78.

persons should be allowed to operate in these areas with a certain amount of freedom. 7. It may be necessary to shift the problem domain, say in response to a realisation of need from administrative/ management/social areas. V. HORSNELL

IMPLICATIONS FOR CURRICULUM DEVELOPMENT

Taking one of the messages of the Forum to heart, we began by restricting our domain to the question: What role should education for research play in the curriculum of a first professional course of study for information science? Again in a pattern of the Forum, we then began to talk about the curriculum of information science in general. Consensus was reached on the following points: 1. The ideal for which education for information science should strive would be akin to programs already established in Poland and Berlin at the Masters level, and at Ohio State. University and Georgia Tech. at the PhD level. In the first case, an integrated 4 year program with two initial years of general, theoretical core courses, followed by two years of elective, practical course work coupled with research and practical laboratory work. In the second case, establishment of a multi-disciplinary faculty whose major emphasis is on basic research on a variety of the phenomena identified as being of interest to information science (with a strong integrating position). The reality of the situation (in the US and UK) limits the possibilities to a one-year course with emphasis on professional training. In view of 2, above, the general format of a few, basic theoretic core courses and a wide selection of practice

2.

3.

and research oriented electives should be as followed. 4. There should be a core course in research for information science. ing structure: Discussion of scientific research in general, and of the significance of methodology. What constitutes scientific research, and the place of information science in the system of sciences. Problems (the phenomena) of information science, how previous methodologies are applicable to them, and in what ways they are unique. The dangers of applying methodologies from problems of one science to problems of another. - Beginning of individual research projects. From this point, the course becomes a seminar, combined with a few lectures on the specific topics of: Data acquisition (eg questionnaires, specific experiment design, specific sampling problems Data analysis (eg specific statistical methods, computer analysis packages) Data presentation (or information generation) Theses Completion It was also noted that the core courses should have as general and basic a flavour as possible, in order to anticipate change as well as possible. Basing a core course on semiotics was a specific suggestion. N. J. BELKIN It should have approximately the follow-