Chapter 18 — Security architecture
Lesson map
Lesson 18 published. Map, example, EX052.
flowchart TD
N0["Learning outcomes"]
N1["18.1 Human factors in architecture"]
N2["18.2 Layered defences"]
N3["18.3 Reducing reliance on vigilance"]
N4["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4
N4 --> EX["Worked example"]
EX --> Q["Exercise EX052"]
Logical drawing
flowchart LR
N0["Learning outcomes"]
N1["18.1 Human factors in architecture"]
N2["18.2 Layered defences"]
N3["18.3 Reducing reliance on vigilance"]
N4["Synthesis"]
N0 --> N1 --> N2 --> N3 --> N4 --> Q["EX052"]
[Learning outcomes]
|
v
[18.1 Human factors in architecture]
|
v
[18.2 Layered defences]
|
v
[18.3 Reducing reliance on vigilance]
|
v
[Synthesis]
|
v
[Worked example] --> [Exercise EX052]
Learning outcomes
You will (1) place human factors in security architecture; (2) design layered defences that tolerate human error; (3) use zero trust ideas to reduce reliance on vigilance.
18.1 Human factors in architecture
Aiko studies how architecture choices shape what people must do. If one stolen password gives access to everything, every staff member must be perfect all the time. If access is segmented and checked continuously, a single mistake has limited impact. NIST SP 800-207 describes zero trust as removing implicit trust based on network location and evaluating each access request. Aiko sees it as a human factors design: it reduces the cost of individual errors, which are inevitable in a large workforce. She explains this view to the IT team.
18.2 Layered defences
Aiko designs layers for Greyford's remote access: email filtering removes most phishing; phishing-resistant authentication with security keys stops stolen passwords from working; device checks block unmanaged laptops; monitoring detects unusual access; and staff reporting provides early warning. Each layer catches some failures of the others. She emphasises that staff are one layer among many, not the last line of defence, a framing that reduces pressure and blame on individuals. She draws the layers as a simple diagram for the board, showing which threat each layer addresses and how a failure in one is caught by another, so that nobody assumes a single control is enough.
18.3 Reducing reliance on vigilance
Aiko lists places where Greyford relies on staff noticing something: checking sender names, spotting fake login pages and verifying payment changes. For each, she asks whether architecture can remove the need. Security keys make fake login pages ineffective because the key will not authenticate to the wrong site. Payment changes require callback verification by a second person. Saltzer and Schroeder (1975) recommended designs that do not depend on users being constantly careful, and Aiko finds that principle still fits modern systems. She lists the remaining reliances for review.
Worked example
Aiko applies zero trust as a human factors design, layers defences so staff are one layer among many and removes several reliances on vigilance.
Synthetic case. Greyford's design is synthetic; SP 800-207 and Saltzer and Schroeder (1975) are real.
Errors and limits
Layers add cost and complexity. Removing vigilance needs may reduce staff engagement if poorly explained.
Exercises
- EX052 — Single password risk. What happens if one stolen password gives access to everything? Rubric: 3 = every staff member must be perfect all the time;; 2 = perfection;; 1 = vague;; 0 = none.
- EX053 — Staff as a layer. How does Aiko position staff in the layered design? Rubric: 3 = as one layer among many, not the last line of defence;; 2 = one layer;; 1 = last line;; 0 = none.
- EX054 — Security key benefit. How do security keys reduce reliance on vigilance? Rubric: 3 = they will not authenticate to fake login pages;; 2 = phishing-resistant;; 1 = vague;; 0 = none.
Synthesis
Architecture that tolerates human error reduces pressure on individuals and makes security more reliable.
Continue studying — chapter 18 of 50
← Chapter 17 (draft) · Course chapter index · Chapter 19 (draft) →