The Timoney Group has been responsible for groundbreaking demonstrations of Google Earth’s usefulness as an analytical tool — Envisioning Jonah Gas, South American Trade (a collaboration with Eicher-GIS) and now Gulf Impact.
In this interview with Brian Timoney, I ask him about what it means to be a Google Earth developer, why he uses open-source tools, what the strengths and weaknesses are of server-side processing, and whether SketchUp has a role in GIS.
Ogle Earth: In a few sentences, what did you do before Google Earth came along?
Brian Timoney: I come from a GIS background working in the environmental, petroleum, and military contracting fields. The primary impetus for starting my own business, interstingly enough, was as much organizational as technological. Specifically, there was a disjunct between the sophistication of the tools being used at the analytical level and the static output (99.9% of the time a paper map) being used at the decision-making level.
Since the learning curve for GIS software has traditionally been forbiddingly steep, it wasn’t realistic to expect overworked managers to tackle those programs when they just wanted answers to some straightforward questions. So I was interested in building simple-to-use, but useful tools: we started building a Flash-based vector mapping engine when all of a sudden the Google technologies hit (GMaps and GE).
OE: You describe yourself as a full-time Google Earth developer. What
does that involve?
BT: The term “full-time Google Earth developer” is a bit of a misnomer in the software development sense since the customization of the interface itself is limited. However, in Google Earth we saw an interface that an average user both understands and enjoys using. So our task is to be able to fortify the great, visually engaging end-user experience with the ability to look at a variety of data that might be stored in a shapefile, Excel spreadsheet, or enterprise database. With GulfImpact.com, we’ve extended the functionality to include basic geoprocessing capabilities that are accessed over the web in a manner that the user isn’t too far out of their comfort zone.
OE: On your site you plug open-source tools like PostGIS and PHP. Why?
BT: At the risk of sounding like a money-grubbing consultant, the interest in open source doesn’t have a lot to do with idealism but more to do with the value proposition. Building on the Apache/PHP/PostGIS platform lets us get our services out to a vast audience in a cost-effective manner. But we emphasize to clients that if they’re in a .NET/Oracle world, great, we can help them; if they have no resources, no problem, we can build something very cost-effective that doesn’t include expensive licensing issues. In short, open-source to us isn’t about ‘free’, it’s about flexibility.
OE: Google Earth (free) doesn’t do Geospatial analysis, so you’ve moved that functionality server-side and send the results to Google Earth. What are the advantages/disadvantages of this approach to the ArcGIS Explorer client-side model?
BT: Having been generally satisfied with the ESRI products I’ve used over the years — starting with command-line ArcInfo — I think much of the Google Earth vs. ArcGIS debate is unfocussed. Those whose job is to collect/create/modify/analyze spatial data have a very different set of requirements than the (much larger) audience of end-users who need a simple, but powerful interface for visualizing and querying data. The fact that Google Earth has been such a hit with a general audience is not only a testament to its technological virtuosity but also to the not-so-user-friendly nature of the web mapping experience before Google/Keyhole.
No longer being an ArcGIS user and not having seen in person the new ArcExplorer, I can’t comment on how all of that technology (along with ArcServer) fits together. However, I’ll simply make the generic observation that with each bit of functionality one adds to an interface, the more complex it becomes for a user to navigate. And I don’t think it’s a stretch to say that GIS has traditionally favored increased functionality over user-friendliness.
Our approach to server-side geoprocessing is that if we can create easy-to-use tools that help the general user answer basic spatial questions while still keeping them in a comfortable interface environment then that has real value in an organization where, as you go up the food chain, managers and VPs have neither the time nor the interest in learning a lot of new software just to get the answers they want.
OE: Is SketchUp of use to you as a GIS developer?
BT: Having met the SketchUp guys at a couple of functions in Colorado, I’m happy for them that they were acquired for a nice chunk of change. The democratization of 3-D models seems to be well afoot, and I can’t imagine any city or county not having downloadable 3-D models of project proposals, etc. on their websites available for public review in, say, five years. Being more of a data guy, I’m interested in the possibilities of dynamically changing the appearance of 3-D models based on user-defined attributes such as “make every building red in city x that hasn’t been inspected for this or that code regulation in the last 3 years…” In short, visually encoding real landscapes in 3-D seems to me to hold all sorts of promise for anyone in the geospatial field.
OE: Thank you, Brian.