States: A survey of qualitative researchers","publisher":"PLOS ONE","url":"https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0261719","source_class":"PRIMARY_RESEARCH","publication_date":"2021-12-31","accessed_at":"2026-08-03","claims_supported":["In a nonprobability survey of 425 U.S. qualitative researchers, 96% had never shared qualitative data through a repository.","The most frequently reported concerns included lack of participant permission, sensitivity, breach of trust, institutional policy, and inadequate de-identification.","Respondents identified covered repository costs, ethics/compliance guidance, anonymization assistance, and repository consultation as facilitators.","The study cautions that its U.S. health-oriented, criterion-based sample limits generalizability, including to ethnography specifically."]},{"source_id":"S2","title":"Enabling qualitative research data sharing using a natural language processing pipeline for deidentification: moving beyond HIPAA Safe Harbor identifiers","publisher":"JAMIA Open / Oxford University Press","url":"https://academic.oup.com/jamiaopen/article/4/3/ooab069/6356665","source_class":"PRIMARY_RESEARCH","publication_date":"2021-08-23","accessed_at":"2026-08-03","claims_supported":["An NLP de-identification pipeline evaluated on two qualitative datasets totaling 1.2 million words achieved an approximately 0.90 F1 score.","Most identifiers encountered were outside HIPAA Safe Harbor categories, demonstrating the limits of generic de-identification rules.","The authors state that some contextual identifiers cannot be reliably detected automatically and that trained human validation remains necessary.","The study explicitly notes that ethnographies and detailed field notes may contain more identifiers and present greater challenges than the evaluated materials."]},{"source_id":"S3","title":"Prepare your data collection for deposit","publisher":"UK Data Service","url":"https://ukdataservice.ac.uk/help/deposit-data/prepare-your-data-for-deposit/","source_class":"OFFICIAL_GUIDANCE","publication_date":"n.d.","accessed_at":"2026-08-03","claims_supported":["Deposits must be cleaned, quality-assured, documented, and accompanied by metadata covering ethics, consent, disclosure risk, access conditions, formats, and rights holders.","Missing required information delays deposit.","Privacy protection requires combined transparency, technical protection, and governance; neither consent nor anonymization alone is sufficient.","The repository requires lawful-basis documentation, disclosure-risk assessment, appropriate access conditions, validation, removal of temporary files, and contextual documentation."]},{"source_id":"S4","title":"QDR Curation Handbook","publisher":"Qualitative Data Repository","url":"https://zenodo.org/records/4672678","source_class":"OFFICIAL_GUIDANCE","publication_date":"2021-04-08","accessed_at":"2026-08-03","claims_supported":["QDR already uses a standardized internal process spanning depositor contact, file processing, Dataverse operations, publication, and post-publication activity.","A first-party handbook exists for qualitative-data curation and digital preservation, making a governed repeatable curation pathway direct prior art."]},{"source_id":"S5","title":"Repository Operations","publisher":"Inter-university Consortium for Political and Social Research","url":"https://www.icpsr.umich.edu/sites/icpsr/about/repository-operations","source_class":"OFFICIAL_ORGANIZATION_DATA","publication_date":"n.d.","accessed_at":"2026-08-03","claims_supported":["ICPSR already operates secure authenticated ingest, rights attestation, checksums, quality assurance, confidentiality review, format transformation, metadata enhancement, and preservation workflows.","ICPSR records transformations through processing histories and syntax files, providing auditability and reproducibility.","Repository staff and repository policies retain selection, curation, preservation, access, and deaccession authority."]},{"source_id":"S6","title":"The DCN CURATE(D) Steps","publisher":"Data Curation Network","url":"https://datacuration.network/outputs/workflows/","source_class":"OFFICIAL_GUIDANCE","publication_date":"2022","accessed_at":"2026-08-03","claims_supported":["The Data Curation Network already provides standardized, revised curation steps and checklists.","Existing steps cover file and documentation checks, understanding data, requesting missing information, augmenting metadata, transforming formats, evaluating FAIRness, and documenting curation activity."]},{"source_id":"S7","title":"Labels","publisher":"Local Contexts","url":"https://localcontexts.org/labels/about-the-labels/","source_class":"OFFICIAL_GUIDANCE","publication_date":"n.d.","accessed_at":"2026-08-03","claims_supported":["Traditional Knowledge and Biocultural Labels encode community-specific provenance, protocols, permissions, and future-use expectations in data infrastructure.","Permission beyond generally approved uses requires direct engagement with primary cultural authorities.","Use of the labels requires community decision-making through community-defined processes; a repository or workbench cannot manufacture this authority."]