Access Level Record for Serials¹ Working Group Final Report

July 24, 2006

Executive Summary

Declining library budgets, competition from Internet search engines and information services, and the escalating costs of cataloging have caused libraries to emphasize ways to meet user needs while decreasing costs. The Access Level Record for Serials Working Group was formed to develop a single CONSER standard record that would apply to all formats (print as well as online2), replace existing multiple record levels, and reduce serials cataloging costs by requiring in serial records only those elements that are necessary to meet FRBR user tasks. Cost savings and user benefits could also be realized by recording the elements in a way that is more straightforward for the cataloger to provide and easier for the end user to understand. As the name of the record implies, the emphasis of this record is on access points rather than elaborate and often redundant description. The record is intended to be a "floor," to which additional elements can be added if such elements are essential to meeting FRBR user tasks for a specific resource or to meet the needs of a particular institution.

The Working Group used a list of AACR2 and MARC 21 elements to identify those data elements of highest value for meeting the FRBR user tasks: find, identify, select, and obtain. The group also sought input from catalogers about record simplifications, and incorporated many of these ideas into cataloging guidelines to accompany the element set. The group tested the element set and cataloging guidelines via pilot projects at 14 U.S. libraries. Pilot projects included a cataloging component and a review component. During the cataloging component, the times it took to create records at both access and control levels were recorded. During the review component, reference, acquisitions, cataloging, and other staff reviewed the records and answered survey questions about the records' ability to meet users' and reviewers' needs. The Working Group then adjusted the element set and guidelines based on this feedback.

Time saved varied by institution, ranging from lows of 5% and 14% (Stanford, Harvard) to highs of 30-35% (UCLA, Georgia, GPO, LC) on the 8 bibliographic descriptions all libraries did in common. The average time saved in the creation of serial bibliographic descriptions was 25%. There was also evidence of a "learning curve" in using the access level guidelines. Cataloging times improved as a cataloger created more access level records, so future savings are anticipated to be greater.

Reviewers of the access pilot records were positive about the simpler, "less cluttered" records and the clearer way some information was presented. Significant concerns were raised about omission of place of publication. As a result, place was made a mandatory element, and guidelines developed for problematic cases. If place had been mandatory in the pilot element set, acceptability by reviewers would have ranged as high as 83%. Some concerns were raised about the overall lack of detail in the records and about other missing elements such as the "distinguishing" uniform titles used on serials, but none of these concerns rose to anywhere near the level of concern over not including place of publication.

Following positive reactions to the preliminary report of the Working Group presented and discussed at the CONSER Operations meeting in April 2006, the Working Group met at ALA Annual in June 2006. The Group worked on finalizing the element set and guidelines, agreed on a target implementation date of October 1, 2006 [likely to be adjusted to November 2006], and developed the additional implementation recommendations found at the end of this report.

Back to Top

Introduction

The dogmas of the quiet past are inadequate to the stormy present. The occasion is piled high with difficulty, and we must rise with the occasion. As our case is new, so we must think anew and act anew. ~Abraham Lincoln~

In the above quote, Lincoln was describing a momentous and stormy time in our nation's history. Nonetheless, Lincoln's words seem appropriate to the subject of this report because we are certainly facing a momentous and stormy time in the history of bibliographic control in the United States. At this time it seems, indeed, that "the dogmas of the quiet past are inadequate..." Development of a single, significantly streamlined standard record for the cooperative cataloging of serials is only the beginning of the thinking and acting anew that will be required for what will likely continue to be a stormy present and an even more stormy future for bibliographic control. It was in this spirit of coping with the present and preparing for the future that the Working Group for the access level record for serials began its work.

"I like the idea of simplifying records; it helps the average patron to have a cleaner, uncluttered looking record; and it helps those who work on and with the records to pare down non-essentials."

"... most users don't look at all the extra stuff we put in [records] anyway."

