Search UW System

UW System Home > Council of University of Wisconsin Libraries > Former Committees & Taskforces > Cataloging Full-Text Electronic Resources > Documents

Date: February 7, 2000

To: CUWL

From: CUWL Task Force on Cataloging Full-Text Journals in Aggregator Databases

Linda Cecchini (Eau Claire)

Steve Elfstrand (Eau Claire)

Cindy Gandrud (River Falls)

Richard Reeb (Madison)

Mary Richards (Stout)

RE: COOPERATIVE CATALOGING PROGRESS REPORT

I. Introduction:

At its September meeting the CUWL Executive Committee created this task force and charged us with "developing shared practices and directions for providing cataloging and holdings information for full-text titles such as those in EbscoHost." In November, the Directors asked us to develop specific recommendations for cooperative cataloging of e-journals in aggregator databases.

The focus of this task force has been on providing access to e-journals from the online catalog rather than from a web page. The Libraries’ OPAC allows for integration with other (print) journals in the collection. Web-based catalogs such as Voyager can enable users to connect directly to the full-text source via hypertext links in the catalog record.

II. Developments since last report:

1. OCLC WorldCat Collection Sets for aggregator databases now available.

On January 14, 2000 OCLC WorldCat Collection Sets (formerly known as the OCLC Major Microform Service) were enhanced to include cataloging records representing full-text journals in aggregator databases. The enhancement also includes the option for periodic updates containing new records that have been added to the set.

Bibliographic records for netLibrary, Project Muse, Academic Press IDEAL, JSTOR, Kluwer, Documenting the American South, and Elsevier are now available through WorldCat Collection Sets. Cataloging is currently in progress for sets of records for items in Springer, Wiley, MCB Press, Royal Society of Chemists, Institute of Physics and American Physical Society. OCLC is looking for volunteers to catalog other aggregator databases, especially Wilson Select, IAC Infotrac, Lexis-Nexis Academic Universe, UMI ProQuest Direct, ABI/INFORM and most full-text OCLC FirstSearch databases.

 

 

B. The PCC Standing Committee on Automation (SCA) Task Group on Journals in Aggregator Databases has issued its January 2000 final report (attached).

This ALA committee is examining ways in which libraries and vendors might work together to develop "a useful, cost-effective and timely means for providing records to identify full-text electronic journals acquired in aggregator databases." Ebsco has been working with this committee since early 1999 to create a test set of bibliographic records to develop a model for other vendors of serial aggregators. In December 1999, California State University, Northridge successfully loaded the record set into its OPAC (Geac) and reportedly is pleased with the results. This committee plans to pursue creation of record sets for major aggregations like ProQuest and Lexis-Nexis Academic Universe. This group also plans to recruit WorldCat Collections Sets contributors from among the OCLC membership.

The PCC/SCA Task Group issued the following recommendations for record creation and maintenance:

PCC/SCA Recommendations for Record Creation:

1. Human-created
MARC records are best for aggregators containing no more than a couple hundred titles.
2. For larger ones, the task group recommends machine-derived MARC records that are produced by computer programs from the content of an existing human-created record, such as the print version, as the basis of an electronic version record.
3. Three other options are available in situations where machine-derived records are not possible:

1. Machine-generated
MARC records that are produced by computer

programs from data elements provided by the vendor.

1. Local scripting
by a single institution to create minimal-level record sets from vendor-supplied title/ISSN listings.
2. Creation of a single, combined index of serial title coverage within all the aggregator databases to which an institution subscribes. One example of this is the Jointly Administered Knowledge Environment (jake), which is hosted by the medical library at the Yale University School of Medicine.

Jake is a searchable database containing information about e-resources and how they relate to one another. They offer the software at no cost. The URL is jake.med.yale.edu

 

PCC/SCA Recommendations for Record Maintenance:

1. The PCC/SCA task group recommends distribution of the entire record set each time the OPAC is to be updated.

The records in any set need to be maintained along the following lines:

• Overall record set distribution/delivery
• Added & dropped titles
• Changes in the volume coverage for each title
• The completeness of the content for the volumes covered
• Currency of the URLs in the aggregator analytic records
• Creation of new records for the conventional changes of title for serials included in an aggregator database
• Cancellation/change of a subscription necessitating complete removal of a record set.

