Monthly Archives: September 2014

Carl Anderson: “Rejecting your own work as the path forward is an important but painful thing”

Carl Anderson
Carl Anderson

Carl Anderson (@candrsn) started hacking on a TRS-80 in the 70s, quickly upgraded to an Apple II+, and has used all sizes and types of computer systems in his career. He is a polyglot and seeks answers using technology instead of seeking specific technology to answer questions. He is dogmatically pragmatic. In the last 30+ years he has worked for and supported local county, state, and federal government, the private sector and universities, and volunteered for too many things. He is the President of URISA, has served in many roles in local and international GIS organizations, and really enjoys working with the people he meets.

Carl was interviewed for GeoHipster by Randal Hale.

How did you get into GIS/mapping?

I have always been interested in the spatial relationships between things, and that interest drew me to measuring and representing those relationships. Specifically, in the early ‘80s I came across a book on microcomputer computational graphics and I was hooked. (Myers, Roy E. (1982) – Microcomputer Graphics).

So that interest drove you to pick a college major? Or was this a learn-on-the-job (as many of us did) career?

The path I took was in no way linear. I had taken programming courses while in high school, started out as an art major in college and later switched to engineering sciences. All the while I kept picking up odd jobs that included measurement, design, spatial analysis and information management components.

I’m currently fascinated with hybrid setups — taking commercial and open source, and building a hybrid system. You did that at Fulton County (Georgia), where you were the GIS Manager?

I was at Fulton County for a long time. I filled many roles while there, from traffic analysis in Public Works, to GIS management in the Planning department and later IT department, to leading a Business Intelligence division of the IT department. That last division included GIS, Database Development, Web Development and Data Integration, and was quite a challenge.

With regard to hybrid setups, I am proud of what we achieved at Fulton County. I had learned that things can really hum along when each component does its own one thing well. Back in the early 90s it took a lot more heavy lifting to connect components. Today, using REST, GeoJSON, WFS, WMS, XAML, OBDC, and others it is really easy to hook things up. We used to run into issues with every piece of software we tried, and had to patch code in Perl, PHP, the Linux kernel, mars NWS (a Novell emulator), TN 5470 (to talk to the IBM mainframe), and too many others. I took a bit of grief over building systems that combined both open and proprietary components, but it was the only way forward as we knew that as long as we did not require a line item in the budget, we could do nearly anything we wanted.

That hybrid approach allowed us to be super responsive to our client base and keep up and leverage the changing technology and internal (county-government-wide) standards. We were able to build custom GIS apps for Planning, the Tax Assessor, the Tax Commissioner, 911 and EMA, the Police and Sheriff, Voter Registation, and more. Using a hybrid approach, the end users were not affected when we upgraded core GIS software, or even when we switched databases.

One particular technology journey was especially interesting. We had to change the data storage system several times, and we were never completely in control of the timing. We started out moving from ARCStorm to SDE 3.0, but had to backtrack due to a need to reload data from original sources. We then moved to Oracle for a few years but were using a license borrowed from a different department that wanted it back. I had been playing with Postgresql95, and we moved everything from Oracle during a stressful week of dump as SQL, refactor as a PostgreSQL95 variant, and load SQL. After migration we managed to keep systems in sync using a mixture of Perl and PHP. PostGIS did not yet exist, so we developed our own spatial storage and predicates. In late 2001 we got turned on to PostGIS. As the 2000s progressed, it got easier to move and model data, so we started to connect live to client data systems and integrate on the fly with data caching. When I left we were connecting systems using MSSQLServer, PostgreSQL, Oracle, Berkley DB, MS Access, SharePoint, and others. The whole journey required us to reject what we had been doing, refocus on what we really needed, and find a solution — any solution — that fit our true needs. Rejecting your own work as the path forward is an important but painful thing.

While you were chugging along at Fulton County, you ended up helping co-write something called FGDC-STD-016-2011? How painful/fun was that?

