Theory and Other Quantitative Factors in Code Design for Document Card Systems*

In the past ten years, the field of information retrieval has witnessed the development of many new systems, devices, and theories. In particular, two opposing “schools” of thought on card indexing systems have developed. One claims that the term card (unit term) or “collating” system is the most desirable. The other advocates the document card (unit record) or ‘‘scanning” system. Dr. Whaley has noted many of the advantages and disadvantages of collating and scanning systems, and I am glad to adopt his terminology and agree with most of his comments. I For the record, however, I wish to remind the proponents of term card systems that theirs was no new finding. Coste1102 says Batten3 anticipated Taube 4 by 15 years. Batten was anticipated by at least another 35 years. One term card system began at the turn of the century at Johns Hopkins Hospital. Subsequently, it went through all the evolutionary stages which the inherent clearly demonstrate similarities between term card and document card systems. This does not mean that the rediscovery of the term card system was an insignificant development. After all, many useful ideas and inventions are redkcovered and we are grateful for these discoveries. However, when appropriate, our precursors ought to be given credit. Even the ten column posting card was anticipated by Paul Otlet, founder of the modern documentation movement.5 Indeed, long ago, the term card system was used in several medical institutions, including Johns Hopkins Hospital and the Mayo Clinic. Texts on medical records management demonstrate such systems.c These consist of one 3 x 5 card for each disease (term). Each card then lists the case history document numbers for all patients so diagnosed. Ultimately, the number of case history numbers grew larger