A world of mistaken notions

If you looked at the past centuries, what stands out prominently are their structures. You look at their temples, churches, mosques, synagogues, gurudwaras, universities, observatories, railway stations, their monuments – the list can go on – and you thought about it, you will realise that none of these would have been possible without a deep understanding of geometry, ratio & proportion, strength of materials, malleability of materials, operation of levers, to mention just a few. Each of these would have demanded years of disciplined study and precise calculations. Each would have called for many decisions and making a choice of which decisions are worth taking – my apologies for the play on words. The contrast between these structures of the past and those of today is stark.

Consider Tyco Brahe. He lay on his back to gaze at the stars and jotted down all his observations, most of which were calculations. In today’s language, it is observational data, not ‘raw data’ but his calculations. As the Library of Congress, USA, puts it “Tycho Brahe collected observational data at an unprecedented scale, and developed his own competing model” (https://www.loc.gov/collections/finding-our-place-in-the-cosmos-with-carl-sagan/articles-and-essays/modeling-the-cosmos/whose-revolution-copernicus-brahe-and-kepler). They say competing model because the article, Whose Revolution? Copernicus, Brahe & Kepler is comparing him, Copernicus, Johann Kepler. My purpose is to show the practice of using data to build a coherent picture, in the 16^th century. Incidentally, this is a fine article, and says, in the early part, that “In 1563, at age 16, he observed Jupiter overtaking Saturn as the planets moved past each other. Even with his simple observations he saw that existing tables for predicting this conjunction were off by a month, and even Copernicus’s model was off by two days. In his work, he demonstrated that better data could help to create much more robust models” (URL as above)

Let us go back farther. Plato’s Laws. This is a conversation among three men, one of who is the Athenian (widely considered a Plato surrogate). The discussion is on the size of the city, which, according to the Athenian, should consist of exactly 5,040 households. Why? As George G Szpiro explains in a delightfully insightful book “Numbers Rule” with an intriguing sub-title ‘The vexing mathematics of democracy from Plato to the present’, it is because 5040 is ‘a convenient number’. Why? “It can be divided by all natural numbers up to ten…. (In fact) “altogether it has 59 divisors”, which comes in handy if you need to partition the city and households have to be whole integers. And we are talking of more than 2000 years ago!

Let us come back to 19^th century to breweries in London. In a book called “The effortless economy of science?” Philip Mirowski, Carl Koch Professor of Economics at the University of Notre Dame when he wrote this book in 2004, has a chapter 9 titled “Brewing, Betting and Rationality in London, 1822-44”, where he describes the practice of forecasting in one brewery run by the family of Truman. Mirowski describes in detail the practice of forecasts made by each partner of the price of barley and the actual price between the period 1822-46, and also that the partners must have been familiar with the principle of extrapolation. Deviations were calculated (standard deviation) for each partner all of which went into the calculation of the amounts due to the partners.

The subject of Operations Research was literally born during the Second World War to help Britain to better use war material. After the war, it began to be extended to industry and other areas and is a major discipline dealing with decision-making. An extremely challenging disciplined, students have dealt with the subject with no calculator let alone the use of any software! You could find a solution only if you understood what the problem was and posed it in an appropriate manner. Today, we just have a proliferation of subjects dealing with data and decision-making driven perhaps more by the way the education system is organized rather than as improved ways of studying the subject.

All the steel mills, automobile factories, to mention the most basic industries could not have been built without using data. It is not just absurd to think human beings have decided until twenty years ago without taking into account data; it reflects a poverty of thinking, a narcissistic attitude about ourselves today and willful ignorance. Or there is just a vast business to be built in and on data and this could be a carefully orchestrated marketing campaign continuously raising decibel levels.

In sum, we are living at a time when there is an enormous amount of computing power available. That’s all. And we make terrible mistakes despite the access to enormous amounts of data. Or perhaps even because of it, because we really are so fascinated with all the computing power that we have forgotten that intelligence has to come first from us, before it could become artificial.

Takeaways

The structures created centuries ago is standing proof of use of ‘data’

The growth of astronomy demonstrates the same aspect

The present times just enjoy enormous computing power