BIBCO Operations Committee Meeting

May 3-4, 2001
Summary

Joint BIBCO/CONSER session
Thursday May 3, 2001

The fourth annual BIBCO Operations Committee Meeting held at the Library of Congress on May 3-4, 2001 commenced with a one-day joint session with the CONSER Operations Committee. A special welcome to the new BIBCO OpCo representatives as well as to new CONSER members was given. The new BIBCO representatives are: Alice Jacobs (NLM), John Sluk (Oberlin), John B. Wright (BYU), Jimmie Lundgren (University of Florida), and Chris Mueller (University of New Mexico). New CONSER members in attendance are: Mary Grenci (University of Oregon), Alwyn Owen (National Library of Wales), Rennette Davis (University of Chicago), and Everett Allgood (New York University). Larry Alford, Chair of the Program for Cooperative Cataloging (PCC), recognized the presence of 3 members of the PCC Policy Committee: Carlen Ruschoff (University of Maryland), Glenn Patton (OCLC), and Robert Wolven (Columbia). Alford extended his thanks to the Cooperative Cataloging Team, John D. Byrum, Ruta Penkiunas, and the participants for the Program's success during the past year.

Metadata in PCC libraries

The agenda commenced with a discussion led by Les Hawkins, CONSER Specialist, Library of Congress' Serial Record Division. The lead-in preface was the definition of metadata to ensure that all participants were opened to the "same page".

For purposes of this discussion metadata "refers to cataloging codes, lists of data elements, or other schema that are used to create records or data that enhances the retrieval and interpretation of bibliographic resources in digital form."

Cornell uses Dublin Core to assist in searching the documents that make up its Library Gateway Help system. They have also used Dublin Core in CORC (Cooperative Online Resource Catalog). Other digital collections projects at Cornell use SGML (with a TEILite mark-up), the Dienst protocol, and EAD (Encoded Archival Description). Cornell has a clearinghouse node in the National Spatial Data Infrastructure called CUGIR. The metadata for these geospatial datasets is the Federal Georgraphic Data Committee's (FGDC) content standard. In other cases, Cornell uses locally created metadata structures, for example to serve numeric data and reports for the USDA Economics and Statistics System.

The following list enumerates only those unique metadata schemes that had not been mentioned heretofore; hence, the list for each is not meant to be exhaustive.

The University of Oregon is using visual resources for slide collections and added that they are working to develop a core list of data elements, regardless of the metadata scheme that is being used.

Columbia is working on specialized projects using a modified AACR2 format and is using a selector input template for digital resources called Digital Scriptorium.

The University of Florida is using finding aids on an access database. The Florida Center for Library Automation is using "NDLDT" (Networked Digital Library of Dissertations and Theses metadata).

The National Library of Wales reports using SGML, XML, and TEI.

The National Library of Medicine (U.S.) reported using Dublin Core for digital resources, and enhance Dublin Core for more permanent resources and for materials that have been archived.

BEAT (Business and Economics Advisory Team) at the Library of Congress reported using Dublin Core and free Web-based resources and have put a template into use designed by reference librarians.

Harvard University uses a union catalog of visual resources (VIA) and OASIS, which is an archival and manuscript finding aid using "EAD" (Encoded Archival Description).

Hawkins proceeded to involve participants by attempting to ascertain the kind and level of staff that are involved with either metadata applications and/or creating records for digital resources. Essentially, responses were quite mixed. Staff involvement seems to be at all levels and is a highly-collaborative function, often requiring the assistance of reference staff. It was pointed out that often the resource itself may require the use of a particular metadata scheme. Naturally, support functions and funding issues are vital for maintenance of the metadata. Abstracting and indexing services are creating the largest number of metadata schemes.

Questioning also focused on the individuals responsible at each institution for the cataloging of metadata. The response was varied; however, many institutions reported that mainly serials catalogers appear to be involved with the cataloging of electronic integrating resources, although there were some institutions that reported the use of monographic catalogers, and those that reported the use of both. As an interesting aside, it was noted that catalogers with an AV (Audio/Visual) background had the least apprehension about cataloging digital resources.

Hawkins did try to get a reading about when and if AACR2 will be used to provide description and access to resources such as Web sites, databases, e-journals, and other digital resources. Responses indicated that AACR2 will be used for such resources when it is published.

