CONSER PUBLICATION PATTERN INITIATIVE:
SHARING PUBLICATION PATTERNS THROUGH CONSER RECORDS
Frequently Asked Questions
prepared by Mary Ann Van Cura and Jean Hirons,
CONSER Task Force on Publication Patterns and Holdings
What is the Problem?
Many libraries are in the process of implementing new integrated library systems that use predictive
check-in and require the creation of publication patterns for each title. Each library is examining the
serial to determine the appropriate pattern and coding this pattern within their local system. Depending
on the system, the pattern may be based entirely or only partially on the MARC Format for Holdings
Data (MFHD). Currently, there is no way in which to share this effort, as there is in the creation of
bibliographic records.
A related problem is that because system usage of the MFHD varies, check-in records often cannot
migrate from one system to another. Thus, when a new system is implemented, staff must start from
scratch creating new check-in records, and often the affiliated holdings records as well.
In the area of holdings, there is also much divergence. Holdings have been created using a variety of
different methods, including free text summaries in local systems, OCLC local data records (LDRs), field
049 in the OCLC bibliographic record, or according to the MFHD. Use of Z39.50 opens up new
possibilities for libraries to share holdings data; however, holdings data cannot be shared unless it is
created and communicated in a standardized way.
A final problem is that there is no easy way for libraries to determine what has been published of a serial,
as opposed to what an individual library owns.
What is the solution?
CONSER's long term goal is to promote standardization and sharing of data. If library staff found
publication pattern data within existing OCLC records, they would have ready access to that data and
would not have to create it themselves. Achieving this goal may involve revising the format, encouraging
system vendors to more fully use the format, educating library staff in the use of the format, and finding
the right partners and vehicles for contribution and sharing of pattern/holdings data. Pattern and holdings
data must be as easy to share and communicate as bibliographic data. Through cooperation and
standardization, libraries will realize significant cost savings and patrons will more readily find their
materials.
Why CONSER?
While CONSER has traditionally limited its scope to the bibliographic record, its success with
standardizing the creation and sharing of bibliographic data makes the CONSER Program a natural
candidate for a leading role in the equally important areas of serials control and holdings. Furthermore,
working with patterns is not a new effort for CONSER. During the early 1990's a CONSER task force
investigated the creation of a publication pattern database (see below). The earlier effort was not feasible
at the time because not enough libraries were using the MFHD. With the current move to predictive
check-in systems and the need to be able to share holdings data via Z39.50, the time seems right for
CONSER to renew its interest in this area.
What is the CONSER Experiment?
The primary purpose of the CONSER experiment is to gather data to assess our current ability and
inability to share pattern data via CONSER records. Selective CONSER and CONSER Enhance libraries
will add MARC holdings data, the captions and patterns (field 853) and related holdings (field 863), to
bibliographic records for serial titles. This data will represent the current pattern for the principal work;
fields for supplements and indexes (854-55/864-65) will not be added due to space limitations in records.
To get the project started, CONSER records will be 'seeded' with holdings data from several large
libraries.
As a result of the experiment we hope to learn more about the usefulness of such data and be able to
project time and cost savings to libraries. We also hope to learn what the stumbling blocks are to sharing
pattern data. These may include different system's implementation of the MARC Format for Holdings
Data (MFHD), inadequacies of the format itself, problems with import and export of pattern data, and
issues relating to staffing and workflow.
What are captions and patterns?
A caption "is the word, phrase, or abbreviation indicating the bibliographic unit into which a serial or
multipart item has been divided by the publisher." Examples include volume, number, part, year, season,
etc. and their abbreviations. A pattern is correlated with the captions to establish publication intervals,
number of lower units per higher unit, and relationship of numbering systems (whether the lower unit
restarts numbering when the higher unit increments or continues to increment independently). These
correlations enable the check-in system to predict the next issue of most titles. In turn, the captions are
correlated with the actual enumeration and chronology to display holdings to the user.
How do you expect people to use these patterns?
If the software were written to do so, vendors could transfer the data from the CONSER bibliographic
record to the serial control record as the bibliographic record is loaded into the local system. The data
from the caption and pattern fields could be transferred into the serial control record or item record with
no human intervention. Currently, most systems require that staff do the initial work manually. Library
staff could retrieve the CONSER bibliographic record, and use the data contained in the publication
pattern field to manually create a serial check in record. As libraries have benefitted from copy
cataloging, they would now benefit from copying publication patterns. The mental effort and skill
required for original record creation would be reduced. Staff could copy and paste or type what they
found in the OCLC record into the local system.
Why embed data in the bibliographic record?
Library staff are used to searching for data in OCLC and importing CONSER records. Thus, housing the
current publication pattern in the CONSER record for import into a system makes the most sense.
In order to facilitate the experiment, OCLC is defining a new field 891 that will be used in bibliographic
records to carry holdings data. Fields 853 and 863 will be embedded within the 891 field. By adding
indicators and subfields, a CONSER library can include all the coding necessary to enable the data to be
imported, with only slight changes, into a MARC holdings record or MARC-based serial check in record
as a regular coded field.
How will the limit on the OCLC record size affect this proposal?
For most records, there is sufficient room to accommodate the additional data, which in most cases will
consist of two fields. OCLC does have a restriction on record size and some records have reached that
length, however, the removal of LC local fields 890, 901, and 911 should provide some room. CONSER
is also investigating the removal of abstracting and indexing services coverage data (field 510) to a
different site. But for the time being, the addition of a single pair of 891 fields containing the 853/863
data should not place an undue burden on record size.
What happens when a new pattern is needed? Where will old patterns be retained?
