[modeless]

The difference is that it's useful for navigating the real world. You could have way better directions displays that show directions in context instead of just schematically. It would make the petabytes of imagery that has already been collected much more accessible and therefore useful, instead of being relegated to a special clunky Street View mode that is rarely visited. It would enable exploring real spaces in a way that provides much better spatial context, to build a spatial memory that helps your navigation when you get to the real place. And yes, it would be fun. At one time, Google was into that sort of thing.

[taeric]

I could see this as an argument for a heads up display. So, good for projecting directions onto a windshield or for having the glasses thing. But this? I don't see how a VR world helps anyone navigate the real world. That is, you seem to be saying the VR data is needed for AR usage. And I just don't see how those are helping each other too much.

I'm fully bought off on the "it would be fun" aspect. I don't see a value proposition for it, though.

[modeless]

A heads up display doesn't need a 3D rendering of the environment around it because the environment is already visible through the screen. The 3D rendering is so you can see what to expect before you get there. If you don't understand why that could possibly be useful then I don't know what to tell you; you'll have to take it for granted that some people's brains work differently than yours and can benefit from seeing places they are about to visit in 3D before they get there.

[encom]

Reading a map isn't that hard. It just sounds like an elaborate way to illustrate navigation with crayons. A cool product demo, but not very useful in practice.

[int0x29]

Google maps has two different versions of this. One of them has a step by step series of street view images and the other does a full animated fly through of every street. The second one may be web only.