},{"source_id":"S8","title":"Preparing Your Data Management and Sharing Plan","publisher":"U.S. National Science Foundation","url":"https://www.nsf.gov/funding/data-management-plan","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2026","accessed_at":"2026-08-03","claims_supported":["NSF proposals require a data management and sharing plan.","NSF-funded investigators are expected to share primary data and supporting materials within a reasonable time and at no more than incremental cost.","Plans address metadata standards, privacy and confidentiality, intellectual-property and other rights, reuse, archiving, and preservation, supplying an identifiable funder-side need for credible deposit pathways."]}],"problem_evidence":{"support":"MODERATE","rationale":"The broader problem is visible and consequential: qualitative data sharing is rare, researchers report permission, sensitivity, trust, institutional-policy, de-identification, guidance, and funding barriers, and UK Data Service guidance confirms that incomplete ethics, metadata, validation, and documentation delay deposit. However, the searched evidence does not quantify the proposal's narrower population of completed ethnographic collections that are already authorized yet remain stranded principally because of repeatable technical conversion work. Unique authority and contextual disputes may dominate some ethnographic cases.","source_ids":["S1","S2","S3"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"Researchers express demand for repository-funded costs, guidance, anonymization assistance, and consultation; NSF imposes data-management and sharing obligations; and QDR, ICPSR, UK Data Service, and DCN are identifiable repositories or repository networks already performing related work. This establishes plausible adopters, authorizers, and funders, but no named institution has expressed intent to adopt this specific external pre-deposit workbench or authorized access to the proposed shadow-test collections.","source_ids":["S1","S3","S4","S5","S6","S8"]},"prior_art":{"proximity":"ESTABLISHED_PRACTICE","closest_analogues":[{"name":"QDR standardized qualitative-data curation process","similarity":"Direct analogue: a qualitative-data repository already standardizes depositor contact, file processing, repository operations, publication, and subsequent curation activity.","remaining_difference":"The proposal relocates and packages more work into a separable pre-deposit workbench and makes object-level authority mapping, secure regeneration, capacity metrics, and a comparative turnover assay explicit; evidence that this configuration improves QDR-like practice is absent.","source_ids":["S4"]},{"name":"ICPSR secure ingest and professional curation workflow","similarity":"Substantial functional overlap in secure intake, rights attestation, checksums, confidentiality review, quality assurance, transformations, metadata, audit histories, preservation, and independent repository authority.","remaining_difference":"The proposal adds an upstream ethnography-specific eligibility and quarantine layer plus repeated ready-state verification; ICPSR's published page does not establish the same shadow-assay or community-authority-to-object mapping design.","source_ids":["S5"]},{"name":"Data Curation Network CURATE(D) workflow","similarity":"Existing reusable checklists already cover inventory, comprehension, missing-information requests, metadata augmentation, format transformation, evaluation, and documentation.","remaining_difference":"CURATE(D) is a general curation practice rather than an ethnography-specific secure automation environment with explicit authority attestations, case purge, saturation monitoring, and safety falsifiers.","source_ids":["S6"]},{"name":"Automated qualitative-text de-identification pipeline","similarity":"Demonstrates that a repeatable computational tool can reduce part of the manual disclosure-control burden while retaining human review.","remaining_difference":"It addresses text de-identification only, was evaluated on health-oriented narratives and interviews rather than full multimodal ethnographic archives, and cannot resolve consent, collective authority, contextual integrity, packaging, or repository acceptance.","source_ids":["S2"]}],"distinctive_claim_remaining":"For deposits whose preservation and proposed access class are already authorized, an institution-side, separable workbench combining object-level constraint mapping, prevalidated transformations, auditable automation, bounded specialist review, quarantine, and verified purge will reduce total preparation labor and elapsed time across repeated