Why this is in my collection
From the publisher:
THE MUCH ANTICIPATED FIRST INSTALLMENT IN THE WALL STREET JOURNAL BESTSELLING SERIES 'THE GREAT MENTAL MODELS'. Solve problems. Think with clarity. Achieve your goals. The secret to better decision-making is learning things that won't change. Mastering a small number of versatile concepts with broad applicability enables you to rapidly grasp new areas, identify patterns, and understand how the world works. Don't waste your time on knowledge with an expiry date - focus on the fundamentals. The Farnam Street latticework of mental models gives you the durable cognitive tools you need to avoid pro
Highlights
- Learn things that won't change — the book's premise is to invest in knowledge without an expiry date; applied to data engineering this is the MDDE bet stated as an epistemic strategy: models and rules are the durable layer, platforms are the perishable one.
- Occam's razor as a design principle — prefer the explanation, and the design, with the fewest moving parts; a modeling grammar of only a handful of construct types is Occam applied to architecture, because fewer structures means fewer failure modes.
- Hanlon's razor for data quality — never attribute to malice what process gaps explain; broken data almost always signals a missing rule or an ambiguous definition, so the fix is systemic, not personal.
- Probabilistic thinking — treat claims as credences to update, not facts to accept; the working habit of verifying against source data instead of trusting stale status documents is probabilistic thinking made routine.
- Thought experiments as cheap prototypes — reasoning a design through hypothetical futures (the reorg, the source replacement, the tenfold volume) costs nothing and catches structural errors before they are load-bearing.
- Models must be checked against the territory — Parrish insists a mental model is only useful while reality confirms it; the same audit discipline applies to a data model or a knowledge vault, both of which drift silently unless deliberately re-verified.
Highlights on this page are generated with the help of AI.