| The single record approach recommended in the CDL Cataloging Guidelines
represents the cataloging standard employed by most serials catalogers
at this date. It is a practice that has worked well for a number
of years in describing variant formats, the primary variant format being
micro-reproductions. Digital resources represent a new paradigm. In the
context of ever-changing content at web sites, a vigorous ongoing national
debate is presently taking place regarding the adequacy of the single record
approach in addressing digital content, this in the larger context of a
wide-ranging, formal consideration of the future of AACR2.
Although the single record approach continues to be preferred by a number of catalogers because it represents the present prevailing national practice for serials, the growing number of content differences in digital resources means an increasingly large number of resources are judged to be "exceptional" and are cataloged separately. More catalogers now support a consistent, more efficient separate record approach, with print records copied and edited to reflect content differences. In the instance of aggregator databases, a recent survey by CONSER reflected the growing divergence of preference for single vs. Separate record (or other) approach: Which access method are you using now? (Check all appropriate)
Which access method would you most like to employ in your institution?
(Check one item)
Comment: Half the people who chose "none of the above" explained that they did so because the survey required a single answer, and they used the comment section to indicate they wanted a combination of 2 or 3 of the first three choices. Combining those people with all the responses in favor of B and C, we have 44 (or 71%) of the respondents saying they want something in the OPAC to represent aggregator serial titles. Although aggregators represent a specific instance of digital content, they reflect the growing tendency for digital content to differ substantively from print or to be packaged and presented differently from print. In those instances a digital resource is initially generated by a publisher or agency as a reproduction of print, it does not normally take long for the digital version to begin veering from the content of the print analog. For this reason, applications of the single record standard now typically embody a large and growing number of exceptions to the general rule, so that catalogers are directed to evaluate prescribed decision points and to catalog specific digital resources on separate records if the differences in content between print and digital versions meet the criteria prescribed by any one of numerous decision points. Advantages of a Uniform Separate Record approach are: 1. The cataloging record can serve users by providing full description and access, whether a user searches MELVYL, a Library's local catalog or the CDL Directory to identify digital resources. Description and access points can be provided either by means of direct access to a MARC record, or by means of cross-walks that transfer data from one metadata record type to another. The separate record approach means that users would always access a record for the digital resource itself. This would be more logical and intelligible to the user than accessing a print record on some occasions and a digital record on other occasions when searching for a digital resource. 2. Describing several formats in the same record can result in lengthy records. Several cataloging listservs have carried streams detailing the potential confusion for users and catalogers of consulting lengthy single records where data in specific fields cannot readily by identified with a specific format. Records are bound to continue to grow in size with the life of a serial or resource, becoming more difficult to read and more difficult to maintain. From a systems perspective, it is not logically possible to parse such records into separates describing specific formats if one wishes in future to generate a file of records for digital, or another specific format. 3. Describing the digital resource directly means cataloging treatment would always be consistent. Each format has its own record, whether CD, print or digital, whereas the current standard results in apparent inconsistencies of treatment. Separate records in all instances mean catalogers need not make judgments about whether a single record is sufficient to manage acquisitions or whether a single record gives enough information to support unambiguous interlibrary loan citations, which judgments would inevitably differ among catalogers. 4. It is simpler to catalog a material directly accessed by the cataloger than one in print that may not be available to the cataloger. The cataloger need not research or surmise the existence of a print version. In some instances, the cataloger may not know that a print version exists, or the print version may differ to such a degree (or differ in future to such a degree), that the single record approach is not advisable. As this would, by necessity, be catalogers' judgment, judgment would inevitably differ. 5. It is simpler to maintain the record in the event the content of the digital version changes, veers from the print, or if the print version ceases. 6. Information about the print (or other) version carried in 530, 740 fields, etc. is not so likely to require record maintenance as a base description of the print. Most changes could be reflected solely in the record for the print (or other) version by those libraries holding that version. 7. The 530 field describes alternative format(s). Individual libraries could make individual decisions about the need for additional records and detail for other formats depending upon whether those tangible formats are held and depending upon the degree of detail required. 8. 7xx Linking Entry Fields could provide links to records in MELVYL for alternative formats and versions. This is an approach that has been successfully employed by UCSB's Alexandria Project, whose collections reflect a great many "family" relationships (parent/child, etc.). One should expect the future of digital resources to represent many more such complex bibliographic relationships. Parent/child relations, etc., can be clearly delineated and accounted for in linking entry fields, while such relations and complexity make a confusing presentation in a single record. 9. The issue of holdings presentation is simplified and the presentation of holdings normalized if holdings statements for digital resources do not need to be constructed in MELVYL. "Access Info" is fully provided for in the CDL Directory display and if a common PURL is carried in the bib record, this obviates the need to manage the present, and anticipate the future, design of MELVYL for holdings, as well as the need to account for difficult issues of holdings conversion where holdings formats differ from one system to another (particularly since not all local systems yet support the MARC holdings format). 10. If the PURL in the bibliographic record is associated and tied to the CDL Directory, libraries could accept and load distributed bib records "as is", without needing to check on the currency of URLs, or edit them into existing print records. Libraries could load revised records "as is", overwriting existing records, without concern for editing or deleting data in the records related to other formats. 11. The "separate records" concept provides for separate STABLE records for archival (preservation) masters. The content and substance of bibliographic description for the archival master would be steady when the content of a "use" copy changes. The practice of employing multiple 007 fields in a single record for a microfilm use copy and preservation master microfilm works only because content of the version does not change. But, content of digital formats changes regularly. 12. The employment of separate records is consistent with the future of systems and of bibliographic control towards object-oriented design and is an approach that does not risk becoming obsoleted in future, requiring the wholesale revision of cataloging records. 13. It is relatively simple to "clone" a bibliographic record into a new record, making the simple editing changes required to compose a new record for a different format. While it may appear to some that less work is involved for the cataloger in creating a single record for all formats, as opposed to separate records, the overall workload for creating and maintaining single records is in fact far greater than that for creating and maintaining separate records. With the single record approach, more work and research is required of the cataloger creating the original record. Records must be revised and redistributed whenever the content and links for any single format changes. Libraries must review existing print records in their local catalogs, edit them, and in some instances re-catalog, to reflect digital counterparts, rather than simply loading a file of new, separate records. The largest workload for us in taking the "separate records" approach would involve the revision of records already constructed, but we are still at a relatively early stage of production, so that that workload seems manageable. A colleague has characterized this as a "you can pay me now or you can pay me later situation - with the later being a whole lot more time consuming and expensive." While the current practice still largely entails the utilization of a print record to reflect the digital version of a reproduced text, standards do provide for the generation of separate records to reflect different formats, where a library or cataloging agency deems this necessary, so that a separate record approach would not be inconsistent with present standards, while it would be consistent with the growing evolution of catalogers' thinking on this subject and the direction most likely to be taken in future. In re-reading the working principles set forward so clearly and wisely
by the TFER report, it seems to me the concept of separate records is quite
consistent with each item of the principles:
This is an opportunity to take an approach to shared cataloging and
access that might serve as a model for employment in other cooperative
databases, and that generates a product that can more
|