{
  "schema": "fictameridian-census-publication/v2",
  "source": "census_work",
  "run_id": "20261018T003131Z",
  "as_of": "2026-10-18T00:31:31Z",
  "domain": "bocg",
  "published_at": "2026-10-18T02:39:38.997327+00:00",
  "prefect_flow_run_id": "db9beba6-c71d-4c3a-bc45-dbb49f731902",
  "metrics": {
    "relevant_units": 3754,
    "relevant_spans": 6772,
    "candidate_spans": 1281,
    "ontology_divisions": 41
  },
  "visuals": {
    "distinct_units": 18017,
    "roots": [
      {
        "label": "Forensic mail",
        "count": 14244
      },
      {
        "label": "Forensic documents",
        "count": 3074
      },
      {
        "label": "Archives",
        "count": 570
      },
      {
        "label": "Working folder",
        "count": 129
      }
    ],
    "total_spans": 35169,
    "unique_spans": 19323,
    "span_classes": {
      "unique": 19323,
      "exact-duplicate": 5257,
      "near-duplicate": 4388,
      "boilerplate": 3761,
      "quoted-reply": 2440
    },
    "density": [
      {
        "root": "archives",
        "in_domain_units": 10,
        "means": {
          "type_token": 0.267,
          "numeric": 0.051,
          "decision_per_1k": 2.443,
          "question_per_1k": 5.885,
          "quoted": 0.001,
          "boilerplate": 0.478,
          "domain_terms_per_1k": 198.933
        },
        "coverage": {
          "type_token": 10,
          "numeric": 10,
          "decision_per_1k": 10,
          "question_per_1k": 10,
          "quoted": 10,
          "boilerplate": 10,
          "domain_terms_per_1k": 10
        },
        "label": "Archives"
      },
      {
        "root": "working-folder",
        "in_domain_units": 42,
        "means": {
          "type_token": 0.427,
          "numeric": 0.082,
          "decision_per_1k": 4.895,
          "question_per_1k": 3.36,
          "quoted": 0.013,
          "boilerplate": 0.54,
          "domain_terms_per_1k": 83.413
        },
        "coverage": {
          "type_token": 42,
          "numeric": 42,
          "decision_per_1k": 42,
          "question_per_1k": 42,
          "quoted": 42,
          "boilerplate": 42,
          "domain_terms_per_1k": 42
        },
        "label": "Working folder"
      },
      {
        "root": "forensic-mail",
        "in_domain_units": 3342,
        "means": {
          "type_token": 0.551,
          "numeric": 0.06,
          "decision_per_1k": 2.879,
          "question_per_1k": 3.365,
          "quoted": 0.056,
          "boilerplate": null,
          "domain_terms_per_1k": 68.066
        },
        "coverage": {
          "type_token": 3342,
          "numeric": 3342,
          "decision_per_1k": 3342,
          "question_per_1k": 3342,
          "quoted": 3342,
          "boilerplate": 0,
          "domain_terms_per_1k": 3342
        },
        "label": "Forensic mail"
      },
      {
        "root": "forensic-docs",
        "in_domain_units": 360,
        "means": {
          "type_token": 0.376,
          "numeric": 0.108,
          "decision_per_1k": 3.199,
          "question_per_1k": 1.767,
          "quoted": 0.001,
          "boilerplate": 0.414,
          "domain_terms_per_1k": 704.109
        },
        "coverage": {
          "type_token": 357,
          "numeric": 357,
          "decision_per_1k": 357,
          "question_per_1k": 357,
          "quoted": 357,
          "boilerplate": 357,
          "domain_terms_per_1k": 360
        },
        "label": "Forensic documents"
      }
    ],
    "raw": {
      "totals": {
        "files": 19992,
        "bytes": 8312069297,
        "text_units": 17147,
        "lines": 2270994,
        "tokens_words": 12526708
      },
      "roots": [
        {
          "label": "Archives",
          "files": 583,
          "bytes": 21584058,
          "text_units": 581,
          "lines": 498243,
          "tokens_words": 2085570
        },
        {
          "label": "Working folder",
          "files": 232,
          "bytes": 38759982,
          "text_units": 227,
          "lines": 787753,
          "tokens_words": 3994769
        },
        {
          "label": "Forensic mail",
          "files": 15499,
          "bytes": 479018926,
          "text_units": 15499,
          "lines": 615240,
          "tokens_words": 4231919
        },
        {
          "label": "Forensic documents",
          "files": 3678,
          "bytes": 7772706331,
          "text_units": 840,
          "lines": 369758,
          "tokens_words": 2214450
        }
      ]
    },
    "relevance": {
      "schema": "fictameridian-relevance-mask/v1",
      "ontology": {
        "name": "bocg",
        "release_tag": "v0.6.2",
        "commit": "99370e504e79d29cd60423e3d643a7482ecf7f0e",
        "sha256": "a388d8625e420934fac3f171085a22ffbc4e5a10b9b520154d7d7cc4f9ccd43b"
      },
