DM Exclusive: Chris Anderson, CEO of 3D Robotics, on the future of GIS
Chris Anderson is CEO of 3D Robotics, founder and chairman of the Linux Foundation’s Dronecode Project, and founder of the DIY Drones and DIY Robocars communities. He spent eleven years as the editor in chief of Wired Magazine and seven years with The Economist. He is the author of the New York Times best-selling books, “The Long Tail: Why the Future of Business is Selling Less of More” and “Free: The Future of a Radical Price,” as well as the new “Makers: The New Industrial Revolution.” In 2007, he was named to "Time 100," the news magazine's list of the 100 most influential men and women in the world. Our distinguished columnist, Bill McNeil, spoke with Anderson, Monday, September 11, regarding his vision of the future of the GIS industry.
McNeil: How do you see big data sourced from drones, satellites, and ground/marine-based vehicles impacting the future of the GIS industry?
Anderson: That’s a really good question. There are really two different ends of the GIS spectrum. The construction world is all about designing structures in relative space. Typically, we don't even have absolute coordinates. It's all relative coordinates, whereas the traditional GIS world has real-world coordinates.
What's interesting is that these two worlds are converging. The traditional CAD, architecture, civil engineering, and construction markets are now starting to move out from abstract coordinate systems to local coordinate systems. Now they're starting to put their designs on a GIS framework. The building lives on a site. The site is in a town. The town is in a region. What was once a building is now part of an entire urban fabric that includes coordinates of infrastructure — things like roads, bridges, power grids, pipelines, and other parts of the surrounding area.
CAD companies like Autodesk will be moving to GIS. Meanwhile the traditional GIS world of maps and latitude/longitude is starting to move down into very granular cities. Buildings, within buildings, and even multiple building layers will have coordinates.
They're starting to move into the high-resolution scanning and digitizing of the physical world. In a sense, you can think of the GIS world as microscoping into the AEC world. The AEC world is telescoping out to the GIS world. What this means is we now have two different industries that are really the same industry. They have different cultures, and languages, and tools, and even coordinate systems. In our AEC world we used to start with no physical coordinate systems at all. It is all relative.
The construction world is going from absolute coordinate systems to local coordinate systems, so now they're starting to synchronize with global coordinate systems. That's adding GIS into traditional CAD tools.
When you start doing that, you end up starting to use the language of GIS, yet it's a very unconventional approach. Just think of the precision difference. CAD lives in a world of one-centimeter grids. GIS lives, at best, in a world of one-meter grids. So you have this really interesting change in accuracy where the construction world's incredibly accurate but not scalable. The GIS world is very scalable, but not incredibly accurate.
As we attain the ability to measure the world with greater precision, we have to ask if we are creating one database or two. What standards are we using? Which culture? Which language shall we adopt? The answer, of course, is going to be a hybrid of the two. It's really the first time in decades that these two industries are getting together. Forget the globe. What is the database of a city? What is the database of a town?
McNeil: How does time or temporal data fit your scenario?
Anderson: What is the database over time, not just the database of the moment or the database of the construction project? The database could be a living city that not only is built but is maintained and ultimately interacts with other databases of the world.
I think this is one of the moments when we're all realizing we're talking about the same thing. It's like the blind man and the elephant. We're all talking about the same things, but we use different words and languages. There's an opportunity to start to pull back and harmonize these grand projects.
When you say “map the planet,” there is really no such thing as a map. The planet changes by the minute, by the second. It's a living database. You can think of it as a geospatial data. There are lots of other kinds of data that can be overlaid on a geospatial coordinate system.
Buildings used to be designed by an architect as a blueprint. It was a building and that was it. Later, if you wanted to make changes, the builder would access the original architecture plans and see if any changes were made. Maybe there were and maybe there weren’t, but it doesn’t matter because the blueprint was not a living document.
This is typical; it happens in academia all the time. Chemists don't talk to the biologists and biologists don't talk to the physicists. They're all solving the same problem. They just use different language, different methods, and have different cultures. This is an interdisciplinary moment for GIS.
The one thing I think automation really allows is constant sampling and constant measurements. In the past, measuring the world has been expensive and difficult and therefore we didn't do it very often. Now satellites and drones are changing that. They are robots in the sky and space.
In the construction industry surveying is still an occasional thing. You might survey twice on a project, once before you design it and once before you build it. What would happen if surveying was free? What would happen if it required no people whatsoever or if surveying was automatic? Then you'd survey every day. Maybe you'd survey every hour. Then the assumption would be that the site is constantly monitored. Then rather than having one or two surveys, you'd have daily or even hourly surveys.
Timing would completely change the culture of monitoring and managing sites. Now you would assume that the site database is up to date. You would start to depend on it for things like managing your workflow and developing supply chains. Customers would start to demand it as a form of auditing and transparency. Regulators and governments would start to require it so they'd have a paper trail. We would start to assume accurate data rather than assume inaccurate data, and we would compensate for that by creating longer schedules with more time for processes. Time, the fourth dimension, is probably the most neglected of the dimensions that we work in, but I think it is our biggest opportunity.
McNeil: These are fascinating concepts. From where will the data to support these ideas come?
Anderson: Satellites and drones will generate a significant amount of data. There's an explosion of micro-satellites in space that is increasing coverage almost on a daily basis. This will generate more measurable data, but 80 percent of the earth is covered by clouds at any point in time. In other words, you're not going to get consistent coverage from satellites, at least not predictably. Again this is the fourth dimension, the time dimension, and the cadence and frequency of satellite data is dependent on when the sky is clear.
You have to have devices that operate under cloud cover. That could be manned aircraft. It can be drones. It can be static sensors and cameras that are on location all the time. This is an issue that is not going to be solved by any one type of vehicle or instrument.
McNeil: Unlike satellites, drones can operate under cloud formations. Is the industry mature enough to be a major data provider?
Anderson: What I see is kind of gradual evolution of the industry. The focus has really gone from drone hardware to the data side. The balance between hardware and software companies is now shifting more towards software. The general assumption is that manufactures like DJI have built a good enough drone. It is no longer about drones; it’s about software applications that capture and process the data. It's all about: Who are the customers? What do they need? Why would they use this? What's the ROI? How will they use it? That is a healthy maturity.
There's still some of the hobbyist types, or the people who are kind of fascinated by the drones themselves. That will probably always be the case. But the balance of the industry has now shifted fully towards the commercial application of the drones.
McNeil: What are some of the applications software developers are creating?
Anderson: Many of the software companies soon realized that simply collecting drone imagery was not enough. You have to deliver something that works and provides a pragmatic solution to real problems in particular industries. You have to work with the tools those industries already use. Some drone developers focus on agriculture, some focus on insurance, others focus on construction and GIS, and some developers build first responder apps.
McNeil: What would you like our readers to take away from this interview?
Anderson: For now, two things: The database created from the massive influx of information will need to have a common coordinate system; and time, the fourth dimension, must be an integral part of the database.