TASK GROUP ON IMPLEMENTATION OF INTEGRATING RESOURCES
INTERIM REPORT
April 23, 2001
Background: (1)
An integrating resource is:
A bibliographic resource that is added to or changed by means of
updates that do not remain discrete and are integrated into the whole. Examples
of integrating resources include updating loose-leafs and updating Web sites.
The concept of integrating resources has been endorsed by the Joint Steering
Committee (JSC) of AACR2. JSC is expected to publish a new Chapter 12, which
includes rules for integrating resources, in early 2002.
Meanwhile, introduction of this new concept immediately raises the question
of how to identify integrating resources through MARC coding. MARBI proposal
2001-05 to define code 'i' for integrating resources in leader/07 (Bibliographic
Level) will be discussed at ALA in June. Catalogers have enthusiastically supported
earlier discussion papers on the topic as it will distinguish integrating resources
from monographs and serials.
Charge
In February 2001, the Program for Cooperative Cataloging (PCC) Policy Committee
set up a "Task Group on Implementation of Integrating Resources" under the
auspices of the Joint Standing Committee on Training and Standing Committee
on Standards. The Task Group was charged to:
- Recommend whether certain types of integrating resources
should be handled by CONSER and others by BIBCO libraries
- Recommend whether certain types of integrating resources
should be handled by CONSER and others by BIBCO libraries;
- Recommend mechanisms to allow for either BIBCO or
CONSER libraries to update records for integrating
resources prior to adoption of Leader/07, value "i"(assuming
that it is passed by MARBI) ;
- Recommend mechanisms to allow for either BIBCO or
CONSER libraries to update records for integrating
resources following the adoption of Leader/07, value "i" (assuming
that it is passed by MARBI);
- Recommend appropriate types of and places for documentation
relating to integrating resources;
- Recommend ways in which to meet training needs associated
with cataloging integrating resources;
- Consider the maintenance and distribution issues associated
with integrating resources (including loose-leafs) and recommend further
action.
PCC further charged the Task Group to submit an Interim report April 30, 2001,
and a final report October 2001. This constitutes the Interim report of the
Task Group.
Summary of Activities
During this initial period, the Task Group on Implementation of Integrating
Resources reviewed the charge, discussed scenarios for authenticating records
for integrating resources, and discussed a survey to poll people on needs and
suggestions.
The remainder of this interim report is divided into the following segments:
- Brief statistical summary
- Task Group work on each charge
- Further recommendations
- Survey
- Next steps
NOTE1: Task Group members spent considerable time discussing one aspect of
record distribution -- redistribution of records from RLIN to OCLC. As the
Interim report was built over time, some segments of the report may still reflect
our struggle to balance the desire for fullest potential participation with
the realities of the current system structures. In the end, most task group
members agreed that -- for now -- further discussion of record redistribution
would apply to CONSER and to BIBCO members who work in OCLC.
NOTE2: The phrasing of the report is somewhat imprecise. Within the context
of this report, cataloging by BIBCO institutions is understood to refer only
to those records contributed to the BIBCO program (i.e., coded 042 pcc). Similarly,
cataloging by CONSER institutions is understood to refer only to those records
cataloged as part of the CONSER program.
TOP
I. Brief statistical summary
On April 2nd, Robert Bremer completed two surveys of records in OCLC to determine
the number of current integrating resource records.(2)
A search of OCLC WorldCat revealed 44,301 records for open-ended
loose-leaf resources. This total only takes into account records that are described
in the 300 field as being loose-leaf (or looseleaf or loose leaf) and that
have an open fixed field date (008/06 DtSt="m"; 008/11-14 Date2="9999"). WorldCat
included another 4, 575 records for closed loose-leaf resources (008/06 DtSt="m";
008/11-14 Date2 ` "9999").
