If you could fly, how high would you fly?

Before I answer that, let’s look at the Maps app in iOS. It currently offers two perspectives:

1. Flat overhead

This is a utilitarian, well organized, metadata-heavy view. It reveals the city grid and contains a lot of non-visual data about nearby businesses, landmarks, and locations of interest.

2. Street-level

This display, currently provided by Google Street View, is more realistic and closer to the user’s own perception and experience. It’s a great way to see what it feels like to be there.

Some attempts have been made to combine the data-richness of the first, flat perspective with the immersion of the second, deeper one. This is typically labeled augmented reality (AR) and it looks something like this, the Monocle view in the Yelp app:

Cute. But in almost every way, a poorer user interface than using either of the two views above. The small angle of view we humans are accustomed to reduces the pool of useful data by quite a lot; there’s far too much overlap on the Z axis, extending away from us; the background image doesn’t correspond to the labeled places in any meaningful way (what does Kabuki Restaurant & Deli have to do with that whiteboard marker or that chart pinned to the wall?)

Apple has purchased three mapping companies in the last few years. Their latest map-tech acquisition was the Swedish startup C3. They developed a way to generate 3D maps from satellite images (and presumably other secret sauces). Here’s a fascinating example on YouTube:

Two more examples: New York City, Hoover Dam.

A map of this sort on your iPhone would be a mighty impressive thing. But is it useful?

3D navigation is notoriously difficult for most people and most input methods. It’s usually solved by locking one of the dimensions and making good assumptions on the user’s behalf. Most 3D games glue your character to the ground plane - as would be the case in the real world - and offer occasional climbs and jumps.

Street View works this way. But even two-dimensional movement in it is slightly clumsy. You’re stuck on the track the Google’s Street View vehicle traveled; hitting those flattened-down arrows in the street to move forward is fussy; once you move, the transition is jolting. It’s understandable that this is not “real”, free 3D, and it probably shouldn’t be if users are to keep their sanity. But I think a better set of limitations is possible.

Let’s say you pull out your iPhone and move about the virtual city of Oslo as shown in C3’s video. Pan with your finger and pinch to pull in and out, rotate to, well, rotate (in aeronautical terms, yaw; pitch and roll would be locked.) This works well enough, I think. Once you get down to the street itself, it switches to an improved street-level view. That transition could certainly be made much nicer than the herpy-derpy rotation and fade we get today.

The other fascinating thing here is the vantage point that exists between these two positions: a sort of astral projection thirty feet above your head. If you could fly, this is the altitude I think you’d use to find your way around town. Any lower, and you’re as good as a pedestrian; any higher, and it becomes hard to tell what building is what. Think about it this way: a helicopter tour of the city mainly showcases its layout and major features, while a double-decker bus tour can teach you about its streets, shops, and even its residents.

Some middles are golden, some awkward. Without seeing it, I don’t know which camp my astral projection idea would fall in. But I’m definitely excited to see what Apple does with 3D cartography.