OpenCyber AcademyLesson map
Lesson 07 published. Map, example, EX039.
flowchart TD
N0["Learning outcomes and connection"]
N1["7.1 Argument structure"]
N2["7.2 Directional limitations"]
N3["7.3 Archive"]
N4["7.5 Abstract surgery"]
N5["Chapter synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> N5
N5 --> EX["Worked example"]
EX --> Q["Exercise EX039"]
Logical drawing
flowchart LR
N0["Learning outcomes and connection"]
N1["7.1 Argument structure"]
N2["7.2 Directional limitations"]
N3["7.3 Archive"]
N4["7.5 Abstract surgery"]
N5["Chapter synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> Q["EX039"]
[Learning outcomes and connection]
|
v
[7.1 Argument structure]
|
v
[7.2 Directional limitations]
|
v
[7.3 Archive]
|
v
[7.5 Abstract surgery]
|
v
[Chapter synthesis]
|
v
[Worked example] --> [Exercise EX039]
Learning outcomes and connection
After this chapter you can (1) structure the dissertation argument, (2) write directional limitations, (3) build a minimal tiered archive, and (4) avoid overclaiming in the abstract.
7.1 Argument structure
Introduction (question, gap); background (seminar); methods (design, ethics, metrics); results (model, sensitivity, rig); discussion (rivals, boundaries); conclusion. Every claim in the abstract must appear with evidence and scope in the body.
A practical writing order: draft the results tables first, then the methods that produced them, then the discussion, then the introduction. The introduction written last is less likely to promise chapters you did not deliver. Keep a claim register — claim, section, evidence pointer, scope, falsifier — and update it weekly in the final semester. When a sentence in the abstract lacks a row in the register, delete the sentence or add the row.
7.2 Directional limitations
Noise-free flow meter in simulation → overstates recall (check on R2). Single-loop model → unknown direction for multi-loop plants. Synthetic operators → unknown. Invented parameters → not calibrated to a real plant. Censoring → understates delay when attacks are weak.
Directional language matters. "Limited sample size" says nothing about bias direction. "Invented orifice coefficient may overstate separation between detectors if real tanks have slower dynamics" tells a reader how to adjust confidence. If you truly do not know the direction, write unknown — that is more honest than a vague apology. Examiners treat directional limitations as evidence of research maturity.
7.3 Archive
Public: code and invented simulator (T0). Synthetic traces (T1). R2 configuration (T2). Never publish real plant maps. Hash-check before release; fail closed on mismatch.
Write a manifest listing every file, its tier and its hash, and run the check on the release candidate rather than on the working directory. Keep T3 material, such as campus contact details, out of the archive entirely. A reader should be able to rerun Labs 3 and 6 from the public tier alone and get the published numbers.
7.4 Common errors and limits
Limitations without direction; abstract claims broader than results; archives that dump everything.
Each error weakens the link between evidence and claim. A limitation without a direction does not tell the reader how to adjust confidence. An abstract that says industrial control systems when the study measured one model and one rig invites rejection. An archive containing everything makes review slow and risks releasing configuration that should stay restricted. Read the abstract against the results tables once more before submission and remove any wording the tables cannot support.
7.5 Abstract surgery
Write the abstract last. Paste every sentence into the claim register. If a sentence has no evidence row, cut it. Include one sentence on class B failure and one on synthetic scope. Prefer "in two synthetic contexts" to "in industrial control systems" when that is what you measured. Readers who need a broader claim will not be fooled by broader wording; examiners who enforce scope will reward precision.
Test the result with a reader who has not seen the dissertation: ask them what the study showed and when the method fails. If they cannot mention class B or the synthetic scope, the abstract still overstates or hides something. Repeat the surgery after every major revision, because results text changes faster than abstracts do.
Exercises — chapter 7
- EX039 — Abstract that includes class B failure. Write an abstract (≤150 words) that includes the class B failure.
Show marking criteria
Rubric: 3 = ≤150 words including class B failure and contexts; 2 = omits class B; 1 = overclaim; 0 = none. - EX040 — Directional limitations table. Build a directional limitations table with ≥5 rows.
Show marking criteria
Rubric: 3 = ≥5 rows with direction labels; 2 = 3–4 rows; 1 = undirected list; 0 = none. - EX041 — Tier eight Harbor Loop artifacts. Assign tiers to eight artifacts and justify two borderline cases.
Show marking criteria
Rubric: 3 = eight tiers + two borderline justifications; 2 = tiers only; 1 = four; 0 = none. - EX042 — Dataset card for Harbor Loop simulator. Write a dataset card for the Harbor Loop simulator (invented vs calibrated).
Show marking criteria
Rubric: 3 = purpose/parameters/invented/prohibited uses; 2 = missing field; 1 = purpose only; 0 = none. - EX043 — Chapter word budgets to a limit. Outline chapter word budgets that sum to your institution's limit.
Show marking criteria
Rubric: 3 = budgets sum to stated institutional limit; 2 = budgets without sum check; 1 = vague; 0 = none. - EX044 — Red-team one overclaim in the abstract. Red-team your abstract: find one overclaim and rewrite it.
Show marking criteria
Rubric: 3 = finds overclaim and narrowed rewrite; 2 = finds without rewrite; 1 = nitpick; 0 = none.
Chapter synthesis
Write the argument so a skeptical reader can find evidence, scope and directional limits without hunting; keep the archive small, tiered and hash-checked.
Continue studying — chapter 07 of 08
← Chapter 06 (draft) · Course chapter index · Chapter 08 (draft) →