The metadata warm-up session seemed to produce the desired effect. Much enthusiasm was generated and participation involved many of the OpCo representatives. When asked how participants envision the role of PCC in helping to define the standards by which digital resources would be cataloged, the consensus felt that a subgroup of the Standing Committee on Standards could be tasked with developing some guidelines and that the PCC should become a "clearinghouse" for metadata standards and that no one particular scheme be preferred. It was also suggested that catalogers begin to look "outside" the box of technical services to help promote access to digital resources.

Action: PCC, via SCS to investigate need to define guidelines for digital resources cataloging. PCC Web site to be used as clearinghouse for various defined metadata standards.

Standing Committee on Automation (SCA)

The discussion moved to an update on automated classification given by SCA Chair, Karen Calhoun (Cornell). Classification is the last cataloging process to be significantly touched by automation and remains essentially an expensive manual process. Calhoun referred participants to the work done by the SCA Task Group on Automated Classification, chaired by Gary Strawn (Northwestern). Their report includes product specifications for ILS vendors to consider and includes three main recommendations; (1) that the ILS be able to produce a list of subject headings associated with a classification number; (2) that the ILS be able to produce a list classification numbers associated with a subject heading; and (3) that the ILS be able to report duplicate classification numbers.

Calhoun then referred participants to the questions prepared by Jeanne Baker (University of Maryland) in advance of the Operations Committee meeting. The questions could be helpful for assessing a library's readiness for and interest in automated classification. Calhoun queried the OpCo representatives if the enhancements suggested by the Task Group were available now, in what ways would changes take place in what is currently being done at each home institution? The overall feeling was that the output of cataloged titles would increase substantially. Encouraging the OpCo representatives to get the word out to the ILS vendors, particularly to the various vendors' Users Group meetings, Calhoun solicited assistance from attendees who would volunteer to publicize the need for classification automation.

Action: John B. Wright (BYU, SIRSI) and Jeanne Baker (University of Maryland, Voyager) will work with Calhoun to help with reaching vendors.

Calhoun turned her attention to OCLC batch processing. The SCA Task Group on OCLC Batch Processing has completed three surveys--one with BIBCO liaisons, one with CONSER OpCo representatives, and one with OCLC Users Council delegates. The Task Group's preliminary report (made available in advance of the OpCo meeting) reported that catalogers at BIBCO/CONSER libraries who are contributing Program records online in OCLC are happy with this contribution method. However, BIBCO institutions that batchload Program records are dissatisfied with the process, and enhancements to batchloading would be a significant help to these libraries. There was some interest in batchloading enhancements among CONSER respondents to the survey, but not at the level expressed among BIBCO respondents.

While the availability of batchloading enhancements does not appear to be a driving factor in who joins PCC or how much is contributed, such enhancements could further the goals of the PCC by making more BIBCO upgrades available for use in WorldCat. Currently, when batchloaded BIBCO upgrades match existing WorldCat records, only the library's holding symbol is attached, and the BIBCO upgrade itself is discarded. If batchload were changed, these upgrades would no longer be discarded.

In the past, OCLC Users Council delegates had expressed some reservations about the replacement of OCLC member-contributed records with BIBCO full and core records. The Task Group's Users Council survey was done to learn more about the delegates' concerns and preferences. Results indicated that over three-fourths of the Users Council respondents were either somewhat in favor or fully in favor of OCLC batchloading enhancements that would facilitate the contribution of Program records. Residual uneasiness about the enhancements may be the result of a general lack of familiarity with PCC core and full record guidelines among Users Council delegates.

Survey participants were asked to choose among a variety of options for replacing OCLC member-contributed records with Program records. A majority of respondents from all three surveys preferred a replacement option in which any batchloaded PCC core record would replace any less-than-full member record, and any batchloaded PCC full record would replace any member record. There was some support for an option in which data from member and PCC records would be merged, but it was clear that merging would be complex, difficult, and costly to implement. Calhoun reminded participants that some merging already takes place; when the OCLC batchloading process "bumps" a member record with an incoming record, subject headings and call numbers in schemes not present on the incoming record are retained.

Calhoun asked OpCo participants to help the Task Group with its recommendations to OCLC and the PCC Policy Committee. There was further discussion of the "replace" versus "merge" issue, with some OpCo representatives from CONSER libraries expressing support for data merging.

Action:The Task Group will make recommendations to the PCC Policy Committee and to OCLC, taking OpCo respondents' comments into account.

Utilities Wish List

The meeting then featured the initial utilities responses (both RLG and OCLC) to the PCC wish list of electronic enhancements designed to facilitate contributions of records. Ed Glazier, RLG, took center stage to tackle the eleven issues for BIBCO/NACO/SACO outlined in the background document. Cynthia Whitacre followed in turn with OCLC's response. For reporting purposes each of the 11 points will be enumerated with the respective responses from the utilities.

