CONSER Operations Committee Meeting

April 27-28, 2006

Summary  

Welcome and Introduction: Joint BICO and CONSER OpCo Meeting

The joint meeting of the BIBCO and CONSER Operations Committees opened with a welcome from the Chair of the PCC, Mark R. Watson (University of Oregon). On behalf of the PCC Policy Committee, Watson thanked the members of the CONSER and the BIBCO operations committees for their expertise and service on the operations committee and thanked their respective institutions for supporting that service. Watson also thanked the Library of Congress and the Cooperative Cataloging Team for hosting the meeting as well as the facilitators and note takers selected to conduct the five strategic decisions breakout sessions later in the morning. Watson then acknowledged the topic uppermost in most attendees minds: the recent Library of Congress decision to cease creating series authority records as part of Library of Congress Cataloging on May 1, 2006 [later changed to June 1, 2006]. Watson stated that despite this decision the survival of cooperative cataloging was not in jeopardy.

Presentation: Sharing Metadata and Future Directions of PCC Cooperation Sharon Tsai (LC); Robert Bremer (OCLC); Luiz Mendes (UCLA)

Sharon Tsai (LC) opened this part of the joint meeting with a presentation of copy cataloging at the Library of Congress, including the historical background and the essential process of copy cataloging at LC; the different types of copy cataloging at LC, encoding level 7 copy cataloging and the pilot project for same; new proposals for copy cataloging; and the use of resource records. Tsai’s presentation can be accessed at: http://www.loc.gov/catdir/pcc/bibco/Tsai_BIBCO_06.pdf.
Robert Bremer (OCLC) spoke on the sharing of metadata, stating that OCLC is based on the MARC format, the information community creates records using many different and often incompatible metadata schemes, including schemes not based on a set of rules or standards. The records thus created meet local needs but because of the lack of standardization may not meet cooperative needs. Bremer suggested that the PCC might wish to define a baseline or guidelines (as opposed to a set of rules) for non-MARC metadata.
Luiz Mendes (UCLA) presented “If it is good for PCC, then it is good for me!” “It” or “cooperative leadership” is comprised of the process, the record, the tools, and the players. Each of those four components presents a series of questions useful to the ultimate question “Where does PCC go from here?” Mendes’s presentation can be accessed at: http://www.loc.gov/catdir/pcc/bibco/Mendes_BIBCO_06.pdf.

PCC Strategic Directions 2010: Introduction, Background, and Process

In preparation for the breakout of the attendees into five groups for the purpose of discussing the five PCC strategic directions 2010 and possible actions to realize the goals, Watson explained that the five strategic directions had their genesis with the PCC Policy Committee (PoCo) at the November 2005 PoCo meeting. The present stage, seeking input from the PCC operations-level representatives on possible action items, is prelude to PoCo meeting in November 2006 where the five draft strategic directions 2010, the goals, and the action items will be discussed and finalized.

PCC Strategic Directions 2010: Breakout Sessions and Action Item Reports

The attendees of today’s meeting then broke into five groups:

Strategic Direction 1: Be a Forward Thinking, Influential Leader in the Global Metadata Community
Action Items SD 1

Strategic Direction 2: Redefine the Common Enterprise
Action Items SD 2

Strategic Direction 3: Build on and Expand Partnerships and Collaborations in Support of the Common Enterprise
Action Items SD 3

Strategic Direction 4: Pursue Globalization
Action Items SD 4

Strategic Direction 5: Lead in the Education and Training of Catalogers
Action Items SD 5

[Consult http://www.loc.gov/catdir/pcc/draftstratdir.html for the complete Draft Strategic Directions for the Program for Cooperative Cataloging (PCC)]

The five breakout groups spent approximately 30 minutes identifying possible action items for their respective strategic directions. All action items were recorded on flipcharts. Another 30 minutes were then spent with each group’s facilitator reporting on the action items so identified. Consult "Action Items" links above for the five strategic directions and action items identified during the breakout sessions.

CONSER Operations Meeting April 27, 2006

Les began the CONSER operations meeting with introductions of new members and representatives:

Shana L. McDanold, Saint Louis University, Pius XII Memorial Library, a new member (Margaret Smith trainer/reviewer)

Amanda Louie, Kara Hyde of Serials Solutions, new member.

