Showing papers by "Eugene Garfield published in 1976"

Significant journals of science.

Abstract: In 1974 the Science Citation Index (SCI) covered about 401,000 articles and communications in 2,443 scientific and technical journals. They cited about 3.2 million different publications an average of 1.8 times each. In this article some results of an analysis of more than 5 million citations in the references of journal articles indexed for the SCI in 1974 are presented and an attempt is made to interpret of those results in the light of an earlier study of 1969 citations.

The Permuterm Subject Index: An Autobiographical Review.

Abstract: The Permuterm Subject Index (PSI) section of the Science Citation Index (SCI) was designed more than ten years ago and has been published both quarterly and annually since 1966. There is, however, no ‘primordial’ citable paper about the PSI. It has been described and discussed from different standpoints in a number of papers (1,2), but none of them provides the formal description usually accorded a new bibliographic tool. This article is intended to provide such a reference point for future workers in information science. The PSI was designed in 1964 at the Institute for Scientific Information (1S1) by myself and Irving Sher, my principal research collaborator at the time. In the subsequent development of the PSI, contributions were also made by others, including Arthur W. Elias, who was then in charge of production operations at 1S1. In the early sixties we were too preoccupied with the task of convincing the library and information community of the value of citation indexing even to consider the idea of publishing a word index. But it was a logical development once we added the Source Zndex containing full titles. The value of the PSI as a ‘natural language’ index is now well recognized and exploited by its users, but this was not the original reason for its development. The PSI was developed as one solution to a problem commonly faced by uses of the Citation index section of the Science Citation Index (SCZ_). While the typical scientist-user could enter the Citation Index with a known author or paper, other users with a limited knowledge of the subject often lacked a starting point for their search. Before publication of the PSI, we told users whose unfamiliarity with subject matter left them doubtful about a starting point to consult an encyclopedia or the subject index of a book. If these failed, we told them to use another index, such as Chemical Abstracts, Biological Abstracts, Physics Abstracts or Index Medicus. Once the user identified a relevant older paper, it could be used to begin a search in the Citation Index. Users of the SC1—and librarians in particular needed some tool with which a starting point, or what used to be called a target reference, could be quickly and easily identified. In those days the information community was pre-occupied with KeyWord-in-Context (KWIC] indexes. The development of the KWIC index, which was subsequently vigorously marketed by IBM, undoubtedly had an enormous impact (3, 4, 5). But I was never happy with the KWIC system for a number of reasons. First, Sher and I felt that the KWIC index was highly uneconomical for a printed index. KWIC’S use of space is prodigious, and it can be extremely time-consuming to use in searches involving more than one term. 546