INTERVIEW WITH AURELIA MOSER

Aurelia Moser is a librarian and developer based in New York. She develops open source mapping software at CartoDB and teaches data visualization courses. Aurelia: “a story is something you can extract meaning from, and when a map is informative, even if static, it provides that story”.

Graphic Hunters: What is CartoBD and what is your role?
Aurelia Moser: CartoDB is an open source engine on mapping software. Our specialty is making the data processing, geocoding and querying process simple and easy. CartoDB is free-to-use for beginners and full of features that support both coders and non-coders alike. We have a hosted GUI interface for people who have data and want to do minimal tech work beyond customizing colors and style. And we have a series of libraries to help people bootstrap their own projects in Javascript.

I’m a Map Scientist on the Community team. We build curriculum for our Map Academy and Tutorials online. We also develop maps and tests of our CartoJS libraries, Editor examples and various APIs. We do a fair amount of support work with our community of journalists, developers and academics. Likewise, we coordinate the CartoDB Ambassador’s Program to showcase our superusers, and the Startup/NonProfit Grants program to support our partners on their projects.

What is the first most important question one should ask before starting making a map?
I would say: “How will this help people navigate?” Maps are built to communicate information and aid in navigation; whether that navigation is physical or conceptual.

Maps often compare or connect something to something else; cartographically it connects geography to a representation. The goal is ultimately to aid navigation and better construct a conception of the world. When you build a map, you need to think about how it helps or hinders navigation, how you advocate for either (help/hinder) with the projection of reality you present in that construction.

What are the most important rules or steps in making an interactive map?
There are four rules or steps.
1. Consider your goal: why is a map important? How does it help you navigate or explore these data? What will the interactive accomplish?
2. Consider your audience: who would interact with this map and why?
3. Consider your data: do you have mappable or geospatial information? How clean, solid or reliable is it? Will it allow you to represent your goal accurately to your audience?
4. Aggregate these considerations: collate them in a tech stack and design implementation. Keep your audience and goal in mind. And represent the data with accuracy and appropriate disclaimer when representational inaccuracies are inevitable.

What’s the biggest mistake often made in making maps?
Applying data to geography that fails to provide a better understanding of place.
What tools or skills are important to learn when you want to work with interactive maps?  
There are appropriate tools for all components of maps; from a tech perspective I would recommend a lot of open source libraries and Javascript. A solid understanding of how to deal with geospatial information is important too; how to query it with SQL/PostGISQL. Much of the magic of online maps these days can be accomplished with some Javascript libraries and some cool geo-APIs to get you the data and the basemaps you need.

How important is it for people to know some basic coding?
It’s quite helpful to have training in coding for making maps in general, and interactives in specific. Developing traditional maps might require some training in GIS desktop software for digital production. Interactive development (where the content displays dynamically) however requires coding outside the typical GIS stack. The good news is that lots of this kind of coding is free to learn and openly documented on the internet, so it’s a bit less-opaque than GIS precedent.

Can you mention a good example of storytelling with maps, and why is this a good example?
A few months ago, some participants at our semi-annual Eco-Hackathon built a map of sea mammal trajectories in the ocean. They had some scientific data about the GPS location points of mammals with sensors attached, that had been monitored for some time to track their migration patterns. But the data were in tabular format; unattractive and largely illegible. The eco-hackers managed to translate those points into glowing and mobile pins on a map, to illustrate the migratory movements of the animals and the typical cluster-patters that they follow at the species level.

It’s a simple map, using Torque (our time-series data library) to track the animals over time. And it tells a good story because it illuminates so much information about how mammals ascribe (or don’t) to the expected trajectory, where they group and diverge in their migrations, how quickly they move and with what patterns they prefer. It’s mesmerizing and informative, like some of the best stories.

When does a map of geographical data becomes a story?
Immediately? I would say that a story is something you can extract meaning from, and when a map is informative, even if static, it provides that story.

How important is the role of design in visualizing a map?
Quite. Poorly designed maps are illegible, or they skew information in a way that makes them difficult to navigate conceptually or physically. A poorly designed map is absent of purpose, or it completely obfuscates the potential for a purpose (which all maps possess).

What is the most frequently asked question in your training sessions and what is your answer to this?
I receive a lot of enthusiastic questions about ‘where to start?’ In recent years we’ve taken additional measures to ensure that starting up with geospatial software is quite easy. We provide online curriculum, free accounts, in-person trainings, online webinars and many ways to learn and self-educate.

Likewise, we developed a base-library of data called ‘Common Data’ that anyone with an account has access to. People can create mashups with their data and more common or clean sets. In this case ‘typical datasets’ like administrative boundaries, census information and natural earth physical geography datasets. We’re trying every day to build more modifications, features and extensions that make answering the ‘starter’ question that much easier.

What lessons do you want participants to take home for the training?
Some hopeful take-aways:
• some comfort with geospatial data
• a fluency in basic coding for making maps
• a strong reflex for building navigable narratives