"I am a fanatic for detail; I need to have, or feel that I have, every possible detail in order to do the best work. Probably in 99 out of 100 reference desk transactions, an access level record would be sufficient; but I still prefer to have as much detail as possible."

~Pilot Project Survey responses from reference staff~

The above quotes illustrate the wide spectrum of opinions reference librarians have about catalog records. Given the fullness of standards such as AACR2 and MARC 21, it is not surprising that some librarians--catalogers and reference librarians alike-think the best catalog record must be the fullest record. On the other hand, this project demonstrated that records that are quicker, easier, and more cost effective for catalogers to create and easier for users to understand can do a reasonably good job of meeting FRBR user tasks.

Back to Top

Background

The project to develop an access level record for serials grew out of two discussions that took place at the May 2005 Joint BIBCO/CONSER Operations meeting. At that meeting, interest was expressed in extending the access level record developed at LC for monographic and integrating remote access electronic resources3 more widely, including to serials. The second discussion concerned the CONSER Operations Group's interest in replacing the multiple and sometimes confusing record levels currently used for CONSER cataloging with one standard record to which CONSER institutions or catalogers could add elements as needed by the institution or as appropriate to the resource.

An August 2005 PCC call for volunteers resulted in the formation of a Working Group as well as a preliminary list of potential pilot project institutions and others interested in the project. The charge given to the Working Group was to build on the work already done at LC and to assess serial user tasks, develop an essential data set and cataloging guidelines, design a pilot project, create a test set of records by applying the data set and guidelines, evaluate the results of the pilot, and if the pilot was successful, to use the pilot results to recommend a single standard record for CONSER. Further background information and the text of the charge to the Working Group can be found in Appendix A (PDF, 76KB).

Back to Top

Record Objectives

The Working Group (13 librarians from 8 institutions) met at the Library of Congress during two two-day sessions in October and November 2005 to assess user tasks and develop the mandatory element set and cataloging guidelines. Subsequent work was done by email and by meeting at ALA Midwinter and Annual. The Working Group began by agreeing on the following objectives:

Functionality: The record should function as a single "CONSER standard," (or "CONSER-authenticated") record, replacing existing levels. It should be a "floor," not a "ceiling," and should include sufficient information to meet FRBR user tasks. Catalogers should be free to add those additional elements essential to meeting FRBR user tasks that might be required by the particular nature of the resource or particular needs of an institution. The record should function in library ILSs, shared catalogs, and metasearch systems.

Cost-Effectiveness (overall cost savings could be achieved if at least two of the following three objectives were met): The cost of access level record

1. creation
2. maintenance
3. cataloger training should be less than the comparable costs for full level records.

Conformity to Current Standards: Rules used to create access level records should conform to-or be acceptably compatible with-current standards for descriptive cataloging (AACR2), subject access (LCSH, etc.), and content designation (MARC 21), and should enable record distribution to MARC subscribers. The Working Group's recommendations regarding cataloging rules, R.I.s, and practices can be found in Appendix M (PDF, 118KB). MARC 21 recommendations are in Appendix N (PDF, 75KB).

The Mandatory Element Set

The Working Group took a user-focused and systematic approach to determining a mandatory element set for the access level record. Rather than starting from a full record and arbitrarily discarding elements based on group members' gut feelings or personal preferences, the access level record for serials--as was the access level record for electronic resources-was built from the ground up. During their in-person meetings at LC, the Working Group scored a "core data set" (the extensive table of AACR2 and MARC 21 data elements found in Appendix C (PDF, 624KB)) according to each element's value (high or low) to a user's ability to meet the FRBR user tasks: find, identify, select, obtain. The core data set the Working Group used was adapted for serials by Ed Jones from the core data set that Tom Delsey developed under an LC contract to support the access level record for electronic resources.