The total of 44,301 is only a rough estimate. Many records in OCLC WorldCat
may in fact represent open-ended loose-leaf resources, but may just lack coding
needed for identification. To give an idea of the magnitude of uncertainty,
72,000 additional records included the term "loose-leaf" (or a variant) in
the 300 field, but lacked the specified fixed field coding (e.g., 008/06 DtSt="s").
Some of these records clearly were not open-ended (e.g., records with a closed
260 field date) or did not represent integrating resources (e.g., a record
for a transcription of some century-old church records). But in other cases,
the nature of the resource was unclear -- and possibly could only be resolved
by a trip to the shelves.
A search of OCLC CORC yielded only 8,308 records for electronic
integrating resources. In this case, the search includes monographs (Leader/07
= "m") with fixed field 008/23 Form = "s" and an open fixed field date (008/06
DtSt="m"; 008/11-14 Date2="9999").
Statistics from the RLIN file could not be obtained, since multiple RLIN records
may represent a single manifestation.
RETURN TO ACTIVITIES LIST
TOP
II. Task Group work on each charge
1. Develop various scenarios whereby integrating
resources are handled by CONSER and/or BIBCO
The Task Group considered four scenarios (developed by J. Hirons) for handling
integrating resources; later, a fifth category emerged. The five categories
discussed by the Task Group are:
- Authorize BIBCO members to catalog and maintain all
integrating resources;
- Authorize CONSER members to catalog and maintain all
integrating resources;
- Define some categories of integrating resources for
BIBCO maintenance; others for CONSER maintenance;
- Authorize anyone who is in CONSER or BIBCO to authenticate
or maintain a record for an integrating resource;
- Define categories of integrating resources for inclusion
in CONSER database; but authorize both BIBCO and CONSER members to authenticate
and maintain all integrating resources (including those in the CONSER database).
After delineating the advantages and disadvantages of each scenario, some
common themes emerged. The group agreed that success of the program will depend
on:
- Cooperation between BIBCO and CONSER members, to make the best use of expertise;
- Timely maintenance of records;
- Clear understanding of the scope of record distribution/redistribution;
- Documentation and training; which should be created through a cooperative
effort of BIBCO & CONSER members.
Specific benefits and disadvantages of each of these scenarios is discussed
in depth below:
Option 1: Authorize BIBCO members to catalog and maintain
all integrating resources.
Pros:
- Convenience for monograph catalogers: BIBCO catalogers currently cataloging
integrating resources could continue to do so, with no change to workflow.
- Inclusiveness: BIBCO catalogers could continue to maintain records, regardless
of utility affiliation.
- Continuity with past practice: No change to CONSER database
- Size of group: The number of BIBCO institutions exceeds that of CONSER
institutions
Cons:
- Cooperation limited: Limits the opportunity of cooperation between BIBCO
and CONSER
- Program goals: The goals of BIBCO have been to create more records faster,
better, cheaper, etc. rather than maintaining a single record for an indefinite
period of time. The concept of ongoing record maintenance would require the
addition of a new responsibility for integrating resources.
- Record maintenance complexities: In cases where integrating resources have
related serial counterparts, serial bib records may need maintenance at the
same time as the records for the integrating resource. Option 1 would entail
more complex arrangements for maintenance of related CONSER serial records.
- Redistribution of records: BIBCO records are not redistributed among utilities
as are CONSER records. Also, if ALL BIBCO libraries were to contribute successfully
to cooperative maintenance of records for integrating resources, then a mechanism
for participation of RLIN libraries would need to be worked out.
RETURN TO SCENARIO LIST
Option 2: Authorize CONSER members to catalog and maintain
all integrating resources.
Pros:
- CONSER "fit": In terms of maintenance, cataloging rules and record distribution,
integrating resources fit into the current CONSER mechanism and workflow.
Since the CONSER database and its distribution currently accommodates ongoing
revision of records, it should be possible to include records for new and
revised integrated resources. Documentation could be added to the CONSER
Cataloging Manual.
- Distribution: CONSER records are currently distributed via CDA to all LC
Serials Subscribers.
