The Scale at Which Everything Changes: Why more of the same eventually becomes something else

I met the threshold version of this in the pump business, without a name for it. For years I ran service jobs the way you run a small crew: verbal assignments, one person holding every site in his head, prices set by feel from the last job. It worked. Then the number of concurrent sites grew past a line I never marked, and the whole thing stopped working at once. Not slowly. The method that had been fine on a handful of jobs produced missed schedules and a margin I could not reconstruct once the sites multiplied. I had assumed the bigger version of my workshop was the small one with more of everything. It turned out to be a different animal.

Growth gets treated as more of the same until a threshold quietly turns it into something else. Galileo wrote the reason down in 1638: as a shape gets larger its volume outruns its surface, so a structure cannot be enlarged, its proportions have to be re-derived. Biology says the same through Kleiber's law, where a cat a hundred times a mouse's mass burns only about thirty-two times the energy. Fred Brooks caught the inversion in 1975 in one line, that adding people to a late software project makes it later. Robin Dunbar put a rough ceiling of 150 on the relationships a person can hold in his head before memory has to give way to a written system.

This is a short book about phase transitions and scale thresholds in plain terms, nonlinearity and power laws, the square-cube law and Kleiber's law, network effects and diseconomies of scale, critical mass and the tipping point, and why the model that fit one size so cleanly can describe a different thing at the next.

This book is written from an Indonesian operator's context, but the threshold it describes shows up just as hard in other emerging-market and small-business settings.