Chapter 7 — Dissertation writing and limitations
Lesson map
Lesson 07 published. Map, example, EX039.
flowchart TD
N0["Learning outcomes"]
N1["7.1 Structure around the argument"]
N2["7.2 Specific limitations"]
N3["7.3 Connecting to practice"]
N4["7.4 The practitioner brief"]
N5["7.5 Writing process"]
N6["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6
N6 --> EX["Worked example"]
EX --> Q["Exercise EX039"]
Logical drawing
flowchart LR
N0["Learning outcomes"]
N1["7.1 Structure around the argument"]
N2["7.2 Specific limitations"]
N3["7.3 Connecting to practice"]
N4["7.4 The practitioner brief"]
N5["7.5 Writing process"]
N6["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> N5 --> N6 --> Q["EX039"]
[Learning outcomes]
|
v
[7.1 Structure around the argument]
|
v
[7.2 Specific limitations]
|
v
[7.3 Connecting to practice]
|
v
[7.4 The practitioner brief]
|
v
[7.5 Writing process]
|
v
[Synthesis]
|
v
[Worked example] --> [Exercise EX039]
Learning outcomes
You will (1) structure a dissertation around its argument; (2) write limitations that are specific and useful; (3) connect findings to architecture practice responsibly; (4) prepare a practitioner brief.
7.1 Structure around the argument
Kenji structures his dissertation so each chapter supports one step of the argument: the question and its gap, the conceptual chapter from the seminar, the method and pilot, the results and rival checks, and the discussion. He writes the discussion first in outline, then checks that every claim in it points back to a result. Claims without support are cut or moved to future work. Dr Aydın reminds him that examiners read the abstract, introduction and conclusion closely and look for any mismatch with the results chapter, so those three sections must be written last and checked line by line.
7.2 Specific limitations
Kenji's limitations are concrete. The model assumes every hop is exploitable, so reach is an upper bound. Networks were generated from parameters abstracted from one employer and may not represent organisations with very different structures, such as hospitals with many legacy devices. Policies were assumed to be enforced exactly except for the misconfiguration parameter. Time, detection and response were not modelled. Rule count is a crude measure of complexity. For each limitation he states its likely direction of effect on his findings, which helps readers judge whether it weakens or strengthens the conclusions.
7.3 Connecting to practice
In the discussion, Kenji relates findings to standards without overreaching. NIST SP 800-207 recommends per-request access decisions; his results suggest that the benefit depends heavily on avoiding broad template rules. SP 800-53 Rev. 5 includes controls for boundary protection and least privilege, and his complexity result suggests organisations should plan for rule management, not only rule creation. He frames these as considerations for architects, not as requirements. He also states that his study cannot say whether real attackers would exploit the paths his model found, which only field evidence could establish.
7.4 The practitioner brief
Kenji writes a two-page brief for Vantor and similar organisations. It explains in plain language that, in simulations shaped like their network, identity-based segmentation reduced how far an attacker could move from one compromised laptop, but that copying broad rule templates removed most of the benefit, and that identity rules were about three times as many to manage. It recommends piloting on one department with rule reviews. It links to the full dissertation and lists the limitations. He checks the brief with Dr Aydın to ensure it says nothing the dissertation does not support.
7.5 Writing process
Kenji writes in short daily sessions, keeping a list of open questions at the top of each chapter file. He sends chapters to Dr Aydın one at a time and records her comments and his responses in a table, as he will later be asked to do for examiners. A classmate reads the results chapter without context and marks every sentence she cannot verify from a table or figure; Kenji rewrites those sentences. This cold-reader check catches assumptions that the author no longer notices because he has lived with the work for months.
Worked example
Kenji outlines the discussion first, cuts unsupported claims, writes five specific limitations with their direction of effect, relates findings to SP 800-207 and SP 800-53 Rev. 5 and drafts a checked two-page brief.
Synthetic case. The dissertation drafts and reader comments are synthetic; SP 800-207 and SP 800-53 Rev. 5 are real.
Errors and limits
Direction-of-effect statements are themselves judgements. Practitioner briefs risk being read without their limitations.
Exercises
- EX039 — Last sections. Which sections are written last and why? Rubric: 3 = abstract, introduction, conclusion, so they match results;; 2 = named;; 1 = first;; 0 = none.
- EX040 — Limitation direction. Why state the likely direction of each limitation's effect? Rubric: 3 = helps readers judge whether it weakens or strengthens conclusions;; 2 = direction;; 1 = not needed;; 0 = none.
- EX041 — SP 800-53 link. What does Kenji's complexity result suggest regarding SP 800-53 Rev. 5 boundary controls? Rubric: 3 = plan for rule management, not only rule creation;; 2 = management;; 1 = vague;; 0 = none.
- EX042 — Field evidence gap. What can only field evidence establish? Rubric: 3 = whether real attackers would exploit the modelled paths;; 2 = real attackers;; 1 = simulation shows;; 0 = none.
- EX043 — Brief recommendation. What does the brief recommend Vantor do? Rubric: 3 = pilot on one department with rule reviews;; 2 = pilot;; 1 = full rollout;; 0 = none.
- EX044 — Cold reader. What did the cold-reader check catch? Rubric: 3 = sentences not verifiable from a table or figure;; 2 = unsupported;; 1 = typos;; 0 = none.
Synthesis
A dissertation persuades when every claim traces to a result, limitations are specific and practitioner advice stays within the evidence.
Continue studying — chapter 07 of 08
← Chapter 06 (draft) · Course chapter index · Chapter 08 (draft) →