Opera’s tech demos implementing device camera stream and sensor orientation in a mobile browser clearly indicate that Blair McIntyre, Alex Hill and the Augmented Environments Lab at Georgia Tech have it right with Argon: mobile sensor-based AR is moving towards open web APIs for content definition, styling and placement. If Opera have any say, soon HTML5 is going to mandate everything necessary to do sensor based AR with a couple of lines of javascript, and then any old web-dev can roll an AR view.
Mobile web browsers aren’t necessarily the most logical or best mobile AR platforms, but being able to initiate an AR mode from a normal webpage without downloading an app is huge. (Somebody needs to come up with an “AR mode” icon.) Having each site be an entry point to its AR content is a readily understood variation on the subscribe-to-layers solution. If you want to see what Yelp says about this restaurant fire up yelp.com. If you want to book it for next wednesday start from opentable.com instead.
However, I feel the real gold here is to be had if someone can figure out search. Providing an AR “search this view” would allow search providers to determine, purely from zero-click camera-view queries not only which businesses or landmarks are people most interested in (in aggregate), but which of the results they return for a spatial search are most selected and by whom. This is the fodder of “spatial page rank”.
The really experiential stuff is still going to require a stand-alone app, as will computer-vision based solutions, but bringing AR to mobile web-browsers is the best way to take AR mainstream. Once it’s there, people won’t call it AR anymore, it’ll still be “the web”, just the web you can see right in front of you.
-
augre posted this
