Maps are for worldbuilders like hammers for carpenters. You always need one to pin something down, be it a log or an idea of a continent.
When I first started worldbuilding I didn’t know what I was doing. I didn’t know what I was doing because I didn’t know what I wanted to build. I just knew I wanted to buld something. I started inventing my first language in 2019 after which I decided I that inventing a language without a place for it to be spoken is not the real deal. I was always just “seeing what comes of it”, just fiddling around - but never investing real effort to make it right. So my first world grew without structure.
I had to learn the hard way that when it comes to creating a planet there is much more to it than just drawing lines inside a rectangle and calling it a map of besaid planet.
Map Projections
When I dug into Artifexian’s Worldbuilder’s Log I got introduced to map projections for the first time in a “serious manner”. Projecting a higher dimension onto a lower dimension is always just a vehicle, but a critical one. So what really helped me with that was thinking about it from the ground up. What would it look like for a 2-dimensional being to look at a 3 dimensional being and vice versa? xkcdHatGuy did a video on the topic. He made his last video 11 years ago and I generally admire his thought process.
We humans came up with a lot of different ways to solve the problem of not always carrying a physical globe with us in many ways.
Coordinate Systems
So you have learned that everything is a data point. And what property does every point in existence have? A position. You just have to give it a position relative to something.
Let me bring to you a smart comment straight out of a Reddit thread.
Maps tend to be split up using grids, which work just like the graphs you’ve seen in maths - A bunch of numbers along the bottom as your X axis and a bunch of numbers up the side as your Y axis. Those grid lines are then subdivided further. The purpose of doing this is so that you can just give someone an X and a Y coordinate on that grid instead of having to tell them to look through the entire map until they find a particular notable feature.
For a quick example, let’s say we were looking at this map. You were trying to find the tourist information centre and you’re so dumb you can’t find it on a one-building map. I could tell you that it’s in the top left square, but that’s not helpful at all. There are four entire squares here! So instead, I tell you it’s in 47,33. That lets you know right away which square it’s in. You just follow the X axis along until you get to square 47, and follow the Y axis up until you get to square 33, and now you know exactly which square it is!
But it turns out you’re actually so dumb that even knowing the square it’s in, you still can’t find it. In that case, I could tell you it’s at 476,334. The last digit of each of those is like a decimal. So you go along to square 47,33 again like last time, but now you know you need to count 6 decimal places along the X axis and 4 up the Y axis, and you now know not just which building the tourist information is, but also the exact part of that building.
The same principle works for massive maps too, only those maps are so big that finding the tourist information without the coordinates could actually be difficult. There might be ten thousand squares on a big map which would be very impractical to search through one by one.
Coordinates only work for specific maps. Each map will have its own coordinate system, which will depend on the scale of the map, the scale of the grid, and where the map places its 0,0 point - one map that puts it on a particular lake will have different coordinates to another map that puts it a mile south of that lake. The humans of Earth happen to have decided on another map system though - latitude and longitude. These work pretty similarly to grind coordinates, but cover the entire globe. The idea of these is that because they use a universal system - angles away from an arbitrary centre-point - as opposed to arbitrary distances away from an arbitrary centre point, anyone can locate anything on Earth just by knowing how angles work and where the arbitrary centre point is, and if more precision is needed, a new decimal place can be added without needing to change any of the scales of the map.
Taken from this thread: ELI5: How do coordinates work in maps Comment by the user Nephisimian
When we look at maps of our world there is always a (0,0). The tricky part is only that you are never working with a perfectly rectangular basemap to begin with. Because every 2D Map of our world is only projection of a 3D sphere.
Mathematically spoken you have a lot of possibilities to transform the area of a 3D-Sphere into a 2D-plane. Take any given point on the sphere and map it onto the 2D-raster.
Mercator Projection
Aitoff Projection
Equirectangular Projection
Healpix Projection
The most important map projection that you will need is the equirectangular projection. Why? Because it uses a mostly straight rectangular raster which is the standard input format for a lot of the tools we use for worldbuilding. Let alone the fact that a rectangular image does not have the need for transparency with non-rectangular edges.
You can always derive other projections of it later.
Madeline James has written a much more extensive post about map projections, find her post here.
Correcting a wrong Map Projection
If you have accidentially gone forward creating your world with a non-equirectangular projection, here is a workflow to convert it to an equirectangular projection.
In my case I found the Robinson Projection to be really pretty. I had already drawn all continents before I realized that I am going to be stuck with this projection because none of the tools I used can handle it.
Fortunately there is a tool that can handle all sorts of geospatial problemsolving - unless you are afraid of the command line of course. I deciphered those commands for you.
GDAL
GDAL is a translator library for raster and vector geospatial data formats that is released under an MIT style Open Source License by the Open Source Geospatial Foundation. As a library, it presents a single raster abstract data model and single vector abstract data model to the calling application for all supported formats.
gdal_translate 250809_conworldI_robinson_v002.png 250809_conworldI_robinson_v002.2.vrt -a_srs "+proj=robin +lon_0=0 +x_0=0 +y_0=0 +datum=WGS84 +units=m +no_defs" -a_ullr -17005831.44 8625157.17 17005831.44 -8625157.17
-
gdal_translate- Converts raster data between formats and optionally applies spatial metadata.
-
250809_conworldI_robinson_v002.png- The input image (a plain PNG).
-
250809_conworldI_robinson_v002.2.vrt- The output file, here in VRT format (a lightweight XML “wrapper” that stores metadata and references the PNG as the pixel source).
-
-a_srs "...projection..."- Assigns a spatial reference system (SRS) to the raster.
+proj=robin- Robinson projection (often used for world maps).+lon_0=0- Central meridian at longitude 0°.+x_0=0 +y_0=0- No false easting/northing.+datum=WGS84- Uses WGS84 as the underlying datum.+units=m- Coordinates are in meters.+no_defs- Ignore defaults from proj definitions.
-
-a_ullr xmin ymax xmax ymin- Assigns bounding coordinates to the image (Upper Left X/Y, Lower Right X/Y) in the projection’s coordinate system.
-17005831.44(left, in meters)8625157.17(top, in meters)17005831.44(right, in meters)-8625157.17(bottom, in meters)
gdalwarp -t_srs EPSG:4326 -dstalpha -of PNG 250809_conworldI_robinson_v002.2.vrt 250809_conworldI_robinson_v002.2.tif
-
gdalwarp- Reprojects and warps rasters into a new coordinate system or extent.
-
-t_srs EPSG:4326- Target Spatial Reference System = WGS 84 geographic coordinates (lat/lon).
-
-dstalpha
Adds an alpha (transparency) band to the output. -
-of PNG- Output format = PNG.
-
250809_conworldI_robinson_v002.2.vrt- Input raster (your georeferenced VRT wrapper around the PNG).
-
250809_conworldI_robinson_v002.2.tif- Output file. Despite
-of PNG, it has a.tifextension.
- Output file. Despite
History of cartography
-
Meridian
-
Graticule
Colors in Maps
The document contains information on land-based classificat
Two ways are possible to identify and characterize existing cartographic styles: existing maps may be first visually grouped to highlight various styles, or second individually analyzed to extract systematically main visual characteristics of their styles. (Christophe, 2012, p. 7)[^1]