For many years, I've been interested in what I could use collected data for. This was way beyond the days of "big data" and data science which are all the flavour now. I don't agree with the way that some multi-conglomerates are using data now. Mapping data is of great interest to me because of the power of visualising that data on a map where there is a geographic need. You can look at numbers over and over but you won't see the big picture until that data is placed on a map.
In the mid 90's, I bought a Garmin GPS device and started experimenting with connecting the device to a PC, downloading waypoints using the NMEA protocol via a serial port (in the days when there were serial ports included on all PC's).
I wrote a program to among other things collect waypoints for displaying on a map. The GPS device only exported Longitude/Latitude. The client wanted Easting and Northing. I found out that someone had some code that took the Long/Lat and converted it to Easting/Northing. They were quite happy to share the source code with me. I took this rather long series of functions that triangulated the earth to achieve this feat and in a language I had never seen before … Fortran.
Never one to shy away from a challenge, I converted the code to Delphi (Pascal). It took quite awhile to get working. I took the GPS round our neighbourhood collecting waypoints, excitedly connected it to my PC and downloaded it into a database table. My download program converted the Long/Lat waypoints into Easting/Northing and both formats were stored side by side. I hooked the data up to Arcview (the mapping software) and my waypoints didn't match where I had walked, they were about 1 km out!!!
I contacted the originator of the Fortran code and he said, "oh … you need a fudge factor"!
Absolutely classic!! Can you imagine today's GPS needing a fudge factor or maybe they do!
How things have changed since then.