In the Report of The Library of Congress Working Group on the Future of Bibliographic Control, On the Record, section 1.1, Eliminate Redundancies, has several recommendations for using externally available bibliographic data and for further automating the CIP process. With this in mind, the Acquisitions and Bibliographic Access Directorate (ABA) is conducting a pilot project designed to make available ONIX data being received from publishers to the Electronic CIP (E-CIP) program. The pilot began in June 2009 and tested ONIX data from two publishers, Cambridge University Press and Wiley. The pilot is testing several aspects:
- The availability of ONIX data for items in the CIP stream
- The usefulness of the data in cataloging
- Any problems or unexpected results from converting the data from ONIX to MARC
- Changes that would be needed to the CIP workflow
- What additional information can be extracted from the ONIX data that would not normally be provided in MARC records
A virtual test section was established in the E-CIP Traffic Manager and incoming CIP applications from the two publishers are being diverted to this virtual section (except for items for the National Library of Medicine and the National Agricultural Library) for descriptive cataloging processing. If an ONIX record is found (based on matching the ISBN of the forthcoming book with ISBNs in the ONIX data), the data is converted immediately and a MARC record created. From here, the catalogers involved compare the resulting record to the publisher-supplied information from the electronic galley to look for differences or any missing/incorrect elements. Should there not be an ONIX record for the forthcoming book, the CIP application is forwarded to its original destination for normal processing.
In addition to the basic bibliographic record, the table of contents, if provided in the ONIX record, is provided in the 505 field of the MARC record with first indicator value '8' (no display constant) and the legend "Machine generated contents note:." No attempt is made to convert the field into a "perfect" contents note, so elements not normally found in a regular AACR2 contents note are in the 505, such as the words "Chapter," "Part," etc. as well as sections of the table of contents like the introduction, bibliography, etc. that are not normally mentioned.
Additionally, ONIX records frequently contain summaries (called descriptions in ONIX), and also on the E-CIP application there is a space for the publisher to provide summary information. If either of these is present, the conversion program includes them in separate 520 fields, quoted to indicate that LC did not create the summary and with "--Provided by publisher." at the end to indicate the source. The catalogers involved read any summaries and if they fit guidelines for including summaries in E-CIP records, the summaries may be left in the record. There is a potential for duplicate summaries to be created. If the summaries are very similar or identical, the cataloger would delete one of them.
Once the rest of the descriptive cataloging process is completed, the catalogers involved will create a report on the results, provide any impressions they wish to relate, and will then forward the E-CIP to the original destination for completion of subject cataloging and end-stage processing. As anticipated that this diversion and conversion has not added significant processing time to the E-CIP. As the pilot proceeds, any needed changes in the conversion application will be made as will any needed changes to the pilot workflow. Reports will be gathered weekly so that the pilot committee can be kept up to date on all of the issues.
Any questions can be directed to David Williamson, Cataloging Automation Specialist, ABA (email@example.com).