Keenan 2013
Keenan 2013
To cite this article: Teressa M. Keenan , Jennie M. Burroughs & Suzanne Ebanues (2013) Partners
in Collaborative Cataloging: The U.S. Government Printing Office and the University of Montana,
Cataloging & Classification Quarterly, 51:1-3, 118-128, DOI: 10.1080/01639374.2012.719073
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
Cataloging & Classification Quarterly, 51:118–128, 2013
ISSN: 0163-9374 print / 1544-4554 online
DOI: 10.1080/01639374.2012.719073
TERESSA M. KEENAN
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
Maureen and Mike Mansfield Library, The University of Montana, Missoula, Montana, USA
JENNIE M. BURROUGHS
University of Minnesota Libraries, Minneapolis, Minnesota, USA
SUZANNE EBANUES
U.S. Government Printing Office, Washington, DC, USA
118
Partners in Collaborative Cataloging 119
Congress to the success of bibliographic utilities such as the OCLC with al-
most 25,900 members in 170 countries,1 collaboration has proven invaluable
in the sharing of bibliographic records. The establishment of the Program for
Cooperative Cataloging (PCC) in 1995 led to the structure of the current or-
ganization, whose activities are devoted to expanding access to information
resources by developing and maintaining bibliographic standards, comple-
menting metadata creation efforts of others and serving as a clearinghouse
for best practices.2 The PCC continues to support collaborative cataloging
practices through its documentation and training efforts.
Collaborative cataloging endeavors are not limited to large-scale inter-
national and national undertakings like the ones listed above. A review of
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
Like many depository libraries, the UM Libraries have been working on pre-
1976 retrospective cataloging for a long time. It has been common practice to
fully catalog items that circulate upon their return to the library. In addition
to adding records for items that circulate, UM used the results of a local user
survey and curricular assessments to prioritize areas of the collection for
more focused cataloging efforts. As the result of a 2006 study, UM focused
on cataloging materials related to forestry and other natural resource areas.9
With the primary goal of facilitating access and discovery for users, the
library performed assessments to gauge the resources needed to tackle the
cataloging backlog more aggressively. Through measuring the collection and
using the Association of Research Libraries (ARL) estimate of 52 items per
Partners in Collaborative Cataloging 121
rough estimate that staff could copy catalog 32 items or .59 linear feet per
hour. Assuming that over 700,000 legacy items require cataloging and that
the vast majority will have cataloging available on OCLC, it would require
21,875 hours to catalog the collection. If one person worked 40 hours per
week on the project, it would take over 10 years to complete.
While there are many ways to distribute the work, it was quickly appar-
ent that ML would need to pursue additional methods to increase efficiency
of the project to improve patron access to these materials. One option was
using available cataloging from commercial sources, such as records from
Readex for the Serial Set; this batch load process effectively shaved two
years off the project. Another powerful option was to collaborate with other
institutions.
One time-consuming portion of the cataloging process (in the aggregate)
is identifying the most appropriate record for an item and assessing its quality.
There is variation in government information collections at federal depository
libraries; however, due to the nature of distribution, many depository libraries
are searching for the same materials. This repeated searching for the same
materials across similar institutions suggests a tremendous duplication of
effort in the depository library community. ML began looking for ways to
collaboratively tackle the identification step, drawing on the library science
expertise of the community and sharing resulting efforts in turn. Identifying
the best available records (and creating records where necessary) for a given
monographic series or Superintendent of Documents stem would provide
the basis for rapid customization and batch loading. Such a project requires
shared tools, trust among partners, and a central distribution point. For ML,
GPO serves as a trusted partner and provides a central mechanism for sharing
expertise and records.