1. Validation of headings on bibliographic records; linked authority control.
   RLG: Plans to perform this function were abandoned years and ago and have not been revisited.
   OCLC: Experimenting with making the authority file available in OCLC's FirstSearch so that references can assist in end-user searching. OCLC already offers linked authorities in CORC.

2. Record distribution between OCLC and RLIN.
   RLG: RLG remains prepared to discuss this issue.
   OCLC: Equity of exchange is important to OCLC. OCLC plans to examine searching of other databases, the first of these will be to a database in the Netherlands; linking to RLG is a possibility to be explored.

3. Batch-loaded BIBCO records to overlay other records in OCLC.
   RLG: No comment necessary.
   OCLC: Is striving to improve on this capability and is working with the SCA Task Group on OCLC Batch Processing to define enhancements to the process.

4. SACO and classification workflow online in the utilities.
   RLG: Development not justified; not cost-effective; however, Glazier noted that RLG does not want to interrupt LC's mechanism of subject heading review.
   OCLC: SACO is under LC's control; OCLC is looking to LC to provide direction in how to streamline SACO proposal submission.

5. BFM: Revision of headings on bibliographic records in the utilities' databases.
   RLG: Not planning to implement and do. See 1 above.
   OCLC: "Good news"! OCLC is doing pro-active clean-up of headings and is experimenting with using reports of newly established heading which would eliminate the need for BIBCO libraries to report BFM. Global replace is not yet possible except in the context of linked authorities in CORC.

6. Series numbering should file/sort in the utilities as a PCC Task Group has recommended for vendors of local systems.
   RLG: No development is currently underway to implement.
   OCLC: OCLC does not currently display the contents of subfield $v in truncated lists of search results, but will consider it as our new platform is implemented.

7. Ability to import records from remote databases into the OCLC database.
   RLG: No comment necessary.
   OCLC: This is part of OCLC's Extended WorldCat strategy.

8. More search capabilities of the MARC21 tagging in records, including the 007/008 fields.
   RLG: Currently available through Web-based EUREKA and RLIN technical processing system.
   OCLC: OCLC responded that it is interested in clarification on these points (e.g., Which fixed fields are most useful?) and will be asking for input on the BIBCO- and CONSER- listserves.

9. The ability to see all in-process authority records in OCLC, as members can do in RLIN.
   RLG: No comment necessary.
   OCLC: Definitively no!

10. Retain a user-friendly interface.
    RLG: We are working to make EUREKA even more user-friendly.
    OCLC: Working hard at retaining a user-friendly interface. OCLC is moving in that direction, particularly with the move to the relational database.

11. Allow for bigger records.
    RLG: EUREKA has essentially no limits on record size.
    OCLC: There are no limits to record size in CORC, and as OCLC implements its new relational database, there will no longer be a limit to record size.

Robert Bremer(OCLC), spoke to the CONSER wish list categories listed as: 1) linking-related; 2) multi-part/multi-dimensional records; 3) maintenance-related; and 4) long-range.

Bremer stated that concerns with abilities to link to external databases is indeed the strategy on which OCLC's WorldCat is based and that links to other databases is a very real possibility. He added that OCLC is working toward making better use of existing links in records to bring together related records in catalogs. Bremer continued that with maintenance-related issues that more discussion at all levels would be required, particularly with input needed from other library users and not only technical services staff and that long-range planning should involve input from a myriad of sources. Bremer announced that the new OCLC platform with its new databases and database models should become available at the end of Summer 2002.

PCC Task Force on Multiple Manifestations of Electronic Resources

Jean Hirons reported on the findings in the final report of the PCC Task Force on Multiple Manifestations of Electronic Resources, chaired by John Riemer. Hirons summarized some of the work done by the PCC on works issued in multiple versions. Efforts have included 1) the development of the single-record approach for providing access to online versions through manipulation of the record for the print version; 2) the development of guidelines in the CONSER Cataloging Manual (CCM) which provide guidance on the use of separate records for electronic versions vs. the single record approach; and 3) working with vendors and aggregators of serials to provide access to bibliographic records contained in the vendor's products.

Hirons announced a promising development in the formation of a JSC-commissioned task force to study cataloging at the expression level, using the OCLC-Europe database for the sample. Jennifer Bowen will be chair of this group; JSC member Matthew Beacom will be a member as well.

