Recently I’ve rekindled my lifelong interest in maps and cartography. Many years ago I worked as a full time cartographer for a couple of years, and this map was the project that consumed most of my time there. Produced back in 1997, before modern online mapping and consumer GPS navigation existed, this was a hand-crafted paper map designed to help bicyclists navigate the Eastern part of Massachusetts.
My involvement in the project ranged from hands on design and cartography to custom coding, as described in the sections below.
This map was drawn by hand in AutoCAD. Roads were traced from public domain sources (to keep file size down), and road names were broken, tilted, and placed by hand. Similarly, place names, feature icons, and all other elements were carefully placed by hand and arranged to maximize the amount of information while maintaining legibility. It was a painstaking process, but yielded a map that was always praised by its users for its extreme usability for its intended audience – road bicyclists.
Due to hardware and software limitations of that era, the coastline data we utilized for our base map contained far too many points and choked the system. I implemented a well-known curve simplification algorithm in AutoLISP in order to reduce the number of points in the coastline polylines by over 50% while preserving the perceptible coastline shape.
A critical piece of information for any cyclist is – where are the hills? Using a grid of public DEM elevation data, I wrote a routine to walk along each road polyline, interpolate the elevation at regular intervals, and derive the slope along the road. Blue hill arrows of varying thickness were then automatically placed to indicate several levels of hill intensity. This became a highly prized feature of our bicycle maps.
These days we are used to being able to zoom in and out to any map scale instantaneously, without considering the drastic impact scale has on everything – what is included or not, how labels are arranged so as not to compete with each other, etc. All of this is worked out ahead of time by the creators of the tiles that make up the map at each zoom level. For this map, scale was tangible in the form of countless small design decisions, ranging from global choices like colors, font face and size, etc. to very local ones like where to position a town name so as to minimize the number of local features it obscures.