The resulting mandatory element set that can be found in Appendix D (PDF, 182KB) is comprised primarily of only those elements that were scored "high" in terms of their value to the user. A few additional elements were added to meet library processing needs (internal control numbers and fixed field coding) and shared cataloging efficiency (e.g., a note on the issue used for cataloging). Some notable omissions from the list of mandatory elements include: place of publication except in limited cases [restored as a result of pilot project findings], many bytes in fixed fields, most "distinguishing" uniform titles for serials, all but a few notes, and many redundant added entries and other redundant elements. It should be emphasized that no changes were made to current "full level" practices for providing name and subject access points. This is one crucial difference from the "core level" record where the expectation was that subject headings would be limited.

Back to Top

Summary Of Mandatory Elements

Selected Leader and Fixed Field codes
Control or Identification numbers: (ISSN, LCCN, CODEN) and 042 code
Main entry
Abbreviated title
Titles: title proper, variant titles, former titles
Edition statement
Publisher
Place (originally in limited cases, later made mandatory)
Extent (field 300) required except for print and online resources
Current frequency
Date/designation (all unformatted)
Specified notes: source of title, DBO, LIC, reproduction, system details (limited), language notes, index notes
Subject and name added entries
Most linking fields
Series added entries (only if SAR is made)
Series statements (required only if no SAR is made)
URLs (as specified)

Back to Top

Summary of Omitted Elements

008 22: except for microforms
008 18-19 (frequency, regularity)
Distinguishing Uniform Titles (except for generic titles, monographic series)
Other title information, statements of responsibility (generally)
Parallel titles from 245 (retained in 246)
Place of publication generally (later reinstated)
Added entries that duplicate linking fields
Series statements (if SAR and added entry are made)
Series added entries (if no SAR is made)
Extent (field 300) not required for print and online resources
Formatted beginning and ending volumes and dates (362); all will be unformatted
Many notes, including 321, 580, 550, 440
006 and 007: all but 1st 2 bytes
730, 740, 787

Back to Top

Cataloging Guidelines

Feedback from LC catalogers assessing reasons for time saved creating access level records during LC's remote access electronic resources pilot indicated that one reason for time saved was "'in case of doubt' rules in guidelines that provided the freedom to decide and move on" 4. Accordingly, part of the charge to the serials Working Group was to develop cataloging guidelines. In order to identify cataloging practices or decisions that take up catalogers' time without resulting in corresponding benefits, brainstorming sessions with catalogers were held at various libraries around the country and input was solicited from catalogers via a survey posted on CONSRLST.

The resulting ideas, compiled in Appendix F (PDF, 267KB) 5, gave the Working Group a foundation for developing a set of cataloging guidelines that provide rules of thumb intended to reduce or eliminate cataloger "agonizing" over certain decisions such as establishing corporate headings, and determining major vs. minor serial title changes. The guidelines also include simplified instructions for how and when to provide most of the elements that are affected by new or changed practices. For example, the cataloging guidelines inform the cataloger to provide beginning and ending volumes and dates in all cases by using a simple, "Began with... ended with" note instead of following the former complex instructions and decision-making steps for recording beginning and ending dates and volumes based on whether the cataloger has the first issue of a serial in hand or not. The Working Group concluded that such a distinction has no meaning for the user and only serves to complicate cataloging and cataloging training, and make the resulting records less predictable and less comprehensible for the user.

The goals set by the Working Group for the cataloging guidelines were:

• Eliminate or minimize redundancies
• Use system and system-display capabilities more fully
• Provide guidance to expedite cataloger decision-making
• Allow for omitted elements (e.g., place) to be supplied by publishers or others [this proved premature and place was made mandatory]
• Make records clearer for users
• Use language consistent with the "floor" approach by stating, "It is not required to..." rather than "Do not..." when giving guidance about elements that are not to be provided routinely

In accordance with these goals and in keeping with the philosophy that this record emphasizes access over description, the Working Group set out to require as few notes as possible and make the required notes as predictable and standardized as possible. For cataloging in a cooperative environment, "description based on" and "latest issue consulted" notes were deemed essential, but to remove the current equation that determines when to use such notes or not, the notes are required in all cases. Although this decision went against the goal of avoiding redundancy, it advanced the goal of simplifying record creation and training. The Working Group felt that the decision-making time and complexity that were saved by this new practice would more than make up for those cases in which one or the other note was redundant. The Group also recommends in Appendix L (PDF, 130KB) (Display Issues) and Appendix N (PDF, 75KB) (MARC 21 Recommendations) that new MARC tags or indicator values be explored to allow "cataloger specific" notes to be suppressed in public OPAC displays.