Melanie Watts is a new representative of Ebsco.

New developments: UC Funnel implementation; San Diego State University project to add LCSH in NLM records.

Limited latest entry for title variants
1) The paper under discussion resulted from the review of RDA part 1 and was intended to gather feedback from CONSER and the Continuing Resources Cataloging Committee (CRCC). If there was enough interest among these groups in the idea, a proposal would be forwarded to the JSC.

Comments:

One pro not mentioned in the paper is that use studies show the current title is cited more often than earlier titles, although that may vary by the subject focus of the serial.There is confusion between practices for integrating resources and serials, moving toward latest entry for serials would make the practices for these resources the same. During RDA discussion, several catalogers mentioned that a limited use of latest entry has been used for sometime in German catalogs, it would be useful to get more details on how they have implemented it and how they have made use of a "benchmark title" for judging title changes.

UCSD reported that its public services staff were consulted about the limited latest entry proposal and didn’t see a big advantage over our current practices of using 246 for current variant titles. Another comment from public services staff at a CONSER institution: we don’t care where the title is given, if the acquisitions display cannot clearly show the current title, it is a problem of the system or system implementation. Indexing can provide access to the title, so making the display clearer to the user expecting to see the current title is what is needed.

Other comments: Couldn’t you use a 246 with an indicator value for displaying the current title with a preface of “current title.” The re-description implied in the examples involves too much re-cataloging.

Minor title change rules, especially where the type of resource is added or dropped within the first five words seem especially problematic. [The operations committee did not make a recommendation for changes in these rules.]

Decision: The group felt that more effort should be focused on improving user displays. The Standing Committee on Automation might be a resource to consult on this issue, whether to explore a means of recommending a “current issue” indicator for the 246 field or some other solution. The group also felt it worthwhile to get more information from the German cataloging specialists to see how the approach is implemented there.

Action: The CONSER group that put the paper together will submit a proposal to the Standing Committee on Automation for exploring ways to improve displays of current title and other elements in serial records. The group will also get in touch with the German cataloging specialists on the RDA-L list to get further information about how limited use latest entry is used.

Discussion of linking practices: (Ed Jones, National University)

Question 1) Should we change our practices on multiple links to stop limiting linking fields for certain classes of relationships? There are two sets of instructions in the CONSER documentation in question:

a. Under CEG, "Linking entry fields, general instructions" Special instructions-Numerous related records: If numerous related titles have a common title, make a 580 note that contains the common title and give the record control numbers of each of the related titles in a single linking field. The example for this practice involves numerous geographic supplements. If the there is no common title, only a 580 note is given, no linking field.

Discussion: This was probably devised as a convenience to catalogers and to reduce the number of fields in a record under earlier OCLC field record limits. The recording of multiple record control numbers in one linking field doesn't provide a machine link to other resources and there are currently now no limits on the number of fields that can be added to the record.

Decision: Make this practice optional.

b. CCM 31.16.1 Multiple linking relationships, gives an example of a print serial that for some issues is published simultaneously with an online version and then is continued by the online entirely. The instructions say to give the link for the primary relationship and explain the multiple relationships in a 580 note.

Discussion: Sometimes giving two links to the same title is confusing in an OPAC display, likely this was the origin of this policy. Is it a problem with local implementation of displays, decisions about indexing certain linking fields or not? It would be nice to limit use of 580 as much as possible and take advantage of the $i in linking fields to make displayable notes.

Decision: Make adding multiple links optional, there is still a need sometimes to use the 580.

Question 2) Should we change guidelines in record requirements for links (including access level records). It was decided to roll this discussion into the access level for serials discussion later in the meeting.

Question 3) Should CONSER restrict use of 580?

Discussion: In many cases, 2nd indicator values for display constants and $i notes can be used instead of a 580 to express the linking relationships. The group felt that this was preferable to using 580 notes, but recognized that there were times when it is necessary to use a note to clarify a relationship. This led to discussion of the practice of coding the first indicator of linking fields 1 when a 580 note is used to explain a complex relationship. The Task Group On Linking Entries Final Report recommends reexamination of this practice because "While this practice prevents redundant notes, it also defeats the goal of hyperlinking." Members decided that it is probably time to vote on weather to change CONSER practice in coding this, that is to prefer to always code the first indicator in the linking entry field as 0 whether or not a 580 note appears in the record.

