Ed Glazier
Draft 3/26/03
Until we come up with different terminology, I fear I have used "universal holdings" and "pattern data" more or less interchangeably below.
In section 12.3, AACR2 Rev. 2002 calls for the use of numbering (ISBD Area 3) "for serials (with the exception of unnumbered monographic series) if cataloguing from the first and/or last issue or part" as part of the bibliographic description. The various rules in this section spell out the u
se of numeric and/or alphabetic designation, chronological designation, alternative numbering, and changes in numbering.Rule 12.7B1 calls for notes on frequency or changes in frequency.
Rules under 12.7B10 call for notes on numbering and issuing peculiarities, including change in numbering.
In MARC21 Bibliographic, this information is generally recorded in field 362 (numeric and chronological designations); in various note fields (numbering peculiarities, etc., also frequency of changes in frequency) or in fixed fields (i.e., for Frequency and Regularity).
The CONSER Publication Pattern Initiative has attempted to record what is essentially this information in encoded form using fields from the MARC Format for Holdings and Locations (MFHL). The stated purposes of the Initiative were to share creation of this pattern data for "implementing new integrated library systems that use predictive check-in and require the creation of publication patterns for each title".
This data attempts to represent the "Platonic ideal of a serial" as it has been published, if you will, rather than the actual holdings of any individual institution. For an institution that holds the entire run of a serial, the pattern data and the local holdings data can be assumed to be one and the same, although for an ongoing serial with no publication gaps, the pattern data is generally restricted to the starting enumeration and chronology, plus frequency and regularity information - the data that is needed to power the predictive check-in process in some local systems - but not the complete holdings, which are recorded in some sort of check-in file.
The pattern data can thus logically be seen as part of the "universal bibliographic description" of a continuing resource. However, just as the simple numeric and chronological designations recorded in field 362 are potentially confusing to the end-user of a bibliographic record, this detailed pattern data is likely to be even more confusing. In both cases, the end-user is apt to mistake the starting date of the resource for the first volume of local holdings. This would argue for the pattern data to be suppressed from an end-user display of a bibliographic record and instead displayed only to acquisitions or cataloging staff, who may need the data for administration and management of the subscription and the holdings.
Any given local system vendor may be able (or not) to use pattern data as input to an automated check-in process. They may be able to use the data if recorded in fields of the MFHL, but may need to map it to proprietary fields. Thus, the information recorded as pattern data is certainly of potential utility to all who hold or subscribe to a particular title, but the specific usability of this format of the data is probably not universal.
There is potential confusion with using fields of the MFHL both for pattern data and for actual holdings. MFHL as currently written does not specifically provide for the communication of pattern data and, in fact, specifically requires the use of at least one 852 field when holdings data is communicated. There are no current provisions in the format for using an 852 field to signify universal data instead of a specific location. The current use of 852$a containing "universal pattern" instead of a real location, while useful during the pilot period of the pattern initiative, should be reviewed, either with the goal of incorporating this usage in the MFHL or, better from some perspectives, finding a better way to identify this data. For an organization that attempts to process both institution-specific bibliographic records with local holdings and institution-independent records with pattern data, it is sometimes difficult to distinguish between the two uses of the same MFHL fields.