deposits versus eligibility-matched bespoke preparation, without increasing missed restrictions, disclosure findings, context loss, corruption, excluded-object concealment, repository revisions, or downstream burden. The claim fails if gains arise from cleaner-case selection, proportional extra curator or community labor, weakened standards, omitted difficult objects, or incomplete recovery between cases.","confidence":"HIGH"},"implementation_evidence":{"support":"MODERATE","rationale":"Component feasibility is well supported: major repositories operate secure ingest, checksums, quality assurance, transformations, metadata, access controls, and audit histories; standardized curation checklists exist; and a qualitative-text NLP pipeline reached approximately 0.90 F1. Legal and governance boundaries can be represented through attestations, access conditions, and Local Contexts mechanisms. Whole-system feasibility remains unproven for heterogeneous ethnographic text, media, field notes, relational identifiers, community permissions, contextual preservation, secure purge, and labor transfer. Human and community review cannot safely be eliminated.","source_ids":["S2","S3","S4","S5","S6","S7"]},"scores":{"meaningful_impact":{"score":4,"rationale":"Responsible preservation and reuse can conserve costly qualitative records and reduce repeated participant data collection, while safer preparation protects trust and confidentiality. The size of the specifically eligible stranded-ethnography population and realized impact are unmeasured.","source_ids":["S1","S2","S8"]},"stakeholder_pull":{"score":4,"rationale":"Researchers report demand for funding, guidance, anonymization help, and consultation; funders require sharing plans; and repositories already invest in curation. Pull for this exact workbench, rather than existing repository services, has not been demonstrated.","source_ids":["S1","S4","S5","S8"]},"incremental_advantage":{"score":2,"rationale":"The proposal integrates useful safeguards and measurement, but most operational elements already appear in QDR, ICPSR, UK Data Service, and DCN practice. Comparative advantage over those workflows is purely prospective.","source_ids":["S3","S4","S5","S6"]},"distinctiveness_plausibility":{"score":2,"rationale":"Object-level authority mapping, verified regeneration, saturation metrics, and a pre-registered turnover/selectivity assay form a contrastive configuration, but the underlying governed curation pathway is established practice and Local Contexts already supports community-specific permission metadata.","source_ids":["S4","S5","S6","S7"]},"technical_implementability":{"score":4,"rationale":"Secure ingest, checksums, transformations, metadata schemas, audit histories, and assisted de-identification are demonstrably implementable. Automated contextual disclosure and authority judgments remain incomplete and require specialists.","source_ids":["S2","S5","S6"]},"adoption_authority_feasibility":{"score":3,"rationale":"Repository and institutional data stewards have identifiable authority over technical intake and acceptance, while communities retain relevant cultural authority. A deployable governance split is plausible, but actual permissions, jurisdiction-specific privacy duties, and a named partner are unresolved.","source_ids":["S3","S5","S7","S8"]},"evidence_readiness":{"score":3,"rationale":"A six-case shadow assay with an archived bespoke comparator is bounded and measurable, but it requires permissioned sensitive records, reconstructable historical labor data, independent reviewers, and institution-specific authority that web evidence cannot supply.","source_ids":["S2","S4","S5"]},"safety_net_benefit":{"score":4,"rationale":"Quarantine and escalation could preserve an ordinary route for ambiguous or unsafe cases while helping eligible collections. Benefit depends on transparent eligibility and on avoiding false reassurance from automated validation.","source_ids":["S2","S3","S7"]},"scalability":{"score":3,"rationale":"Templates, checklists, automation, and repeated workflows can scale routine steps, but heterogeneous formats, contextual identifiers, community-specific protocols, specialist capacity, and repository receiving capacity constrain turnover.","source_ids":["S2","S4","S5","S6","S7"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"Pre-register and conduct the offline six-deposit shadow assay, including controlled storage, workbench configuration, labor reconstruction, independent archival/privacy/governance review, security verification, and analysis.","confidence":"LOW","assumptions":["No repository release or alteration of archival masters occurs.","Evaluation use of all controlled copies is already authorized.","The study uses approximately 0.5-1.0 FTE of combined archival, research, engineering, privacy, and project-management effort for three to six months.","Community-governance review is