Cataloging & Classification Quarterly

Volume 32, Number 1 2001


CATALOGING NEWS
Sandy Roe, News Editor  

Welcome to the news column.  Its purpose is to disseminate information on any aspect of cataloging and classification that may be of interest to the cataloging community.  This column is not just intended for news items, but serves to document discussions of interest as well as news concerning you, your research efforts, and your organization.  Please send any pertinent materials, notes, minutes, or reports to: Sandy Roe; Memorial Library; Minnesota State University, Mankato; Mankato, MN 56001-8419 (email: mailto:skroe@ilstu.edu; phone: 507-389-2155).  News columns will typically be available at the CCQ website (http://catalogingandclassificationquarterly.com/ccqissue.html) linked from their corresponding volume and issue numbers prior to their appearance in print.

We would appreciate receiving items having to do with: 

Research and Opinion

 Events

 People

 EVENTS 

Festschrift for Pauline Atherton Cochrane published by the Graduate School of Library & Information Science, University of Illinois at Urbana-Champaign – Saving the User’s Time through Subject Access Innovation 

Pauline Atherton Cochrane has been contributing to library and information science for fifty years.  Her work on indexing, subject access, and the user-oriented approach had immediate and sustained impact, and she continues to be one of our most heavily cited authors and most beloved personages.  Each paper addresses a critical element of subject access.  Taken together, these articles inform historically, discuss the present state of affairs, and project into the future, and as such cover ground that has not before been so clearly articulated. 

William J. Wheeler, editor

Yale University 

1999 Amendments for Anglo-American Cataloging Rules, 2nd ed. 1998 Rev.

Amendments to AACR2r which were approved at the Joint Steering Committee (JSC) meeting in Brisbane in October 1999 have been published.  They may be obtained from the American Library Association at http://www.ala.org/editions/updates/aacr2/ in either PDF format or on CD-ROM (to current AACR2-e customers) free of charge. 

Meeting Minutes of the ALCTS Technical Services Directors of Large Research Libraries Discussion Group (“Big Heads”), held during the American Library Association Meeting, Chicago, IL, July 7, 2000 

Chair Judith Nadler (University of Chicago) opened the meeting at 9:30 a.m. by acknowledging the work of Karen Muller, the ALCTS Executive Director, in setting up the Big Heads electronic mailing list and of Judith Hopkins for taking the minutes and placing them on the Web.

Lee Leighton (University of California at Berkeley) was elected Chair to replace Mike Kaplan who had left Indiana University and thus had left Big Heads.

Larry Alford (University of North Carolina at Chapel Hill) was elected Chair Elect.

Carol Pitts Diedrich (Ohio State University), Chair of the Membership Committee, gave the committee's report. The Committee's charge had been to develop membership criteria which would:  1. result in a group size which would facilitate discussion and interaction at ALA and between ALAs; 2. ensure a reasonable continuity; 3. provide a selection process which is objective and seen to be fair; 4. not be burdensome to administer.

Based on these criteria the Committee recommended the following: 1. Enlarge the membership from 26 to a total of 28 including 24 regular members and 4 standing members (LC, NAL, NLM, and NYPL); 2. Base regular membership on ARL membership criteria index rankings. The index is calculated using 5 data elements: number of volumes held; number of volumes added (gross); number of current serials received; total expenditures; and number of professionals plus support staff.  3. Review the membership every three years. Calculate rankings by averaging the most recent three ARL rankings and taking the top 24.  4. Distribute the ranking every three years prior to the annual meeting and announce changes. Invite the new member institutions to send an appropriate representative to the upcoming meeting. Member institutions falling below 24 in the rankings would continue in Big Heads and continue to send a representative through the next annual meeting at which time they would leave the group.  5. Hold the first review prior to the 2001 annual meeting.

Judith Nadler commented on the second recommendation. What made the most sense to her was that both Big Heads and the Collection Development Officers of Large Research Libraries Discussion Group would use the same ratings to choose their members. This consistency would validate the fairness of the approach. She characterized the present system of choosing members as backward looking; the proposals seemed to her to be forward looking.

Catherine Tierney (Stanford University) said that if the grandfathering provision were omitted Big Heads could remain an effective size for discussion; with it the group ran the risk of becoming too large.

Joyce Ogburn (University of Washington) spoke in favor of the recommendations.

The recommendations were approved unanimously on a voice vote.

