It is fairly widely accepted that the first true data visualisation was made by the Dutch cartographer, mathematician and engineer Michael van Langren. Langren served as royal mathematician to King Philip IV of Spain, and became interested in one of the most important mathematical problems of the day - how to accurately determine longitude. Being able to determine longitude was hugely important for navigation, especially at sea. There were various methods used at the time but most were known to be estimations at best.
To demonstrate the wide difference in the estimations, in 1644 Langren published a simple, single dimensional chart to demonstrate the significance of the problem to the Spanish court.
|Langren drawing for the Spanish court of 1644|
On it he plotted the estimations made by 12 astronomers of the difference in longitude between Toledo in Spain and Rome in Italy. His drawing clearly shows what we would now refer to as the statistical distribution of the 12 estimates, making it not simply a picture but a chart, and marking Langren as the farther of modern data visualisation. Although it is extremely simple by modern data visualisation standards, it could certainly be argued that is it still the best possible way to visualise the data.
With the technology available to us today, we can now not only see the huge variation in the estimates, but the fact that even the smallest is significantly too large. The below is an overlay of the above chart on a modern Google map scaled to match.
|Langren drawing of 1644 overlaid on a Google map|
It is worth noting however that the story of Langren's chart dates back even earlier. The chart he published in 1644 is now known to have been first drawn much earlier. In a letter he wrote in 1628 to Isabella Clara Eugenia, daughter of Philip II of Spain, Langren included a drawing, that whilst showing fewer longitudinal estimates, is no doubt the same.
|Section of a letter by Langren to Isabella Clara Eugenia, 1628|