Action: CONSER documentation will be changed to reduce usage of 580 uses in favor of using “i” and display constants.

Action: A statement based on the proposal of the Task Group on Linking Entries Final Report will be developed and CONSER members will be polled through email on accepting it or not.

Who: Les, with help from other members for wording.

OCLC e-Holdings Record Authentication Discussion

Adolfo Tarango (UCSD) and Robert Bremer (OCLC) gave some background and an update on the project. It grew out of a reference enhancement service OCLC developed and the records that were a bi-product. Distribution of these records through the CONSER file was proposed as an enhancement for subscribers using the file to build e-resource knowledge bases. In April, a group was put into place to look at the first hundred or so records that were authenticated by OCLC. OCLC is in final stages of enhancing records created earlier in the year with information from the Openly Informatics knowledge base, so there will be a delay in authenticating more of these records. OCLC will provide another set of authenticated records for review later in May.

Some of the issues discussed by the group included 856 fields that do not work or do not link directly to the resource it cites. The group considered recommending that the 856 field not be provided in these records. 856 fields were provided originally to be sure that at least one 856 was on every record, this is something that can be reconsidered if the 856 fields are not useful.

For resources where there is not an existing record or resource representing a 780 or 785 relationship, the title of the related record is given in the $t of the linking field without record control numbers. This is a common practice with other CONSER authenticated e-serial records.

Robert mentioned the possibility of being able to provide reciprocal links in the records when performing the authentication.

Decision: We need to document our practices with the records authenticated through this project, the record elements and expectations for CONSER members in terms of consolidating and upgrading them. This documentation should be publicly available on the CONSER web site.

Action: Les and the group will draft guidelines for describing the project and the record elements.

ISSN Revision

ISSN Revision

Regina Reynolds (LC) gave an update on the revision of the ISSN standard and explained the concept of the“linking ISSN” (ISSN-L). ISSN-L is a new name for what has most recently been called the Medium-Neutral ISSN. The purpose and function remain the same: ISSN-L is a function or mechanism for providing a collocating ISSN to link among different media versions of a continuing resource. It provides a way to support OpenURL services and interoperability with other identifiers, DOI, URN, etc. To enable ISSN-L to function as intended, the ISSN Network is exploring the development of look-up and distribution services so that OpenURL knowledge bases, utilities, and other databases can be populated with current and correct ISSN, ISSN-L, and ISSN relationship information. The business model for these new services is still being determined, but it is likely that the service will be a subscription product since the ISSN Network derives much-needed revenue for its operations from product sales.

For newly-assigned ISSN, the ISSN-L will be the same as the first medium-specific ISSN assigned to a continuing resource. The ISSN International Centre will also retrospectively designate an ISSN-L for every continuing resource in the ISSN Register. Because of the way ISSN are distributed to the ISSN centers around the world, and because of the rules for designating ISSN-L when only some titles in a cluster of medium-specific ISSN change, no algorithmic way to determine the ISSN-L is possible.

Each ISSN record will contain its own medium-specific ISSN as well as its ISSN-L, as separately tagged elements. At this point a determination of where the ISSN-L will reside in the MARC record has not been made. Use of field 024 and development of a new subfield in field 022 are both being considered; other options could also emerge. Proposals will have to be made to the appropriate bodies for both the MARC 21 and UNIMARC formats.

A revised version of the standard has been submitted to ISO for registration as Draft International Standard 3297. It is expected that the 6-month voting period on DIS 3297 will begin by the end of May 2006 and it is hoped that publication can take place in the first half of 2007.

RDA discussion

Kris Lindlan (University of Washington) and Paul Weiss (UCSD) talked briefly about PCC review of RDA part 1. There were two PCC channels for comments, the CONSER review group and the PCC Standing Committee on Training. Some international members of PCC felt that it would be inappropriate to try to compile a comprehensive PCC response, as many international members were bound to submit comments through their own country’s constituent review structure. There is some uncertainty about how many of the 400 pages of CC:DA comments were included in the final response by the CC:DA. The draft of part 2 is expected in July.

