dessalles

Google Earth 4.3 was released yesterday which, with the introduction of photo-realistic 3D buildings and a sunlight simulator, inches Google closer to being the proprietor of impressively realistic models of every major city on the planet. Above, a view of the San Francisco skyline just before sunset. Below, the skyline from the soon to be released Grand Theft Auto IV. I wonder how much longer until Google manages to catch up with the realism of today’s videogames? Right now, they seem to be at about Flight Simulator 2004…so, 4 years behind.

GTA IV has modeled its game world of Liberty City after New York City (replete with a “Statue of Happiness”). And what I’m eager to learn is how the game experience will differ for local New Yorkers who’ll be playing with their own existing mental maps of the city. Will we find ourselves with a distinct advantage over our online, multi-player opponents, unfamiliar with the broken grid of lower Manhattan? Will they know to take the West Side Highway of 10th Ave?

But what excites me about this more realistic Google Earth isn’t transforming it into a game as much as it now seems appropriate to plug some real-time data feeds into these urban models. With a suitably recognizable topology and it’s new understanding of time/sunlight, what the next version of Google Earth needs isn’t three-dimensional trees.

Dan Hill’s excellent survey of Tansport Informatics reminds us that Helsinki’s public transportation system is already using Google Maps to display the location of its buses & trams in real-time; so that might be a good place to start. And of course, there’s also Cabspotting which tracks San Francisco Yellow Cabs. Air traffic is probably doable as well.

And then, as I hinted in a previous post, CCTV systems and webcams feeding into Microsoft’s Photosynth could populate a city model’s sidewalks with pedestrians (among other moving things). But I suppose, we could just as easily map cellphones if we want to insert people into Google Earth.

It’d also be nice to see Google grab data from the electrical grid and get some lights turning on in these buildings at night.

There’s something new and different about watching representations of real, physical objects move across a city map, and I wonder how that experience would differ when presented in 3D. Obviously, we loose the ability to oversee everything at once; but that’s what the maps are for. What we do gain is the ability to experience events from the perspective of an individual. We gain partial information. And if the idea is to design a digital, urban infrastructure for individuals to efficiently act on, that perspective is probably a fairly valuable design resource. Once we can digitally experience the complexity of choosing between a cab, the oncoming bus or the nearest subway station, it’ll be much easier to design tools and interfaces that expose the right amount/kind of information to those in need of it.