I built a smaller version of that in my own yard. To save the cost of a second transformer and an automatic transfer switch, I ran a pumping installation off a single utility feeder. On the drawing it looked clean. In the field it meant every pump on that site hung on one line, so a single feeder fault would have stopped all of them at once, not slowed them, stopped them. I only saw it when I traced what each output actually passed through, because the single-line diagram showed a backup that did not exist. When I finally costed a second feeder against one day of total stoppage, the redundancy came out cheaper than the outage I had been quietly risking for years.
This book argues that efficiency carries a hidden bill: it concentrates dependence. Every optimization strips out slack and routes more of the system through fewer points, until the most efficient version hides a single node whose failure is not a setback but an ending. A single point of failure, by definition, has no backup, so the system runs perfectly right up to the moment it does not run at all. Charles Perrow saw the deeper version in his 1984 book Normal Accidents, where tight coupling between parts means a local fault spreads before anyone can catch it. Lean and just-in-time made this worse by design, because the inventory that used to absorb a shock was the first thing optimization deleted. The answer is old engineering, not cleverness. Duplicate the critical node, or build the fault tolerance to contain it, and pay for the redundancy before the cascade sends the invoice.
This is a short book about the single point of failure and cascading failure in plain terms, efficiency versus resilience as a tradeoff, tight coupling and the slack that optimization removes, the bus factor and the hidden critical node, and buying back redundancy and fault tolerance where a failure would be total rather than partial.
This book is written from an Indonesian operator's context, but the concentrated dependency it describes hides in other emerging-market and small-business settings too.
Ibrahim A. An operator who runs three small businesses in Indonesia: industrial pump distribution, engineering and welding workshop, and handmade leather craft. Based in Cileungsi, Bogor. Writes from the field-operator perspective, not a consultant's.
A practising zettelkasten user since 2019. Publishes for readers who want to read from an operator with skin in the game.
ORCID: 0009-0006-0425-4923