PCC and expanded use of non-Latin scripts

Dave Reser (LC) gave an overview of the impact of OCLC’s implementation of Cyrillic, Greek, and Hebrew on PCC members. The PCC implementation date for these scripts is June 1, 2006 to allow time for notification to CDS customers. Since LC and CDS continue to distribute MARC-8 records, scripts outside of the repertoire are still a problem for LC, so more work needs to be done before Thai and Tamil CONSER records can be distributed. Reser's presentation is available from: http://www.loc.gov/catdir/pcc/bibco/Reser_BIBCO_06.pdf

Joint BIBCO/CONSER meeting April 28, 2006

Series Authority Control and LC’s Decision to Cease Creating Series Authority Records as Part of Library of Congress Cataloging

On Friday morning, Maureen Landry, acting PCC secretariat and chief of the Library of Congress Serial Record Division, provided background information on the decision-making process that led an LC policy advisory group to recommend that LC cease creating series authorities records as part of its bibliographic records, effective May 1, 2006. The determination was made based on the results of various scenarios presented by LC’s Cataloging Policy and Support Office; the results of an analysis of the number of series and bibliographic records created in fy05; and consideration of budgetary and staffing limitations. One finding of the analysis was that a substantial amount of cataloging time could be saved in the area of series control. The desire to implement a policy that would be “unambiguous and unexceptional” led to the decision to discontinue tracing all series at LC as part of its cataloging.

Audience members asked about delaying the implementation in light of the need to revise the LC Rule Interpretations (LCRIs) and Descriptive Cataloging Manual guidelines (DCMs) and the utilities’ need for time to determine if a different workflow is necessary for handling LC records. Participants expressed concern about LC records overlaying PCC members’ records, especially when the copy cataloging workflow at LC will call for 440s and 490/830s to be changed to untraced 490s. Some wondered how this decision affects LC’s relationship with and participation in the PCC.

As a result of the discussion, participants voted to make the following recommendations to the PCC Policy Committee:

The BIBCO and CONSER operations committees recommend to the PCC Policy Committee:

1) That it [the PCC Policy Committee] officially communicate to the Library of Congress the concerns that have arisen on our email lists and in discussion at our meetings about LC’s new series policy;

2) That the PCC evaluate series practices in the following areas:

1. Simplify series authority creation and documentation to support it. Areas that should be covered include:

a. Bib maintenance
b. Unambiguous distinction of strong versus weak series
c. Selected series authority control or total

2. Assess importance of series authority control
3. Cost-benefit analysis of both original series control work and copy cataloging
4. Establish a discrete time frame for this evaluation process

OCLC Report

Robert Bremer and Cynthia Whitacre (OCLC) reported that the 100 millionth record will be added to WorldCat in the near future. Other news included the implementation of Bibliographic level “i” for integrating resources, additions to the MARC character set (degree sign, phonogram copyright mark, musical sharp, inverted question mark, etc.), the inclusion of both 13-digit and corresponding 10-digit International Standard Bibliographic Numbers (ISBNs) in the 020 field, and the addition of Thai and Tamil scripts for use with Connexion client 1.50. More details are available at http://www.loc.gov/catdir/pcc/bibco/OCLC_BIBCO_06.pdf.

RLG Report

Ed Glazier (RLG) reported on the implementation of conflict checking of authorities in the RLIN21 environment. Reporting of conflicts to the Library of Congress is being resumed as a result. Conflict checking of LC/NACO authority files will occur daily as each new distribution file is loaded with weekly checks of the LC/SACO authority file. More information about Glazier’s remarks can be found at http://www.loc.gov/catdir/pcc/bibco/RLG_BIBCO_06.pdf.

Standing Committee Reports

Standing Committee on Automation (SCA)

Gary Charbonneau (chair, Indiana University) reported on the progress of the Monograph Aggregator Task Group, which is charged with developing a reference guide that lists required date elements for machine-derived and machine-generated monographic records. The committee has produced a draft of the vendor guide. SCA’s other task group, the Task Group on Normalization, has completed its charge to produce a draft report on all aspects of the normalization issue. The full report is accessible at http://www.loc.gov/catdir/pcc/archive/SCA2ndQ2006.pdf.

