What is covered, week by week, and which topics each lecture draws on.
Every lecture the course claims to run gets a dated page and a teaching-team reference. Each lecture develops one part of the investigative method and connects it to the week’s practical work and assessment preparation. Week 1 also includes the course’s lecture deck.
Observation Before Explanation
What this course is, why it is shaped as an investigation rather than a syllabus, and what a student needs to have ready before Week 1's session.
Week 1 · 22 February 2027
Proving a Hypothesis Wrong on Purpose
Why a hypothesis needs a disconfirming test to count as one, and how Debugging Field Notes turns that habit into the semester's first graded work.
Week 2 · 1 March 2027
A Model Is a Claim, Not a Map
How to build a working model of a system's intended behaviour when there is no one left to explain it, and why structural reading has to do the work documentation used to.
Week 3 · 8 March 2027
Instrumentation Is a Hypothesis Too
What runtime evidence can tell you that reading the code cannot, and why instrumentation only counts as investigation when a structural model decided where to place it.
Week 4 · 15 March 2027
Timing Is a Lead, Not a Verdict
Why a system's history counts as evidence, and why the change that sits closest in time to a failure is not automatically the change that caused it.
Week 5 · 22 March 2027
Discriminating Between Explanations
How to choose between two explanations that both fit the evidence you already have, by asking what each one predicts under a change you control rather than what each is merely consistent with.
Week 6 · 29 March 2027
A Fact Needs a Source Before a Verdict
What to do when two pieces of evidence contradict each other — and why that is a different problem from two hypotheses that both fit the same, trustworthy evidence.
Week 7 · 12 April 2027
The Blast Radius of a Fix
Why removing a symptom is not the same as knowing a fix is safe, and how to reason about what an intervention might disturb before applying it.
Week 8 · 19 April 2027
What Does This Test Actually Prove?
Why a disappeared symptom and a passing smoke test are evidence for a narrower claim than "the diagnosed cause is fixed," and how to verify the claim that actually matters.
Week 9 · 26 April 2027
Choosing Where to Begin
How to plan an investigation into a large, unfamiliar system when nobody tells you which component, file, or evidence source to start with.
Week 10 · 3 May 2027
How Much Should You Trust the System?
What changes when the system's own evidence cannot be trusted at face value — and why the answer is calibrated confidence, not blanket distrust.
Week 11 · 10 May 2027
The Weakest Link in the Chain
What it means to defend an investigation rather than merely present one, and how to find the weakest link in a reasoning chain — including your own.
Week 12 · 17 May 2027