      "ontology_rule": "in-domain = >= 2 strong (df < 0.05) + >= 1 seed term + >= 3.0/1k words; generic df >= 0.25 excluded",
      "human_rulings": 0,
      "metrics": {
        "relevant_units": 3754,
        "relevant_spans": 6772,
        "candidate_spans": 1281,
        "ontology_divisions": 41
      },
      "divisions": [
        {
          "id": "buyside_investment_compliance",
          "units": 0,
          "spans": 1,
          "candidates": 1
        },
        {
          "id": "buyside_investment_operations",
          "units": 2,
          "spans": 19,
          "candidates": 5
        },
        {
          "id": "buyside_ldi_overlay",
          "units": 1,
          "spans": 13,
          "candidates": 0
        },
        {
          "id": "buyside_performance_risk_reporting",
          "units": 0,
          "spans": 1,
          "candidates": 0
        },
        {
          "id": "buyside_portfolio_management",
          "units": 42,
          "spans": 43,
          "candidates": 8
        },
        {
          "id": "buyside_trading_execution",
          "units": 148,
          "spans": 321,
          "candidates": 50
        },
        {
          "id": "capital_markets_origination_combined",
          "units": 19,
          "spans": 89,
          "candidates": 15
        },
        {
          "id": "client_lifecycle_kyc",
          "units": 1,
          "spans": 1,
          "candidates": 0
        },
        {
          "id": "collateral_margin_management",
          "units": 1,
          "spans": 7,
          "candidates": 2
        },
        {
          "id": "commodities_physical_operations",
          "units": 2,
          "spans": 10,
          "candidates": 1
        },
        {
          "id": "commodities_trading",
          "units": 487,
          "spans": 516,
          "candidates": 96
        },
        {
          "id": "credit_trading",
          "units": 367,
          "spans": 881,
          "candidates": 156
        },
        {
          "id": "cross_asset_structuring",
          "units": 3,
          "spans": 12,
          "candidates": 4
        },
        {
          "id": "custody_asset_servicing",
          "units": 4,
          "spans": 29,
          "candidates": 3
        },
        {
          "id": "dcm_origination",
          "units": 66,
          "spans": 97,
          "candidates": 22
        },
        {
          "id": "ecm_origination",
          "units": 55,
          "spans": 97,
          "candidates": 28
        },
        {
          "id": "electronic_execution",
          "units": 4,
          "spans": 29,
          "candidates": 4
        },
        {
          "id": "em_trading",
          "units": 36,
          "spans": 153,
          "candidates": 22
        },
        {
          "id": "equities_trading",
          "units": 561,
          "spans": 716,
          "candidates": 161
        },
        {
          "id": "equity_derivatives_structured",
          "units": 91,
          "spans": 321,
          "candidates": 50
        },
        {
          "id": "ficc_combined",
          "units": 1,
          "spans": 3,
          "candidates": 0
        },
        {
          "id": "fx_trading",
          "units": 291,
          "spans": 550,
          "candidates": 94
        },
        {
          "id": "institutional_sales_coverage",
          "units": 4,
          "spans": 6,
          "candidates": 3
        },
        {
          "id": "listed_derivatives_client_clearing",
          "units": 121,
          "spans": 180,
          "candidates": 44
        },
        {
          "id": "ma_advisory",
          "units": 1,
          "spans": 3,
          "candidates": 1
        },
        {
          "id": "macro_combined_rates_fx_em",
          "units": 15,
          "spans": 73,
          "candidates": 18
        },
        {
          "id": "market_risk_management",
          "units": 6,
          "spans": 41,
          "candidates": 5
        },
        {
          "id": "market_risk_valuation_control",
          "units": 0,
          "spans": 10,
          "candidates": 2
        },
        {
          "id": "model_risk_management",
          "units": 9,
          "spans": 27,
          "candidates": 8
        },
        {
          "id": "municipal_public_finance",
          "units": 0,
          "spans": 1,
          "candidates": 0
        },
        {
          "id": "prime_brokerage_financing",
          "units": 564,
          "spans": 749,
          "candidates": 138
        },
        {
          "id": "product_control_ipv",
          "units": 16,
          "spans": 62,
          "candidates": 20
        },
        {
          "id": "rates_trading",
          "units": 388,
          "spans": 698,
          "candidates": 112
        },
        {
          "id": "regulatory_transaction_reporting",