An informal show of hands revealed what some BIBCO and CONSER members were doing to handle multiple manifestations. Some institutions are loading record sets from various aggregators. Attendees expressed having experience with other monographic record sets including Books 24 x 7, Lexus/Nexus Tripod, proQuest records, and sets from NetLibrary. Cornell is experimenting with harvesting data from digital resources to automatically create its own record sets for loading into its catalog. This alleviates a resource drain by relieving pressure on cataloging staff in that at least a brief record is created. The University of Florida is also experimenting with creating its own record sets for its digital collections.

Action: PCC Web site to be used as a clearinghouse for existing record sets that are available, including lists of commercially available products.

Hirons pointed out that CONSER assumes that the print record represents the first tangible manifestation of an item. Attendees agreed that sometimes the electronic version is actually produced as the first or primary manifestation and other format versions are secondary. In a case where a record for the print or other tangible format version does not exist, elements describing the availability of the print or other versions could be added to the record for the electronic version.

ISSN

Several issues relating to ISSN and multiple aggregator records were raised. Under current CONSER policy, separate records are created for online version of a single title simultaneously distributed by different aggregators. The ISSN network is assigning one ISSN to the print version and a separate ISSN for the online version; however, it is not assigning a separate ISSN to the online versions of a single title offered by multiple aggregators. If there is only one ISSN for the online version of a print title, which among multiple aggregator records should be used for recording the ISSN?

Regina Reynolds (LC, NSDP, the U.S. ISSN Center) suggested the ideal of being able to assign the ISSN for an electronic version based on the original electronic text produced by the publisher, creating a "master" record for the electronic version. This approach emphasizes identification of the electronic version of the work over description of individual aggregator online versions which might differ from one another in completeness, coverage, etc. It would allow libraries to maintain specific information about aggregator versions (subscription details, coverage, etc.) at the holdings level rather than the bibliographic level.

Some attendees described the local practice of using multiple ISSNs in multiple 022 fields (one for the print, one for the electronic version) when using the single record approach.

Action: CONSER to consider the impact of incorporating the use of multiple ISSNs into its single-record approach for national level records.

AACR2, MARBI, and Integrating Resources

An update on AACR2 and MARBI actions relating to Integrating Resources (IR) was provided by Hirons as well. The JSC has given basic approval to the revised chapter 12 of AACR2, which will have the title "Continuing Resources" with publication expected in 2002. Chapter 12 will provide rules for both successively issued serials (including series) and integrating resources.

An integrating resource will be defined in AACR2 as "a bibliographic resource that is added to or changed by means of updates that do not remain discrete and are integrated into the whole."

Hirons presented issues with MARC coding for IRs and reported on several MARBI documents to be discussed at ALA in June, including 1) Proposal 2001-05: Bibliographic level (Leader/07) code 'i' for integrating resources; 2) a discussion paper dealing with which 008 to be used with textual materials: serials or books; and 3) Proposal 2001-04: Repeatable 260 fields. Hirons noted that the decisions concerning bibliographic level 'i' that will identify a record as an integrating resource may give PCC program members more opportunity to perform maintenance on records for IRs and for distributing them. Should code 'i' be implemented for IRs, the determination would need to be made whether it should be used with the Serials or Books/008 field.

The Books/008 is currently in use because these records are coded as bib level 'm'; the continued use of Books/008 would not require that records be converted; however, the bytes seem less useful. Additional possible codes to be added if the Serials/008 is used to help describe IRs would be the code 'k' to identify a frequency of "continuously updated," the code 'l' to identify looseleaf publications, and the code '2' for identifying latest entry, similar to an update for an IR, as when a new title is recorded in the record. In her description of the MARBI proposal 2001-04, the possible use of the 247 field in records for IRs to record earlier titles was suggested. Hirons asked for a show of hand on which 008 should be used and there was unanimous approval of the serials 008.

Action:Compile a chapter on the cataloging of Integrating resources. Who: Volunteers will include: John Sluk, Naomi Young, and legal catalogers, Judy Kuhagen, etc. (depending on whether loose-leafs are included or given separate chapter.)

Note: PCC participants are encouraged to visit the JSC Web site for the latest updates and progress on changes to AACR2 beyond those discussed at this meeting.

Interim Report of the Task Group on Implementation of Integrating Resources

Valerie Bross (UCLA) gave an overview on the Interim Report of the Task Group on Implementation of Integrating Resources. The group focused on the needs for training, documentation, maintenance of records, and the distribution of records for IRs. It seems that the PCC is in a good position to create training materials and documentation for the cataloging of integrating resources. Since catalogers in both the BIBCO and CONSER Programs could conceivably be involved in creating records for integrating resources, it was suggested that documentation be available widely, perhaps as a module of the CONSER Cataloging Manual (and consequently available through Catalogers Desktop) and as a chapter in the BIBCO manual.

