{
  "title": "Other metal ore sample method records for AI",
  "source_page": "https://fiund.com/industries/other-metal-ore-method-records-data",
  "authorship": "Original fictional Fiund editorial exercise. No actual operating records or third-party source assets are included.",
  "example": {
    "title": "The same field name can conceal a method change",
    "context": "A mock ore review presents a repeated field called response R. Method labels are abstract identifiers rather than descriptions of proprietary analytical procedures.",
    "records": [
      [
        "OM1",
        "Practice report H1",
        "Specimen Z1: response R=10 units, method M-A."
      ],
      [
        "OM2",
        "Practice report H2",
        "Specimen Z2: response R=12 units; method not recorded."
      ],
      [
        "OM3",
        "Method-change notice",
        "Later Z1 reports use M-B. No statement establishes equivalence between M-A and M-B, and no notice names Z2."
      ]
    ],
    "question": "Can H1 and H2 be described as a same-method comparison, and does the Z1 change notice identify Z2's method?",
    "answer": "H1 names M-A for Z1, while H2 leaves Z2's method unknown. A shared field name and matching unit do not establish the same analytical method. OM3 concerns later Z1 reports only; it neither supplies Z2's missing method nor establishes equivalence with M-A. Preserve the two observations, identify the missing method and request the relevant report-level evidence before comparison.",
    "failureModes": [
      "Filling Z2's missing method from a notice about Z1.",
      "Treating a common response label as proof that measurements across method versions are interchangeable."
    ],
    "fields": [
      [
        "Report-level method identifier",
        "Makes method attribution specific to the observation."
      ],
      [
        "Method effective scope",
        "Limits a change notice to the reports it actually covers."
      ],
      [
        "Field definition and unit",
        "Retains a shared label without assuming shared semantics."
      ],
      [
        "Equivalence evidence status",
        "Shows whether a comparison has supporting validation."
      ]
    ],
    "holdout": "Supply Z2's method as M-B and a validation note limited to a different material class. Require the model to identify that the note's scope still does not establish equivalence for these specimens.",
    "boundary": "Use arbitrary response values and fictional method codes. Include no actual assays, analytical recipes, reserve data, equipment or operating details."
  }
}