11.3 Time for a New Approach
231
The emergence of the “Internet of Things” means that we will soon be able to
measure almost anything in real time, using networks of sensors that can communicate with each other in a wireless way. Interestingly, to support self-organization,
it is not necessary to store the measured data for long. Therefore, the collection of
as much data as possible, which is at the core of today’s Big-Data paradigm, can be
replaced by a superior Smart Data approach, where tailored measurements are made
to produce temporary data for specific uses. Such real-time measurements would be
sufficient to provide the information needed for the self-organized structures, properties and functions which we may want to produce. Moreover, as I have underlined
before, it is anyway impossible to process all of the data currently available. Storing
more data does not necessarily mean better results—it’s just more expensive. So,
why should we retain more data in the first place?
Thus, to fully unleash the power of information, we will need to go beyond
the brute-force machine learning approach of contemporary Big Data analytics. We
must learn to combine knowledge from computer science, complexity science and
the social sciences to get the measurements and interactions right. So far, we have
rarely combined the knowledge and skills from these disparate disciplines effectively.
Silicon Valley is probably too technology-driven, while the social sciences tend to
underutilize technology. Both don’t pay enough attention to complexity science, but
it’s value for solving real-world problems will soon be obvious. The golden age of
complexity science is near.
11.4 How to Make the “Invisible Hand” Work
Interactions between the components of a complex dynamical system produce “externalities”, i.e. external effects such as reputation, happiness, or wealth, emissions,
waste, or noise, or other consequences that affect the environment or others in a
positive or negative way. These externalities can be altered by introducing or modifying feedback loops in the system, for example, by introducing value exchange.
Such feedbacks allow the system components to adapt to the local conditions in
ways that produce or restore the desired functionality. In economic systems, feedback mechanisms are often produced by financial costs or rewards, while in social
systems it is common to use incentives or sanctions. However, certain kinds of
information exchange and coordination mechanisms can be even more efficient
(“altruistic signaling”, for instance). It is also important to consider that the use
of a single feedback mechanism (such as money) is usually too restricted to let a
complex socio-economic system self-organize successfully, and therefore we need a
multi-dimensional incentive and value exchange system, as I have proposed before.
To allow for real-time measurements of our world, my collaborators and I have
started to work on a distributed Digital Nervous System as a participatory citizen web.
With this enabling technology, called Nervousnet, one could measure externalities
231
The emergence of the “Internet of Things” means that we will soon be able to
measure almost anything in real time, using networks of sensors that can communicate with each other in a wireless way. Interestingly, to support self-organization,
it is not necessary to store the measured data for long. Therefore, the collection of
as much data as possible, which is at the core of today’s Big-Data paradigm, can be
replaced by a superior Smart Data approach, where tailored measurements are made
to produce temporary data for specific uses. Such real-time measurements would be
sufficient to provide the information needed for the self-organized structures, properties and functions which we may want to produce. Moreover, as I have underlined
before, it is anyway impossible to process all of the data currently available. Storing
more data does not necessarily mean better results—it’s just more expensive. So,
why should we retain more data in the first place?
Thus, to fully unleash the power of information, we will need to go beyond
the brute-force machine learning approach of contemporary Big Data analytics. We
must learn to combine knowledge from computer science, complexity science and
the social sciences to get the measurements and interactions right. So far, we have
rarely combined the knowledge and skills from these disparate disciplines effectively.
Silicon Valley is probably too technology-driven, while the social sciences tend to
underutilize technology. Both don’t pay enough attention to complexity science, but
it’s value for solving real-world problems will soon be obvious. The golden age of
complexity science is near.
11.4 How to Make the “Invisible Hand” Work
Interactions between the components of a complex dynamical system produce “externalities”, i.e. external effects such as reputation, happiness, or wealth, emissions,
waste, or noise, or other consequences that affect the environment or others in a
positive or negative way. These externalities can be altered by introducing or modifying feedback loops in the system, for example, by introducing value exchange.
Such feedbacks allow the system components to adapt to the local conditions in
ways that produce or restore the desired functionality. In economic systems, feedback mechanisms are often produced by financial costs or rewards, while in social
systems it is common to use incentives or sanctions. However, certain kinds of
information exchange and coordination mechanisms can be even more efficient
(“altruistic signaling”, for instance). It is also important to consider that the use
of a single feedback mechanism (such as money) is usually too restricted to let a
complex socio-economic system self-organize successfully, and therefore we need a
multi-dimensional incentive and value exchange system, as I have proposed before.
To allow for real-time measurements of our world, my collaborators and I have
started to work on a distributed Digital Nervous System as a participatory citizen web.
With this enabling technology, called Nervousnet, one could measure externalities