Action: A decision will be name whether there should be one large document for all IRs or whether separate CCM-like modules should be developed for looseleaf publications and electronic IRs. Other issues to be resolved include who should prepare the documentation, and when the documentation should become available.

Maintenance of records for IRs was discussed, including record distribution questions. Generally it was agreed that record maintenance for IRs would be important and should be shared among BIBCO and CONSER institutions. However, it was noted that BIBCO records are not simultaneously shared in both OCLC and RLIN, thereby making it difficult to share maintenance of all records for integrating resources. There was also uncertainty expressed about the mechanisms that would trigger the need to perform maintenance on IRs. Would routine staff scrutiny be required to monitor changes or would automated means (such as link checking software) be sufficient?

Action: Maintenance guidelines will be developed to give BIBCO and CONSER libraries an idea of the resources required to monitor and make maintenance changes to IRs.

Record distribution and sharing among the utilities also brought up the use of LCCNs for record authentication and record distribution in the CONSER database. It was suggested that automatic generation of LCCNs in OCLC may be a possibility. Perhaps this automated mechanism would allow more libraries (BIBCO and CONSER) to share in record authentication and the maintenance of IRs.

Standing Committee on Standards

Ann Caldwell (Brown), Chair, Standing Committee on Standards (SCS), followed with a report on the work of the SCS. Caldwell focused on the review of the core records for all formats and has made note of any discrepancies among the various core standards. As she reported in the previous day's meeting they have been grappling with the core record elements included in the record for cartographic materials. At issue is whether a single core record for all types of cartographic materials should be sufficient, which the SCS has indeed chosen to endorse. It has been determined that certain elements of the cartographic core do overlap with elements of other core records, most notably, the core records developed for computer files and serials core. Caldwell commented that the SCS may look to develop a core record with standard elements to be used across the board for all materials, regardless of format, with footnotes to supplement the core record with additional elements based on the type of material. No negative feedback was generated; however, it was noted that each core record also be maintained as it currently is.

Action: SCS to maintain separate core record standard, one with its complete array of elements accompanied by footnotes to supply additional elements based on the material, and a core record with all elements for each type of material without reference to the others.

Kay Guiles, (Senior Policy Specialist, CPSO, LC) presented a follow-up report concerning recommended changes to LCRIs from the SCS, Cross Reference Task Group's final report that had been submitted on December 17, 1999. The proposed LCRIs drafted in response to recommendations #8, #9, #10, and #13 are available on the CPSO web site. Comments on the proposed changes are to be sent to the CPSO e-mail account at cpso@loc.gov by July 20, 2001. LC's disposition of the remaining 8 recommendations can also be found in the LC report.

Standing Committee on Training

Carol Hixson, Chair, Standing Committee on Training (SCT) was the last speaker for the day. She began her report with the announcement that the SACO Participants' Manual authored by Adam Schiff (U. Washington) had been completed and was at the LC print shop and would soon also be available in print, on LC's Cataloger's Desktop, and via the SACO home page in PDF form. The PCC is deeply indebted to Schiff for this achievement which will greatly help to facilitate contributions to the SACO Program.

Next, Hixson applauded the work of the four SCT Task Groups ([1] the Task Group on Educational Needs of the Cataloging Community: Final report; [2] the Task Group on NACO Continuing Education: Final report in which one recommendation called for a revision to the NACO Training Manual to parallel the CONSER Editing Guide and the CONSER Cataloging Manual, which are both sources rich in examples; [3] the Task Group on PCC Participant and Training Documentation: Final report; and [4] the Task Group on Web-Based Training and Distance Education: Interim report. Hixson also noted the work of the joint SCS-SCT Task Group on Implementation of Integrating Resources and their Interim Report, which had been presented by Bross earlier in the day. Hixson wanted feedback from OpCo representatives whether or not it falls within the purview of the SCT to conduct basic cataloging classes.