Otherwise known as the FGDC Address data standard, it was a lot of work and a lot of fun in a geeky kind of way. The most important thing I learned is that developing a standard takes a long time. The Address Standard took 5 years from kick-off to adoption. Reviewing addressing practices in the US revealed many odd things that you might not suspect exist. Part of that work was informed by a custom geocoding engine we wrote at Fulton County. It used a more natural linguistic pattern to locate candidate addresses instead of the more literal matching engines available to us in commercial GIS software. Currently I am helping on an international address framework standard (ISO 19160-1), likewise lots of fun in an internationally-geeky way.

So I’m going to slow-pitch one question, then stick you on another one. You are now President of URISA. Given the whole idea of GeoHipster (people who work/think outside the box) I’ve been questioning as of late how effective the big organizations are (URISA, ASPRS, GITA) at maintaining a connection with the industry when it seems to be changing at a rapid pace. First off — what is URISA? Secondly — how is URISA adapting to the needs of its members (assuming it is)?

URISA, officially the “Urban and Regional Information Systems Association”, got started in 1966 as an organization to help share ideas and results using a (then) new-fangled thing (computers) to solve problems in urban and regional planning and government.

Our tagline is “Fostering Excellence in GIS”, and we take that to heart; we exist to help GIS practitioners succeed. One of the big roles URISA continues to fill is to help make newer, useful techniques and technology feel safe for GIS practitioners to implement. As a community of GIS practitioners, URISA allows people to see what their peers are doing in GIS, what is working, what is not, how people can repeat successful ideas. It is also working to identify practices in setting up and managing GIS systems that are particularly useful, effective or efficient.

I was at a conference a few months back and someone said “Blah blah I know a guy who maps caves” and I responded “Blah blah blah I know a guy who maps caves”. We both knew the same guy. I know how busy we get and it’s been a while since I sat in a canoe. Are you still a member of the Athens Speleological Society? How did you end up mapping caves?

At the moment I am only an armchair caver, but I do feel then need to get underground again. I have always felt a stronger affinity to the Dogwood City Grotto, in Atlanta GA. than the Athens Speological Society near the University of Georgia.

When I was at Georgia Tech I ran into several great people who used the mapping and survey of caves as a way to investigate the spatial relationship of caves. People like Bill Putman, Ed Stausser, Steve Attaway, Marion Smith, Jim Smith — they all stirred up my curiosity and got me mapping caves. Cave surveying is quite different from modern above-ground survey techniques. The equipment has to be super rugged, be able to get completely wet, and fit into a small backpack. At Georgia Tech and later, some super-cool armchair caving projects were undertaken, like running 100s of paper topo maps through a scanner and digitally automating the plotting of 1000s of caves. Did you know that paper topo maps shrink over time, and that they do not shrink in a uniform way? Figuring out how to account for that was a problem.

You’re currently working with Spatial Focus and are working with the U.S. Census Bureau. What are you doing with Census?

I am supporting the LEHD program that produces statistics on jobs, workers, workplaces and the connections between them. I focus on making sure that we always get the geography right.

Finally — I haven’t asked you about skinny jeans, your favorite capa mocha half caf latte (I have no idea what I’m saying), or your bike or whether it’s a fixie or a flexie (I think I made that word up) or how your record collection is coming along. We define hipsters as people who think outside the box and often shun the mainstream (see visitor poll with 1106 responses). Would you consider yourself a geohipster?

I am pretty sure that I do not entirely fit in anywhere. A friend recently mentioned that if I am hip, it is entirely by accident and not planned. As some examples, in the ‘80s I frequented a PBR bar in Atlanta and I have had long hair and a beard most of the past 30 years, including a fabulous mullet in the ‘80s. Possibly cool now, but at the time — not so much.

With regard to the mainstream, as a polyglot I think that I am pretty good at making choices for software / language / tool for each task I encounter. I don’t think that I choose what I already know, but what will cause the best outcome.

Oops, almost forgot… Favorite Linux distribution?

