Inspired by the #YearOfOpen project, I’m trying to get into the habit of talking more about what happens behind the scenes at Krakatoa Design. We’ve been spending a lot of time on the cork maps these past few months, so this seems like a good place to kick things off. If you ever wanted to know how we create the artwork that get’s etched onto the cork, this is for you.
To paint in broad brushstrokes, the maps start as digital mapping files, go through an extensive filtering process, then get cleaned up, and finally, converted to a filetype that’s laser friendly.
That’s cool, Jeff, but I’m here for the nitty gritty details. OK, you asked for it. Let’s get into it.
The internet is a playground for anyone who loves maps. There is more data out there than you could imagine, but much of it is copyrighted or in a format that is very difficult to manipulate. We use public data from either OpenStreetMaps.org or governmental data that is compiled by various municipal agencies. This mapping data may be a GIS (Geographic Information System) or vector (think PDF or Adobe Illustrator) file.
GIS files are ideal because components (roads, buildings, water, etc) are linked to longitudinal and latitudinal coordinates. In other words, every component knows it’s place which is a huge help when you’re making spacial files. The only problem with GIS files is that there’s so much data. We need to have clean, minimal artwork for etching, and that takes lots of filtering and editing.
An interface like Overpass Turbo, or program like ArcGIS, allow you to specify what data you would like to filter out. On our maps, we usually stick to roads, parks, and water. We don’t need sidewalks, fire hydrants, driveways, parking lots, drainage ditches, or any of the other stuff that makes up 98% of GIS files. What we filter for depends on the scale of the map. On a neighborhood level, we might want to show every road. On a regional level, maybe we only want interstates, US and state highways, and primary roads.
If you're new to the GIS world I strongly recommend using the Wizard function in Overpass Turbo. It allows you to filter map data without knowing the technical query language.
Regardless of what filtering criteria we use, once we’re happy with how it looks, we need to export it. I've had the best luck exporting as a geoJSON file type.
With OpenStreetMap.org, we’ll need to involve another program in the exporting process. QGIS, a great piece of open source software, allows us to take the Open Street map export (the geoJSON file) and convert it to something that’s laser friendly. We work in Illustrator, so exporting as a DXF does the trick. QGIS is capable of a whole lot more than just converting files, but for this project it's just a pitstop.
This is where the bulk of our time is spent. Once we have a vector file that can be read in Illustrator, assuming all the data made it, we’re ready to start cleaning things up. The laser can only etch one color and one stroke weight, so we need to be creative with how we represent data, otherwise it turns into an indistinguishable jumble of lines. Roads need to be be smoothed out, water needs to get filled with wave symbols, borders need to be turned into dotted lines, etc. How much cleaning is dependent on what laser you’re using.
Illustrator is geared towards print work and lasers aren't exactly printers. Smarter lasers are more successful at translating exactly what you’ve drawn into an equivalent etching, but there’s almost always some additional manipulation needed to get the laser to ‘see’ what you’re seeing on the screen. Using Illustrator in Outline Mode (Command+Y) is a good starting point to understand what the laser will see.
Now that we've got the file as clean as possible it’s time to power up the laser and let the testing begin. Easy, right?
If you’re excited about learning more, here are some good resources:
- How to download QGIS - It's surprisingly tricky, especially on a Mac.
- QGIS overview - great starting point on how to use the software
- Open Street Map tags - in order to filter, you'll need to learn how everything is classified. This is a great reference for all the types of classification tags
- Overpass Turbo Wizard - good overview of how to the Wizard feature, the easiest way to start filtering map data