Secondary_English_OS := a closed-loop reliability system that upgrades:
  Vocabulary (P3 under load),
  Comprehension (literal + inferential stability),
  Summary (compression without meaning loss),
  Writing (situational + continuous) under time pressure,
into exam-stable output with low variance.

P3(Secondary English) = student can:
  - score consistently across unseen passages
  - infer without guessing (evidence-backed)
  - write with clarity + tone control + variation
  - summarise accurately within word limits
  - perform under exam timer with minimal collapse

Secondary_English_OS {
  Inputs:
    - reading exposure (nonfiction + narrative)
    - explicit vocab/chunk/structure training
    - timed exam-format practice
  Engines:
    - Vocabulary_OS (Z0–Z3 upgraded for tone + precision)
    - Comprehension_OS (inference stability + evidence discipline)
    - Summary_OS (compression + paraphrase control)
    - Writing_OS (situational + continuous, audience + purpose)
    - Structure_OS (sentence automation + paragraph coherence)
  Sensors:
    - inference error tags
    - tone/register errors
    - summary leakage (meaning loss) + repetition
    - timed performance variance
  Repairs:
    - micro drills + rewrite loops
    - timed sets + post-mortem correction
  Outputs:
    - O-Level grade reliability with reduced variance
}

P0: cannot extract meaning reliably; writing breaks down
P1: can answer when guided; vocab is recognition-only
P2: can do well sometimes; inconsistent under exam conditions
P3: stable output under time pressure + variation across topics

Secondary English is a variance-control problem.
High performers are not “more talented” — they are more stable under variation.

Load = {
  TextDensity↑ (more ideas per paragraph),
  InferenceDemand↑,
  PrecisionMarking↑,
  TimePressure↑,
  Tone/Audience constraints↑
}

Goal: P3 vocabulary that survives exam load + supports precision.
Key: chunks + tone + paraphrase families

- RecallTime <2s for targets
- correct collocation + tone
- can paraphrase without meaning drift
- upgrades appear naturally in timed writing

Goal: literal accuracy + inference stability + evidence discipline.

- answers are consistent after re-read
- inference is supported by text
- avoids “over-infer” mistakes

Goal: reduce text to essentials within word limit without distortion.
Summary is not copying — it is controlled paraphrase + selection.

- hits word limit
- keeps all required points
- paraphrases with minimal repetition
- no meaning leakage

Goal: purpose + audience + tone + structure under time.

- addresses task fully
- maintains tone/register
- coherent paragraphing
- stable language variation

Goal: grammar + structure become automatic so brain can focus on ideas.

- low grammar error rate
- connector diversity
- clarity: reader knows actor/action/intent

FM1: Passive vocab (P1) — “knows but cannot use”
FM2: Inference guessing — answers not anchored in text
FM3: Summary leakage — meaning lost or points missing
FM4: Tone/register mismatch — informal when formal, wrong audience
FM5: Weak paragraph coherence — idea drift, repetition
FM6: Exam variance collapse — does well at home, drops in exam

V_Sensors {
  RecallTime < 2s (targets)
  CollocationErrorRate low
  RegisterErrorRate low
  ParaphraseSuccessRate high
  VariationIndex trending up
}

C_Sensors {
  WrongAnswerTags:
    - misread detail
    - vocab gap
    - inference leap (over-infer)
    - missed implication
    - evidence not cited
  StabilityCheck:
    - same answer after re-read? (Y/N)
}

SUM_Sensors {
  PointsCaptured / PointsRequired
  WordLimitCompliance
  RepetitionRate
  MeaningLeakEvents (distortion)
}

W_Sensors {
  TaskCompletionScore
  ToneMatchScore
  CoherenceScore (topic sentence → support → close)
  TimedFinishRate
  UpgradeCountCorrect (>= 12 per full piece)
}

VAR_Sensors {
  ScoreStdDev across 5 timed sets
  CollapseEvents under timer
  TopicSensitivity (drops on unfamiliar topics)
}

Loop_Vocab_Secondary {
  choose 8 words/week + 4 chunks/week
  for each:
    - 2-sentence anchor
    - collocation bank (3 pairs)
    - paraphrase family (2 alternatives)
  timed use:
    - 8-min paragraph with 6 upgrades
  repair:
    - rewrite weakest paragraph
}

Loop_Evidence {
  unseen passage set
  for each question:
    - answer
    - highlight evidence line(s)
    - rewrite answer in own words
  tag wrong answers by type
  fix the dominant tag each week
}

Loop_Summary {
  take 1 passage segment
  steps:
    1) list required points (no paraphrase yet)
    2) compress points into 1–2 sentences
    3) paraphrase with synonym families
    4) check meaning drift
    5) final word count trim
}

Loop_Tone {
  pick 1 situational task
  write 2 versions:
    - formal neutral
    - warm persuasive
  compare:
    - tone markers
    - register errors
  repair:
    - rewrite introduction + conclusion only
}

Loop_Timed_Set {
  do timed paper section
  post-mortem:
    - tag failure modes
    - identify weakest component (vocab / inference / summary / tone)
    - apply one targeted repair loop immediately
}

- build chunk bank + paraphrase families
- start evidence discipline tags
- timed paragraph drills 2×/week

- summary compression drills 3×/week
- situational writing tone practice weekly
- rewrite weakest paragraph every session

- weekly full timed set + post-mortem
- eliminate dominant wrong-answer tag
- increase upgrade density while maintaining accuracy

- timed practice across varied topics
- focus on consistency: reduce score variance
- final polishing of introductions/conclusions + summary trimming

More reading → recognition rises (P1)
→ no paraphrase training
→ summary leaks meaning
→ inference becomes guessing
→ tone errors appear
→ exam variance increases
→ grade stagnates despite effort

Secondary English mastery requires:
  evidence discipline + compression training + tone control + load drills,
not just “read more” or “memorise more words.”

If the student “knows words” but writes basic English,
they are stuck in P1 (recognition) or P2 (slow recall).
Use Vocabulary OS loops to force P3 retrieval under timed writing.

Weak writing is usually a system leak:
planning → structure → vocabulary retrieval → load collapse.
Use a closed-loop repair plan, not random practice.

O-Levels OS is variance-control: stable performance under topic variation and exam time pressure.
Secondary English OS is one of the core pipelines feeding that reliability.