          "units": 134,
          "spans": 151,
          "candidates": 40
        },
        {
          "id": "repo_secfin_collateral",
          "units": 11,
          "spans": 91,
          "candidates": 26
        },
        {
          "id": "securitised_products",
          "units": 68,
          "spans": 188,
          "candidates": 42
        },
        {
          "id": "sellside_research",
          "units": 57,
          "spans": 170,
          "candidates": 26
        },
        {
          "id": "trade_lifecycle_operations",
          "units": 140,
          "spans": 203,
          "candidates": 42
        },
        {
          "id": "trade_surveillance",
          "units": 5,
          "spans": 29,
          "candidates": 7
        },
        {
          "id": "treasury_balance_sheet_funding",
          "units": 0,
          "spans": 4,
          "candidates": 1
        },
        {
          "id": "xva_counterparty_risk",
          "units": 33,
          "spans": 177,
          "candidates": 24
        }
      ],
      "unit_stages": [
        {
          "label": "Raw source units",
          "retained": 19992
        },
        {
          "label": "Distinct source units",
          "retained": 18017
        },
        {
          "label": "Eligible text units",
          "retained": 11781
        },
        {
          "label": "Personal/home excluded",
          "retained": 10598
        },
        {
          "label": "Ontology-matched units",
          "retained": 3754
        }
      ],
      "span_stages": [
        {
          "label": "All recorded passages",
          "retained": 35169,
          "excluded": 0
        },
        {
          "key": "nonpersonal",
          "label": "Exclude personal/home records",
          "retained": 33037,
          "excluded": 2132
        },
        {
          "key": "unique",
          "label": "Keep unique passages",
          "retained": 18166,
          "excluded": 14871
        },
        {
          "key": "ontology",
          "label": "Match the configured ontology",
          "retained": 6772,
          "excluded": 11394
        },
        {
          "key": "time",
          "label": "Have an assigned or assignable date",
          "retained": 5508,
          "excluded": 1264
        },
        {
          "key": "complexity",
          "label": "Have medium or high complexity",
          "retained": 2058,
          "excluded": 3450
        },
        {
          "key": "unmeasured",
          "label": "Be outside the measured base",
          "retained": 1281,
          "excluded": 777
        },
        {
          "key": "not_bulk",
          "label": "Exclude bulk samples",
          "retained": 1281,
          "excluded": 0
        }
      ],
      "ablations": [
        {
          "key": "nonpersonal",
          "label": "Exclude personal/home records",
          "retained": 1340,
          "additional": 59
        },
        {
          "key": "unique",
          "label": "Keep unique passages",
          "retained": 2780,
          "additional": 1499
        },
        {
          "key": "ontology",
          "label": "Match the configured ontology",
          "retained": 3161,
          "additional": 1880
        },
        {
          "key": "time",
          "label": "Have an assigned or assignable date",
          "retained": 1313,
          "additional": 32
        },
        {
          "key": "complexity",
          "label": "Have medium or high complexity",
          "retained": 4389,
          "additional": 3108
        },
        {
          "key": "unmeasured",
          "label": "Be outside the measured base",
          "retained": 1732,
          "additional": 451
        },
        {
          "key": "not_bulk",
          "label": "Exclude bulk samples",
          "retained": 1281,
          "additional": 0
        }
      ],
      "density": [
        {
          "root": "archives",
          "in_domain_units": 10,
          "means": {
            "type_token": 0.267,
            "numeric": 0.051,
            "decision_per_1k": 2.443,
            "question_per_1k": 5.885,
            "quoted": 0.001,
            "boilerplate": 0.478,
            "domain_terms_per_1k": 198.933
          },
          "coverage": {
            "type_token": 10,
            "numeric": 10,
            "decision_per_1k": 10,
            "question_per_1k": 10,
            "quoted": 10,
            "boilerplate": 10,
            "domain_terms_per_1k": 10
          }
        },
        {
          "root": "working-folder",
          "in_domain_units": 42,
          "means": {
            "type_token": 0.427,
            "numeric": 0.082,
            "decision_per_1k": 4.895,
            "question_per_1k": 3.36,
            "quoted": 0.013,
            "boilerplate": 0.54,
