When asked what he had learned about the changing geospatial analytical landscape, Patrick Biltgen, technical director for analytics at Vencore, challenged the question. It lacks context, he said.
“We talk a lot about intelligence analysis,” Biltgen said Sunday during a panel at GEOINT Foreword in which participants in the pre-conference science and technology agenda reflected upon the day’s events. “The word analysis means to break apart. … I’m going to gather data about all the parts, and I’m going to look at all of those parts.”
What anyone who needs intelligence analysis really wants is synthesis, Biltgen said: “Synthesis … literally means to compose.” They want pieces of the puzzle assembled into a picture they can use to generate a forecast.
Is that the job of an analyst? It can be.
“We shouldn’t be afraid of [automation] as analysts,” Biltgen said of earlier references to emerging technology. “It’s like, ‘the machines are coming for your jobs.’ So? We’re going to be intelligence ‘synthesists’ instead of intelligence analysts.”
Sue Kalweit, the National Geospatial-Intelligence Agency’s director of analysis, agreed.
“What we’re hearing is that automation is going to take our jobs away,” she said. “For years, our analysts have wanted nothing but time back—the time to do more predictive analysis, the time to do research, the time to really delve into a problem. With the outset of more and more imagery, they’ve lost that time. Now, with the power of automation, we’re able to give them back time to do that [predictive] analysis.”
Also on the panel were Adam Maher, co-founder and president of Ursa Space Systems, Dr. Cordula Robinson, associate teaching professor at Northeastern University, and Dr. Amanda Ziemann, an Agnew National Security Postdoctoral Fellow with Los Alamos National Laboratory.
They offered candor in responses to questions about the changing face of analysis. For example, asked about the temporal element of geospatial analysis, particularly as applied to activity-based intelligence, Maher pointed to time as an enabler of forecasting. “The question you’re trying to answer is what activity is happening,” he said. “But what the boss really wants to know is what is going to happen.”
Biltgen spoke of young analysts trending toward the immediate. Real-time data doesn’t awe them as it sometimes does their elders.
“I work with a lot of our interns and our new hires and … real time is normal for them,” Biltgen said. “If you say to them, ‘I have some data from yesterday, a paper device,’ it’s like ‘yawn.’”
He continued, “The workforce that is entering has tremendous mastery of real-time data. They expect everything now. … Motion is normal. Time is normal. Stuff happening right now is normal. I think as quickly as possible, we in the geospatial industry want to get those things into our workflows.”
Although panelists represented government, industry, and academia, they all shared a common thread.
“You have to listen to your customers,” said Maher, who added that communicating results to those customers could be difficult.
That’s particularly true when customers are on the tactical edge.
“We live in two different worlds,” Biltgen said. “There’s the suit-wearing, highly networked, work in a big building … and then there’s all my stuff is full of dust. This is a constant challenge, not just from a data but from a tool standpoint. … There are parts of the world where you can take a flash drive and tie it to a carrier pigeon and get it to the front faster than you can send it over the local coms.”
Understanding these nuances is part of the changing analytical landscape as well.