UM cataloging staff began concentrating on Forest Service materials due
to the strong forestry program at UM and interest by researchers in that
subject area. Even though bibliographic records were available for a large
majority of the materials needing to be cataloged, some records would re-
quire original cataloging. Additionally, some of the existing copy records
were very brief. While users would benefit from enhancement and authority
122 T. M. Keenan et al.
work on such brief records, the Government Documents Librarian and the
Metadata Librarian determined that a minimal record was more beneficial to
library users than no record at all, so local guidelines were established based
on the Bibliographic Record Cooperative Program (BIBCO) core record re-
quirements11 and OCLC Bibliographic Formats and Standards.12 Copy cat-
alogers would accept pre-existing bibliographic records as they are, only
editing them if they did not meet the core record guidelines. If no record
was found, copy catalogers would create minimal records. These minimal
records would contain complete descriptive information following the core
record guidelines but would not require the addition of subject headings,
classification, or authority work beyond validation of pre-established head-
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
ings. Not requiring subject analysis and SuDoc classification at the time of
initial cataloging permitted additional paraprofessional staff with varying de-
grees of experience in cataloging government documents to be involved in
the project. The outcome was a larger number of materials processed in
a shorter period of time without the need for additional staff and without
causing an overload of work for the faculty cataloger. We plan to consider
these records for enhancement at a later date. Joining the GPO in the collab-
orative cataloging project enabled ML to enrich these records sooner rather
than later and allowed the enhancement to proceed without the need to
reprioritize faculty cataloger efforts.
At the same time, GPO was beginning its largest effort to address the
gaps in electronic cataloging as part of the agency’s goal to create a com-
prehensive and authoritative national bibliography of U.S. government pub-
lications. The project to convert the cards in GPO’s historic shelf list, which
includes bibliographic information on publications published from the 1880s
through 1992, into MARC (Machine Readable Cataloging) records began in
January 2010 with contract staff transcribing the information on the cards
into GPO’s Integrated Library System (ILS). As of the end of January 2012,
over 123,000 records for pre-1976 or previously uncataloged publications
had been added to the U.S. Catalog of Government Publications (CGP).
However, the GPO shelf list is not a complete listing of all publications dis-
tributed through the Federal Depository Library Program (FDLP). As noted
earlier, GPO recognized that collaboration with depository libraries is criti-
cal to filling in the gaps in bibliographic access to this historic material and
to making the records available widely through the CGP. As a result, GPO
developed guidelines that outlined the evaluation, acceptance, and imple-
mentation of cooperative cataloging partnerships.13
As the first step to forming a partnership with GPO, ML wrote a pro-
posal based on local collection strengths, the area of greatest cataloging
effort, and likeliest unique contribution: Forest Service materials. Montana
is home to Region 1 of the Forest Service, and, as noted earlier, The Uni-
versity of Montana has a strong forestry program. As a result, ML has a
deep forestry collection that includes a sizable amount of Forest Service gray
Partners in Collaborative Cataloging 123
eral strategy was designed. ML would create the minimal core bibliographic
record to which GPO would add a formal SuDoc classification as well as
subject, series title, and name headings, as needed. The completed records
would then be made available to other libraries and the public through
GPO’s CGP and its record distribution programs. Finally, ML would add the
enhanced records to OCLC, thus providing an additional avenue of access
to the bibliographic information. Specific workflows were then designed by
both ML and GPO to address their respective portions of the project.
Once individual institutional procedures were established, two small test
files, each containing 25 bibliographic records created by ML, were sent to
GPO. The processing of these test files enabled both institutions to examine
their local practices, make any necessary adjustments, and create a timeline
for processing future files. Once both institutions were satisfied that the
general process would work the project began in earnest.
The ML Metadata Librarian worked with staff within the Technology
& Systems Services Department of the library to develop a query resulting
in a report containing a list of all of the documents within a specific call
number range that included the library’s MARC organization code in the
cataloging source field (040 subfields a and/or d). Other data represented
in the spreadsheet included the title (245 subfield a), Government Docu-
ment Classification Number (086 subfield a), publication date (260 subfield
c), OCLC control number (035 subfield a), and the Voyager bibliographic
number. Data in the spreadsheet was sorted by classification and date, al-
lowing initial focus to be placed on the pre-1976 titles cataloged by ML. Post
1976 materials in the list were later examined to select any gray literature not
already included in the CGP.
The ongoing partnership follows a monthly workflow. With specific
titles identified, records are grouped into sets of 50. Earlier tests indicated
that files containing 50 titles were the optimal size for the GPO to process
each month. ML creates the file of selected records by using VgerSelect to
find and extract the bibliographic records from the library’s local catalog.14
The following naming convention is used to facilitate easier tracking of files:
GPO_Partnership-[YYYY]-[MM], where YYYY and MM are the year and the
124 T. M. Keenan et al.
month that the batch will be sent to GPO. A new record set file is prepared
and sent to the GPO at the beginning of each month via e-mail.
