Monday, July 9, 2007

The Future Evolutions of Google Earth

Having been bored and with nothing better to do, I settled in for the evening and poked around the planet a bit with Google Earth. I checked out Baqubah, then flew over to Machu Pichu, where I turned on terrain and followed the trail up to the ruins. Then off to Vancouver, a tiny nest of concrete and steel packed into the rocky mountains; from there, I followed the border to Toronto. After that, back to Tokyo, where I explored the urban warrens of Shinjuku and Shibuya. That's more or less a tour of where I'm thinking about living, where I used to live, and where I live now, respectively.

One thing pops out when you compare Japanese cities to Western cities. Turn on 3D buildings and you'll see what I mean. Tokyo and Osaka are exhaustively, thoroughly, obsessively mapped. Vancouver, Toronto, and London (the three Western cities I chose for comparison) are almost entirely not, the sole exceptions being for the most part skyscrapers. Those skyscrapers are largely rendered (not photographically, but some effort has been made to create cartoonish models of the buildings as they actually appear.) Japanese buildings, however, are almost entirely blank, giving only an indication of proportion.

I'm not sure what this means. I suspect it is possible that the Japanese government might have invested in having the blueprints of every building in the country fed into a database and then layered the information in google maps as a survey tool. Western governments have not, I suspected, bothered to trouble themselves with such activities; as a result, models will generally be made by citizens (who will tend to concentrate on the larger, flashier structures, for obvious reasons) or corporations wanting to show off their HQ to the world (ditto.)

It's easy to see a lot of ways that Google Earth, or successor software, might evolve to take advantage of greater information and greater processing power. If an automated way of interpolating photographic data onto unrendered 3D structures could be found - assisted by photos with timestamps and GPS coordinates - a simple trawl through the photosharing networks would gather enough data to render every building in every part of the world in very high detail, full-color and fully textured. Refined a little more, some feedback from the photos on the actual shapes of the buildings would allow models of even higher accuracy to be built. It would be natural from there that the interface be extended to easily enable a walking pace through a city, so that areas might be explored in more detail. The same photo-mapping algorithms would enable rooms to be easily added, a feature for which businesses and citizens could likely find a million uses. Natural features would be amenable to the same process: zoom in far enough on a forest, and eventually you're traveling underneath the canopy, marveling at the trees; find the mouth of a cave system, and you can go spelunking; take a dip off the beach and swim through a reef, or in the deep ocean and plumb the depths of a continental trench.

That's what you could do with good photo-mapping. Video-mapping is the next logical step. Cameras are proliferating everywhere. London is perhaps the most extreme, with it's Ring of Steel, built of concrete and surveillance cameras (no steel, somehow not so surprising in this metaphorical age we live in.) England as a whole has millions of them. Follow David Brin's advice, connect all those cameras to the internet in real time, and that network becomes less the scary tool of a postmodern police state but a true public resource (the police state isn't necessary anyway: citizenry will do all the security work, millions of eyes being so much better at catching criminals and terrorists than are thousands that it's pretty much a law of nature.) Now feed all that video - stationary, wearable, portable, remote or autonomous - into google earth, along of course with any other open web-connected cameras on the planet, and voila: the earth rendered in real time, complete with all moving parts including people. The world would, for humans at least, take on complete transparency.

Privacy enthusiasts are probably squirming at this point, but lets face it: privacy has basically been an illusion ever since Google. Though I imagine things will remain similar to the present in at least one respect. Whatever company provides the software would almost certainly provide property-owners the ability to shut their dwellings out of the system, if they so chose, thus keeping their domestic lives as private as they've ever been. In public, no such thing as privacy would exist. Anyone who wishes, anywhere in the world, at any time from now to the end of history, can see what you're doing. They could follow your entire public life if they so chose (though most people would likely just fastforward to the interesting bits.) If privacy is possible in such a world, it is only in the sense it is now: through obscurity.

Google Earth, at this point, has morphed into something less like the ultimate cartographic tool and more into a kind hyper-geographical earth: a linked, tagged, fully searchable skin of information that clings to reality like a shadow. It doesn't have to reflect reality precisely of course; in addition to privacy holes, people might throw up projected buildings as architectural proposals, or show cities as they were in the past (it would be interesting to rewind, say, Beijing or London for a few thousand years and watch them grow in fast-forward.) In addition to such straightforward fantasies, people might well populate the hyper-geography with avatars, interacting with other avatars or real people all over the world. I'd imagine that people would generally prefer to see whatever avatars are about; hiding one's avatar from real people is the kind of thing I could see becoming a gross breech of etiquette, or even being made a criminal offense. It's nice to be able to see how many people are about, and it would be a handy way to tell, for example, if someone is following you (indeed, privacy concerns might largely be allayed if people were able to see a list of exactly who was performing searches on them.) No one would have to turn on the avatar layer if they didn't feel like it, but if they did - especially in any area of particularly high interest (commercial, cultural, political, scientific, or military) - they'd witness a ghostly blur of people, teleporting in and out and around too fast for the eye to get a good fix on any but the few who aren't just popping in for a look-see, split-second sightseers on a lunch-break world tour.

I should note I started assuming the proliferation of contact-lens displays and other wearable devices about halfway through that last paragraph. They'd render this hypergeographical earth something that wasn't just there when you're sitting in front of your computer, but essentially a permanent feature of the human sensory experience. Not a bad culmination for a little program that currently maps out an uneven, high-ish resolution version of the planet.

Sphere: Related Content

No comments: