The modern world of today consists, to a great extent, of an artificial milieu created by human beings and structured into large technological systems. The global telecommunications network belongs to one of the most complex of these systems, even if one only considers the physical network with its telephones, exchanges, optical fibre cables, satellites, etc. If, as well, other parts of the system, such as development laboratories, institutes of technology and general technological knowledge are included, this system takes on gigantic dimensions. Despite its magnitude, it holds together and forms a working entity with an impressive level of reliability in service. This reliability in turn, may to a great extent be ascribed to ‘self-healing qualities’ with online, automatic fault isolation and call rerouting.
One of the most interesting properties of electronic networks (apart from the fact that they work so well!) is that they are constantly undergoing change. They are modified by different phenomena, such as technical development, investment, changed legislation, merging of carriers, implementation of new services and connection of new users. The system thus develops according to its own logic, a logic which is itself subject to change and thus changes the whole system yet again. Like many other big technological systems, the electronic network is therefore characterized by an interplay or combination of change and stability. Significant for all communication networks is that as their value increases, the more users it has. This is called Metcalfe’s law, after the Internet pioneer Bob Metcalfe. The law tells us that the value of the network for the individual user increases with the square of the number of users.
Another interesting feature of the modem electronic network is that it neutralizes the age-old debate regarding centralization versus decentralization. All the capacity of extreme centralization and extreme decentralization can exist simultaneously in the same network. With regards to this quality, the modern electronic network parallels the human brain, which must be considered both centralized and decentralized at the same time.
Networks consist of links connecting the numerous nodes. Links are of various kinds, from simple copper wires to transmitting and receiving satellites. The nodes are exchanges, today mainly computerized and often both processing and storing the flow between the links. Distinctions can be made between the following networks:
- Information networks
- Communication networks
- Computer networks
- Relational networks
- Hybrid networks (of 1, 2, 3, 4)
Information networks are characterized by nodes which are data- bases, while communication networks connect information processors (human beings or computers). The communication system enjoys a development which is relatively independent of the information system that it supports.
Two basic styles of network exist: the hierarchical in which one node or exchange is designated to manage the other, and the peer-to-peer, where every node manages itself. In the second case, all devices are of an equal status and there is no hierarchy for communication. With the combined intelligence of all nodes in the network, a peer-to-peer configuration is able to cooperate and automatically to determine the best routes and also reroute around problems. This capability is the result of a continuous sensing, feedback, feedforward and adjustment among all subsystems communicating within the network.
Each kind of network has its own morphology and evolutionary pattern with a different appearance of nodes and links. Complexity, reliability, vulnerability and dependability vary according to the type of network. Evolutionary growth, rather than planned development, is a characteristic property of nearly all networks and has been studied by several communication researchers, among others, Samuelson (1977). The growth is subject to the needs of its users and is largely independent of existing technology.
The various stages of this growth can be considered a life-cycle with the steps shown in Figure 10.2. The stages are:
1st stage Differentiated loci exist randomly distributed within all dimensions of a medium
2nd stage Formation of interconnections by ‘Brownian’ motion
3rd stage Choices are being made; the dynamic is no longer Brownian
Figure 10.2 Life-cycle of a telecommunication network.
(Reprinted with permission from K. Samuelson et al., Information Systems and Netivorks, North- HoIIand, 1977.)
4th stage Limiting the alternatives
5th stage Limiting the number of connections and the emergence of decision bias
6th stage Increasing structure and differentiation of structure System
7th stage fragmentation prior to renewal
Figure 10.3 Comparison between information and communication network growth.
(Reprinted with permission from K. Samuelson et al., Information Systems and Networks, North- Holland, 1977.)
Figure 10.3 shows evolutionary growth and orderly expansion in an information network as compared with that in a communication network. Note that there are only six-stages, as the initial networks already exists.
No matter how the architecture is designed, all electronic networks function according to one of the following methods:
- fixed connection point to point via a subscribed link;
- circuit switching physical linking on demand (e.g. telephone line);
- packet switching logical connection is established via the nodes of commercial carriers which relays the communication content, consisting of small electronic message modules or packages.
The advantages of networks may be seen in many applications. Some of those that exist on a world-wide basis are:
- Personal communication (telephone, fax)
- Information retrieval (knowledge databases)
- Meteorological information (weather forecasts)
- Geological information (earthquakes, )
- Electronic mail (message storage and forwarding)
- Sale of tickets (air-carrier tickets, )
- Electronic banking (all kinds of commissions)
- Tele-medicine (exchange of patient data, )
- Tele-conferencing
- Police information (INTERPOL)
- Financial information (REUTERS)
- Customs information
Source: Skyttner Lars (2006), General Systems Theory: Problems, Perspectives, Practice, Wspc, 2nd Edition.