Scientific information: Problems and prospects

Abstract

ing or indexing services, 64 per cent.; from compendia, 4-3 per cent.; from casual conversation, 22*6 per cent.; by asking colleagues, 84 per cent.; other methods, 11-5 per cent. Closely similar results have emerged from other inquiries into the problem. They can be roughly summarised as follows: current scanning of journals, 27 per cent.; citations in other literature, 19 per cent.; oral recommendation, 21 per cent.; previous knowledge (e.g. in own index), 8 per cent.; index, abstract, bibliography and catalogues, 25 per cent. Discussion with colleagues takes up at least as much time as consultation of formal information. Some of this discussion would occur in any case. It arises out of the need for intellectual conviviality; from the sheer intrinsic value of speaking about vitally interesting subjects; some of it is however necessitated by the difficulty which all scientists now face in mastering the literature of their subjects.5 The institutionally organised communication system of science is slow. Eighteen months may elapse between the time a research reaches the reporting stage and its publication in a journal. It may be up to a year before the article is recorded in an abstracting serial, and yet another nine months before it is mentioned in an annual review. Because of these delays, the institutional system is bypassed in some cases. In fast-moving fields of basic science, scientists learn of new work at colloquia and conferences; they circulate manuscripts, duplicated reports and preprints to colleagues. This informal system of communication harks back to the sixteenth and seventeenth centuries when new results were communicated by personal correspondence. It may be a satisfactory substitute for a minority of scientists those with good personal contacts in their field. But it is a retrograde step for science as a whole, creating more problems than it solves. Informal communications of this kind are often unedited and unrecorded bibliographically, and the general community of scientists finds 4 Glass, B. and Norwood, S. H., " How scientists actually learn of work important to them Proceedings of International Conference on Scientific Information (Washington : National Academy of Sciences National Research Council, 1959), Vol. I, pp. 195-198. 5 Since the 17th century an exponential growth in the total number of journals has been maintained. Sooner or later the rising curve must keel over into an S-shaped form, since the world's population of authors cannot indefinitely expand, but there is no sign of any early slackening of the pace. We might reasonably assume that the scientific and technical manpower and hence literary output of the most developed quarter of the world could be doubled in a decade, that the remaining three-quarters of mankind will reach an equal state of development within a century or less; and that the world population will by then be doubled (Urquhart, D. J., " The National Lending Library Journal of Documentation , XIII (1957), 1, pp. 13-31). In such circumstances the world's output of scientific and technical literature would be 16 times the existing output. There is little reason to think that this is an overestimate. This content downloaded from 207.46.13.57 on Sat, 10 Sep 2016 05:14:20 UTC All use subject to http://about.jstor.org/terms