I installed my first Linux distro in 1993, it fit onto two 1.44MB floppies, and was released by a Portuguese telco company. Ironically, I do not speak Portuguese. It had exactly what I needed. Later I have used Debian, Redhat, SUSE, and all sorts of other distros. Using a polyglot parallel, my favorite distro is the one that has the tool that I need that day. As an aside, I really loved the Enlightenment Window Manager in the late ’90s. For its time, it was really creative and challenged the boundaries of what window decorations and widgets could be.

Finally — any parting words of wisdom for the good-looking and smart readers of GeoHipster?

Try to live life one, or less, mistakes at a time. Try to make the successes big and the mistakes small. Try nearly everything once — if it is not good, don’t do it again. Lastly, living in the present is much more fun than living in the past.

Bill Morris: “There’s a lot of value in questioning the establishment”

Bill Morris
Bill Morris

Bill Morris is a passable developer, a derivative cartographer, and a GIS refugee. Having cleared a decade as a geospatial professional and founder of Geosprocket LLC, Bill is now mapping renewable energy markets as the Lead Visualization Engineer at Faraday Inc., where he has yet to pay for a software license but is getting nervous that the streak can’t possibly hold forever. Bill is a lifelong Vermonter, with furtive dashes into the outside world.

Bill was interviewed for GeoHipster by Mike Dolbow.

Q: How did you get into mapping/GIS?

A: I was a music major at Middlebury College about 15 years ago when a friend convinced me to take a geography class. Fortunately that was about the time I realized that I was a pretty bad musician, so it made a lot of sense to shift into a field that seemed to offer both a series of structural worldviews and a technical skillset. I keep running into awesome Middlebury geography grads in the wider world; I know I’m lucky to have stumbled into that department and be launched into the world with the uncontrollable desire to map stuff.

Q: Earlier this year you put your own business, GeoSprocket LLC, on hold, to join Faraday. After about six months, what is different today from when you made that transition?

A: I’m a lot less stressed.

In all seriousness, as a freelancer I grew accustomed to reaching critical stopping points – letting documentation searches drag on way too long – before putting out a question on StackOverflow or begging help from someone via Twitter. But the Faraday team seems like a hive mind most days. Pretty much any block in my technical knowledge can be covered really quickly by one of my colleagues, and I know I can offer the same to them. The efficiency that comes from a complementary team can’t be understated, and I know this because I’ve been the squeaky wheel a few times elsewhere.

I’m also a bit more pragmatic about the umbrella of GIS technology. Learning how to optimize PostGIS with a hundred million data points – in tens of thousands of configurations – has given me new perspective on limits. I’ve started to understand the database admins who reflexively scoff at spatial; whenever there’s a choke point in our data processing, it’s usually a buffer or a point-in-polygon operation. Removing the abstraction of the desktop GIS platform speeds things up a lot, but geospatial analysis is still the slow donkey bringing up the rear of the wagon train.

Q: What are some of the more interesting projects you’ve been working on lately?

A: Faraday is letting me go a little crazy with visualizations. Some things are sticking (MOAR HEXAGONS) and others aren’t (not all datasets look good as a pulsar), but it’s an amazing iterative environment for trying out ideas. We’re aiming for a distinctive, map-centric design in our platform, and over the past few months Mapbox Studio has been invaluable for tying the cartography to the app design. Our clients are also looking to us to make sense of some pretty abstract statistical concepts, so I’ve been getting into the weeds of practical information design, then emerging and hammering something together with D3. Combined with our goal of increasing renewable energy’s market share, this fulfills most of my “dream job” prerequisites.

My side projects have slowed down this year, but I’m hoping to get back to a greater level of involvement with the Humanitarian Openstreetmap Team. Crisis and development work are really motivating for me as hard-edged examples of the power of maps.

Q: You once said: “I didn’t know the first thing about code when I got into this world, but it was amazing how easy it was to adapt a little bit” with help from free resources like CodeAcademy. I have found the same thing, but also found others in the “traditional GIS space” reluctant to take the plunge into things like Javascript and HTML. What advice would you give geographers who aren’t sure if coding is for them?