System details notes, formerly required for all electronic resources (e.g., Mode of access: World Wide Web) are required only in non-routine cases. Notes justifying added entries or documenting that a resource is the official organ of a body are no longer required. Notes that duplicate a linking entry field are no longer required. Appendix L (PDF, 130KB) lists possible changes to MARC 21 that could allow for even fewer notes.

Back to Top

Pilot Projects

A crucial part of the project's charge was to test the element set and guidelines via pilot projects. Unlike the access level record for electronic resources that preceded it, the serial record was designed in the context of a cooperative program and thus pilot project volunteers were sought from the CONSER community. Fourteen institutions participated in pilot projects that took place in spring 2006 to test the creation of access level records by catalogers and evaluate the records using reference, acquisitions, systems, and other library staff. Pilot participants were: Columbia University, Harvard University, Library and Archives Canada, U.S. Government Printing Office, Library of Congress, National Agriculture Library, National Library of Medicine, Oklahoma State University, University of Florida in a joint project with Stanford University, University of California at Los Angeles, University of Chicago, University of Georgia, and University of Washington. Each institution was asked to assemble a project team consisting at least two catalogers plus reference, acquisitions and other staff who could review and compare the completed records.

Catalogers at the participating institutions used the mandatory element set and cataloging guidelines to catalog a group of serials at both access level and at a control level that represented the institution's usual practices. Each institution was asked to create a total of 28 records: 8 common titles that were cataloged at both access and control levels and 12 institution-specific titles, of which 6 were done at access level and 6 at control level. There was a requirement that no cataloger should catalog the same title at both levels. In order to keep the focus on the new type of description, and to simplify record creation, no classification, subject headings, or authority records were added to or created for the 8 common titles. Instructions used for the pilot projects are in Appendix G (PDF, 63KB).

A Web-based survey instrument (Survey Monkey) was used to collect both structured (time needed to create records) and unstructured data (mostly free-text comments) from pilot catalogers and record reviewers. Data collected from catalogers included information about the job category of the respondent, the language and format of the record, and the time it took to create each access level record and each control record. Data collected from reviewers included the job category of the respondent, the respondent's assessment of the reviewed records' ability to meet FRBR end user tasks and the respondents' own job task needs. For the full text of the survey questions, see Appendix H (PDF, 101KB).

Back to Top

Pilot Project Results

Detailed pilot project survey data can be found in the following appendices: Appendix I (PDF, 114KB). Cataloging Survey Data, Appendix J (PDF, 158KB). Reviewer Survey Data, Appendix K (Excel, 253KB). Survey Data Excel Spreadsheets. Sample pilot project records can be found in Appendix P (PDF, 62KB).

38 catalogers at 14 institutions created 327 records over 5 weeks as follows:

• 167 access level records
• 160 control records (done according to the institution's usual practices)
• 256 records (78%) were original
• 71 records (21%) used OCLC member copy
• Bibliographic descriptions (without subject analysis or authority records) were created for 8 titles that all institutions cataloged in common, to facilitate comparisons
• Complete records were created for 12 titles of each institution's own choosing
• 8.6 = average number of records created per cataloger
• 8 = median/mode number of records created per cataloger

Back to Top

Time Study Results

Bibliographic descriptions for all titles

• Average time for 167 access level records: 23.7 minutes
• Average time for 160 control records: 30 minutes
• Time saved: 21% (6.3 minutes/record)
• Time saved varied by institution with time savings ranging from lows of 5% and 14% (Stanford, Harvard) to highs of 30-35% (UCLA, Georgia, GPO, LC)

Bibliographic descriptions for the eight common titles

• Average time for 88 access level records: 23 minutes
• Average time for 87 control records 30.6 minutes
• Time saved: 25% (7.6 minutes/record)

Complete records (including subject analysis and authority record creation)

• Average time for 86 access level records: 35.1 minutes
• Average time for 84 control records: 49 minutes
• Time saved: 22% (9.8 minutes/record)

Cataloging times improved as a cataloger created more access level records.