- ISSN: Some will receive ISSN and have ISSN records in the CONSER database
Cons:
- Workload: the major difficulty in having all integrating resources cataloged
by CONSER only is unmanageable workload
- Expertise: CONSER catalogers may not have expertise with these materials.
Also, this option excludes BIBCO catalogers having expertise to contribute
records.
- Cooperation limited: Limits the opportunity of cooperation between BIBCO
and CONSER
RETURN TO SCENARIO LIST
Option 3: Define some categories of integrating resources
for BIBCO maintenance; others for CONSER maintenance.
Pros:
- Fits with current division of labor between BIBCO & CONSER: Promotes
the concept of "Continuing resources"
- Fits with other programs: Would mirror the ISSN assignment to a great extent.
- Less disruptive: Would retain status quo for loose-leafs, while allowing
databases to be cataloged by CONSER
- Expertise: Both CONSER and BIBCO expertise would be utilized
Cons:
- Unclear boundaries: May force artificial lines between BIBCO resources
and CONSER resources
- Complex: Could negate the advantages of a single bib level code 'I' (if
approved) which doesn't require the distinction be made based on continuance
- Less predictable: Would mean less consistency and predictability as to
what would be included in the CONSER database.
- Distribution: Under current policies, records not defined as part of CONSER
would not be distributed via CDS to LC Serials Subscribers.
RETURN TO SCENARIO LIST
Option 4: Authorize anyone who is in CONSER or BIBCO
to authenticate or maintain a record for an integrating resource
Pros:
- Ease of implementation: Would require no change to current workflows
- Flexibility: Would allow all catalogers equal access to records
- Expertise: Makes use of expertise in both BIBCO and CONSER
Cons:
- Effect on CONSER database: There are two possible ways to implement option
4. The first method would be to exclude integrating resource records from
the CONSER database. But in this case, serials that become integrating resources
in electronic format would be excluded simply based on cataloging convention.
The second method would be to permit CONSER libraries to authenticate records
freely. But this could lead to uneven definition of the CONSER database.
- Effect of unstructured program on records for related serials: Some integrating
resources are first published simultaneous and later continue serials issued
in paper, microform, or CD-ROM. How will CONSER records for tangible serials
related integrating resources be closed and the appropriate links made?
- Distribution: Under current policies, records not defined as part of CONSER
would not be distributed via CDS to LC Serials Subscribers.
RETURN TO SCENARIO LIST
Option 5: Define categories of integrating resources
for inclusion in CONSER database; but authorize both BIBCO and CONSER members
to authenticate and maintain all integrating resources (including those in
this database).
Pros:
- Predictability: Allows predictability for catalogers as to what gets into
CONSER and what doesn't
- Expertise: Makes use of expertise in both BIBCO and CONSER
- Cooperation: Could foster cooperation among BIBCO and CONSER catalogers,
with wider exchange of cataloging knowledge and wider distribution of documentation
developed.
Cons:
- Limitation on involvement: The integrating resource records authenticated
for CONSER would be limited to OCLC. How would RLIN BIBCO libraries participate?
- Guidelines: Reaching a consensus for and developing clear, unambiguous
guidelines for the type of records included in CONSER may not be swift or
easy.
- Responsiveness: More bureaucracy could mean longer development time. Right
now, CONSER produces high-quality, timely documentation, in response to issues.
RETURN TO SCENARIO LIST
RETURN TO CHARGE LIST
2. Recommend appropriate types of and places
for documentation relating to integrating resources
Discussion:
Documentation needs to be widely available and jointly produced. Loose-leafs
are significantly different from electronic IR and should probably be given
separate procedures. The documentation method could be a revised version of
Hallam that could be made available via CONSER Cataloging Manual, PCC Web site,
separate purchase from CDS, etc. While most loose-leafs are legal, not all
are, and thus, this cannot be relegated to the legal literature.
Electronic IR (web sites and databases) pose their own problems and could
be treated together. We should recognize that many web sites will be cataloged
using different standards (e.g., Dublin core) and probably via CORC.