Upon receipt of each record set submitted by ML, GPO begins a multi-
step process resulting in enhancement of the individual bibliographic records
for each Forest Service title. First the records are loaded into GPO’s integrated
library system (Aleph 500), and a report is run to determine if the file contains
duplicates of records already in the database. ML records with the same OCLC
number as existing CGP records are excluded from the load. An automation
librarian then globally suppresses the new records from public view and
adds a 955 field with the text “Cataloging Partner; University of Montana.”
Next, a list of the Aleph system numbers for the records is submitted
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
to staff trained to assign SuDoc classification numbers. This step in the en-
hancement process is the most time consuming as classification practices
have changed over time and multiple resources must be consulted. If the
record includes a SuDoc number in the 086 field, the specialist attempts to
verify that number using resources such as the GPO shelflist, the Monthly
Catalog of U.S. Government Publications, and the Guide to U.S. Government
Publications. SuDoc numbers that do not match the information found in
these resources are deleted, and a new GPO verified number is added to the
086 field of the record. Exceptions are made if the record contains a valid
class for a series but the publication was originally distributed to depositories
under a different class number. In these cases, two 086 fields, one for each
class, are added to the record. If ML had assigned a locally created SuDoc
number, identifiable by the addition of an “X” to the end of the number, the
locally assigned numbers are retained and moved to a 086 subfield z. For
records that do not contain a SuDoc number, the resources listed above are
consulted to determine if a number was previously assigned. If no class can
be located, the specialist will create a new SuDoc number for the publication.
The classification staff are also responsible for including notes in the
record regarding the publication’s distribution through the FDLP. If there
is positive identification that the title was distributed to depository libraries
by consulting the GPO shelflist and the Monthly Catalog of U.S. Government
Publications, then the GPO item number is added or retained in the 074 field.
If there is no indication that the title was distributed to depository libraries,
no item number is added to the bibliographic record and a 500 note with
the text “As of (date), GPO could not validate the distribution status of this
title for the FDLP” is added to the record.
Following the completion of classification upgrades, all records are re-
viewed for subject analysis and authority control. A technical services librar-
ian validates any ML supplied subject headings and upgrades them. If no
subject heading was supplied in the record, GPO adds one authorized LC
heading to the record based on the title information. The technical services
librarian also validates any ML supplied Name Authority Cooperative (NACO)
headings. If ML provides a higher level name authority for the agency, and if
Partners in Collaborative Cataloging 125
a lower body is required based on the information provided in the 260 field,
GPO replaces the 710 access point (for the higher level corporate body)
with a 710 access point with the name authority for the lowest corporate
body. If no corporate name authority is available in the record, GPO adds an
authorized NACO heading to the record based on the information in the 260
field or the SuDoc number. GPO does not validate or assign NACO headings
for personal names. A similar process is followed for validating series state-
ments. If the series statement is not established, GPO will request a surrogate
from ML in order to create the heading and submit it to the name authority
file.
The newly enhanced records are then unsuppressed and become avail-
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
able for use by depository libraries and the public via the CGP. Users can
easily identify these records by performing a keyword search for “Cataloging
Partner; University of Montana.” The completed records are included in the
monthly record loads to federal depositories participating in GPO’s Record
Distribution Project. Libraries are also welcome to download the records for
use in their local catalogs. Depository libraries have the ability to download
records via GPO’s Z39.50 server, while other users may send up to 20 records
at a time via e-mail.
Once the enhancement process is complete, ML retrieves the updated
records from the CGP. A tab delimited text file of corresponding OCLC num-
bers from the record set file sent to GPO is created from the same Excel
spreadsheet used to create the original list of titles. Using the OCLC number
as a match point, ML then retrieves the enriched bibliographic records from
the CGP by using the Z39.50 client available with the MarcEdit program.15
Once retrieved, the new records are added to the library’s local catalog
and merged with the earlier records using the bulk loading process available
within the Voyager system. In an effort to expand access to the bibliographic
information to institutions beyond federal depository libraries and those fa-
miliar with the CGP, ML then exports the newly enhanced records to OCLC
where they are merged with the older, less complete records.
CONCLUSIONS
Even though this project is still in its infancy (at the time of this writing),
initial evaluation indicates that it will be a success. Both institutions are
seeing positive movement toward achieving their goals. The addition of
over 12,000 bibliographic records to the online catalog, for example, has
moved ML closer to the goal of cataloging its entire government documents
collection, while GPO has been able to make good progress on its goal of
creating electronic cataloging records for all U.S. government publications.
