Outlines the reasons why the International Standard Book Number (ISBN) is proving to be an unreliable identifier for records in bibliographic data banks and suggests how the Universal Standard Book Number (USBN) could be used as an effective substitute. Results are given of an analysis of the 4% duplicate rate found by Beale and Lynch, together with the details of the work in progress at the University of Bradford.

A method is described for the construction of a control number for a bibliographical record which could provide a universal standard book number for that record. Since the new number relies on logical processes for its construction it could be obtained either manually or on the computer. The USBN consists of nine elements. The two most important of these are obtained from the title, and use a code for their construction which is based on the frequency of distribution of alphabetic letters.