Kristen Grady is a GIS Specialist at NYC Emergency Management and has over ten years of experience working in GIS. Prior to working at NYCEM she spent about six years working in academia trying really hard - but eventually failing - to avoid working a 9-5 office job. (Although saving the city from the apocalypse turned out to be a pretty cool job, so it’s OK). She’s loved airplanes even longer than geography and hopes to combine her two passions into an actual paying job someday. But for now, she makes maps and writes python code by day and stares at her airplane emergency card collection by night, which currently stands at an impressive 139. (And yes, they all very clearly read “PLEASE DO NOT REMOVE FROM AIRCRAFT.” She says she’s sorry!) Find her on Twitter and LinkedIn.
Q: How did you get into GIS?
A: I think it happened about six or seven miles up, somewhere near where the troposphere meets the stratosphere. I was flying from New York to California in June 2006 — my first time flying jetBlue — and I had never seen a live flight-tracker map before. I was enthralled! It was a perfectly clear day all the way across the country, and my 8 megapixel camera was pointed out the window for the entire six-hour flight. I’d take a picture of something neat on the ground and then immediately snap a photo of the map. When I got back to New York a few days later I sat with these photos and Google Earth, which I had just downloaded for the very first time, and spent hours trying to figure out what was in my photos…
A few months later, my Weather and Climate instructor was giving a lecture on remote sensing. He was going through slides of satellite images and having us guess what they were of. I knew them all! At the time I was a philosophy major, but I immediately went and switched my major to geography. The next semester, while taking the required geo-technologies course for the geography major I finally got to play with desktop GIS software and made some (pretty terrible) maps. But I knew that this was what I wanted to do. Someone in the geography department back then had made a comment that geography was a perfect discipline for “someone with ADD” because the variety of projects and aspects of GIS that you could focus on were truly infinite. That sealed the deal for me.
Geography and mapmaking were always passions of mine. I had my face buried in atlases and had been making pretty intricate maps since I was a kid, like this one that I made at age eleven. But I took a *very* circuitous route through college, losing myself about a hundred times, before finally taking that life-changing jetBlue flight that reminded that, at my core, I was born to be a geographer. So after seven long years in undergrad, I finally got that geography degree, and found GIS, and I’m so glad I did!
Q: You have a Master of Science degree in Geographic Information Science. What is the one most important (or most valuable) thing you got out of your course of study?
A: I graduated with a B.A. in 2008, probably the worst year in recent history to start looking for a “real” job. So I ended up mashing together some part-time GIS research jobs and continued taking graduate-level GIS and cartography courses for fun. This eventually led me into a PhD program, which I was in for two and a half years before deciding to call it quits with a Masters. (A story for another time!)
So unlike in an undergrad program, where you’re essentially just learning how to use tools (at least in my experience), in a graduate program you are also being taught how to think critically about those tools, as well as how to think critically about the disciplines of GIS and geography themselves. You have to think about the consequences of your analyses, the ethics of your maps, the ethics of your tools. You have to think about things like the effects of aggregation, the cultural implications of using a certain color on a map…
Then there’s learning about different geographic “paradigms” and critical geographies, such as feminist geography or Marxist geography… I had no idea while I was in undergrad that there was such a rich philosophy of geography. I feel lucky to have been exposed to that. Having that experience at the graduate level has definitely made me a better, more critical map-maker.
Q: You work for NYC Emergency Management. Is your job stressful? Last week Amazon S3 went down because of a typo. If *you* make a typo lives are at stake. Do you ever think about that? Does it stress you?
A: Oh totally. I put a lot of pressure on myself because I am a perfectionist. But the paradox of working in emergency management, where your maps and your data really ought to be showing the most correct information, is that when sh*t hits the fan, there is rarely any time to go over everything in painstaking detail. It is not my nature to work this way *at all* so it’s been an interesting challenge for me.
There is always a struggle between balancing the quality of your work and being efficient. This is why I try to automate things using Python and by using map templates that I created a while back. This way we can spend more time on making sure the map and data are accurate and less time on things like creating a layout from scratch, or worse, creating an Esri scale bar from scratch (it’s the worst!). I have actually written a Python script that automates that process for us. Hooray!
Q: I interned for Manhattan Borough President’s Office in 1992. We used MapInfo then. Have things changed in NYC? What technology do you use these days?