A: I take it for granted that this question is settled. That obviously everyone should learn how to code, what’s the big deal? But it doesn’t take much self-awareness to realize that it was my good fortune to have both the need and the resources to learn programmatic approaches to mapping. There are plenty of GIS analysts who can keep on working without javascript or python, and I think many more who simply don’t have the time or the support. I’d like to help those in the latter group and not alienate those in the former, but my patience is waning for the anti-developer reactionary set.

To the geographers who only know the GIS desktop and feel its limits: ask for help. It’s more readily available than you think. Hell, ask me for help.

Q: Your Twitter handle is “vtcraghead”. I get the VT part, but I had to Google “Craghead”. Is that a reference to the village in England, or something else?

A: I wanted to have a unique email after college, and I was climbing like a madman at that point so I registered “vtcraghead@hotmail.com” and joined the brave new digital world. The handle stuck with me, but by the time I registered it with Twitter it was more of a joke about how I used to tie in a lot.

Although it’s curious to see that Craghead is in Durham, which reminds me of my favorite song about surveyors and the broader impact of mapping lines in the dirt . . .

Q: We define hipsters as people who think outside the box and often shun the mainstream (see visitor poll with 1106 responses). Would you consider yourself a hipster? (Who else would aspire to play in a “low profile funk band”?) How do you feel about the term hipster?

A: As with most of the previous interviewees, I subscribe to the middle ground. I admire the geohipsters (none would self-identify, I’m sure) who helped me break out of incumbent technologies, and those who are innovating geospatial tools in ways we could only dream about a decade ago. But I’m not a fan of the brash contrarian hipster archetype, either in real life or as a straw man.

As far as my own identity? I ride my bike constantly, but it has ten gears. Skinny jeans on me would be a war crime. This is Vermont, and inside these borders PBR is outlawed. However, I think there’s a lot of value in questioning the establishment.

Q: Geohipster (and geohipsterism as a concept) is sometimes criticized for being exclusive and/or attempting to foster divisions within the industry. Or sometimes for being different for the sake of being different. You once rolled your own basemap tileset (using Mapbox’s guidelines). Did you do that to be different?

A: Oh jeez – that sounds like metahipsterism.

I did that as an experiment in self-reliance. I feel so poisoned by my experience with a single-vendor-technology career track that I’m always watching the exits. I love Mapbox, but I wanted to know if I could make an attractive web map without paying them anything, which is the occasional promise of open source tools.

Geohipsters fostering divisions? I see this as the current manifestation of an endless social dynamic: A new group enters a space, with new ideas. The old group finds it easier to feel threatened and defensive than to adapt. The new group can always do a better job of assisting the adaptation. </overlysimplisticparable>

Q: Like me, it’s pretty clear you’re an active dad. Loving your kids comes second nature, but let’s face it, they also require a lot of attention. What’s more tempting to compare to your kids: your projects or your customers?

A: Projects for sure. Mostly adorable and exhausting in equal measure. Thankfully, my customers neither throw legos at me nor tell me they love me.

Q: I’ve always had a theory that New England states are like siblings from the same family: they have rivalries and unique characteristics, but when challenged will band together and “defend their identity” to other states. As a fellow geographer from New England, what’s your take on that?

A: New Hampshire is definitely Vermont’s evil twin, but we’ll take it over Texas. Don’t even get me started about Sox-Yankees.

I can be a bit of a Vermont nationalist, but I’d say our industry (probably not uniquely) has flattened the cultural obstacles to collaboration. The folks I interact with on Twitter are everywhere, and it’s almost a non-issue for my career that I don’t live in D.C. or the Bay Area. That’s why I’m a technophile, in a nutshell.

Q: Admittedly, it was over 25 years ago, but Vermont is the only place I’ve observed this phenomenon. Have you seen this, and can you possibly offer an explanation?

A: Witch windows were a cheap alternative to dormers for venting and light on the upper floors of old farmhouses. I worked on a house years ago that had one, but I admit this is the first I realize they’re just a Vermont thing 🙂

Q: Any final words for GeoHipster readers?

A: I don’t personally want to be defined by my struggles against Esri. That comes up a lot in projects that I’m passionate about, but for better or worse they are the “incumbent” in this space, and they are the portal through which many of us enter the world of mapping. I’m probably just mellowing with age, but I’d rather emphasize the positivity of flexible skillsets and robust community in mapping than rant about vendor lock-in. We’ll probably all get more done with that perspective.