Conclusion:
The Task Group recommended that documentation like that in the CONSER Editing
Guide/CONSER Cataloging Manual be developed to support cataloging of integrating
resources. A joint committee or task force comprised of members from BIBCO,
CONSER, and LC could be formed to work on such documentation.
An important priority will be to identify the appropriate groups of people
to approach in developing and testing documentation. For loose-leaf works,
the law-library community has long been recognized for expertise; business
library catalogers may also have an interest in revision of Adele Hallam's
guidelines.
A similar community of e-resources catalogers has been developing, as libraries
define positions for Integrating Resources Catalogers. Two groups likely to
include IR catalogers as members are: CORC (Cooperative Online Resource Catalog)
and OLAC (Online Audio-Visual Catalogers).
RETURN TO CHARGE LIST
3. Recommend ways in which to meet training
needs associated with cataloging integrating resources
The Task Group recommends that training be given to both BIBCO and CONSER
members at the annual meeting in 2002. Additional training may be needed for
BIBCO libraries not currently represented on the Operations Committee. The
development of SCCTP or SCCTP-type training in integrating resources could
be done separately or in conjunction with the SCCTP electronic journals course
(a one day course currently under development). This might provide a nice opportunity
for BIBCO members to get into this kind of training development and the two
courses could be combined as a two-day session on electronic resources. Again,
we may want to split out loose-leafs.
Some questions that need to be considered in conjunction with training include:
- What are incentives for members to join and commit to IR record creation
and maintenance?
- What workflows are effective?
- How should BIBCO/CONSER create & maintain IR documentation?
- How can we identify IR "resource people" and establish this expert group
in our community for ongoing training and consultation?
RETURN TO CHARGE LIST
4. Consider the maintenance and distribution
issues associated with integrating resources (including loose-leafs) and
recommend further action
Several ideas for maintenance of integrating resource records were proposed
for consideration. Here are a few:
- Develop a method for identifying currency of bib record description. Two
ideas presented by members of the group were: (1) expanded use of 936 field
for update information; or (2) use of subfield "5" to identify currency of
data within specific fields of a record: e.g., 856 data. (3)
- Work within BIBCO/CONSER libraries for better monitoring of certain resources
(e.g., high-profile resources). There may be several strategies for achieving
this goal. One strategy would be for the authenticating library to have responsibility
for maintenance.
Until publishers, vendors, and suppliers can be persuaded
of a responsibility to customers and researchers, this
may be the only way to ensure some correspondence between
e-resources and the records.
- Revisit the issue of record distribution/redistribution between utilities(4)
RETURN TO CHARGE LIST
RETURN TO ACTIVITIES LIST
TOP
III. Further Recommendations
1. Maintenance: CONSER libraries that catalog integrating resources should
agree to maintenance of these records, and BIBCO libraries will be encouraged
to do the same. A separate task group should be constituted to consider strategies
for monitoring aspects of e-resource integrating resources. The work of this
task group should be expedited, given the immediacy of need. (Two possible
methods are mentioned above.)
Discussion: Some members of the Task Group expressed concern about
the relation of RLIN to cooperative maintenance efforts. RLIN is rich in records
in non-Roman scripts, for example; catalogers might benefit from cooperative
maintenance of vernacular records for integrating resources.
2. Authorizations: The capabilities of both CONSER and BIBCO/Enhance authorizations
should include Bibliographic Level "i" records.
3. CONSER database: Some integrating resource records should be authenticated
and distributed in the CONSER database. Potential candidates for authentication
are:
- Records with ISSN assigned (NSDP and ISSN/C will be authenticating some
databases and web sites)
- Titles whose other physical manifestations (e.g., print, CD) are serials
- Electronic journals treated as integrating resources because they do not
retain earlier titles
These will be the records distributed to other utilities from OCLC unless
we want to find a way to distribute all integrating resource records. This
would require LCCN assignments and may not be feasible or wise (further discussions
needed with CDS).
