Name all 16 laws and their groups from memory. For each law in Group 1 (Mathematical Constraints), state the formula or precise quantitative claim that makes it a constraint rather than a guideline.
A team says: “Our service is slow — we should add more application servers.” Apply Amdahl’s Law (L1) to diagnose whether this will work. What specific measurement do you need before recommending horizontal scaling, and what does the answer tell you about where to invest instead?
Conway’s Law (L10) and Brooks’s Law (L6) interact. Explain the interaction: if you restructure a team to produce a better architecture (the Inverse Conway Maneuver), why does Brooks’s Law predict the transition will get worse before it gets better? What does this imply about when to make team restructuring decisions?
Goodhart’s Law (L9) and Hyrum’s Law (L14) both describe how observation and measurement change system behaviour. What is structurally similar about them? Give one concrete production engineering example where each operates and explain what breaks in each case.
You are advising a team building their first public API. Three laws from Group 3 (Design Wisdom) are most critical to get right before the first external consumer. Which three, and why does the window for applying each law close once enough callers exist?
Concept: F9 — The 16 Engineering Laws
Thread: T12 (Tradeoffs) ← Mathematical proof (Book 1, Ch 6) → Strategic constraints (Book 6, Ch 6)
Core Idea: Sixteen laws — five mathematical constraints (Amdahl, Moore, CAP, Little, Metcalfe), four human behaviour patterns (Brooks, Hofstadter, Parkinson, Goodhart), five design principles (Conway, Gall, Postel, Leaky Abstractions, Hyrum), and two statistical heuristics (Pareto, Linus) — describe the real constraints on what engineering can achieve. The type of law determines how to use it: constraints cannot be circumvented; tendencies can be mitigated.
Tradeoff: Correctness vs Performance (F4 #9) — mathematical constraints (Amdahl, CAP) are correctness constraints; heuristics (Pareto) are performance guides; conflating them produces either unnecessary caution or physics violations
Failure Mode: FM9 (Silent Data Corruption) — Goodhart’s Law is the most common cause: the metric is correct, the system is optimising it, the underlying objective is degrading silently
Signal: Technology planning (Moore, Amdahl); team sizing decisions (Brooks, Conway); API design (Hyrum, Postel); estimate validation (Hofstadter, Parkinson); capacity planning (Little, Amdahl)
Maps to: Reference Book Ch 8 (F4 Tradeoffs — each law creates or modifies a tradeoff); Book 6 Ch 6 (applying the laws in leadership context); Book 7 Ch 3, 7 (Goodhart and Metcalfe in product design)