Saturday, November 6, 2010

The Internet of Missing-Some-Things, Part 1

The Internet of Things buzz seems to be all over the place. It is certainly on the cusp of becoming a big market, and there's no doubt in my mind it will be huge. But before it can really rocket to the #1 technology revolution that it promises it can be, someone needs to solve a couple of key issues. Until then, it will still be the "Internet of Missing-Some-Things."

The IoT revolution was technically possible decades ago. When you can buy a device that can send data to the Internet, the technical possibilities expand greatly. But it wasn't economically or practically possible. It was too expensive to buy lots of devices like that. Even if you could afford the devices, the bandwidth was expensive and difficult to find, and you needed a pretty hefty power source to power it. So applications of devices that connected to the Internet were typically (1) relatively expensive, like PC's and web servers, and (2) limited by proximity to high bandwidth, like...well...PC's and web servers installed on a wired network, and (3) limited by high power demands, guessed it.

The three big drivers of IoT as it stands today are the huge downward cost pressures on "smart" device hardware, huge downward cost pressures on the cost of connectivity, and a relatively large increase in availability of network connectivity. Now that you can buy a cheap device that is smart enough to talk on the Internet, and the Internet is available in lots more places, IoT can become a reality.

But the real IoT revolution is still out of arm's reach. There's still two missing pieces of the puzzle: True ubiquitous connectivity, and ubiquitous electrical power for small devices.

As a server-side software developer who lives in a metropolitan area, it seems like there is power and connectivity everywhere. But talk to anyone who is building an embedded device, and you'll see that power and network connectivity are still relatively difficult to come by. Rural areas are an obvious example here. Farmers are still relatively limited in the monitoring devices they can use, simply because cell networks and power lines often don't cover their fields.

But even in places where you'd think it'd be easy to plop down a device, it isn't. It is still a relatively difficult conversation to convince a building owner to run power and Ethernet to the bowls of the elevator shaft, or into every corner of the warehouse.

Until power and connectivity is ubiquitous, the IoT revolution will always be somewhat limited. Fortunately, people are already working on those problems.

1 comment: