ISR11 Design Consideration for Time Shared Automatic Documentation Centers chapter M. E. Lesk Harvard University Gerard Salton Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. x-lO One could of course develop a current a[OCRerr]reness system that would slowly grow into a general retrieval system, but such a system would have few users. 2) It should be a field in which the literature is published primarily in English. Although the foreign language problem can unquestionably be overcome, there is no point in tackling the problem immediately if it is not necessary to do so. A purely English collection will also be easier to use, and this may help in attracting users. 3) It should be a field in [OCRerr]`hich very little of the material carries a security classification, since the initial system will be used initially for experiments in documentation systems, and it will thus be desirable to be able to publish the results of these experiments. A system which does not include classified docu- ments will also be easier to use by the customers. It should be a field with a good abstracting service and with strong American scientific societies, which might cooperate with the system designers. 5) It should be a field with at least one hundred active researchers at the institution where the documentation center is established, to insure an adequate user population. 6) it should be a field for which the conventional library has an adequate and easily accessible shelf collection. [OCRerr]othing would be more frustrating than to have the ccmputer select a document which is only available through a slow inter-library loan procedure. A field that satisfies most of these criteria is physics, concentrating on the journal articles only, and avoiding much of solid state physics. Physics papers are published primarily in English and Russian, and the major Russian journals are available in English translation. [OCRerr]terial becomes out-of-date extremely quickly. As a basic science, it is largely unclassified (avoiding, of course, applied nuclear physics). One predominant publishing