New models for electronic/digital resource description (Calhoun, Nadler, others) 

Karen Calhoun (Cornell University) said she was working on a research project to look at innovative workflows for distributed resource descriptions. She described traditional workflows as being self-contained and linear, with one person able to do each step of the workflow. The Internet resource process, on the other hand, is an iterative, looping, and collaborative workflow, involving players from different functional groups in the library--selectors, acquisitions, cataloging, information technology staff—as well as groups outside the library—vendors and publishers.

Calhoun commented that it might be more helpful to use a systems analysis approach rather than a workflow (flowcharting) approach to gain insight into the electronic resource description process. She provided a diagram in which data flowed from various groups and data sources (selectors and reference staff, catalogers, resource databases such as CORC, metadata knowledge bases, and Internet resources) to feed the electronic resource description process. Inquiries and responses ran bi-directionally between selectors and catalogers, between selectors and acquisitions staff, and between selectors and vendors, publishers, authors, and other providers of resource descriptions.

Sally Sinn (National Agricultural Library) asked if Calhoun's research might handle the recognition that the catalog must include all levels of materials --journal articles as well as books, AV materials, and technical reports-and must show the relationships between the various levels. Karen Calhoun agreed on the need to manage and make sense of different levels of granularity in resource descriptions. If this is not done, searchers are confused. Cornell had done a user study about 18 months ago, using focus groups of students, faculty, and library staff. They found that, contrary to their expectations, searchers of their networked resources database often didn't clearly understand what they were searching nor have a grasp of what kind of networked resource would be helpful to them. What is needed is a middle layer beyond the search box to assist users to better understand the contents of the catalog (including its granularity) and to assist them in choosing an appropriate database for their information need.

Duane Arenales (National Library of Medicine) asked about the relationships, if any, between these approaches and search engines such as Google and Yahoo. Lee Leighton said that the answer depended on several factors. The Berkeley catalog is not searched by any of those engines but various databases to which Berkeley provides access are.

Sally Sinn commented that lots of workflow decisions are based on such factors as what one's local management system provides and what barriers it imposes.

Cindy Shelton (UCLA) commended the recommendation of an article on digital portals that Roxanne Sellberg (Northwestern University) had made over the Big Heads electronic mailing list prior to the meeting. ["The case for creating a scholar's portal to the web" / Jerry Campbell, made at the 136th ARL membership meeting, May 2000. The URL is: http://www.arl.org/arl/proceedings/136/portal.html.]  She said that UCLA has a system for the management and tracking of digital resources and wondered if other libraries have such a system.

Judith Nadler said that one result of their recent reorganization is the Information Resources Management Division which encompasses the activities involved in acquiring and processing materials in all formats. The reorganization 1) gave the Head of Acquisitions responsibility for ordering and licensing digital resources as well as for maintaining them; 2) amalgamated serials cataloging and electronic materials cataloging; and 3) established a Digital Library Development Center which will address issues related to digital library architecture.

Karen Calhoun asked if any libraries get resource descriptions from sources other than catalogers.  Sally Sinn said that scientists at NAL are assigning record data to their publications; these can serve later as primary resources for catalogers. Duane Arenales) said that NLM had come up with a consistent set of Dublin Core elements which researchers can assign to their publications.

Arno Kastner (NYU) said he was interested in finding out how people are handling electronic journals that are NOT received through aggregations.  Joan Swanekamp (Yale) said that Yale has tried to provide URLs to the hard copy records for those titles. They loaded EBSCO records and added local information.  When the items described in the records change, they replace those records with newer EBSCO records.

Lee Leighton said that the University of California libraries have introduced a collaborative process (based at UC-San Diego) for serial titles held by all nine of the University of California campuses.  They prefer to use one record rather than multiple copies of a record. Libraries with very few paper resources find it very easy to implement.

Judith Nadler said that the University of Chicago believes that the single record approach provides better service to users. To her question asking how many Big Head Libraries use the single record approach, about eight libraries indicated that they did. Perhaps one used the multiple record approach; all others used a mixed approach. Arno Kastner (NYU) said that while he favored the one record approach NYU had to use the multiple record approach. More research is needed on whether or not a single record approach is best for users. Duane Arenales said that NLM uses the single record approach. Users find a title and are then led to a fuller display of that title which shows that it is available in two versions.

OCLC update: Kellar 

According to Lynn Kellar of OCLC, CORC is one model for describing electronic resources. It went into Production status at OCLC on July 1, 2000 and as of today (7/7/2000) the CORC Resource Catalog contained 323,143 items plus 659 pathfinders. The CORC Founders had added about 5,000 records a month.