The Working Group's preliminary projection is that the access level record will result in a saving of 20-25% of the time needed to create complete serial records.

A more extensive presentation and analysis of cataloging survey data can be found in Appendix I (PDF, 114KB). Complete raw cataloging survey data, including all cataloger responses can be found in an Excel spreadsheet included in Appendix K (Excel, 253KB).

Back to Top

Record Acceptability

• Records were reviewed by 88 reviewers at 13 institutions
• Reviewers represented: reference: 41%; acquisitions: 23%; systems: 13%; "other":23% (cataloging staff, supervisors, cross-functional positions, etc.)
• Biggest success: access records' beginning/ending data was more understandable
• Biggest concern: missing place of publication (over 40% of reviewers objected)
• If place were included in records, overall acceptability would rise from 48% to between 71% and 83%
• Place is now included in the mandatory element set, with limited exceptions and simplification guidelines
• Display issues were mentioned often as affecting how users view records: improved OPAC displays could save cataloging time and improve user acceptability

A more extensive presentation and analysis of reviewer survey data can be found in Appendix J (PDF, 158KB). Complete raw reviewer survey data containing all reviewer responses can be found in an Excel spreadsheet included in Appendix K (Excel, 253KB).

Back to Top

Other Survey Data

Cataloger results and reactions: For almost every cataloger, access level records were faster to create. For serial catalogers already proficient in creating full records, there was a learning (or "unlearning") curve, such that records created later in the project took much less time than early records. Some catalogers noted that the documentation used in the pilot should be fleshed out for production use and would be improved by the inclusion of examples of access level records. However, augmenting the documentation will need to be done carefully, since one of the project goals was to have greatly simplified documentation. See Appendix O (PDF, 249KB) for sample documentation pages.

A significant cataloger comment that was repeated at several different institutions was that the access level record was "liberating!" Apparently, creating a record according to CONSER documentation and standards is felt by some catalogers to be a heavy burden. Other selected cataloger comments: "There's no question, access level records take less time to create." "Not having to create uniform titles was great. Not having to identify and transcribe place of publication was a time-saver." Regarding work with existing copy, one cataloger wrote,"Doing an access level update to a non-access record can result in an odd mix of inclusions and exclusions."

Reviewers' positive reactions: There was much positive reaction from reviewers to simplifying catalog records and making them more "user-friendly." E.g., "There is less clutter on the Access record and most users don't look at all the extra stuff we put in anyway." Spelling out abbreviations was viewed positively. One comment read, "easier...since it does not require extensive notes." Many catalogers and trainers agree that composing notes takes many catalogers a significant amount of time. Most appreciated was the change from varying ways data about beginning and ending dates and volumes were presented in control records to the access level record's simple "Began with... ended with..." statement.

Reviewers' negative reactions: By a small majority (52%), reviewers felt that the access level record as exemplified by the pilot records they examined was not sufficient to meet their needs. More than half of those reviewers expressed concern about place of publication not being a mandatory element. If place were made mandatory [which has been done subsequently], it would appear that a significant majority (between 71% and 83%) of reviewers would find the record adequate. Some reviewers missed uniform titles in cases where titles were otherwise identical. This seemed somewhat dependent on local system displays. Others missed indications of illustrations, maps and size.