Standing Committee on Standards (SCS)

Paul Weiss (chair, UC, San Diego) announced that his group had completed its report on the draft of Part I of Resource Description and Access (RDA) and submitted it to the Committee on Cataloging: Description & Access (CC:DA) for review by the Joint Steering Committee (JSC). The complete report is available at http://www.loc.gov/catdir/pcc/scs/SCS2ndQ2006.pdf.

Standing Committee on Training (SCT)

Caroline Miller (chair, UCLA) noted that the Joint PCC/CCS Task Force to Develop Series Training had completed drafts of all modules for a training-the-trainer program that will be held in April 2007. The Task Group to Update the SACO Participants’ Manual Work was expected to complete its work by April 2006, but is slightly behind schedule. The online Introduction to SACO course is expected to be finished by July 2006. More details on Miller’s report can be found at http://www.loc.gov/catdir/pcc/archive/SCTQRApr06.pdf.

Repeatable 260, coding guidelines (Les Hawkins, CONSER Coordinator)

Practices surrounding the repeatability of $c drew many comments. It was suggested that publication dates be given only in the first/earliest publisher statement or that the date be given only in the first and the last. The use of bracketing dates on issues other than the first or last and use of angle brackets in some cases was questioned. Serials cataloging practices for bracketing differ from multi-part monograph, rare serials, and integrating resources cataloging; promoting one practice over another will mean that one cataloging community’s practices will need to change. Some wondered how ILS vendors will implement the changes and whether RDA will give guidance for providing clear user displays for changes in publishing information.

Decisions: We’ll have more information soon about RDA from the recent JSC meeting and will look at this to see if it has any impact on the draft 260 guidelines. Also since a number of operations meeting discussions involved OPAC displays and how vendors and libraries implement them, it would be worthwhile to ask the SCA to consider some of these issues in exploring guidelines for improved user displays.

Integrating Resources discussions with BIBCO (Les Hawkins, CONSER Coordinator)

Distribution of Integrating Resources:

Status of implementing bibliographic level “i”: An LC implementation plan has been put into place for distribution of serials and integrating resources in one combined CDS distribution file. BIBCO and CONSER members contributing to OCLC will be able to cooperatively maintain records for integrating resources. The next steps towards this distribution plan will happen after OCLC implements the bibliographic level code “i” in June. Several implementation issues remain for making distribution possible, including consideration of authentication codes and the development of documentation.

Provider neutral record for electronic integrating resources

Members were asked to respond to the proposal made at last years CONSER operations meeting. At first, when this was sent to various lists in August 2005, there was not a high level of response, but recently there has been more and more interest expressed by several institutions in moving toward a provider neutral record for IRs.

Action: Les will ask the PCC Steering committee to convene a group to find out what it would take to implement the change.

CONSER Operations Meeting April 28, 2006

Access level record for serials (Regina Reynolds, LC, Diane Boher, NLM)

Regina Reynolds noted that there were differing views from reference staff on the evaluation of the access level records. Some clearly embraced the advantages of the pared down record which gave most users an uncluttered view of the data. Another reviewer, recognizing that while the access level record would work “in 99 out of 100 reference transactions” prefers a record with as much detail as possible.

The guidelines included eliminating or minimizing redundancies, making fuller use of system-display capabilities, and providing for the possibility of vendor or publisher supplied data being added at a later time. An appendix to the final report will include advice for implementing displays from the records. There may be a recommendation for new linking field functionality so that 580 can be omitted.

The record requirements are designed to work with a serial resource in any format. Distinguishing uniform titles are omitted from the mandatory elements. In future guidelines, uniform titles for translations and language editions will be removed. Many notes, 321, 580, 550, 787 are not required.

There were significant savings of nearly 20% in the time it takes to create the access level record compared to control records. The sample used for analysis included mostly original records with some copy. Catalogers did not remove data from cataloging copy. A learning curve was identified and it’s expected that more time and costs savings will be realized as catalogers become used to the guidelines.

Training is an important consideration and Diane Boher noticed that while experienced catalogers needed to hold themselves back in providing a fuller level of information, new catalogers easily applied the guidelines. There is the question of how new staff trained on these guidelines will recognize when it’s appropriate to apply fuller elements when appropriate.