Concerns had been expressed that the CORC system is somewhat 'flaky'. Over the last six months the OCLC staff had done one install every two weeks, and even more frequently in the last two weeks as Production Day had approached. Her team had worked over 30,000 hours in the last six months. The most important thing they had done was to integrate the records in the CORC Resource Catalog into OCLC's WorldCat.

Lynn encouraged people to continue using the CORC-L electronic discussion list as well as the CORC webpage to keep up-to-date on CORC activities.  They hope to continue installing new versions every two weeks to clear up the bugs. New functions will be added on a monthly or quarterly basis. They will be changing the 040 field on the CORC records to show the specific institution which entered the record. And, finally, OCLC hopes to do more Unicode work.

Joan Swanekamp asked if OCLC planned to add the Visual Materials format to CORC.  Lynn Kellar said they are planning to add other metadata formats and asked people to let her know which formats they were most interested in.

Barbara Stelmasik (University of Minnesota) said that Minnesota has been very interested in pathfinders, which their reference librarians had been creating. Eric Childress (OCLC), speaking from the audience, said that pathfinder development was presently on hold during the integration of CORC records into WorldCat.

Larry Alford asked a question about the need to validate URLs. Eric Childress said that OCLC has established the 956 (Local Electronic Location and Access) field (described in OCLC Technical Bulletin 239). It has the same indicator and subfield codes as the 856 field. Like other local fields it will not be included in the master record.

Larry Alford said he was disappointed in the small number of pathfinders. The approach taken of having libraries create pathfinders and for OCLC to maintain them, charging a storage fee to do so, has discouraged their creation and use. He preferred the resource-sharing model used for other materials in WorldCat. Eric Childress said that feedback from libraries had indicated that they wanted to retain control of their pathfinders. It is possible to clone and locally edit them. Larry Alford responded that the need to have a philosophy of sharing resources is critical.

Judith Nadler commented that pathfinders are not the work of catalogers; more individual pride is taken in their creation. They are more institution-specific than are catalog records. A cultural change is probably necessary before pathfinders can be shared the way catalog records are.

Lynn Kellar concluded by saying that OCLC was planning to run focus groups at ALA conferences on pathfinders and their future development.

Present applications (Nadler, Diedrichs, Ogburn, others)

Carol Diedrichs expressed interest in the idea of having selectors provide resource descriptions; this was something they hoped to implement at Ohio State. She was concerned, however, about giving full cataloging authorizations to bibliographers and wondered if OCLC could establish new authorization levels.  Lynn Kellar (OCLC) replied that OCLC could look at different sorts of authorizations that might be needed.

Karen Calhoun said that at Cornell, as an experiment, selectors have been providing CORC level records which have then been upgraded by catalogers. Both reference librarians and selectors were providing resource records. Each group chose different types of items. The CORC at Cornell team discussed, but did not resolve, policy issues related to the roles of reference librarians versus selectors in selection. Catalogers found the summary notes provided by selectors to be very helpful, especially since selectors know why they are selecting a resource, and catalogers tend to lack this information when they are framing summary notes.

 Judith Nadler referred to her Round Robin report http://www.acsu.buffalo.edu/~ulcjh/bh72000rr.html#chicago for a test project in the sciences founded by public services staff, not technical services. She commented that the CORC database is richer in recall but not yet a rich enough database to harvest records; elsewhere in the project there is more precision.

Long term potential (Sellberg, others)

According to Roxanne Sellberg the bibliographic utility is surely the single most important library labor-saving idea since the card. However, the shared cataloging system we all participate in, using the bibliographic utilities, still is marked by extreme levels of redundant human energy because libraries maintain separate, institutional-specific databases in addition to helping to maintain shared databases. Sometimes we participate in consortia that, depending on how they are set up, either mitigate or multiply the basic problem.

 The reason we have all felt we had to maintain separate catalog databases, up to this point, has been that we need to control institution-specific accounting and inventory information. The original OCLC model, which included the production of customized catalog cards, made it easy and relatively cheap for each library to have its own database telling its patrons about the resources mostly in its own buildings. We have been able to develop complex integrated library systems to combine acquisitions, circulation, and inventory information with cataloging records.

The thing is, that rationale does not apply to Web Resources--at least not quite in the same way that it does to other kinds of resources that libraries have collected in the past. My copy of a web site is no different from your copy, because there is only one copy which neither of us owns or usually has any control over.

Now it would be possible in the short term to apply this kind of model we have used for shared cataloging and bibliographic control of traditional library materials to web resources as well.  Many libraries are trying to do that-- to provide one-stop shopping in their online catalogs for books, serials, media of all kinds, and electronic resources, including web resources. I believe that that is a strategy that is bound to fail in the long run, however, because it is just too labor intensive.