            "domain_terms_per_1k": 83.413
          },
          "coverage": {
            "type_token": 42,
            "numeric": 42,
            "decision_per_1k": 42,
            "question_per_1k": 42,
            "quoted": 42,
            "boilerplate": 42,
            "domain_terms_per_1k": 42
          }
        },
        {
          "root": "forensic-mail",
          "in_domain_units": 3342,
          "means": {
            "type_token": 0.551,
            "numeric": 0.06,
            "decision_per_1k": 2.879,
            "question_per_1k": 3.365,
            "quoted": 0.056,
            "boilerplate": null,
            "domain_terms_per_1k": 68.066
          },
          "coverage": {
            "type_token": 3342,
            "numeric": 3342,
            "decision_per_1k": 3342,
            "question_per_1k": 3342,
            "quoted": 3342,
            "boilerplate": 0,
            "domain_terms_per_1k": 3342
          }
        },
        {
          "root": "forensic-docs",
          "in_domain_units": 360,
          "means": {
            "type_token": 0.376,
            "numeric": 0.108,
            "decision_per_1k": 3.199,
            "question_per_1k": 1.767,
            "quoted": 0.001,
            "boilerplate": 0.414,
            "domain_terms_per_1k": 704.109
          },
          "coverage": {
            "type_token": 357,
            "numeric": 357,
            "decision_per_1k": 357,
            "question_per_1k": 357,
            "quoted": 357,
            "boilerplate": 357,
            "domain_terms_per_1k": 360
          }
        }
      ],
      "mask_effect": {
        "personal_spans": 2132,
        "personal_relevant_spans": 92,
        "personal_candidate_spans": 59
      },
      "interpretation": "Recorded lexical ontology matches and heuristic candidate criteria, not human-ratified relevance. Personal/home is the producer's classification, not a guarantee that all remaining correspondence is professional. Ablation drops one gate with all other classifications fixed; effects overlap and are not additive or causal. Unit and span populations are separate: a passage can match even when its whole source unit does not.",
      "source_sha256": {
        "technical-report.json": "cd9af315ac7f3aea51fa0efe6968453ad1508a13c12d76a618186615d9f9fb66",
        "executive-report.md": "4bdeeaf4341b606dede163610e86e0eca5f0a26ef67e40e73a1cba32889322a9",
        "SUMMARY.md": "4b73860be04fb5df690350aa990e5bb88b80c3fc01d7bf77ffe8ce247c594df0",
        "units.jsonl": "dfe1332454be16c129828051fec4892ce13dc004ad6a219f5d1782df99f6669f",
        "spans.jsonl": "7c16afd01752bff58ba671d6e128ed87c281df274efe29d5eccee2c4dfd800ad"
      }
    }
  },
  "interpretation": "Recorded lexical ontology matches and heuristic candidate criteria, not human-ratified relevance. Personal/home is the producer's classification, not a guarantee that all remaining correspondence is professional. Ablation drops one gate with all other classifications fixed; effects overlap and are not additive or causal. Unit and span populations are separate: a passage can match even when its whole source unit does not.",
  "source_sha256": {
    "technical-report.json": "cd9af315ac7f3aea51fa0efe6968453ad1508a13c12d76a618186615d9f9fb66",
    "executive-report.md": "4bdeeaf4341b606dede163610e86e0eca5f0a26ef67e40e73a1cba32889322a9",
    "SUMMARY.md": "4b73860be04fb5df690350aa990e5bb88b80c3fc01d7bf77ffe8ce247c594df0",
    "units.jsonl": "dfe1332454be16c129828051fec4892ce13dc004ad6a219f5d1782df99f6669f",
    "spans.jsonl": "7c16afd01752bff58ba671d6e128ed87c281df274efe29d5eccee2c4dfd800ad",
    "bocg-taxonomy.json": "aaf08fc0d8a9aab2ffc7d226a37a09b59a497fefe000c5260becc335c43027ae"
  },
  "publication_policy_sha256": "b200b848b9f9e8c9cee4e4074f0df5ecc8a0d3360c7e3b81e55a344cd7619892",
  "artifacts": {
    "index.html": {
      "url": "/census/20261018T003131Z/index.html",
      "bytes": 43894,
      "sha256": "991fe9c1a14a52f9a94259b4b4e9035da6c3779728c7b980542dcf0dd5a314be"
    },
    "executive-report.md": {
      "url": "/census/20261018T003131Z/executive-report.md",
      "bytes": 10852,
      "sha256": "4bdeeaf4341b606dede163610e86e0eca5f0a26ef67e40e73a1cba32889322a9"
    },
    "summary.md": {
      "url": "/census/20261018T003131Z/summary.md",
      "bytes": 777,
      "sha256": "4b73860be04fb5df690350aa990e5bb88b80c3fc01d7bf77ffe8ce247c594df0"
    },
    "technical-report.public.json": {
      "url": "/census/20261018T003131Z/technical-report.public.json",
      "bytes": 254738,
      "sha256": "fedf5ff6a45e4dc0f075266c9cc0c4daa91e9237a99d362a653f4940ae70f4d6"
    },
    "provenance.json": {
      "url": "/census/20261018T003131Z/provenance.json",
      "bytes": 6190,
      "sha256": "e703dbfc299b626173f61f9fdd70eb2b8b2a854d9d05c626e86ab8a8659089c6"
    }
  }
}