RETURN TO ACTIVITIES LIST
TOP
IV. Survey
Draft questions for contacts (rev. 3/20/01)
Name:
Institution:
Position (expertise):
Email:
Background. Integrating resources are bibliographic resources
that are updated over time by additions or changes that integrated into the
whole. This category includes updating loose-leafs, updating databases, and
many Web sites. Currently, because none of these fit the definition of "serial" all
are cataloged and coded as monographs. When the revised chapter 12 of AACR2,
Serials and Integrating Resources, is published in 2002, a new category of
resources will be officially recognized.
The PCC Task Force on the Implementation of Integrating Resources is considering
how these resources can best be handled within the Program for Cooperative
Cataloging. Consideration is being given to who catalogs and maintains the
records, authorizations for maintenance, distribution of records within the
CONSER database, and the documentation and training that will be needed.
1. What do you think that PCC could bring to the cataloging of integrating
resources that is not currently being accomplished?
1.1. What do you think is working now that you would not want to see changed?
1.2. What is not working very well now that you would like to see changed
or improved?
2. Is your institution currently a member of BIBCO ___ CONSER ____?
3. Who in your institution catalogs integrating resources?
3.1. Who catalogs loose-leafs?
3.2. Who catalogs electronic integrating resources?
3.3. Do you think this will change at all in the future?
4. How would you rate the following: Maintaining records for integrating resources
on OCLC is important. __ Strongly agree __ Agree __ Disagree __ Not sure
4.1. If important, how can such maintenance be best accomplished?
4.2. What is PCC's role in this maintenance?
5. What kinds of documentation and training would you like to see for the
various types of integrating resources?
5.1. Can you suggest persons or groups with the expertise to do this? For
non-OCLC users. Do you think that all or some integrating resources should
be part of the CONSER database? (The CONSER database resides on OCLC and is
distributed to RLIN and other databases)
6. Further comments?
RETURN TO ACTIVITIES LIST
TOP
V. Next Steps
During the past two months, the Task Group concentrated on the first point
of the charge,
"Develop various scenarios whereby integrating resources are handled by CONSER
and/or BIBCO."
Task Group members presented pros and cons based on four scenarios; and in
ensuing exchanges, a fifth scenario emerged. While none of the scenarios seemed
perfect, the Task Group generally preferred the scenarios which involved cooperative
maintenance of integrating resources by both BIBCO and CONSER participants.
At this point, Task Group members realized the need for consultation, both
with BIBCO/CONSER participants and with other catalogers who are creating records
for integrating resources. Task Group members are sharing draft versions of
the Interim Report with colleagues for comment and have prepared a survey for
more general distribution.
RETURN TO ACTIVITIES LIST
TOP
ENDNOTES
(1) Additional historical information:
Over the past few years, several discussions on AACR2 revision have resulted
in the rapid development of new models for describing continuing resources.
In 1997, Jean Hirons and Crystal Graham opened these discussions with a paper
titled "Issues Related to Seriality," presented at the International
Conference on the Principles and Future Development of AACR in Toronto. Following
the Conference, four task forces worked on issues related to the definition
of serial, the possible use of successive and latest entry cataloging conventions,
descriptive cataloging of serials, the cataloging of electronic journals, and
possible improvements to the rule for title changes. http://www.loc.gov/acq/conser/serialty.html
Finally, in spring 1999, Jean Hirons, Regina Reynolds, and Judy Kuhagen issued "Revising
AACR2 to Accommodate Seriality." Among the many recommendations in this paper
is a presentation of a new model for considering types of publications; and
in particular, a definition of "integrating resource" as a category of publications
with certain distinctive characteristics. http://www.nlc-bnc.ca/jsc/ser-rep0.html
RETURN
(2) The search for records describing loose-leaf resources
could be phrased:
In OCLC WorldCat, find all records with monographic Bibliographic Level (Leader/07
BLvl='m'), and with a 300 field containing the word "loose-leaf" or "loose
leaf" or "looseleaf," and with a Date/Publication Status (008/06) coded "m," and
with a Date2 (008/11-14) of "9999"
The search for records describing electronic integrating resources could
be phrased:
In OCLC WorldCat or CORC, find all records with monographic Bibliographic
Level (Leader/07 BLvl='m'), and with Fixed Field Form (008/23) coded "s," and
with NO 300 field, and with a Date/Publication Status (008/06) coded "m," and
with a Date2 (008/11-14) of "9999"
RETURN
(3) Serious concerns have been raised regarding both
ideas:
- 936 field: "Note that this field is not a USMARC field but rather an OCLC-defined
field so it would not be appropriate for national-level information"
- field-level control in subfield $5: One Task Group member commented, "Right
now, the use of the subfield $5 at a field level seems overly complex to
me." Another pointed out, "The definition of subfield 5 is 'institution to
which the field applies.' It is not currently a free-text subfield to include
information about what changes have been made and why."
