INFORMATION COMPRESSION
Charles François (2004). INFORMATION COMPRESSION, International Encyclopedia of Systems and Cybernetics, 2(1): 1657.
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(1) |
| ID | ◀ 1657 ▶ |
| Object type | Epistemology, ontology or semantics, Methodology or model |
J. S. WICKEN writes: “Basic to the definition of order is informational compressibility” (1979, p.139).
Unless we find a way to condensate the information content of a system, the only model of it is itself, which means that our rational understanding of it is nil.
WICKEN explains: “Two modes of information compression are possible. One is algorithmic. This applies to cases in which patterned relationships appear in a sequence that allows element-by-element specification to be replaced by a program of instructions for its assembly” (WICKEN adopts thus G. CHAITIN's concept of algorithmic compression of information, also known as KOLMOGOROV-CHAITIN complexity) (1975, p.47-52).
- “For example, the sequence (1,2,3,4) possesses a higher degree of order than does the sequence (1,3,2,4) because a briefer algorithm is required to generate it.
- “The second is probabilistic, and the compression is measured by the relative values of real complexity and maximum conceivable complexity. This mode of compression applies to stochastic processes in which (a) the symbols are not equiprobable and/or (b) the occurrence of one symbol in a sequence influences the probability of occurrence of other symbols” (p.139).