You'd think it would be tough for a 1,200-foot long oil tanker to get lost in the shuffle, but it happens, especially, say, in the blinding bustle at the Port of Singapore, where half of the world's crude is shipped every year.
Keeping tabs on the tankers is big money for hedge funds who try to keep up using the ships' movements to predict where the price of oil is going. Kenneth Singleton, a Stanford economics professor who studies the oil market, tells of firms flying helicopters overhead to gauge inventories. Some funds, he notes, have gone as far as hiring full-time port observers on the ground to radio back intelligence, an attempt to counteract the nautical practice of ships turning off their tracking sensors to intentionally obscure their location.
"But what if investment firms had an eye in the sky?" asks John Fenwick, the co-founder of Skybox Imaging. A former Air Force officer, Fenwick is engaging in one of his company's favorite pastimes: Dreaming up use-cases for its technology -- low-cost satellites that could make images of any location on Earth accessible to the mass market.
Skybox is launching its first satellite into space this September on a Russian rocket, and hedge funds with an interest in monitoring major ports could be among its first customers. Oil futures may never be the same.
Ching-Yu Hu, another Skybox co-founder, prefers to ponder the potential impact on humanitarian aid and emergency response. If the company realizes its goal of sending 24 satellites into orbit by 2018, first responders could see hourly snapshots of disaster-stricken areas that could help identify which roads have collapsed after an earthquake or how quickly a forest fire is spreading.
Camera-wielding drones take too long to dispatch, Google Earth's visuals can be several years out of date, and existing satellite imaging companies can only afford a few eyes in the sky because conventional imaging satellites cost hundreds of millions of dollars. This means they can't capture images frequently enough to monitor rapid changes on Earth, like the position of ships or the unfolding impact of natural disasters.
But the dream applications that keep 27-year-old Julian Mann, Skybox's youngest co-founder, up at night may be even more ambitious.
Imagine the ability to make sweeping, macro-economic predictions based on crop growth across the globe. Or gleaning a country's consumption from images showing the number of cars in every one of its Walmart parking lots. How about all that information being fed through even more macro algorithms, combined together, and used to create an economically predictive power that could yield fortunes and prevent collapses?
These bits of information are "hardly trivial," David Samuels wrote in Wired in a profile of the company in June. "They are digital gold dust, containing clues about the economic health of countries, industries, and individual businesses."
To extract any of these things from pictures is no small task, though, and that's where Mann thinks the company's true mission lies. He says Skybox aims to create a platform that can turn its images -- and eventually those from other satellite companies -- into highly valuable data that's easy to consume. Call it "the Facebook of satellite imagery."
With $91 million in venture-capital funding and staff that counts former standouts from NASA and the Pentagon in its executive ranks, Skybox may be well on its way.
The company's co-founders met while they were grad students at Stanford. Fenwick arrived at the school for his MBA in 2008 after working as a liaison in Congress for the National Reconnaissance Office, the US agency that builds and maintains America's intelligence satellites.
One day after class, he wandered across the street to the engineering school and asked a professor for the smartest kid there. He was immediately introduced to Mann and Dan Berkenstock, another space cadet getting his PhD in aeronautics.
Turns out they had a lot in common. Fenwick had built and launched airborne imaging cameras for the Air Force. Mann and Berkenstock had done work with NASA and built similar cameras for Stanford's Space Systems Development Lab.
The lab's satellites -- called CubeSats -- were small and cheap to build, tinkered together with off-the-shelf electronics. And the images they beamed back to Earth were attracting the attention of NASA. "We realized that we were competing with Boeing and Raytheon for the same government contracts," Mann says.
But Mann and his partners had their sights set on buyers outside the beltway. Google founders Larry Page and Sergey Brin had just announced their sponsorship of the Google Lunar X Prize, promising a bounty of $30 million to the first privately-funded team to land a rover on the moon and send back pictures. "It got us thinking," Mann recalls, "that there were other sources of revenue out there beyond the government." *
A Skybox-like business model is a relatively new idea for the commercial space-based imaging industry, now pegged at roughly $2.3 billion a year. The tiny number of companies that currently sell pictures from space rely on government contracts for around 80 percent of their revenue, according to Mann.
So when Uncle Sam comes calling, they respond as best they can, with high-res imagery that sits in huge files on a server, requires $20,000 worth of software to open and can take weeks or even months to deliver.
Plus, conventional imaging satellites are designed for government work. They're the size of an SUV, weigh two tons, and typically cost around $500 million each.
Skybox says that's overkill for the business world. Its satellites cost under $50 million a piece, thanks in large part to the use of less expensive cameras. The trick, Mann says, is the company's software, which stitches together strings of lower res images into one visualization.
At roughly the dimensions of a mini-fridge, Skybox's 200-pound satellites are also far smaller than conventional models, another cost-saver enabled by an intense company focus on software and on-the-ground computing.
Even the launch mechanics are on a budget: Skybox is shooting their first satellite into orbit with rented space aboard a Russian DNEPR rocket. Hitching a ride with the Russians is apparently vastly cheaper than American alternatives, though it does mean navigating the murky bureaucratic waters of Russian politics (the launch was originally scheduled for last September).
All these savings make it conceivable that, after a few more venture-capital infusions, Skybox could afford to have a lot more satellites in the sky than existing players, allowing them to capture imagery really fast, multiple times a day. Customers could request images and data online. And once the photos are captured, and the data crunched, it could land right in their inboxes, all in a matter of hours.
But before getting its entire fleet star-side, Skybox must demonstrate that its cost model can hold up. In outer space, even the slightest hiccup can push operations into the red. Competitors warn of the vulnerability of low-cost satellites. "From what I've heard, they can stop working abruptly," says William Pora, a salesman at the Satellite Imaging Corporation, a Houston-based reseller of satellite photos.
Cloud cover is also an issue for any satellite imaging outfit. "Companies will often charge clients for each attempt at capturing an image," Pora notes, "so they're still making money off a photo even if a cloud gets in the way." Skybox, for its part, plans to charge for data derived from such photos, suggesting that the cost of failed attempts would be something it eats. "That could break their cost model," Pora adds.
Even if they can prove competitors wrong, Skybox will likely have to deal with a number of domestic obstacles. For all of the company's brainpower and optimism about the future proliferation of photos from space, the concerns are obvious, and not only to those already shielding themselves with tinfoil hats.
There's danger from the potential political ramifications of widely available global data. To keep with the tanker example, China accounts for 9 percent of the world's oil use, but releases no actual data about that consumption. "We're patriotic," Mann insists. "There are certain countries we will never sell to."
But the very existence of such photos might be enough to fan paranoia on all sides.
There's corporate espionage: the idea that rival companies might use these images to get the upper hand on one another, tracking their shipping methods, construction, and the like. Then there's the matter of personal privacy. Sure, Skybox can't zoom close enough to see you, but they can sure zoom in close enough to see where your car's parked.
For its part, Skybox acknowledges its data could be put to unsettling use, but seems to approach the trade-off with the sort of open-source, hacker mentality common in many of today's tech companies. "If you look at companies that truly transformed industries," Mann says, "they're the ones that facilitated a new level of access."
The dissemination of data and increasing global transparency is a net positive, Mann argues. "If there's not a capacity to exploit something for evil, it's probably not that revolutionary."
* Correction: An earlier version of this story wrongly said that NASA was a sponsor of the Google Lunar X Prize and terminated its sponsorship. NASA never sponsored the Google Lunar X Prize, according to the agency and individuals involved with fundraising for the prize.
More From The Atlantic