A follow-up article today by Fast Company provides a more detailed transcript of its interview with Apple Senior Vice Presidents Eddy Cue and Craig Federighi discussing Apple’s challenges surrounding the introduction and ongoing development of its Maps platform. The in-depth comments provide some additional insight into how Apple has been rising to the challenge of improving the quality and accuracy of real-time data in Maps, highlighting some interesting approaches, such as determining where new businesses are opening by tracking geolocated app usage on iPhones, and determining road closures by analyzing traffic activity from users’ iPhones.
Cue explains that Maps is an ongoing project that “just never ends” as businesses are opening and closing every day, and new streets and highways are constantly being built around the world.
Rather than resorting to “old-fashioned” methods of driving around taking photos and analyzing satellite images, Cue explains that Apple has taken the approach of using data that Apple is already collecting from iPhone usage patterns. Cue cites the example of a golf course, explaining that Apple can discover that a new golf course has been opened up simply by looking at app usage: “If we see that all these golf apps are being used at a particular location, and we don’t show that as a golf course, we probably have a problem.” Staff on Apple’s Maps team would then use that data as a trigger to start doing web searches and looking at photos to confirm what’s actually there, and if absolutely necessary, sending somebody to drive by and check it out. Similarly, Cue explains that bridge and road closures are handled in a similar way: “the truth is we don’t really need anyone to tell us that the bridge is closing.
The moment a bridge is actually closed, you can immediately see the effect” simply by iPhones ceasing to move across the bridge and beginning to move in a different direction. Federighi, however, goes on to explain that while Apple collects this data, it’s anonymized and the company “draws a distinction between that [traffic data] and personal information,” such as a user’s daily commute.
.
