The bit economy is fueled by sensors, converting atoms into bits. The data economy is fueled by virtual sensors. A sensor may be passive, you need to look at its value to get it, or it may be active, it can transmit the new value as soon as it detects a change.
Same goes for a Virtual Sensor. It can be the result of a computation that needs to be inspected or such a result can be transmitted to potential users.
The change in data value as well, and often crucial, the correlation among streams of data generates meta-data that can be analysed and in turns generate information.
As an example, the location data obtained by a cell phone analysed over a period of time provides information on the displacement of the phone, hence its moving speed. A phone that moves at speed over 10km per hour is clear worn by a person that is not walking but using a vehicle. The analyses of the type of movement can tell if the vehicle is a bike or a motorised vehicle (higher speed), if it is a public transportation means (the movement halts in certain places for a short time then it resumes…). The analsyses of location data from several phones provide information on the overall traffic in an area, whether it is fluid or if it is slowing down/there is a traffic jam.
We can extend these consideration to the analyses of a complex environment, like a city.
Here, the variety of sensors deployed by many parties are creating a virtual image of a city and of its constituents: its cars, its infrastructures, its people. You can learn quite a bit about people living in a city just by tracing the movement of their cell phones, something that is very easy to do.
We have safety cameras that can be used to observe the way people move about, how much time they spend in front of a window, how that changes as merchandise changes and so on. Taxi cabs can be equipped with pollution detector sensors providing hundreds of thousands of samples per day from every part of a city …
The data captured by sensors are the raw material used by third parties to extract meaning, and value. This can be encapsulated into new data that are, to all effects, the output of virtual sensors.
Applications can be created using these data fueling new biz.
What is important in the creation of value is to move from data to information, to meaning, and more specifically to a meaning that is valuable to a target market.
The analyses of data, in particular, can answer four types of questions of increasing value:
- What happened? As an example the analyses of the location data from cellphones related to many vehicles in an area can point to a traffic jam
- Why did it happen? Following on the example, relating the information that there is a traffic jam with the information that there is a soccer match in the vicinity and that that match finished fifteen minutes ago points to an increase of traffic flowing out of the stadium that is generating the traffic jam.
- What will happen? Knowing the amount of people at the stadium and the time it takes to them to exit the premises and get their car, plus the expected traffic converging at that time of the day from other areas in the vicinity one could foresee a worsening of the traffic jam in the following hour.
- What can we do to avoid the problem? By decreasing the flow of cars from the stadium parking, and by rerouting potential incoming traffic by sending messages to follow alternative routes one can decrease the traffic load and resolve the traffic jam.
These are clearly examples, and many more can be found. The point is that by analysing data we can create value.