The point was made that library schools are not addressing the current needs for technical services librarians; Hixson proposed that the PCC become a training and education hub for technical services and that the training/education be used as a recruiting tool, both to this aspect of the profession as well as to the Program. Hixson summarized the work of the TGs as calling for standardized documentation, following the CONSER model; a NACO Coordinator and a strong BIBCO Coordinator (equivalent to the CONSER Coordinator); a collection of local cataloging documentation with links from the various PCC homepages; and a personnel resources inventory for BIBCO participants, in which strengths of individual BIBCO members could be tapped when needed. Hixson stated that such an inventory had been alluded to during ALA Midwinter and was perplexed that it has not been pursued. A CONSER participant commented that the CONSER documentation does not obviate the need for consulting AACR2, the LCRIs or other official policy documentation. John Byrum (LC) commented that a restructuring in the Cooperative Cataloging Team was in the pipeline that would help strengthen the various PCC programs.

Action: Hixson will develop a survey instrument to garner the personnel resources of BIBCO participants. The responses would then be keyed into a database and maintained. A decision on whose responsibility it is to maintain such a resource will need to be made.

Follow-up on Integrating Resources issues:

The second day of the BIBCO OpCo meeting opened with a follow-up on the integrating resources discussion from the previous day's joint session with the CONSER OpCo. This discussion was designed to develop a response to the recommendations for the cataloging and maintenance of IRs and to provide further discussion points for the ALA Annual BIBCO-At- Large meeting.

GPO stated that as a result of the change in how a serial publication will now be defined, that many of the materials in the collection will fall under the new IR definition. GPO is moving to the cataloging of a virtual collection.

UCLA reported that they are using CORC to catalog their Web resources collection the maintenance of which is an ongoing issue.

Northwestern has been a member of CORC since its inception and maintains that the growing number of electronic resources has increased communication between serials and monograph catalogers. Catalogers there have no time for maintenance of Web resources on a record-by-record basis, and a concern was voiced about knowing when/if the record in the OCLC database represents the latest. Northwestern uses no link checker.

It was pointed out that OCLC does provide the following service: If the 856 field in the master record be modified, the subsequent record will be re-distributed to those libraries that subscribe to that service. Carol Hixson, University of Oregon, pointed out that the maintenance discussion is not limited to the 856 field.

Queensborough Public Library does not catalog loose-leaf publications. Maintenance for IR records is only done when it comes to the attention of the cataloger; however, concern was expressed at the increasing number of updates that may need to be done as the volume of IR records grows.

The University of Florida has a maintenance unit that runs a monthly link checker program; if the original record was created by the University of Florida, then the record is updated in OCLC; however, if the record is done as copy, then it is only "fixed" locally. The University of Florida also expressed ambiguity in deciding when the item in hand is indeed an IR; for example, if an item is a not-regularly issued piece, is it to be considered as an IR?

The National Library of Medicine (NLM) provides cross training to cataloging staff at time of hire. NLM is eager to provide as many links as possible in an effort to be perceived as a "virtual library"and is committed to maintenance in providing for a position of a systems cataloger employed to check and verify links. NLM pointed out that if managers want links in the catalog, then they must "buy-in" to the maintenance it entails.

The BYU spin on this issue is that BIBCO participants must be willing to update all records, regardless of the record's originator or owner. BYU wanted it noted that as an RLIN member that feedback is superfluous.

Brown University asked if para-professionals could be employed in link checking. Brown also reported the hiring of "proctors" for this purpose. The cataloging of IRs are done by monograph catalogers where changes are perceived as "mistakes" and maintenance issues are not deemed to be terribly troublesome.

At Queensborough Public Library para-professionals are used in addition to 2 full-time catalogers who work on cataloging Web resources half-time, particularly resources in Chinese and Korean, with some in European languages as well.

Brooklyn School of Law reported using graduate students to maintain links.

At this point, Carol Hixson, refocused the OpCo attendees to look at the gamut of integrating resources, and not just Web sites and electronic resources. Are BIBCO participants willing to accept the task of updating IR records.

Harvard University places responsibility for the cataloging and updating of loose- leaf publications with the monograph catalogers, while serials catalogers are tasked with the responsibility for electronic resources.

At this point in the discussion, a question about CORC and cataloging standards was raised. It was also noted that not all Web resources are defined as integrating resources.

The University of New Mexico expressed the feeling that the responsibility for maintenance be shared equitably across the board. UNM has recently re- named its serials unit the Electronic Resources Section. They find a challenge in linking across formats and determining what are considered related versions, especially if the frequency of issuance is not shared across versions.

For purposes of this discussion, it was emphasized that Chapter 12 of AACR2 will be reversed and expanded.