The decision to omit place of publication is being rethought based on feedback from reviewers. It is clear that with some types of resources, place is easily omitted, especially online serials where its often difficult to find a place of publication and serials from multinational publishers that list multiple places. There are other situations where place is useful for matching and identifying the resource, particularly if no distinguishing uniform title is provided, so it will be important to include a place of publication for these serials.

In discussing the importance of improving OPAC displays, Regina asked how we get vendors on board to help us design and implement displays that are clearer? One thought was to ask reviewers in the access project to design sample displays or provide requirements in a display appendix.

Comments: Often a distinguishing uniform title with place qualifier is out of synch with later place changes, so its value is questionable. Similarly, use of the term online as a uniform title qualifier isn’t used for all online serials, only in cases where the title conflicts with a print counterpart.

The uniform title however is also a place to control the title access points for many serials in added entries on other records, etc., where will that function take place without the uniform title? There is the possibility of using a citation title or key title for this purpose.

What about other resources, can the access level record be applied to monographs? This is a good way to think about any resource, what record elements do users of various types need to access the record?

On not coding frequency/regularity in the fixed field, OCLC uses the these bytes for matching, though its true that they are not coded in all records. Will there be an impact on batch manipulation of records and matching? OCLC will take a look at this.

Some members expressed discomfort with the loss of detail in the record. The CONSER standard record is a floor and can be added to and enhanced if needed.

For the most part, there were favorable comments on the access level record from the operations representatives. Most representatives were comfortable with using the term “CONSER standard record” rather than to continue to call it the access level record for serials. The final report is due after ALA and this will be sent to the PCC policy committee for approval of as a new CONSER standard. There are some recommendations to be made to MARBI and the JSC for RDA, and a number of implementation issues, including encoding level, authentication code, use of copy, and training/documentation issues.

DLF/OCLC Registry of Digital Material (RDM)

Decision: Since the RDM guidelines for recording precise holdings notes that “… As OCLC does not use MARC21 format for holdings data, fields 533, 534, 583 and/or 856 may be used…” and now OCLC has implemented the MARC21 format for holdings, it was decided that the CONSER group discussing this with the DLF will develop two scenarios, both based on the provider neutral approach to take back to the DLF:

1) One would provide for multiple 533 fields to show precise holdings

2) One provides for showing precise holdings in the MARC21 holdings format

It was felt that later, CONSER should pursue a provider neutral approach for microforms.

Update on SCCTP (Hien Nguyen, CONSER Specialist)

All of the workshops were updated except the holdings workshop in 2005. Since OCLC has implemented the MARC21 holdings format, there is an opportunity to add some additional examples from OCLC to the holdings workshop.

Hien Nguyen (LC), Linda Geisler (LC), Lisa Furubotten (Texas A&M), and I-chene Tai Le (Moyne College) participated in a pilot course offered by Amigos for developers of online training workshops. Linda Geisler mentioned her interest in testing the learning management platform used by Amigos to provide training for LC’s overseas offices.

Steven Miller (University of Wisconsin, Madison) has developed an online version of his cataloging integrating resources workshop and is piloting it the first three weeks of May.

Documentation discussion:

Bruce Johnson (CDS) talked about the CDS/CPSO survey results. Updates to several publications will be made available for free in PDF download beginning with the 2006 subscription year. Some comments from the survey pointed to the desire to combine the CEG and the CCM.

Comments: The CEG has been in the same format for quite some time, there is some information primarily of interest to members in sections A-C that could reside on the Web to make them available. Some of the specific field sections are largely repeated in the MARC documentation. Perhaps some CONSER and LC decisions or practices could be incorporated into the MARC documentation. Making links between sections of the two documents in Catalogers Desktop would be a way to begin to merge the electronic versions of the CCM and the CEG (and could probably be done sooner rather than later.)

The impact of implementing the CONSER standard record and RDA are going to call for us to change CONSER documentation and training material. The CONSER standard record points to the need for simpler, more direct guidelines. RDA may also provide the opportunity to make simplifications in our documentation.

The next CONSER Operations meeting will be held on May 3-4th.