Monthly Archives: April 2014

Bill Dollins to Geohipster: “Programming feels very similar to writing a poem”

Bill Dollins
Bill Dollins

Bill Dollins (Twitter, blog) is a programmer and partner at Zekiah Technologies, responsible for leading Zekiah’s geospatial consulting business.

Bill was interviewed for Geohipster by Atanas Entchev.

Q: You are a Senior Vice President at Zekiah Technologies. Do you consider yourself a mapper, a coder, a businessman, or a social media guru?

A: I tend to think of myself as a programmer first and then a businessperson. I have been programming for a very long time so that’s primarily how I think of myself. I’ve been at Zekiah since 2001 and I take the responsibility of keeping a stable flow of work for our staff very seriously so my role as a businessperson ranks high in my identity. As far as mapping is concerned, I can use my code to make maps but I am definitely not a cartographer. I had no formal training in geography prior to getting into GIS and learned a lot from some very patient professional geographers early on. I have a lot of respect for cartographers and geographers because the knowledge required to do what they do well is very complex and I’m not certain I would be doing them proper justice to hang my hat on that peg.

Social media is an interesting question. I don’t consider myself a guru with it. All of my presence on social media has its genesis in my blog, which was my first social media “property.” That really is an outgrowth of another component of my identity not mentioned above; which is that of a writer. I have written from an early age and programming, for me, is actually a creative experience that feels very similar to writing a poem. Writing is as core to me as programming.

Q: You do contract work for the US Navy, which we probably can’t talk about. So let’s talk about your extracurricular geoactivities which you document on your blog geoMusings. You write about integrating open source with Esri technologies. Tell us more about this. Do you do it for fun?

A: I have been programming in one way or another since I was ten years old. I am exceedingly blessed to be able to make a living at something that I truly enjoy. So, yes, I do it for fun and recreation. That said, very little of what I blog about is purely recreational. I, like many people, started in the geospatial world with Esri technologies. It will come as a shock to no one, especially Esri, that Esri tools alone rarely meet all of a user’s needs. So I have always been involved in integrating various technologies with Esri tools. I’ve gotten fairly adept at abstracting concepts and techniques out of my customer-focused work and turning them into free-standing examples for posts. That abstraction process is very recreational and keeps me mentally flexible.

Since the mid-2000s, I’ve been working more and more with open source geospatial tools. Given that most of my customers are Federal, they also tend to be long-standing Esri shops. As a result, my initial work started out focused on integrating open-source with Esri. My first visible effort with this was participating in zigGIS, which enabled direct read of PostGIS by ArcMap. PostgreSQL and PostGIS were of great benefit to one of my Navy customers and zigGIS was a natural fit. Since then my work has evolved to a point where about 50% of my work is purely with open-source tools, including some current Navy work. Part of that is due to the fact that open source tools are making significant inroads, and part of it is due to my intentionally seeking such work. As a consultant, I think proficiency with a diverse toolset benefits my business and my customers. As a programmer, it’s just damn fun!

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? How do you feel about the term hipster?

A: I think I’ll answer this through the prism of Geohipster. One common thread I have noticed in everyone you’ve profiled so far is a high level of energy, commitment, and enthusiasm for the work that they do. In that regard, I identify with them. I genuinely love what I do and can’t wait to solve the next problem.

The term “hipster” is a passing fad that is already losing its meaning. It is ultimately harmless.

Q: Is there a mainstream of geospatial data handling/representation? Who/what is part of it?

A: There is a mainstream and we are all part of it. The mainstream of handling and representation of geospatial data is, has been, and continues to be the layer. Regardless of technology provenance, geospatial data, especially vector, always distills down to layers. It is the most basic representation in GIS and also its continuing greatest limitation.

Given that GIS descends from map-making software, the continued prevalence of the layer is understandable. Maps were compiled from mylar separates which became layers in our software. We structure our data as layers. This is a function of both schema and common limitations of our visualization software.

I never really thought much about this until a project I worked on in 2005. It was an R&D project focused at modeling and analyzing infrastructure interdependencies. The system used an agent-based modeling approach and my role was to to provide some ArcObjects interfaces to access the geospatial data. The relevant features were used to instantiate objects in model space that began to interact with and respond to each other. The layer constraint did not exist and each object’s relationships to other objects, regardless of type, were more easily modeled.

