Organizing my materials in my office has always been a problem. This was true in the paper age, the PC age and even in the Internet age. Samuel Arbesman in The Half-life of Facts says "Forget about it." Most of them are not worth remembering anyway. Many are quickly outdated and (the fun part of his story) hang on because we do have our cognitive biases.
The author is a fine writer and takes us through an amazing amount of material in very few pages. There are many knowledge growth paths that follow a logistic curve trajectory, but phase transitions kick in whereby we jump to the next logistic path. This is relevant when Arbesman quickly considers the Santa Fe Institute work on cities. There are important empirical links of many urban phenomena to city size (he mentions gasoline stations) but there are scale economies whereby the number of stations per capita is not fixed. But there are also phase changes whereby we jump to the next logistic path.
Arbesman does not elaborate, but new facts of life signaled by new prices prompt the phase transitions. Think about how cities have changed as autos and highways came into use. This phase transition rendered established relationships moot. Costs of Sprawl are a case in point. The Santa Fe data are fitted across phase changes that are always lurking because price signals are always at work. Those fickle facts.