Cornell has cataloged IRs for several years and reported statistical findings about the use of electronic resources at Cornell. Electronic resources are likely to comprise less than 10% of the Cornell collection materials budget and less then 3% of the number of titles; however these resources account for more than 30% of the use of the collection (based on circulation statistics plus e-resource connections). Clearly, these are very important resources to library users and they deserve catalogers' best efforts. At Cornell there is no staffing dichotomy between serials and monographic catalogers; the operation is "format blind." This has been an advantage for cataloging electronic IRs. Electronic IR records are routinely maintained by technical services staff. Cornell's use of CORC is in a hiatus because of the implementation of a new ILS (Voyager), but they do plan to resume using CORC, and they are considering the possibility of using Dublin Core for the cataloging of electronic resources.

Calhoun (Cornell) in her role as Chair of the SCA, proceeded to give her perspective on the integrating resource discussion. Calhoun felt that the IR model proposed by the TG on Implementing Integrating Resources was based on assumptions that are losing their validity. She postulated that it is unrealistic to assume that the concept of a single, highly standardized, centrally distributed file (the CONSER database), managed by a small group of contributing institutions (the CONSER and BIBCO libraries), would be a viable one for IRs, given the current and projected growth of electronic IRs, their volatility, and the growing array of sources and choices for metadata describing these resources. The proposal ignores the presence of CORC and the emergence of library portals and products like Endeavor's ENCompass and Ex Libris' MetaLib. These developments can be expected to add to pressures that are driving substantial changes in the role of MARC and AACR2-based utility databases and library catalogs and eroding the catalog's place as the single tool for library resource discovery and access.

Calhoun suggested that the TG abandon the CONSER model for authenticating and maintaining electronic IR records. A better model for IR record maintenance in WorldCat might be one that includes OCLC Enhance libraries. She noted that the "PCC stamp" on a record, although a sign of quality, limits the librarians permitted to lock and replace the record in OCLC. Futher, the proposed IR model leaves RLIN users out; this is unacceptable and should be reconsidered. Concluding her remarks, Calhoun said she felt the PCC's role should be centered around the development of training/education resources for the bibliographic control of electronic IRs.

Action: Cristán will forward Calhoun's concerns to the TG on Implementation of Integrating Resources for their consideration.

Core Record Revision

Ann Caldwell, Chair, Standing Committee on Standards (SCS), followed-up the previous days' discussion with the BIBCO OpCo participants on the core record standards revision. She again stated that the work of the SCS is focused on the review of all core record standards. She looked for direction in deciding if elements of some core records be included in other core records and whether or not footnotes should be used. Attendees expressed dissatisfaction with the cartographic materials core record. It is difficult to integrate elements from different core records into what a cataloger might need to describe a particular item, but the cartographic core, as currently defined is much fuller than what had been originally anticipated. Hixson felt that it was useful to have all the needed elements in one record and moved to include all elements in to the core for cartographic materials.

BIBCO Survey Preliminary Results

Calhoun (Cornell) provided the preliminary findings of the BIBCO survey that has been conducted by David Banush (Cornell) for the PCC Policy Committee. The study was designed to look at the way that the BIBCO program is presented to the world with the central research question focused on "why aren't there more core records?" A distinction that was easily discernible is that the perspective about BIBCO between managers vs. catalogers is quite different. The managers expressed a high level of satisfaction with the core record and are open-minded about its use in cataloging; however, with an attitude of "not in my back yard" when it came to its use in the home institution's catalog. Catalogers expressed happiness when core-copy was found; however, catalogers were reluctant to produce or create the records at core level. An apparent misunderstanding still persists about what a core level record requires, particularly in relation to the one subject heading and one subject heading only mentality. NLM was not satisfied that the core-level record is perceived/defined as the "default."

Calhoun asked OpCo participants to consider how the BIBCO program had been marketed in the past, and how it should be marketed in the future, taking into the account the results of Banush's study. She summarized the "four Ps" of marketing (product, price, position and place) and the two main reasons that new products/services fail--that is, the "claim" of the marketing program is either 1) not true or 2) not believed. Calhoun suggested that the initial marketing for BIBCO was centered on the core record and based on the claim "more, better, faster, cheaper." BIBCO marketing, which targeted both technical services managers and practitioners, promulgated the values shared by the founders of the Program--values often shared by library managers, but not necessarily by practicing catalogers. The Banush survey results suggest that the BIBCO marketing claim has not been believed, especially by practicing catalogers, resulting in continuing confusion about and reluctance to accept core records.