I will confess I got a little obsessed with this concept and began delving into it more. Most geospatial databases allow you to remove the geometry constraint to store heterogeneous geometries in a table, including ArcSDE. The biggest limitation was with visualization. In the case of ArcMap (at the time), it would crash if you tried to add such a layer. At a minimum, it is inconvenient in terms of symbology and geometry collision. Layers make that easier.

If I were ever to get the opportunity to dedicate myself to a problem, it would probably be this. I find my mind wandering back to it these many years later. I think that we will probably not get past this until, as an industry, we recognize that map-making is a distinct use case from modeling and analysis and we allow our tools to diverge accordingly, similar to the way CAD and GIS diverged long ago. I could go on about this topic ad nauseam but your readers would probably fall asleep.

Q: Geohipster (and geohipsterism as a concept) is sometimes criticized for being exclusive and/or attempting to foster divisions within the industry. On the other hand, the just-ended State of the Map US (SOTMUS) conference in Washington, DC looked like a huge geohipster lovefest. Where is the industry going? Further fragmenting into tinier factions, or consolidating into a homogeneous whole?

A: The idea that geohipsterism could foster divisions in the industry could possibly be valid if it were approached without irony. I think the direction you have taken Geohipster should allay any such concerns. I was skeptical of it at first but have come to find it quite informative. I appreciate the Q&A format with other-than-the-usual suspects.

I did not attend SOTMUS myself, due to prior family commitments, but there was a photo tweeted from it that I think sums up the current direction of our industry: https://twitter.com/ajturner/status/454809703315668992. There’s Esri, Boundless, and Google at MapBox, all in one photo. It represents the flowering of innovation across our industry from numerous sources, whether traditionally proprietary or fully open source or in between. I see integration as the rule for at least the next few years. With the exception of Google, that photo represents the spectrum of technologies that I am currently using in my consulting work to support customers.

I am integrating MBTiles into a mobile situational awareness system, I am part of a contract team that is placing Boundless technology at the core of a major solution for a civilian Federal agency, and my company is using Esri technology to produce maps and automate infrastructure analysis for defense and homeland security users. This is all current work and tracks with diversification seen by others I talk to.

I see absolutely no evidence that our industry is consolidating to a homogeneous whole. I suppose the risk of fragmentation is there but, right now, each tool suite has its strengths and all of the players have been great about implementing de facto and/or de jure open standards so it’s very easy to pick the right tools for the job and integrate them all.

As a programmer and integrator, I hope our industry never returns to days like the early 2000s, when Esri had little to no credible competition and the whole industry just seemed stagnant. I actually considered leaving the industry at that point. The current level of innovation and competition seems to be pushing everyone forward and even Esri is responding. I’m not sure that would have happened without the competitive stimulus of the likes of Boundless, MapBox, Google, and the wider, independent open-source geospatial community in general.

Q: You own a John Deere and georedneck.com. Do you consider yourself a (geo)redneck? Any plans for georedneck.com?

A: I will confess that my Deere is a baby one; a 17-horsepower lawn tractor. My father owns several farm tractors that would put mine to shame. I bought mine several years ago and it has been a tank. I wouldn’t hesitate to recommend one if you are in the market.

It has become stylish to tack the prefix “geo” onto the front of just about everything, so I parked georedneck.com to head off any irrational exhuberance. I haven’t taken time to decide what, if any, concept may arise from it.

As far as actually being a redneck, I’d say I probably don’t quite fit the bill, but I will say that I am very comfortable with the culture and ethos. It is more nuanced than it is often portrayed and there is a lot to respect about it, if one takes the time to scratch the surface. Labels such as “hipster” and “redneck” can quickly descend into caricature and make it easy to forget we are just talking about people from different backgrounds who are trying to live their lives.

Q: You are always very nice and cordial online. Almost too nice and too cordial. Do you ever say anything bad about anyone?

A: Yes. Myself. I am my own harshest critic.

I was raised by a Southern mom who taught me to praise in public and criticize, directly to the person, in private. That practice has served me well. I strongly believe that a person or company should not initially learn about any negative opinion I may have via social media. I sincerely hope that others would extend the same courtesy to me.

Q: Is there anything else you would like to share with the Geohipster readers?

A: Share your work. Share your thoughts. Share your experience. Share your talent. It has more value than you know.

All one planet

I am just going to leave this here while I work on my tractate (Working title: “Is (geo)hispterism exclusive?” (Thesis: “No”)).

Matt Richards, Josh Livni, Andrew Turner at SOTMUS 2014
Matt Richards, Josh Livni, Andrew Turner at SOTMUS 2014