Moments in Analytics #1: Charles Babbage's Broken Windows
Are you suffering from information overload in a data-driven world? Have you ever asked yourself how we may have gotten to this point? Travel back in time to the Big Bang of analytics and learn about its humble beginnings. It might be older than you think!
It was a fateful Saturday weekend in the 19th Century that two interesting stories were published next to each other in Mechanics' Magazine.
On page 82 of the issue, there was a short article titled, "Table of the Relative Frequency of Occurrence of the Causes of Breaking of Plate Glass Windows", written by Charles Babbage. Over 10 months, the man who had previously worked on the world's first computer, counted 464 incidences of broken windows. Some results:
- 55 windows were broken by throwing stones at windows; an additional 16 were broken by children throwing stones at each other - while the whole 'don’t throw stones at your neighbors, if your own windows are glass' thing had already been around for centuries.
- 14 windows were broken by drunken people 🍻 🥴
It's the last sentence that sticks out the most, though:
It will be of value in many respects, and will, we hope, induce others to furnish more extensive collections of similar and related facts.
Babbage was a pioneer of analytics: collecting, storing, analyzing and sharing an interesting dataset - and encouraging others to do the same. Sure, his analysis was basic (ranking the causes by frequency) - but analytics had to start somewhere. And it started with data small enough and calculations basic enough such that humans did it all by hand.
Which leads us to the next article, "The Swedish Difference Engine". It had been about 20 years since Babbage had worked on his initial computer designs, but others had picked up the torch and continued to develop them. A father-and-son team from Sweden had just sold its second computer (a Scheutz No. 2) to "an enlightened merchant of the city of Albany." It was installed at the Dudley Observatory, but may have never been used.
It wasn't until a few years later, and the next version of the computer (Scheutz No. 3), that it proved to be useful in calculating and printing statistics. It was used in the production of the 1864 English Life Table, which contains statistics calculated from data on 6.5 million deaths.
So it's around 160 years ago, in the midst of the steam age, that the world of analytics got started with its own big bang of laboriously collected datasets, and machines that could finally automate and calculations too long and complicated for humans. There was hype, and there were setbacks - but humans were well on their way to doing more with data.