{
  "administrations": {
    "N": "Frozen reference conventions. Not reading ability truth.",
    "M": "Automated operator on exactly the N selected spans.",
    "F": "Automated operator on the entire exact prepared transcription. SMOG is adapted density-normalized."
  },
  "formulas": {
    "Flesch": "206.835 - 1.015*(words/sentence) - 84.6*(syllables/word), separate selected pools retained.",
    "SMOG": "3.1291 + 1.0430*sqrt(30*polysyllabic_occurrences/proxy_sentences). N has exactly30reference printed sentences. M/F are adaptations."
  },
  "units": {
    "Flesch exact_index_lo_hi": "Exact Flesch output envelope, rational.",
    "SMOG exact_index_lo_hi": "Exact normalized polysyllable density30P/S, NOT final score.",
    "SMOG score_lo_hi": "Final formula output,80significant decimal digits.",
    "Flesch score_lo_hi": "Display decimal from exact rational; rational governs comparisons.",
    "changes exact_density_or_flesch_change_lo_hi": "Flesch output difference or SMOG normalized density difference; not commensurate between formulas.",
    "changes score_change_lo_hi_decimal80": "Within-formula change in its own score units.",
    "words": "Graphic word token occurrences under declared conventions, not unique vocabulary.",
    "sentences": "N reference thought units for Flesch or printed SMOG sentences; M/F finite punctuation-proxy counts.",
    "syllables_lo_hi": "Sums of token occurrence finite alternatives.",
    "polysyllables_lo_hi": "Counts of occurrences with all/any accepted alternatives>=3syllables.",
    "fallback_occurrences": "Automated graphic-token occurrences requiring unverified orthographic surrogate; N blank means not measured here.",
    "zero_nucleus_occurrences": "Automated occurrences with unresolved zero-nucleus CMU source, including components.",
    "sample coordinates": "Half-open Unicode code-point offsets in prepared source, not byte offsets.",
    "unresolved": "Touching/overlapping nondegenerate conditional envelopes. Not equality/equivalence."
  },
  "not_covered": "Unknown orthographic surrogate error, dialect/history, source-edition uncertainty, comprehension and reader characteristics are outside conditional envelopes. They are not confidence intervals.",
  "fixed_order": [
    161,
    1342,
    141,
    158,
    121,
    105
  ]
}