As for the rest of the city, I think it varies. My sense is that it is largely Esri-based. But I am familiar with a few agencies that are moving toward open source technologies, like DoITT, who I believe is using QGIS for their desktop mapping. A colleague of mine at DOHMH uses R, D3, Leaflet, and PostGIS for her mapping projects, and DCP’s new Capital Planning Division has just used all open source technology to create their Facilities Explorer, which I love and was just released to the public.
Q: Tell us about a cool project you work on right now.
A: As I’ve said, things happen really fast in emergency management. A typical work day for me is pretty calm and laid back… until of course, something happens. One of the big ideas last year in the Public Safety Data Development Center (the group I work in within NYCEM GIS) was to create a dataset that answers the question, “What is there?” Meaning, if there is a sudden event, such as a building collapse or an explosion, we immediately want to know all of the facilities that exist in the affected location. Is there a hospital there? A nursing home? A restaurant? A school? We used to do this by adding a bunch of datasets one by one — that we had to think of off the top of our head — to an ArcMap document. But that is both inefficient and prone to oversight (like forgetting a dataset, for example).
Answering this question sounds easy enough (“Why not just use Google?!”) but what is so challenging is bringing all of these disparate datasets, most from different sources and with very different schemas, together into one dataset. The City of New York cannot simply rely on Google’s databases alone for its spatial awareness. We cannot verify the accuracy of their data.
Many of us worked on this project, but my job was to write an ETL in Python that would extract as many datasets as we could (currently 23, but eventually 50 or more) from our database, transform them — perform selections, map the fields, etc. — and then load them into one singular dataset. We still have a long way to go, but at least now, we can pull in this one dataset, which we call “Facilities Master,” select all the points that fall inside a building or within a given radius, and know an awful lot about the facilities in an area, with just a few mouse clicks. And this way you don’t have to think too much, which is always my goal. Plan and prepare when times are calm (think!), and then respond quickly when things get hectic (do!).
Q: You are a Pythonista. What advice will you give to someone who is just getting started with Python in GIS?
A: Wow. What a great word, Pythonista. Can I use that on my resume?!
Learning to code can follow a totally different path for everyone and really depends on your learning style. Some people can start copy/pasting other people’s code right away and fairly quickly manage to build something new that actually works. This approach didn’t work for me. I wasn’t easily able to break through the wall that stood between me wanting to learn to code and unshackling code from abstraction, and so I was a little paralyzed at first. But now I know that in order to learn code, you have to just start writing it and stop pussyfooting. You have to have faith that all those neural connections that you’re creating in your brain will eventually result in some pretty spectacular “eureka!” moments.
As for the more practical aspect of learning to code, you need simply to start out by learning the basics (variables, lists, conditional statements, loops, etc.), and then start playing. If you aren’t able to take a class, there are a million online code-learning sites, most of which are free. Once you know some really basic stuff and have learned what a module is, play around with the Python turtle module, which was originally created to help kids learn to code. It’s a great way to make really cool things happen pretty quickly, and it’s included in the Python Standard Library.
If you want to write scripts and create tools for ArcGIS, you’ll need to learn ArcPy, the Python site package that lets you interact with ArcGIS. Esri has pretty good documentation on how to use arcpy, and GIS Stack Exchange is also a great arcpy resource.
Here are a few rules I think the budding Python coder should follow:
- Know that coding requires incredible self initiative and self learning. Learn how to ask the right questions and become a master Googler. GIS Stack Exchange is indispensable, but users and moderators will publicly shame you if you haven’t done your homework before posting a question. I love that.
- Errors are learning tools, you’ll never stop getting them, and they will only get more complicated over time. Accept them. When you’re comfortable, learn about debugging and error handling.
- Pleeeeease comment your code. You will forget what you have written if you haven’t looked at your script in two weeks. More importantly, if someone else has to read it, explanations in a human language are key. Don’t be lazy. Don’t write sloppy code. Include script headers.
Q: Enough about work. What do you do for fun? Being a Brooklynite, whatever it is surely must be hipster, no?
A: Brooklyn is a pretty special place to live. It is also very hipster. One of my favorite things to do, and fortunately for my budget and my liver I don’t do this too often, is try to find really good craft cocktails. There are some amazing ones to be found in this borough, but obviously also in Manhattan. I have not yet ventured to the other three boroughs in search of craft cocktails, but I should! One of my favs in Manhattan is Amor Y Amargo. They are the standard to which I hold all other craft cocktail bars. A place I love to go to in Brooklyn is Blueprint. They also have incredible bar snacks. Yum!