C. Results of loading Ebsco Records into the UWEC Catalog

Stephen Elfstrand has been in contact with Ebsco staff and has loaded several of their records into the Voyager OPAC at Eau Claire. These are vendor-provided derived records prepared according to the Task Group, multiple record model.

No holdings record (i.e. MFHD) was loaded. The holdings statement is in the 856 $3.

Bulk importing went smoothly.

Updates and Deletes

This may be a concern due to the lack of a bulk delete program in Voyager. PCC/SCA Task Group is currently suggesting deleting the entire record set and reloading for updates. One possible alternative being explored is to have vendors deliver deleted records with delete status set and then identify deletions via Report Writer.

Advantages:

Can get started: Ebsco will have its record sets ready within a few weeks.

Multiple record model makes maintenance easier than the single record model.

Deletes require only deleting the bib not MFHDs also.

Updates using a "replace unconditional" profile easy to do

Will be retrieved by "Journal Title" search

 

 

 

 

Disadvantages

Each title will have a separate record from each vendor that includes it (plus any for the paper and microforms now in the OPAC)

Multiple records are confusing to users and make catalog searching less efficient. (are they more confusing than multiple MFHDs?)

Holdings are in the 856 rather than a MFHD

Holdings are not ANSI formatted

Will need to load updates regularly for each file from each vendor

Will need to do deletes occasionally

May not mesh with current campus practice

III. Options for Access

1. Vendor-derived record sets

Not all vendors currently provide record sets. See PCC/SCA Task Group final report, online at http://lcweb.loc.gov/catdir/pcc/aggfinal.html, and Elfstrands test with Ebsco derived records (Section IIC).

2. OCLC WorldCat Sets

These record sets do not include holdings data. OCLC would notifiy libraries when new titles are added to a collection set, but not when titles are deleted from an aggregation.

3. Local, cooperative, non-OCLC approach

Divide up the databases and have each library contribute cataloging for a section, then share records. (Discussed different ways to do this: shared work files, use of an Access database). Advantages: Could produce records that are more user friendly for patrons: single bib record approach, holdings statements in standard format. Disadvantages: Very labor intensive, slow process - how much would the data change before we finished the first round.

IV…Committee’s Recommendations

• UW libraries should purchase record sets from vendors and utilities (i.e. OCLC) as they become available.
• UW libraries should catalog or buy WorldCat Collection Sets for JSTOR and Project Muse titles now.

The journals in JSTOR and Project Muse are archived and complete and records are already on OCLC.

• UW libraries should start asking for records from aggregator vendors during the negotiation process
. Availability of records should be given sufficient weight during the negotiation process.
• UW libraries should lobby OCLC to include holdings data in WorldCat Collection Sets e.g. 856 subfield 3.
• UW libraries should lobby aggregator vendors to conform to the JSTOR practice of notifying custormers of changes in holdings data.
• UW libraries should lobby Endeavor to provide bulk delete functionality in Release 2000 Voyager software.

 

V…Next steps:

• Fully test loading & maintenance of one record set using Voyager’s bulk import program in release 99.1 [Steve Elfstrand has offered to load the Masterfile record set and one update into the UWEC OPAC. Holdings and locations will not be loaded. Multiple record approach will be taken, i.e. records will not be overlayed.]
• Ask the CUWL CDC to make a list, in priority order, of desired record sets for the aggregators purchased cooperatively.
• Poll the aggregator vendors to find out where progress is or isn’t being made regarding creation of record sets. If possible, enlist the help of Lorie Docken since she already maintains contacts with each vendor.
• For the smaller aggregator databases like JSTOR and Project Muse, explore the possibility of exchanging holdings data..
• Monitor developments in OCLC WorldCat Collection Sets and PCC/SCA Task Group on Journals in Aggregator Databases.
• Raise awareness in the UW library community of the issues pertaining to cataloging journals in aggregator databases.