Other concerns that were raised in the review of the records were that complex resources require more information than that specified for the access level record; that not providing some information on the record might result in more need to retrieve material from collections to distinguish one resource from another; that this record might not provide sufficient information to support scholarly research; that lack of cross-checks, e.g., justification of added entries, might allow some errors to go undetected; and that training of future catalogers to a lower standard could impede their knowing when to go beyond the access requirements. One UCLA reviewer summarized his concerns in this way, "I'm not concerned about UCLA as long as we have Melissa [Beck] and Valerie [Bross], but I hope implementation of access level does not lead to poorer serials cataloging elsewhere."

Back to Top

Pilot Study Factors

It should be kept in mind that the pilot projects were artificial in ways that could have affected the outcomes, particularly in a negative way. The study was designed so that each institution cataloged the same 8 "common" titles, plus 12 titles drawn from their own workflow. The Working Group felt that by asking each institution to catalog the same 8 titles, time comparisons would be more valid. However, the common titles included types of resources with which some catalogers were unfamiliar such as online resources or medical titles. In other cases, catalogers were working in an unfamiliar environment (e.g., NLM catalogers worked in OCLC, not their usual practice). Finally, the project design and instructions, as well as the element set and guidelines for the access level record, were completely new to all catalogers in the project. Nonetheless, significant time savings did occur.

Some possible reasons for the time savings might be found in the survey data. For example, 22.8% of access level records omitted a "distinguishing" uniform title that was required on the control record. 32.9% of records were for online serials where place can be time-consuming to find. Although place is now mandatory, if the cataloger cannot locate a place of publication within one page of a home page or issue page, "[S.l.]" will be recorded. One intriguing possible explanation for time saved is the removal of the "fear factor" and agonizing that some catalogers experience when creating full CONSER records in the CONSER database. The lifting of this burden was evidenced by the catalogers who called the access level record "liberating."

Back to Top

Applicability to Copy

Since OCLC member copy is a common factor in CONSER cataloging, the Working Group deliberately included guidelines for the use of copy in the cataloging guidelines and included resources with copy in the pilot projects. It was anticipated that the use of copy might pose significant challenges since it might not be easy to determine what to add to existing copy, what to update, or what (if anything) to discard from copy. For some of these reasons, copy was excluded from the preliminary implementation phase of LC's access level records for electronic resources.

In practice, copy proved to be less of a challenge than anticipated. Catalogers were instructed to retain all correct (or presumed to be correct) data whether such data were part of the access level mandatory element set or not. All data known to be incorrect were removed and only replaced with correct data if that data were required for access level. Outdated data that would be misleading (former frequencies, former system requirements) were removed and only replaced if required for access level. Some reviewers observed that full records used as copy at access level can result in "odd mixtures of included and omitted data." However, the Working Group felt that as more records begin at access level, or will be maintained at access level, inconsistencies should be minimized over time. During the pilots, few questions arose about the use of copy.

Back to Top

Display Issues

Throughout all phases of the project, one word seemed to surface frequently: "display." The word came up in the context of brainstorming sessions with catalogers, in analyzing elements from a FRBR perspective, in drafting the mandatory element set and cataloging guidelines with an eye to eliminating redundant elements, and in comments from reviewers of the pilot project's records. Many reviewers commented about the elements that were displayed in full, brief, and index displays, even when the access level record had no effect on those elements or displays. Display of records in local systems seemed to have a great impact on the acceptability of access level records (in fact, on catalog records in general). When catalogs provided displays that incorporated elements that reviewers felt were needed for identification and suppressed other elements, reviewers were much more positive about the access level record. Clearly, much work is needed to develop a better relationship between data in catalog records and how that data is used to generate useful and user-friendly indexes and displays.