When I’m not consuming spirits, I am doing much healthier things like snowboarding, taking pictures, hanging with any number of my enormously huge family, including my two little nephews whom I adore, seeking out some top-of-the-line self-serve froyo with my other half, or geeking out hard on airplanes…
Q: You also like airplanes. How did you develop that passion (for it is a passion, right)? Tell us more about it.
A: I could spend hours answering this question! There are so many amazing spatial things going on with airplanes. But to be honest with you, I’m not really sure why I became enamored with them as a kid. I’d give anything to go back to early 1991 and ask that 8-year old girl, who just found out that she was going to be flying Continental Airlines from Newark, New Jersey to Orlando, Florida, why she instantly became so obsessed with them (and with the airline itself).
I think there are a few things going on. For one, I just think the airplane is a beautiful machine. But it’s also a symbol of escape, adventure, and change, and I have always liked all of those things. Also, the airplane affords anyone lucky enough to sit in a window seat an incredible and rare view of the surface of the Earth, which is a pretty spectacular experience for anyone who loves geography, although I didn’t have that particular experience until I was a bit older. My initial obsession mainly involved planespotting, which is at its most basic simply identifying aircraft types and airline liveries.
As I’ve gotten older and as technology has allowed for easy access to all kinds of flight-related goodies, the passion has evolved into an actual hobby. An #avgeek session for me might include using multiple flight-tracking apps (Flightaware, PlaneFinder, Flightradar24) and live ATC feeds to track a single flight or multiple flights that satisfy certain criteria. Sometimes I like to freak people out by “planestalking” them. (I actually coined the term Planestalker in the Urban Dictionary, and as of the time of this writing, it has 4 likes! ha!) Recently, I was planestalking my cousin’s flight from EWR to DEN, and it made a go-around in DEN. They were only feet off the ground before they aborted their landing due to wind and flew around to land on another runway. Nowadays you can go to Flightaware and just download a KML file of your flight. I sent him a picture of his go-around, and he thought it was hysterical (but also pretty cool!).
Some of my favorite airplane “games” or challenges are trying to catch and then follow my pilot talking to ATC from one feed to the next (e.g. from Ground to Tower or from Departure to ARTCC), or predicting where an airplane overhead is coming from or going to and which runway it either took off from or is about to land on (which I am a total expert at predicting, btw!). I had a lot of fun making an animated map of some “Flights over Queens!” a few years ago, but unfortunately it got a little (irreparably) messed up when Carto switched from Editor to Builder.
Even more recently, I’ve developed an affection for aviation-related maps, like VFR sectional charts, arrival and departure procedures, and IFR Enroute High Altitude charts. I mean, talk about not being able to make a mistake! And having to think critically about the implications of your cartographic choices! Who makes these wonderful maps?! I am convinced that they are made by sweet little garden gnomes, working tirelessly in the night, running their maps from tree to tree… There is just so much magic, and a bit of mystery, in flying… it’s fun to uncover it all.
Q: Do you consider yourself a geohipster? Why / why not?
A: You know, at first I didn’t think I was at all, but then I realized that maybe I was a little bit when I was completely unable to answer the first question in the interview — which is a pretty simple and straightforward question: “How did you get into GIS?” — without launching myself into a paralyzing debate on my feelings on the word “GIS.” Did I want to be associated with such a contentious word, what seems now to be a target for people who don’t want to be boxed in and who instead feel that they are part of something bigger than GIS, something geospatial? Just the fact that I was freaking out about the connotation of a word, in a very academic way… that must be somewhat geohipster, no? (Fortunately for the geohipster readership, I decided to scrap the eight-page essay that accompanied that manic thought spiral and instead tell you all a nice little story about flying… hee)
Q: On closing, any words of wisdom for our global readership?
A: I wish I could take credit for this perfectly succinct and beautiful advice that I’m about to give, but I can’t, as it was offered as a suggestion to me by my partner when I read this question to him out loud…
“If you’re on the ground, look up, and if you’re in the sky, look down.”
It’s exactly what would have taken me multiple paragraphs to articulate, but he did it in just one sentence. He knows me so well. I think I’ll just leave it at that.