'What is SP?': A Reply
James Gerard Wolff · The Computer Journal · 1995
My interest in information compression as an explanatory principle in information processing was sparked originally by lectures given by Horace Barlow in Cambridge University describing phenomena in the workings of brains and nervous systems which could be understood in terms of economical coding of information (see, for example, [1] and [2]).Later, an accumulation of evidence led me to the view that many aspects of language learning by children could be understood in these terms [3].About 1987, I began to realize that several aspects of 'mainstream' computing might also be understood as information compression.These insights suggested the possibility of developing a new kind of computing system with more flexibility and 'intelligence' than conventional computers.An article describing this thinking was published in this journal.Subsequent research has attempted to resolve the many questions which naturally arise from the 'SP' conjecture that all kinds of computing and formal reasoning may usefully be understood as information compression by pattern matching, unification and metrics-guided search.Recent publications in this programme of research include [4], which describes the background and motivation for the research, [5] which describes how the design of software and its execution may be seen as information compression, [6] which describes an application of these ideas in best-match retrieval of information, [7] which is an overview of the research at mid-1994 and [8] which describes a recent model of the proposed new system.The latest thinking is described in [9], [10] and [11].