For example, uniform titles that serial catalogers create with much agonizing in order to distinguish same and similar titles could be replaced by index displays that pull publisher, place, and date from records and add this information to title displays, such as is done in some (e.g., OCLC), but not all, systems. If coded data in fixed fields (e.g., language, frequency, etc.) could be translated by systems into user-friendly notes, many redundant elements in catalog records could be eliminated. If it were possible to suppress notes of interest to library staff from public displays, shorter, more understandable records could be displayed to the public. If data from linking fields (earlier and later titles, language, geographic, and other medium editions) could be used to create combined displays from multiple related records, users would be more likely to find the serial title and format that best matched a citation or met their other needs.

To summarize, if OPACs had better display capabilities, including better abilities to fully utilize data that catalogers create and tag, cataloging could be further simplified, while at the same time, user needs for "cleaner," more user-friendly catalogs and catalog records might be more readily met. A list of display issues is included in Appendix L (PDF, 130KB). This list will be made available to the PCC Committee on Automation with the recommendation that these issues be investigated further and the most promising be pursued with ILS vendors.

Back to Top

What's in A Name?

Although the proposed new record has been referred to throughout this report as "the access level record" or "the CONSER access level record for serials," the Working Group does not recommend retention of that name or label because the of possible negative effect such a label will have on record acceptance. The charge to the Working Group was that following a successful pilot and subsequent adoption, this record should become "the single standard CONSER record." Thus the record does not need (and should not have) a name that implies that this record is a lesser record or a "shortcut" record of any kind. Such a stigma will haunt the record, just as the "core" label and the continuing presence of a "full" level have seriously impeded acceptance and use of the core standard. Anecdotal evidence indicates that even pilot participants and those who have attended presentations about this record in which the goal of a "single CONSER standard" record was stated, still wonder how they will be able to distinguish "access level" records from "real" or "regular" CONSER records. Both the name and the coding of this record need to make emphatically clear that this is "The" standard CONSER record6.

A related issue concerns the encoding level and authentication codes that will be used for this record. As we have seen with the core record, once a record can be labeled as less than full, many questions arise about how to determine the correct encoding level and authentication code under various circumstances, and when to change the encoding level and authentication codes if cataloging from copy, or when subsequent modifications are made to records. In creating a single CONSER record, the Working Group's intention is to eliminate the cataloging time, discussion time, documentation pages, and "errors" that result from issues relating to encoding levels and authentication codes so that catalogers can focus on those aspects of the record that involve meeting FRBR tasks. See Recommendation 3 below.

The Working Group understands well the resonance that the word "access" has at the present time: many administrators are focusing on access issues; the word figures prominently in name of the future cataloging code, Resource Description and Access; and the word is useful in communicating a new philosophy to catalogers. However, we feel that these goals can be achieved by embedding the philosophy of access into the introduction and training for this record without risking the negative reactions that might result from any label that implies the record is yet another attempt to arbitrarily "dumb-down" or "downsize" catalog records.

Back to Top

Recommendations

1. The access level record for serials should be adopted as the single, standard "CONSER record," replacing the existing full and core standards. (See recommendation 3 below regarding minimal level records.)
2. The record should not require a distinctive name or label but simply be referred to (and become) the CONSER record (lower case "r").
3. PCC/CONSER encoding levels and authentication codes should be re-examined and simplified in light of this record and to accommodate other needs (e.g., common work by both programs on integrating resources, the LC series decision). This examination should include an assessment of MARC 21 national level record requirements for full level records. Retention of minimal level as it is currently used in CONSER (e.g., for ISSN records, and by some institutions, including LC, for certain categories of resources) is recommended. It is anticipated that the PCC Policy Committee will develop a further charge to pursue this task, appoint additional BIBCO members to the current Working Group, and that the resulting group will replace the task force originally formed under the old PCC Tactical Plan objective 2.1.2 that has been "on hold" pending the outcome of this serials project.
4. The record should be implemented as soon as feasible. October 1, 2006 was recommended but work on revising encoding levels and authentication codes may delay implementation until November 1, 2006. Several institutions have expressed interest in as immediate an implementation as possible.
5. Documentation should consist of a "CONSER Metadata Application Profile" (See Appendix O (PDF, 249KB) for sample pages) that combines the mandatory element list and the cataloging guidelines for each element in a single table. The documentation should also include a short introduction containing the philosophy and rationale for the record and some appendices, e.g., the decision-making guidelines used in the pilots, and examples of complete records (requested by several pilot project catalogers).
6. Documentation should be freely available from the CONSER Web site, and should also be included in Catalogers Desktop; CONSER documentation should be adjusted as part of regular CONSER updates.
7. Training should take place locally before implementation at each CONSER institution. Additional training to accommodate non-CONSER institutions that desire to implement this record will include a session (already planned) that will take place at ALA Midwinter meeting of the ALCTS Continuing Resources Cataloging Committee. A new SCCTP class and distance learning modules could also be developed or current classes and modules could be adapted for the new standard.