By capitalizing on the individual resources and expertise available to
ML and GPO, ML was able to complete the preliminary recon of the library’s
126 T. M. Keenan et al.
Forest Service Collection. This includes all of the pre-1976 historical materials
and much of the gray literature. Utilizing the cataloging expertise provided by
GPO facilitated the successful completion of this portion of ML’s retrospec-
tive cataloging project without the need for additional staffing or increased
budgets. Copy catalogers with varying experience working with federal gov-
ernment documents were able to focus on the descriptive aspects of original
cataloging, leaving the subject analysis, authority, and classification work for
GPO to complete at a later time. Additionally, assistance with enhancement
of bibliographic records from GPO allowed the original cataloger to continue
working on additional subject areas within the government documents col-
lection without neglecting other library materials needing original cataloging
Downloaded by [USC University of Southern California] at 11:08 22 July 2013
or metadata creation.
Likewise, with its strong Forest Service collection, ML was able to pro-
vide the base descriptive record upon which GPO catalogers, who do not
have access to a tangible collection, are able to build. Using the ML record,
GPO staff were able to apply to their knowledge of government information
to subject analysis, name authority work, and SuDoc classification. Working
with ML also allowed GPO to take on a project to fill in a record gap while
not removing staff resources from the cataloging of current U.S. government
publications.
Not only has ML been able to increase visibility of the Forest Service
collection for local users via the catalog, but this project has increased access
to the bibliographic data to a wider audience via the CGP, GPO record
distribution programs, and OCLC. This wider availability of bibliographic
data will provide better interlibrary access to the collection at ML and the
collections at other depository libraries. Members of the federal depository
library community frequently request of GPO access to records for historic
publications, and this project is part of GPO’s ongoing efforts to address
that call. Increasing access to this bibliographic data also makes it possible
for other institutions pursuing indexing and cataloging projects related to
historical government documents to use copy cataloging procedures and
reduce the need for time consuming original cataloging practices.
While it is too early in the process to have collected and analyzed
circulation statistics with respect to the newly cataloged Forest Service ma-
terials, ML fully expects to see an increase in collection use in the future.
Local analysis of usage statistics after increasing exposure to collections via
the addition of bibliographic data to the library’s catalog for similar projects
has corroborated similar results discussed in cataloging literature.16 That is,
that increased access to bibliographic data via online catalogs results in in-
creased usage and/or access to the physical and digital collections that data
represents.17
The collaboration between ML and GPO has allowed each institution
to tackle a project that each felt would be too time and resource intensive
to complete on their own. Although a thorough cost–benefit analysis of this
Partners in Collaborative Cataloging 127
The early success of this pilot project suggests that GPO and ML should
consider continuing their partnership in the future as well as seeking out and
exploring other avenues for collaboration in the future. Using the knowledge
gained from the ongoing work with ML and the workflow developed for
this project, GPO is already undertaking additional collaborative cataloging
projects with other federal depository libraries that have strong collections.
This project and others like it have clearly demonstrated that collaboration
is the key to providing better access to once-hidden collections.
NOTES
available from the FDLP Desktop website. “Cooperative Cataloging Partnerships Guidelines,” FDLP Desk-
top, August 5, 2010, http://www.fdlp.gov/outreach/partnerships/757-catalogingpartnershipguide
14. VgerSelect is a software program written by Gary Strawn for the purpose of finding and extract-
ing information in Voyager records. “Files for the Public (Mostly Voyager Programs) | Northwestern Uni-
versity Library | www.library.northwestern.edu.” http://www.library.northwestern.edu/public (accessed
January 17, 2012).
15. MarcEdit is a free Windows-based MARC editing tool created by Terry Reese. “MarcEdit
Homepage: Your Complete Free MARC Software,” January 8, 2012, http://people.oregonstate.edu/
∼reeset/marcedit/html/index.php
16. Teressa M. Keenan, “Why Purchase When You Can Repurpose? Using Crosswalks to Enhance
User Access,” The Code4Lib Journal, September 21, 2010, http://journal.code4lib.org/articles/3604
17. Bolner, “Documents to the People,” 51–64; McKay, “Sharing Expertise,” 57–74.
Downloaded by [USC University of Southern California] at 11:08 22 July 2013