Calhoun then asked the OpCo representatives to consider the current marketing model for BIBCO and help her provide feedback to give David Banush as he develops his recommendations for the PCC Policy Committee. In the ensuing remarks, participants noted the misunderstandings that persist about BIBCO core and full records. The definitions of each, but especially of full records, are slippery. The quality of BIBCO records continues to be questioned. There was some support for the suggestion that BIBCO communications be revised to moderate the emphasis on core, replacing it with more emphasis on catalogers' judgment and on a key value-added aspect of PCC records--that is, authorized headings supported by NAF and SAF authority records. There was also support for revisiting and revising the BIBCO trainers manual. Bob Wolven suggested that having the answer to the question "what are the unique benefits of the BIBCO program to working catalogers?" would help a great deal toward a successful redefinition of BIBCO's marketing strategy.

Action: Banush will be invited to present his findings and draft recommendations at the BIBCO-At-Large Meeting at the upcoming ALA.

BIBCO Working Group on Series Numbering

Andreas Stamm (Northwestern), next provided an update on the BIBCO Working Group on Series Numbering recommendations. Recommendation no. 3, that MARBI and vendors work together on the subfield $v sorting possibility has been ignored. There appears to be a possibility with programmers for this sorting capability. Recommendation no. 4 is being developed into a discussion paper which is to be sent to Rebecca Guenther, LC, MARC Standards and Network Development Office; however, as yet no proposal has been prepared.

Judy Kuhagen, CPSO, Senior Descriptive Policy Specialist, also was present to convey the follow-up on WGSN's recommendation no. 2. She noted that the JSC has allowed LC to implement early the new wording in AACR2 Appendix B.5. The LCRI will allow for the abbreviation for 'volume' or 'vol.' to be consistently abbreviated to 'v.' in series statements which will help avoid the need for 490/830. Kuhagen reported that because the amount of bibliographic file maintenance (BFM) would be vast, LC has decided not to perform the retrospective BFM that this change would entail, resulting in a split file.* Kuhagen went on to note that the LCRI would contain cut-off dates for these files. Alice Jacobs (NLM), strongly objected to the implementation decision without the BFM portion.

*LC has since determined that the BFM will be performed and no split file will be created; however this will mean that all NACO participants will be asked to perform some amount of maintenance when handling SARs. Guidelines will be issued to all series contributors with implementation of this LCRI to begin August 31, 2001.

BIBCO Participants' Manual

BIBCO OpCo members provided excellent feedback on the draft of the BIBCO Participants' Manual (BMP). A summary of the contributions were issued on the BIBCO list immediately following the meeting and comments were solicited from the BIBCO membership at large. The draft manual can be found on http://darkwing.uoregon.edu/~chixson/bibco/outlinerev.html

Some specific suggested changes to the BPM included: 1) addition of a table in MARC21 field tag order listing where help is available for values defined for that field; 2) expansion of the section on BIBCO Documentation (A4.5.2) to include more detail and examples and to cite the documentation used, especially those using MeSH, NLM, SuDocs, Sears, Dewey, etc.; 3) need to reiterate principles regarding authority control; 4) need to reiterate principles of subject/language knowledge, i.e., it is expected that BIBCO catalogers will not contribute records as BIBCO when they lack sufficient knowledge of the subject or language to be able to do appropriate content analysis; 5) reiterate that the definition of being independent in NACO is language specific, i.e., it is expected that BIBCO catalogers will not contribute NACO records in languages they do not know.

Action: GPO will provide an example of SuDocs classification and NLM will help edit section on their tools. Hixson will solicit help from BIBCO members using other classification and subject schemes, via the BIBCOlist, for more details and examples to fill out Section A4.5.2.

FAQs

Ana Cristán asked the group to look at "to-do" lists from previous BIBCO OpCo meetings in an effort to discern that the BIBCO Operations Committee was carrying out what had been asked. It was suggested that all BIBCO trainers keep a record of problems that arise during each training and that these questions be added to the BIBCO FAQ.

Action: Cristán in consultation with the appropriate groups to develop BIBCO FAQ based on previous years' discussion and suggestions. Trainers listserv to be used as a testing ground for questions in a quiz show like format.

Action: Cynthia Whitacre to consult with OCLC on availability of funds to provide a reward to winner of BIBCO quiz with opportunity to attend OpCo meeting.

Length of committee membership also was a topic of keen interest to the OpCo representatives. A consensus of opinion maintained that it is consistent with other offices that BIBCO OpCo members be elected for a three year period in an effort to maintain continuity and integrity of the Committee.

Action: Cristán will ask the PoCo to change the governance document to change length of OpCo member incumbency from 2 years to 3 years.

The last item for the meeting was the announcement that next year the BIBCO/CONSER Operations Committees meetings would be held on May 1-3, 2002 in Washington, D.C.