RETURN
(4) During the course of the Task Group discussions,
some members of the Task Group asked about the distribution/redistribution
of RLIN records for integrating resources. As one member wrote, "As for the
distribution of IR records, we may want to look at the NACO model of distributing
authority records, in which NACO authority records are created from both OCLC
and RLIN, distributed to LC and redistributed by CDS back to the NAF file in
OCLC and RLIN in a timely manner. Can a similar arrangement made among CDS,
OCLC and RLIN for integrating resources records?"
Ed Glazier summarized the structure of the RLIN Name Authority File and Bib
File:
The RLIN NAF file and the RLIN Bib files are organized on two very different
principles with two different structures.
The Name Authority File has one copy of each record that is the LC Name authority
file, plus new records or updates that have not yet been contributed or were
just contributed today and are awaiting the distribution of the "official" version
of the record. Each authority record contains an 005 field for time and date
of latest transaction. Since there are copies of the LC file at OCLC, the British
Library, and
RLG, the 005 field is used in combination with the record control number to
assure that any update is based on the latest distributed version of a record.
Here is an example. Let us say an authority record is updated on the same day
in OCLC and RLIN. The latest distributed version is from 1/1999. Both an OCLC
user and an RLIN user update copies of the record with the 1/99 date. Let's
say the OCLC record gets to LC first. LC's processing compares the 005 in the
OCLC record with the 005 in the version in the LC file. If they are the same,
the OCLC update replaces the previously distributed version and becomes the
newest version with an 005 for 4/2001. When the RLIN update of the 1/99 record
gets to LC, its 005 is compared. But now, the version in the LC file is the
4/2001 version, while the RLIN update is of the earlier version. The RLIN Update
is rejected and RLIN is notified.
For bibliographic records, there is no such thing as a master record, version
control, etc. For each institution that catalogs a record in RLIN or who uploads
a record from a local system there is a separate and complete physical record
in the RLIN database. Records for the same edition of the same item are gathered
together in what is called a cluster. One record in the cluster is designated
as the primary cluster record, using a complex set of criteria. Generally speaking,
the initial display of records in result sets shows the primary cluster record
and an indication of what institutions own records in the same cluster. Records
with "pcc" in 042 are flagged in this display, though there is no way to indicate
which of them is the most recent version of a record. Any individual record
in the cluster can be displayed and exported or used as a source for copy cataloging.
A methodology for determining what records should be distributed as PCC records
does not exist at this time. There are various ways this could be done and
we would want to make sure we included batch-loaded records as well as records
input or updated online.
In light of Ed Glazier's explanation, most members of the Task Group agreed
that redistribution of RLIN bibliographic records is a long-term ideal. Currently,
the RLIN structure does not support the master record concept on which record
redistribution depends; when/if the structure changes, discussion could be
re-opened. In the meantime, the Task Group will need to table discussion of
this issue. For now, further discussion of record distribution/redistribution
will apply to CONSER and to BIBCO members who work in OCLC.
RETURN
TOP
|