May 14th, 2008In Web3.0, the mobile web browses you
Mobile Computing may force us to rethink our operating definition of a web browser.
For a while folks have been wrestling with the debate of browsers vs native apps for mobile. Browsers are wonderful because they make few assumptions about the underlying hardware (aside from guessing at screen resolution and which browsers a handset may support).
The trouble with browsers is that they miss out on all of the rich hardware and other physical features a mobile device is capable of. What we need are common application layers/frameworks (basically an uber-browser) that can do both.
You shouldn’t blame the browsers of today, they come from a desktop world. What’s worse they come from an 80s/90s desktop world that assumes every terminal is just an immovable terminal of static screen, with few likely peripherals other than a mouse and a keyboard. Even on the desktop world we rely on extra software layers above the browser like flash and silverlight to do anything fun but hardware-dependent like capturing or streaming video and audio. To your usual xHTML standard, the hardware layer, the physical environment is completely unimportant.
The trouble with mobile is that is shatters this assumption. Almost *everything* that is uniquely interesting about a mobile platform, by default, outside the traditional document model of the browser.
The interesting thing about mobiles is that they exist and operate out in the real world. Meanwhile every year, these same devices are becoming increasingly aware and able to interact with this real world. Unlike desktops, mobiles move around, they know where they are, they’re stuffed full of sensors for orientation, movement, touch, they are capable of vision, of hearing, they can (in theory) directly sense each other and communicate through a set of radios with ranges from a few centimeters to a few kilometers.
So how will mobile web-apps avail themselves of these features? How do we build a stack that cleanly and easily interfaces physical presence with virtual. What’s needed is a consistent cross platform set of tools that enables a run-almost-anywhere webap to connect on end to the cloud with AJAX and on the other hand just as easily to the hardware features of it’s platform. Existing apps like google’s mobile maps, safari’s gesture recognition, or NFC contactless applications just feel like early signals of what should be possible.
It used to be the web browser was thought of as your machine’s exciting portal into the virtual world of cyberspace. Today’s more interesting challenge is: how to give cyberspace a portal back into the real world surrounding you and your mobile machine. Anything less is not really mobile computing at all.
There is an underlying big idea here. is that our devices should be / could be / will be, the billion mobile roofing nails that connect and anchor the virtual world to the real world. That seems like a hell of a concept. Who is out there working on it?






