Wind, Water, Work Ancient And Medieval Milling Technology Book

A remarkable monograph from the father of bibliometrics that is just amazingly ahead of its time (scale free networks thinking! in 1960! my god!). mandatory reading for anyone studying citations and the academic literature, and perhaps even just for scientists in general. unfortunately, finding a copy is nearly impossible (although the chapters were all previously published as academic papers).

chapter by chapter summary:

'prologue to a science of science' argues that science has been growing at an exponential rate (according to pretty much an sane measure of the size of science), but that this growth phase is about to transition for exponential to logistic, and ponders the consequences of the resulting saturation of science within society

'galton revisited', after some brief hagiography of galton, builds upon galton's eugenic measurement of human attributes to assess the quality of scientists, how often scientists publish, and the growth in the number of scientists. the basic descriptive statistics are eye opening but the chapter otherwise follows a yucky academic tradition

'invisible colleges and the affluent scientific commuter' investigates one consequence of the growing size of science: namely, that communication via papers in journals is replaced by face to face interaction at research institutions organized around unofficial groups price calls invisible colleges. he also argues that primary purpose of the academic paper is to make priority claims

'political strategy for big science' points out that, as science approaches its saturation level (in terms of labor force and funding) in a society, incentivizing marginal participation in requires additional financial and status motivators. due to differing international progress toward the saturation level, there is a resulting saturation gradient that drives brain drain.

'networks of scientific papers' and 'collaboration in an invisible college' are perhaps the first citation network analyses, finding connected components in a coauthorship network (by hand! using index cards!!) and noting that most citations are to recent papers. old papers get cited approximately uniformly at random. every once in while a bunch of knowledge get compacted into review articles and textbooks.

'measuring the size of science' considers the exponential growth of science more carefully, and concludes that majority of scientists to ever exist are currently alive, and additionally that the majority of papers ever published were published recently. further, this has always been the case since science has been growing exponential for nearly ~300 years according to price. note that these essays are from the 60s, and price predicts scientific saturation sometime around ~2000 in the US. another interesting observation is that national scientific productivity correlates well with wealth (perhaps per capita, i forget), and also national energy use (perhaps also per capita)

'citation measures of hard science, soft science, technology and non-science' points out technologists amusing aversion to citation and argues (with data) that STEM fields are more cumulative and tend to cite much more recently papers than the humanities. price proposes the 'price index', the percentage of citations to papers within the last 5 years.

'studies in scientometrics, part i: transience and continuance in scientific authorship' describes author trajectories through the scientific literature and should be mandatory reading for new grad students to understand the reality of how academic works. interestingly, price argues that scientific institutions (such as phds and tenure) where subsequent to a publishing cycle that most people are transient participants in, rather than a cause of such transience

'of sealing wax and string: a philosophy of experimenter's craft and its role in the genesis of high technology' argues that instrumentation proceeds both basic science and technology, rather than basic science leading to applied science leading to technological innovations

i've omitted several other essays from the index or accidentally blended some summaries together, but, all told, this is a truly groundbreaking collection of work to understand scientific publication and the scope of the scientific enterprise, rooted in interesting and easy to read applied data analysis. it is also full of the alternately charming and jingoistic academic style of the 60s