Jonah Adkins: “Engaging with other like-minded geo people via Twitter provides a great unfiltered look at what’s working and what’s not”

Jonah Adkins
Jonah Adkins

Jonah Adkins (@jonahadkins) is a Sr. Geospatial Analyst with GISi out of his home office in Newport News, Virginia. He has been in GIS since 1999 working for local governments, federal agencies, and most recently as a consultant. Jonah is a published cartographer who enjoys time with his family, maps (duh), Disney, Pro-Wrestling, has a tattoo of Esri North Arrow 51 and was told by Pharrell Williams that he looked like Freddie Mercury.

Jonah was interviewed for GeoHipster by Christina Boggs.

Q: Hey Jonah, thanks for taking the time to sit down with me! I know you from Twitter but many of us know you from your incredible Lost Map or they fall in love with It’s A Small World all over again through your cartography. What do you do for your daytime job?

A: For the last few years I’ve worked as a Sr. Geospatial Analyst for GISi. The majority of my work entails cartography, graphic design, UI/UX for applications, and traditional GIS work for many of our clients. I’ve spent a good portion of last year doing some awesome things with the Navy Shore Energy Program.

Q: Your maps are beautiful, beyond just being a method to convey information; they’re art. Do you have a background in design or some other art media?

A: Nope. I was always a doodler growing up, but never took any classes. I wouldn’t call myself a designer or anything like that, but I think I can tell when something looks “good” — at least to my tastes.

Q: I stumbled upon a great video where you presented great cartographic design elements to keep in mind. Have you thought about teaching this to other GISers? Beyond the occasional conference talk?

A: It’s a funny thing. I’ve always thought that you can’t really teach “making a pretty map.” Books and the like can give you helpful info, but I feel it’s something that’s unique to the person creating the map. Your current emotions, likes and dislikes, all that stuff is in anything you create, for the most part. Then someone told me, “yeah that’s great and all but we want you to show us how to make a pretty map”. So I had to get my thoughts on paper and decided I could at least give some pointers on guiding the creative side to a desirable output.

Q: Gretchen Peterson wrote an incredible book on cartography, have you thought about writing?

A: Never. I have Gretchen’s awesome book “Cartographer’s Toolkit,” a signed first edition!!! It’s been my starting point on several projects where I’m struggling to find my inspiration and I need something to get me started. It’s very good because it’s a tool to guide you, it’s not, to me, an instruction manual. Water doesn’t have to be blue, it’ll be ok, and I’ll be damned if you use Comic Sans as your title font!

Christina: Haha I really did spend some time selecting what font to do this interview in.

Q: You have a robust github, tell us about some of your projects you’ve got going up there…

A: I really love maps, mapping tech and all that goes with it. Basically any of my free time, and some of my not so free time, I’m usually trying out something new or working on an idea. This usually happens with one or both of my daughters watching Saved By Bell in my lap. GitHub is great, it’s collaborative and social, which is something I tend to thrive on.

Weekend Update was one of my first github projects, it’s a riff on Project Linework, a not-so-standard basemap themed after Saturday Night Live’s Weekend Update segment.

Amazing-er Maps is a cartographer’s plea to make online maps better — not that I know better, but with all the open/free map technology out there, certainly some of these “Amazing” maps on the web could be improved.

Custom Backgrounds In AGOL – Probably one of the more popular ones, is a guide to adding another provider’s maps in ArcGIS Online projects. I’m not the first to write about it, but I think it’s safe to say that people are ready, and want control over every aspect of online maps.

Q: Speaking of your daughters, we just had the 10th anniversary of OSM. Your daughter contributed to OSM with you, how did you do that? If you could give advice to other dads out there, how can you inspire your daughters to get involved with mappy stuff? What did she map?

