The Internet of Things, and what we can learn from cars

The phrase "Internet of Things" gets thrown around a lot these days and at first it sounds like a typical bit of management-consultant-speak: a meaningless label without much depth behind it. Whatever the merits of the actual choice of words, the revolution it's referring to is actually very important indeed, and likely to change all of our lives in a significant way over the next few years.

What it ostensibly implies is that everything is connected to the Internet, so your fridge can check the price of orange juice and your thermostat can check the weather forecast. On a more fundamental level, what it actually means is that everything has software it in — so the fridge will keep count of the number of milk bottles it's holding, the thermostat will calculate the correct temperature adjustment, and so on.

And when everything has software, you have to start managing these things like software systems rather than dumb objects. Take the example of cars: they've always been finely tuned and calibrated machines, but even so everything is a lot more complicated and clever than it used to be. Volkswagens are intelligent enough to cheat the testing regime, while the Nissan GT knows when it's on a racing track as opposed to a normal road.

This extra layer of smarts can make our lives behind the wheel that little bit easier but it also means the 'user experience' is going to be harder to get right. Automobile dashboards aren't known for their strong design aesthetic, even on the most expensive cars in the world, and it's no wonder that the manufacturers have been sounding slightly panicked at the thought of Apple and Google bringing all their software and hardware skills to the market.

Software (and Internet) everywhere has security implications too, and again cars provide a prime example — consider Wired's report of a jeep that was remotely hacked by security experts and immobilised while halfway down a busy highway. Super-smart cars may well be the way of the future but they bring a whole new set of implications to consider (not least how all the data that we're going to be generating is stored and used).

It's fortunate then that the rise of software-enabled devices and vehicles has another consequence: nothing is ever "done". A car can be patched or upgraded in the same way as your smartphone; your blender can work with a new range of ingredients without you actually having to go out and buy a new blender; and your fridge can get updated as often as Gmail is.

With these intelligent devices — the "Internet of Things" — comes a need for agile, iterative design cycles to keep them safe and functional, and to ensure we can get the best out of them. It happens to be the same methodology we use at D4: a system of constant refinement and improvement until our customers get exactly the software solution they were looking for.