compensated and counted rather than treated as free labor.","The band is a resource-equivalent estimate; none of the eight sources supplies a 2026 price quote."],"source_ids":["S1","S2","S4","S5"]},"initial_deployment_startup":{"band_2026_usd":"250K_TO_1M","scope":"Design and build a single-institution secure pre-deposit environment, intake contract, authority and restriction schema, transformation recipes, validators, audit logging, purge controls, integration, threat assessment, governance documents, and regression fixtures.","confidence":"LOW","assumptions":["One repository family and a bounded set of supported formats are targeted.","Existing repository, identity, storage, checksum, and preservation infrastructure is reused.","Approximately two to four technical, archival, privacy, and governance staff-equivalents contribute over nine to twelve months.","No attempt is made to automate authority determinations or all audiovisual de-identification.","The scope is inferred from the functions documented in existing workflows, not vendor quotations."],"source_ids":["S3","S4","S5","S6","S7"]},"operational_launch":{"band_2026_usd":"250K_TO_1M","scope":"Launch a limited production service with trained steward and specialist coverage, depositor onboarding, security monitoring, quality sampling, incident response, capacity dashboard, repository handoff, and an independently reviewed initial cohort.","confidence":"LOW","assumptions":["Launch is limited to one institution or repository partnership.","One to two archival/specialist FTEs plus fractional engineering, security, privacy, and governance support are required.","Existing repository appraisal and acceptance remain independent and are not replaced.","Community-authority work is compensated case by case.","Direct salary, insurance, storage, and audit quotations remain unavailable."],"source_ids":["S3","S4","S5","S7"]},"annual_recurring":{"band_2026_usd":"250K_TO_1M","scope":"Operate and maintain the bounded service: stewardship and exception review, compensated governance input, secure compute and storage, testing, policy and recipe refresh, audits, incident readiness, training, and software maintenance.","confidence":"LOW","assumptions":["Annual volume remains within one bounded specialist lane rather than a national-scale service.","At least one archival steward and fractional specialist, engineering, security, and privacy capacity remain necessary.","Automation assists but does not replace contextual disclosure and authority review.","Repository receiving capacity is separately funded.","The estimate excludes long-term repository preservation costs after acceptance."],"source_ids":["S2","S4","S5","S7"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"Primary research and official deposit guidance demonstrate rare qualitative-data sharing and recurring permission, sensitivity, de-identification, metadata, validation, and documentation barriers. The evidence is weaker for the exact prevalence of already-authorized stranded ethnographic collections.","source_ids":["S1","S2","S3"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"QDR, ICPSR, UK Data Service, institutional data stewards, repository directors, and NSF-funded institutions are credible adopters, authorizers, or funders of deposit-preparation capability. No specific adoption commitment has been found.","source_ids":["S3","S4","S5","S8"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The remaining claim fixes eligibility and repository standards, compares total labor and elapsed time with bespoke preparation, measures repeated cycles and recovery, and specifies safety, context, selection, and labor-transfer falsifiers.","source_ids":["S2","S4","S5","S6"]},"bounded_next_evidence_step":{"status":"YES","reason":"An offline assay capped at six previously accepted deposits, with controlled copies, an archived bespoke comparator, independent review, fixed outcomes, and hard failure thresholds is bounded and reversible.","source_ids":["S2","S4","S5"]},"no_unresolved_safety_or_authority_stop":{"status":"UNCERTAIN","reason":"The proposed shadow design can avoid release and preserve repository authority, but no partner has confirmed evaluation permission for sensitive archived records, and community or cultural authority must be established case by case rather than inferred from institutional custody.","source_ids":["S3","S5","S7"]},"credible_cost_scope_and_range":{"status":"UNCERTAIN","reason":"The operational functions are scopeable from established repository workflows, but none of the eight sources provides current labor, security, software, audit, governance-review, or integration prices. The bands are transparent resource-equivalent estimates with low confidence.","source_ids":["S1","S4","S5","S6"]}},"next_evidence_step":"With a named repository and institutional data steward, pre-register an offline shadow assay using controlled copies of no more than six previously accepted ethnographic deposits for which evaluation use is explicitly authorized. Select cases before observing workbench results, stratify routine and complex formats, and include at least one predesignated case expected to be quarantined. For each case, reconstruct only the preparation evidence available at the original deposit date. Compare the workbench against archived bespoke preparation under the same eligibility, proposed access class, acceptance standard, and retained-object set. Measure total depositor, archivist, privacy, technical, specialist, and compensated community-governance labor; elapsed time; revisions; excluded objects; file integrity; metadata completeness; restriction fidelity; contextual preservation; disclosure findings; repository review burden; and purge/recovery effort. Use blinded independent repository and relevant governance reviewers on permitted manifests and samples. Falsify the claim on any authority mismatch, incorrect access class, new disclosure, unexplained content or relationship change, hidden excluded object, failed purge, inability to reroute an ineligible case, higher total labor, or no repeated-cycle time advantage. A pass supports only a separately authorized limited live pilot.","blocking_evidence":["No direct prevalence estimate was found for completed, already-authorized ethnographic collections stranded primarily by repeatable deposit-preparation work.","No named repository or institutional steward has committed to adoption, supplied six permissioned comparator deposits, or approved the shadow assay.","No comparative evidence shows that the integrated workbench reduces total labor or elapsed time relative to QDR-, ICPSR-, UK Data Service-, or DCN-like curation while holding eligibility and standards constant.","Existing de-identification evidence covers health-oriented text and does not establish safety for full ethnographic field notes, audiovisual media, relational identifiers, sacred material, or contextual reidentification.","No evidence establishes reliable automated mapping from consent and community protocols to individual archival objects; such mappings require human and authority-holder validation.","No direct 2026 cost quotations or institution-specific estimates were found for secure infrastructure, integration, specialist review, community-governance labor, audit, or recurring operations.","World novelty, patentability, freedom to operate, market size, and realized impact remain unmeasured."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"The search establishes dense, direct prior art for standardized qualitative-data curation, secure repository ingest, checksums, confidentiality review, metadata and format transformations, audit histories, assisted text de-identification, and community-specific permission metadata. It does not establish whether the exact combination of an institution-side ethnographic pre-deposit workbench, object-level authority mapping, verified case purge/regeneration, capacity controls, and a turnover/selectivity assay exists everywhere or was first proposed here. World novelty, patentability, freedom to operate, market size, and realized impact were not measured.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":false,"progress_targets":["Secure written authorization from a named repository and institutional data steward for evaluation use of up to six previously accepted deposits, including any required community or cultural-authority approvals.","Pre-register the matched shadow-assay protocol, retained-object rules, labor-accounting method, blinded review procedure, recovery test, and hard safety and efficacy falsifiers.","Obtain institution-specific 2026 resource estimates or quotations for secure infrastructure, implementation, specialist and community review, audit, launch, and recurring operation.","Measure whether the workbench produces a repeated-cycle advantage over the institution's existing QDR-, ICPSR-, UK Data Service-, or CURATE(D)-like workflow without shifting labor, selecting easy cases, weakening standards, or hiding exclusions.","Demonstrate restriction fidelity, contextual preservation, safe quarantine, complete purge, and verified ready-state recovery on permissioned controlled copies before any live deposit is processed."],"reason":"Web research verifies the broad problem, credible repository and funder actors, component feasibility, and a bounded comparative claim, but it also shows that governed qualitative-data curation is established practice. The only material remaining advantage is an empirical whole-workflow claim requiring permissioned sensitive deposits, proprietary workflow and labor records, independent human and community review, and secure recovery testing. Those facts cannot be resolved by further bounded web search. Because this is a STOP recommendation, repairable is false as required."},"proposal_index":2}