By SUSAN MORRIS
The Library of Congress recently completed a project to implement online cataloging of materials in Arabic, Hebrew, Persian and East Asian languages and scripts directly in its integrated library management system, the LC ILS. Staff in the Library’s Regional and Cooperative Cataloging Division and the African/Asian Acquisitions and Overseas Operations Division now create catalog records for materials published in the “JACKPHY” languages (Japanese, Arabic, Chinese, Korean, Persian, Hebrew and Yiddish) online in the LC ILS. Library catalog users can view the records immediately, whether they consult the catalog in one of the Library’s 20 reading rooms or on the Library’s Web site at catalog.loc.gov.
Users can search and retrieve JACKPHY-language records using either romanized or original script search terms and may use defined or keyword searches. This improvement delivers easier and timelier access to a significant part of the Library of Congress collections. Last year, 10 percent of the Library’s total cataloging of new books was for JACKPHY languages.
The Library has provided cataloging in machine-readable form for JACKPHY materials for more than 20 years through collaboration with the former Research Libraries Group (RLG). Because the Library of Congress’s legacy cataloging system (MUMS, the Multiple Use MARC System) could accept and display only the roman script, Library of Congress cataloging and information technology staff worked with RLG from 1979 to 1983 to develop a cataloging capability for nonroman scripts in RLG’s bibliographic utility, RLIN.
In June 1983, a Library cataloger created the first online JACKPHY catalog record—consisting of a romanized full record in the machine-readable catalog for a Chinese title and a corresponding catalog card with both romanized and original characters. The record was not verified or completed in the RLIN utility until September of that year. Even after the Library implemented its integrated library system in 1999, online cataloging of JACKPHY script materials involved a complicated workflow in which the Library’s cataloging staff created catalog records with both original-script and romanized characters in RLIN, and information technology staff imported those records into the Library’s catalog the next day. The original-script characters in the records could not be viewed in the Library of Congress Catalog.
Several milestones paved the way for the transition to cataloging of JACKPHY materials directly in the LC ILS. In November 2005, the Library completed an upgrade of its integrated library management system to Voyager with Unicode software (a commercial software from Ex Libris Group). The upgrade converted the Library’s database of bibliographic information to Unicode, a character encoding standard designed to support the interchange and display of all modern languages and scripts. The Unicode-compliant software allowed library users to view records containing East Asian, Arabic and Hebrew characters in the Library’s Online Catalog in the original scripts.
In the upgrade, the Library migrated 13,807,655 bibliographic records, 15,031,357 MARC holdings records and 6,858,768 authority records to the Unicode-compliant system. Users were thus able to search and display JACKPHY characters and scripts in the Library’s Online Catalog, but staff still did not create or update JACKPHY records in the Library’s local system.
In May 2006, RLG merged into OCLC, Inc., to form a single bibliographic utility. OCLC, Inc., announced that it would close the RLIN utility on Aug. 31, 2007. This change necessitated a transition for JACKPHY cataloging from RLIN to either the LC ILS or OCLC WorldCat. The Library’s Acquisitions and Bibliographic Access Directorate conducted comparison tests of cataloging in the two systems and concluded that JACKPHY cataloging in the LC ILS was more efficient and offered timelier service to catalog users.
In consultation with OCLC, the Library’s cooperative cataloging partner institutions, the Council on East Asian Libraries and the Library’s reference staff, the Library decided to implement JACKPHY cataloging in the LC ILS for monographs (books, microforms and electronic resources). Cataloging of serials in the JACKPHY languages continues to be performed in OCLC because the latter is the database of record for the Cooperative Online Serials program (CONSER), which the Library supports through cataloging, training and administrative oversight.
In preparation for the move of JACKPHY cataloging from RLIN to the LC ILS, the Library’s information technology staff updated all staff workstations. The LC ILS Program Office developed a software application, “Transliterator,” which automatically pairs romanized character strings with their original-script counterparts. The Transliterator program also automatically renders romanized Hebrew into the original script, greatly reducing the labor needed to upgrade older records that contain only romanized character strings. The current version of Transliterator also renders original Chinese characters into their romanized equivalents.
As the Library of Congress continues its efforts to improve service to users of JACKPHY catalog records, it welcomes comments and suggestions from end users and other libraries. Comments may be e-mailed to Beacher Wiggins, director for Acquisitions and Bibliographic Access, at bwig@loc.gov.
Susan Morris is assistant to the director for Acquisitions and Bibliographic Access.