I believe that the great long term potential of CORC lies in its potential ability to change the pattern of shared cataloging so that we need not have multiple separate local catalogs. Each item should really only be described when it is created or changed, not every time some library discovers it. We can do it with freely available web resources first, for the reason I mentioned-- with web resources there is no library-specific information to keep track of. However, I hope that once we have made a new model work for free web resources we'll be able to extend the new model to other kinds of resources as well.

In order to make that kind of model possible it is necessary that two things happen. First, that the CORC database be designed not primarily for the convenience and use of catalogers, but just as much for the convenience and use of end users. We must make the CORC database the place our users know they can find out about web sites. It needs to be as fast and as user-friendly as Yahoo, as well as more complete and more reliable and better.

Second, we need to create interfaces between databases that really work and that do not require redundancy. We need to be creating systems that seamlessly search across databases and combine data from multiple databases in a way that is useful.

Just before the break Larry Alford showed a poster that Big Heads were going to present to Mike Kaplan as a departing gift and invited those interested in participating in the gift to sign it.

Holistic approach to acquisitions and cataloging: model used at Stanford, Chicago, others? (Tierney, Nadler)

Catherine Tierney (Stanford) described the Stanford model which was devised in order to reallocate existing Technical Services resources to needs emerging in other library units. The objective was to partner with book vendors for mutual economic advantage. They found out what services vendors already offered or were considering as new opportunities. They attempted to design products and services that other libraries would likely also want. With their domestic vendor they concluded that the Edifact standard was the best choice for Electronic Data Interchange, as general observation was that X.12 was on its way out; expectation was that other libraries and ILS vendors would adopt Edifact. MARC records were the standard for bibliographic record transmittal. LC classification was the appropriate "standard" for a U.S. vendor who wished to charge for classification services.

Materials and records coming from this vendor have met the expectations for quality and service. In fiscal 1999, all incoming materials from this vendor had records and classification (94% LC or CIP; 6 % Provisional Plus) provided; this represents 19,000 items that did not have to be copy cataloged by staff at Stanford and 22% of all incoming material. It took only .75 FTE to do the final quality check and occasional fund adjustments when the materials arrived. What of this model can be generalized across large research libraries so that vendors' investments in systems development and new services have sustainable returns?

What products and services can we agree that we need in essentially a standard way and are willing to pay for?

Judith Nadler said the Stanford model is similar to one used at the University of Chicago; they have three streams: no copy (original cataloging);  full copy (acquisitions does the processing);  mixed.

A vendor provides about 45% of their monographs and also provides bibliographic records with call numbers.

A small cataloging staff at the University of Chicago takes care of other materials. Are there enough libraries needing such services to make it possible for vendors to provide them without increasing their costs? Other vendors are exploring similar model for other materials, e.g., an Italian materials vendor and a vendor in Israel. These vendors provide description and subject headings but not classification. Subjects can be overlaid in Marcadia but classification is essential for speedy internal processing.

Other models

Lee Leighton said that Berkeley uses OCLC's PromptCat service rather than going directly to vendors for cataloging. Berkeley's past experience with storing backlog materials offsite with low level records and a local shelving device was unsatisfactory because the materials have not been retrieved for complete cataloging. Berkeley now prefers to fully catalog its materials before they leave Technical Services for shelving.

Karen Calhoun said that Cornell had been asked to reduce its backlog to a working backlog (one year's cataloging) without new staff. They did it by reducing the number of exceptions to fast cataloging (i.e., accepting more copy as is), buying bibliographic records to accompany approval books, and deploying the resulting staff time savings to work on the backlog.

Arno Kastner (NYU) asked if there was a critical mass needed to make this shelfready processing economically effective? Judith Nadler said Chicago has not yet gone to end processing. They triage materials to determine which unbound materials need binding.

Catherine Tierney (Stanford) said their return rate was 3%. They needed to work with selectors to help them decide that it was better to reject material not needed than to shelve and maintain material for 100 years.

At Stanford, determination for binding shelf-ready monographs is done after receipt from vendor. The percentages of shelf-ready books needing binding before they go to the shelves according to the usual guidelines is low enough to allow for the double end-processing. The science branches in fact decide whether to bind on ad hoc basis, usually after circulation.

Sally Sinn asked what the cut-off point is for vendors to make providing such a service viable for the vendors. Judith Nadler asked for input on this question from the Big Heads via e-mail. With smaller vendors might be able to take a team approach. She noted that it is important for us to to make our priorities known to the vendors in order to help them adjust their priorities and provide a better service to us.

