Abstract
In September of 1999, the Mars Climate Orbiter, built at a cost of $125 million, crashed and burned on its approach to the planet. NASA engineers were devastated by the failure and sought to find its cause. Surprisingly, it was not an equipment failure in the complex spacecraft but a rather pedestrian arithmetic error. It seems that the Jet Propulsion Laboratory (JPL) used the metric system in its navigation calculations but Lockheed Martin Astronautics, the designer and builder of the spacecraft, used the English system to calculate critical acceleration data. As it happened, nothing good could come from this and perhaps the key takeaway from the disaster was that “measurement matters.”
James Vincent, a science and technology reporter for Verge, the Vox media website, has written a fascinating history of how and why humans came to measure just about everything and what this means for how we think about, and interact with, the world. What I found most interesting as I read through the book was how much about metrology (the science of measurement) I take for granted. As someone who has spent more than half a century engaged with engineering and technology, I “know” what someone means when they talk about acre-feet of water, cubic yards of concrete, or pounds per square inch of pressure but does everyone share my understanding of what a foot, yard, pound, or inch actually is? And what does this mean to someone raised and trained in the more universal metric system, who thinks in terms of liters, meters, and kilograms? Early on in human history, standardization in measurement was recognized as a necessity if people were to interact in a meaningful and equitable manner.
Mr. Vincent tracks this history from the earliest human civilizations as they sought to define quantities ranging from the depth of the Nile River’s annual floods, to the length of the seasons, to the amount of harvested grain. Such information was vital to survival on both a personal and societal level. Buyers and sellers of grain, for example, needed to share a common understanding of the quantities exchanged to arrive at a fair price. At the same time, there needed to be implicit trust that the measurements were fair, even if not always precise. Someone “putting their thumb on the scale” is now a cliché but once carried severe penalties. As commerce expanded and trade was no longer just a local practice, the need for more standardization was required but culmination of this effort comes a bit later in the story.
Before this, the nascent age of scientific discovery led to the definition of many units, initially for astronomical observations but later in the physical and natural world as inquisitive minds sought to replicate empirical observations through experimentation and demonstration. The lack of standard tools and a common language to describe the results of measurements were an impediment to sharing this work. For example, without an agreed upon standard of length or weight, how could anyone interpret the conditions and findings of an experiment where they were not present? Without a thermometer and a temperature scale, how hot is “hot?” Although great strides forward were made during this period, it took a revolution to bring real standardization to measurement.
As the French Revolution gained momentum, among the grievances voiced by the peasantry was the call for standardization in weights and measures. In fact, this issue placed higher than the fairness of the courts or infringement on personal liberties. Given the many ways that the nobility used such measures to their advantage in their dealings with the peasantry, this is not surprising. Of all the changes wrought by the French Revolution, the metric system is perhaps the most universal and enduring.
Initially developed by the savants as the French intellectuals were known, the metric system introduced the meter and the kilogram as universal standards for length and weight. The meter was defined as one ten millionth of the length of the meridian from Dunkerque in France to Barcelona in Spain, the kilogram as the weight of a cubic decimeter of water. Both were physically embodied in cast platinum artifacts stored in the French archives and endured for two centuries as the universal standards. However, as it turned out, the meter bar was a bit short, and the platinum kilogram was losing weight as it aged. These physical artifacts have since been replaced by a meter standard based on the speed of light and a kilogram defined from Planck’s constant. More accurate perhaps, but certainly less romantic.
The final element of metrication was the move to a decimal system which made scaling more rational than the prevailing base 12 and base 16 systems of feet and pounds. Do we ever wonder why we buy a dozen rather than 10 eggs? However, despite the logic of the metric system and its almost universal acceptance, the United States, together with Liberia, and Myanmar still have not officially adopted it. In the case of the U.S. at least, this is now mostly an ideological and political issue that has little real impact except in outlier cases like the Mars Orbiter.
Although the technical aspects of the book are certainly interesting, I found Vincent’s views on the societal impacts of measurement far more thought-provoking. From topics ranging from the colonization of the U.S., to the pseudo-science of eugenics, to work productivity, he builds a strong case that whoever gets to measure gets to control what they’re measuring and the results are often not pretty. In the case of the westward expansion of the U.S., the federal government used land surveying to delineate parcels that could be given to or claimed by settlers to populate the vastness of the new nation and push native peoples ever westward from their traditional lands. Treaties with native peoples were violated regularly when their claims to lands were inconsistent with the settlement objective and these conflicts were seldom resolved in favor of the original occupants.
The pseudo-science of eugenics, which promulgated a belief that only the “best” people should be allowed to reproduce and pass on their “superior” genetic makeup was grounded in a belief that such characteristics could be determined by measuring physical and social attributes. That theses “measures” inherently favored those of high social standing who also happened to be mostly white skinned was far from coincidental. Lost in their fascination with measuring, the proponents of eugenics apparently never noticed the underlying racism and immorality embodied in their beliefs.
The ability to collect and analyze data with newly developed statistical tools also supported numerous efforts to improve the efficiency of mass production processes. Henry Ford was an early adopter and his automobile assembly lines effectively turned workers into part of a mechanical process. Each step in the process could be measured and analyzed with the goal of maximizing worker productivity and increasing profits. Although there is nothing inherently wrong in striving for efficiency, dehumanizing the workers involved is not an acceptable tradeoff and this raises a broader point that Vincent addresses.
Just because we can now measure just about everything, it doesn’t mean that we should or that at the least, we should exercise considerable care in how we interpret the results. Management literature abounds with articles that speak of Key Performance Indicators (KPI) and Objectives & Key Results (OKRs) as though they can explain everything that needs to be known. The search for metrics that convey useful information to improve decision-making is certainly a worthy endeavor, but do they tell us everything we need to know? If and where does human judgement come into play? Are we comfortable placing our trust in measures that we don’t fully comprehend?
The book also introduces the 17th century work of John Graunt, a little-known London haberdasher who basically invented social statistics and paved the way for the broad use of statistical measures to structure and guide our lives today. Working from data collected in the Bills of Mortality which compiled births, deaths, and christenings throughout England, he was able to compute many previously unknown facts such as the estimated population of London, infant mortality rates, and other public health trends. Modern metrics such as unemployment and inflation rates are derived in a similar manner and their impact can be seen in the somewhat wild gyrations of the stock and bond markets around their weekly estimates and their anticipated impact on economic policy making.
Although not specifically discussed in Beyond Measure, the infrastructure management community lamented for many years the lack of reliable data on the structural health of buildings and infrastructure to aid in investment decision-making for maintenance and repair (M&R). However, recent advances in sensor and telecommunication technologies have greatly increased the amount of data that can be accumulated on the state of "smart" infrastructure systems and the emergence of the “Internet of Things.” The problem now appears to be not an absence of data but rather, a lack of actionable information. What this suggests is that improved decision-making for M&R investment (and other areas as well) will perhaps be found in better algorithms rather than more data.
Mr. Vincent has much to say about both the history and ethics of measurement and he says it well but I wish he had written more about algorithms and their often very selective impact. Whether “Wins above replacement” tell us more about the value of a baseball player than their batting average is little more than a topic for tavern arguments; algorithms, that intentionally or not, redline minority neighborhoods or reduce the competitiveness of women in the workplace are unjust and indefensible. Although I would have liked to have seen more of what Mr. Vincent had to say about this, at the end of the day, I found Beyond Measure a treasure trove of useful and interesting information and came away with a greater appreciation of all the work that has gone before to make our world a more rational and comprehensible place.