Back to Top

Conclusion

In times of change, learners inherit the Earth, while the learned find themselves beautifully equipped to deal with a world that no longer exists.~Eric Hoffer~

There can be little doubt that we are living in times of change, both within and outside the library universe. It is hoped that the development and implementation of this new standard record for serials will lead serials cataloging down the path of learning and change required to remain cost effective while providing significant value (e.g., controlled name access and controlled vocabulary subject access) that cannot as yet be provided in any other way. The Working Group recognizes that the changes required by this new record may challenge the flexibility of some of our colleagues and may seem to some to devalue the meticulous work and documentation that past practices required. However, many of the ideas embodied in this record were proposed by practicing catalogers, and almost half of the Working Group consisted of catalogers. Many of the catalogers' proposals were supported by the analysis that the Working Group performed to evaluate elements according to FRBR user tasks. While there can be some justifiable sadness at leaving behind what might one day be regarded as a "golden age of cataloging," such a move seems almost overdue in light of the current economic and digital environment. The Working Group therefore urges the speedy implementation of the recommendations in this document as a crucial step towards avoiding the possibility that catalogers might otherwise "find themselves beautifully equipped for a world that no longer exists."

Back to Top

Appendicies

• A: Background and Charge (PDF, 76KB)
• B: Participants (PDF, 89KB)
• C: Core Data Set (PDF, 624KB)
• D: Mandatory Element Set (PDF, 182KB)
• E: Cataloging Guidelines (PDF, 171KB)
• F: Brainstorming Results (PDF, 267KB)
• G: Pilot Project Instructions (PDF, 63KB)
• H: Pilot Survey Questions (PDF, 101KB)
• I: Cataloging Survey Data (PDF, 114KB)
• J: Record Review Survey Data (PDF, 158KB)
• K: Survey data spreadsheet (Excel, 253KB)
• L: Display Issues (PDF, 130KB)
• M: Cataloging Rules/R.I.s/Practices Recommendations (PDF, 118KB)
• N: MARC 21 Recommendations (PDF, 75KB)
• O: Sample draft pages: CONSER Metadata Application Profile (PDF, 249KB)
• P: Sample records (PDF, 62KB)

Back to Top

Notes

1. Because "access level record" was part of the charge , this name will be used throughout the report. However, the Working Group strongly recommends that the record be called simply "the CONSER record. [Back]
2. The access level record developed for LC by Dave Reser only applies to remote access electronic resources. [Back]
3. LC Implementation Plans for Access Level MARC/AACR Records. http://www.loc.gov/catdir/access/accessrecord.html[Back]
4. Reser, David. Presentation "Defining an Access Level Record for Remote Access Electronic Resources" (PPT) made at the 2005 ALA Annual Meeting in Chicago for the ALCTS/CCS Continuing Resources Cataloging Committee. [Back]
5. This compilation will be made available to the PCC Standing Committee on Standards and the CONSER ad-hoc RDA comment group as a source of ideas that might be pursued in order to realize further benefits and simplifications.[Back]
6. There will still be a need for a minimal level. See recommendation 3 [Back]

Printer Version (PDF, 329KB)

To view PDFs

Back to Top