Larry Alford said he was increasingly concerned that it may cost more to get such services from vendors than to do it ourselves, especially as there are only two domestic vendors now.  Judith Nadler said it is better to help vendors do what they can do in a better way.

Duane Arenales asked if libraries have done cost assessments of use of these services and have others used third parties? Catherine Tierney said that by using vendor-provided services they have gotten more things processed even though their internal staff has been reduced by 10%. Their vendor says that in order for its LC classification services to cover expenses, more libraries than Chicago and Stanford need to use this product.

Joyce C. McDonough (Columbia) said that Columbia found it was a loss as to whether they or vendor did classification. She commented that vendors need to be more business-like in pricing these services.

Carol Diedrichs said that Ohio State has no Western language backlog. Material with copy goes through a fastcat process or more complex copy cataloging usually in one week but can take longer to go through labeling. A recent survey of public services staff indicates that the current turnaround time is acceptable. However this speed has come as a result of making choices to do less checking of the cataloging record.

Joan Swanekamp said that Yale's cultural climate finds it acceptable to keep things in the backlog for 2-3 years.

Concern about non-standard vendor records in utilities (Leighton, Swanekamp, others)

Lee Leighton said that non-standard vendor records tend to stay in utilities for a long time without being enhanced. If local non-enhance libraries update a record the update is not permanently made.  Joan Swanekamp said these records complicate triage. Access points, if present, don't match those used in the United States. It is easier to create new records or put items on backlog than to deal with them. She has been converting copy cataloging positions to professional positions because Yale is finding that it needs to do more original cataloging, in part at least because many of these vendor records are not being upgraded to full cataloging records available in the utilities.

Barbara Henigman said that the University of Illinois at Urbana-Champaign does its upgrades in its local catalog and does not send the upgrades to OCLC. As a Dewey library they need to spend lots of time classifying.

Karen Calhoun said their approach is similar. They change records in the local system, then batchload to the utilities. When their records are batchloaded to OCLC, their changes to existing records are not reflected in the corresponding OCLC master records, so the upgrades are lost to other OCLC libraries. Sometimes they upgrade vendor records to PCC records, with authority records for all headings.

It is a shame that these upgrades can't be shared with other OCLC libraries. But, since Cornell can't afford the double keying it would take to make the changes once to records in the local system, then again online in OCLC, they do not expect to change their workflow.

Karen Smith-Yoshimura (RLG) said from the audience that RLG doesn't find much collection overlap.

Quality and treatment of BIBCO records in WorldCat (Alford, Calhoun, others)

Judith Nadler said that the University of Chicago gives high priority to BIBCO but finds lots of problems with other libraries' BIBCO records, e.g., records for foreign language materials often seem to be prepared by people without adequate language knowledge.  Joan Swanekamp added that a few BIBCO libraries don't do the required authority work. Catherine Tierney said for BIBCO and NACO to work we have to speak to the offending libraries about the problems found with their records.

Larry Alford said such problems also have to be reported to PCC so that they improve training.

OCLC governance issues and their effect on us (Alford, others)

The meeting ended before this topic was reached.

Judith Hopkins, State University of New York at Buffalo

PEOPLE

Kewal Krishan, Serials Librarian at University of Saskatchewan Libraries, Receives CACUL Award for Outstanding Academic Librarian.

Kewal Krishan terminated his 34-year career as a Serials Librarian at the University of Saskatchewan Libraries with a bang.  He was honored with the CACUL (Canadian Association of College and University Libraries) Award for Outstanding Academic Librarian at a ceremony held by CACUL at the Canadian Library Association CLA)Annual Conference in Edmonton, Alberta, June 2000.

Prior to his coming to Canada, he worked as a professional librarian in India for nine years, and was honored with Meritorious Library Service Award by the Punjab Library Association in 1966.  He started working at the University of Saskatchewan Libraries in 1967 and got promotions through the rank from Librarian I to Librarian IV (top rank).

In the CACUL Award Citation, he was recognized for his excellent professional achievements through his tireless contributions to CLA as Convenor of the Serials Interest Group (1985-2000), as CLA Liaison Person with NASIG (1996-2000), as well as through his publication record.  His landmark publication “A Hundred Years of Serials Librarianship in Canada, 1893-1993: A Bibliography” was extremely well received within the Canadian library community.

Vinh-The Lam

Head, Technical Services Division

University of Saskatchewan Libraries


| CCQ Homepage | Tables of Contents | Return to Vol. 32, Nr. 1 | Informaworld |

Comments to: Jeffrey Beall at
© Haworth Press, Inc.