New patterns and their associated holdings field will be added by participants as soon as they are aware
of a change. There is currently no place to store the former 853/863 fields and this data will be retained
in the CONSER record as long as there is room. Eventually a different storage location will be devised
and old data will be removed to that location.
One alternative would be to store holdings and patterns on a separate database with a link to the
CONSER record. Another alternative would be a separate screen attached to the OCLC record that
would display the "master" holdings record. Both of these alternatives would require significant work to
create and are not currently feasible.
What is the relationship between pattern data and holdings data?
In the MARC holdings format, 853, 854, and 855 fields define the publication pattern of a work. (Field
853 is for the basic work, while 854 and 855 are for supplements and indexes.) Fields 863, 864, and 865
fields define the holdings, i.e. the enumeration and chronology (numbers and time span) that apply to the
published parts (issues, volumes, etc.) that use that pattern. For purposes of the CONSER experiment,
the holdings given in field 863 for the basic work will probably be limited to a single issue that will
indicate when the pattern became applicable to the serial.
Following are an 853 field for a journal and the associated 863 field for the first issue:
891 00 $9 853 $8 1 $a v. $b no. $u 12 $v r $i (year) $j month $w m $x 01
891 30 $9 863 $8 1.1 $a 1 $b 1 $i 1987 $j 01
In this case, the journal has two levels of enumeration: v. ($a), no. ($b) and the chronology consists of
year ($i) and month ($j). It is monthly ($w) with 12 numbers per volume ($u) and begins in January ($x).
The first issue for the which the pattern is appropriate, as indicated in the 863 field, is v. 1, no.1, January
1987.
If the pattern/holdings data is being added to a record and the issue is not the first for which the pattern
would be applicable (e.g., the library is newly-establishing a predictive check-in record as part of new
ILS implementation), the issue in hand can be given in angle brackets. For example,
891 10 $9 853 $8 1 $a v. $b no. $u 12 $v r $i (year) $j month $w m $x 01
891 40 $9 863 $8 1.1 $a <14> $b <2> $i <1999> $j <02>
(Angle brackets are routinely used in cataloging records to indicate an issue to which the data applies that
is not the first or last.)
How will this experiment help the sharing of holdings data?
While facilitating the sharing of local holdings data is not the focus of the CONSER task force effort, it
could be an important by-product. If, because of the CONSER effort, more libraries begin using the
MFHD to create their holdings in order to take advantage of data in CONSER records and the potential
associated Web site, the impact on standardization could be very positive. In addition, CONSER could
use its ability to share documentation and training in order to promote the use of the MFHD.
What is the relationship of this experiment with the earlier CONSER task force effort to create a database
of publication patterns?
If we were to examine a representative sample of publications, we would find that there are a large, but
still finite, number of publication patterns, some which occur more often than others. It would be possible
to provide a list or database of the possible alternative patterns that might be encountered when dealing
with updated publications. In fact, some integrated online library system vendors have serial control
systems that now provide an extensive list of publication pattern templates to choose from during setup of
check in records. From the list provided, staff select the pattern template that best describes each work in
hand. These patterns templates do not include information specific to any one serial, such as which
issues are combined/omitted or when the calendar years begins. Creating such a database was the focus
of the CONSER task force during the early 1990's. However, if only a database of pattern templates was
created, staff would have to determine which template is appropriate and then add data to make that
pattern template fit the particular serial. With a database of CONSER records containing MFHD
publication patterns, the pattern data is more complete and is connected to the title to which it applies.
Under the current scenario, the CONSER records would contain the complete MARC 853 pattern for
import into the local system.
What if the MFHD is not sufficient?
CONSER Task Force members will be working with system vendors, the ALCTS Serials Section
Committee to Study Serial Standards, and the Network Development and MARC Standards Office to
determine where the format needs improvement and to recommend changes.
Our staff working on publication patterns don't edit OCLC. Can we participate?
It is recognized that check in workflows differ from one institution or library to another, and that in
many, the staff who create publication patterns are not the same staff that catalog or update CONSER
records. It is also probable that in large libraries, this differentiation or specialization is greater than it
usually is in smaller libraries. This differentiation may mean that staff other than CONSER catalogers
are given the authorization to update CONSER records under the guidance of the CONSER catalogers.
This may be essential in order to assure timeliness of data input.
Our library is not using the MARC Format for Holdings Data. Can we participate?
It is certainly possible to input this data even if your system does not use the MARC Format for Holdings
Data. It may be advantageous to participate, especially if your library is using an automated serials
control system which is capable of predicting receipt of issues, because that system is probably based on
similar values. It will unquestionably be harder to participate under these conditions and possibly more
difficult to make local use of the data as readily. We encourage all libraries contemplating participation
to consult with the Task Force about what participation will entail and how it may be accomplished.
What kind of training or documentation will we receive?
Frieda Rosenberg is currently developing an editing guide for pattern/holdings data that will be available
on the CONSER Web site for use during the experiment and could eventually be added to the CONSER
Editing Guide. There is no specific training scheduled at this time. However, it is possible that holdings
training could be added to the roster of courses in the Serials Cataloging Cooperative Training Program.
Who do I contact for more information?
Please feel free to contact the following people to discuss your institution's involvement in the CONSER
experiment.
Carlen Ruschoff
Director of Technical Services
University of Maryland Libraries
Room 2209 McKeldin Library
University of Maryland
College Park, MD 20742-7011
tel.: 301.405.9299
ruschoff@umd.edu
Les Hawkins
CONSER Coordinator
Library of Congress
(202) 707-5947
lhaw@loc.gov
Go to:
- CONSER Publication Pattern Initiative Home Page
- CONSER Home Page
- Program for Cooperative Cataloging Home Page
- Library of Congress Home Page
Library of Congress
Library of Congress Help Desk
(July 20, 2005)