The theory of convergent evolution in the life sciences is used as a metaphor to illustrate the heightened levels of integration achteued by technical services, public services, and collection development in academic and research libraries. The vehicle for this development is automation, which has evolved to the stage where migrations from one automated system to another are more common than new system implementations, and a complete rethinking of information processing is emerging. Organizational charts have been slow to reflect these changes, but workflow patterns are showing more reliance on matrix organizational theories than on traditional hierarchies. It is difficult to see clearly the direction technical services librarians should be moving and to plot the steps necessary to move in that direction. What is clear, however, is that new information needs and changing library environments are revitalizing and restructuring traditional internal relationships.
A practice cataloging collection of 716 books was cataloged using Anglo-American Cataloguing Rules, second edition, 1988 revision, with each rule use recorded. A total of 20,247 rule uses was required, but of the 818 rules in the code, only 232 (28.4%) were used. Most frequently used were rules for choice of name (22.1A and 22.1B) and entry under surname (22.5A1). When rules are ranked by frequency of use, the distribution is best described by an exponential curve. When compared with other studies of rule use, the findings suggest that introductory cataloging instruction and expert systems can identify and focus on a core set of rules, safely ignoring those that are seldom used.
In 1986 an empirical study was conducted to examine the extent of bibliographic relationships as reflected in their frequency of occurrence within the 1968-July 1986 machine-readable database of the Library of Congress. Frequency of occurrence was determined by counting the incidences of specific codes associated with each relationship type within the machine-readable records. Also examined were characteristics of bibliographic items exhibiting particular relationships. Characteristics of interest were language, place of publication, publication date, subject, and bibliographic format, as it was thought such factors might prove useful in predicting particular types of relationships for future cataloging systems. Such information can be of potential use to decision makers and system designers in assessment of appropriate methods for designating specific relationships in both future catalogs and future cataloging rules.
An analysis of data elements in the United States Machine-Readable Cataloging (USMARC) formats for bibliographic, authority, and holdings data is presented. Criteria for the examination of USMARC, derived from the theory of the conceptual schema for databases, are (a) types of entities and data elements that are included, (b) the function of each data element, and (c) the rules of form that govern each data element in the USMARC formats. The major finding of the analysis is that there is a high degree of redundancy present in USMARC, and a more rigorous conceptual plan for USMARC is desirable.
Random samples of main entries on monograph catalog records created by the Library of Congress and the British Library were compared for the year 1989 in order to examine one aspect of the potential usefulness to the Library of Congress of British Library monograph cataloging, especially regarding the extent of agreement in cataloging practice between the two national bibliographic agencies. It was estimated that between 27.1% and 31.5% of the monograph entries in the printed 1989 annual cumulation of the British National Bibliography had been cataloged by both the British Library and the Library of Congress. It was further estimated that, for both choice and form of main entry, agreement was achieved between 60% and 70% of the time, and for choice of main entry alone, between 96% and 99% of the time.
The kinds of variations appearing in name access points found in OCLC bibliographic records were explored by examining a sample of records taken from the OCLC Online Computer Library Center, Inc., Online Union Catalog. Name access points were searched in the Library of Congress Name Authority File (LCNAF], and all records for each name were examined in the OCLC bibliographic file to identify variant forms. For 64.4% of the sample records, an authority record was found for one or more names on the record. Of these, 23.4% (1.5.1% of the entire sample) had one venture names that differed in form front the LCNAF form. Another 5.8% of soniple records had personal monies that had no LCNAF record but differed from the majority of records using that name. An average of 3.4 forms of access points per name was found. Prolific authors figured prominently. Only 38 names accounted for 84,5% of all records examined for the 457 personal names. Nearly 44% of variants were identified as being a "near match" to the standard form. Nearly 29% were found to be only a single typographical error away from. matching the standard. Over two-thirds off all variations occurred in dates. It might he possible to correct many variations with machine assistance. Single typographical errors and near matches could be identified by machine for human review.