A: My sweet Sophie 🙂 She really thinks what I do is cool, and after a few years of talking to her school classes about maps, she’s been bugging me to teach her something. What better way than through OpenStreetMap? She’s 10 and starting 5th grade, the iD editor was really easy for her to pick up. We talked about what she wanted to do, she decided on schools and parks, because “I think kids like me would like to know where playgrounds and schools are.” She loves the fact that everyone can see the work she has done. She wants to teach her friends at school once she goes back next week.

Q: So recently you started up a Hampton Roads, Virginia #maptime, how’s that going?

A: Like most I’d seen some ramblings of #maptime over the last several months on Twitter, and after hearing Lyzi Diamond talk about it (twice) at the Esri UC, I was really inspired to get one started. I like helping and sharing about maps, and have been struggling with a way to do that locally with more than the typical GIS crowd. I’d started attending the local Code For America Brigade meet-ups earlier this year and @maptimehrva is a great extension of that hack night concept. Come hang out, and let’s talk maps. Doesn’t matter what you know, you’re gonna learn something tonight.

Q: I find the cartography you do with ArcGIS Desktop breathtaking but honestly, I get excited when I toss in a drop shadow or I do a transparency mask to highlight my area of interest. If you were to put three pieces of low-hanging fruit out there for other Desktop users to implement in their maps, what would they be? (READ: a couple of cool “tricks” in Arc that have good cartographic payoffs?)

A: Regardless of what software you use, practical knowledge of concepts helps greatly, experience helps too! To me it’s fun to learn the ins and outs, push its limitations and figure out ways to do things it easily can’t. Something I tell people who ask for cartographic help: GIS people make GIS maps — which is a bad way of saying GIS people tend to make maps they can read and understand. A friend made a great analogy of that — people would rather read “SF Earthquake: 6.1” than be shown a seismograph.

My three tips would be:

  1. Use a color palette. Colors are better when they are not fighting with each other for paper space supremacy. But also be mindful of too much color. Nothing stands out when everything is on the same color ramp. Finding the right balance pays off in the end.
  2. Never accept the defaults. Things I’ve seen overused for the last 10 years: ArcMap Yellow (hex #fcfbab) in the legend or graphic background, Esri North Arrow 9, Layer_and_field_names_like_this_in_the_legend… Change something, change everything. You’ll gain experience just by exploring the options.
  3. The One “Thing” — before you start a map, determine what is the one “thing” the viewer should take away from the map. Is it the neato font? Keep going. Is it the pattern on the water? Keep going. Keep going until it’s the purpose of the map. All of those design elements should only help tell the story, they shouldn’t be the story.

Q: I love the term geohipster, I take it as a playful comment. How do you feel about the term, do you self-identify as a geohipster? What does it mean to you?

A: It’s a playful comment I can relate to. I get regular comments about my mustache, and I’m like, I’ve had this thing since sixth grade, and I’ve only been without it once. Then all these people started growing ‘staches out of novelty, or because it was the new cool thing, and I get lumped in with that crowd. Maps have been around. It’s only natural for those of us that have been around with them to say “yeah, but I’ve been mapping since before Google.”

Q: Five awards at ESRI UCs, first place at Virginia’s GIS Conference last year, Runner Up, Best Cartographic Design at last year’s FOSS4G, 2nd Place in this years GISCI-GISP Map Contest — do you have any award winning pieces in the works?

A: I try to do several personal projects a year. It helps me keep my skills sharp, and gives me a chance to just have fun mapping something for myself. All shrouded in secrecy of course, sorry.

Q: Some weeks I see you at #gistribe, here’s your opportunity for a #shamelessplug — do you have anything you would like to share with geohipster readers?

A: #gistribe, #geowebchat and others provide such a great social resource. My #shamelessplug would be to invest your time in social discussions like those. Company feeds are good for updates and examples, but only provide one view of the technology. Engaging with other like-minded geo people via twitter provides a great unfiltered look at what’s working and what’s not. It’s the biggest downer to me when I talk to someone in our field who’s never heard of OpenStreetMap, Github, Leaflet, Tilemill, etc. And it happens often. There’s so many great things happening in Geo that you’ll never know about unless you step outside of your bubble and explore.