October 5, 1995

Executive Summary
Harvard/CONSER Task Force
I. Harvard's Investment in Serial Publications
II. CONSER Cataloging
III. The CONSER Office
IV. Room for Improvement
Task Force Membership

Executive Summary

"The library's database should be the bedrock of service to its users. Essentially, productive use of a database involves the expenditure of time - time spent by a librarian at the beginning of the process, or time spent by those thousands of users at the end of the process as they seek to make sense of incoherent or even random bibliographic records." (Michael Gorman in Library Journal, September 5, 1995)

Ownership of information in serial publications is one of Harvard Libraries' greatest assets. Our financial investment in serials has been, and will continue to be, our most significant collection development cost. The last half of this century has produced an ever increasing number of serial publications - many of which cover topics that did not exist during the first half of the century. Researchers are now faced with the need to access serial information published in multiple formats - in addition to print and microform holdings, serials may now be acquired on CD-ROM, on tape, or from a remote online system. In an increasingly complex environment with a staggering volume of publications produced, continued adherence to CONSER standards is the simplest and most cost effective guarantee that our users are provided the means to access the information they seek.

The CONSER Task Force recommends that:

  1. the long-standing policy requiring CONSER level cataloging of all current serial receipts be reaffirmed,
  2. new technologies be used to streamline workflow when feasible, including
  3. the libraries investigate the use of the CORE cataloging standard for appropriate serials
  4. the CONSER Office actively work with libraries to improve workflow effectiveness and efficiency.
  5. a three-tiered system of CONSER participation be established with:

    These recommendations are intended to increase the efficiency of serial cataloging operations without sacrificing quality of access. They will result in fuller information available to users as soon as possible, in greater throughput, and in increased flexibility for libraries to reorganize internal operations and to choose the most effective processing model for them. They take advantage of current technology and position the libraries to take advantage of developments still underway.

    The success of a three-tiered system will depend upon: the use of OCLC software scheduled for release in January of 1996; the potential for HOLLIS serial record upload to OCLC - tentatively scheduled for testing late this fall; and an analysis of the program cost including OCLC credits earned for CONSER production. Therefore, we recommend that a second CONSER task force be convened in January to study and recommend whether our proposals are economically viable for individual libraries and for the HUL CONSER Office.



    Harvard has been an active participant in the CONSER program for nearly 20 years. During that period, the university has made a major contribution to the building of the CONSER database, a premier national library resource. One of the underpinnings of our CONSER participation has been the University Library Council's policy that all current serials cataloging will be processed through the CONSER program. Support for CONSER participation has come both from individual libraries (who do much of the bibliographic work) and from the University Library (which provides staff and funding for the HUL CONSER Office).

    Much has changed in cataloging operations and automation since Harvard first joined CONSER. The migration of bibliographic work to local systems, new options for electronic record transfer and pressure to achieve maximum productivity all affect the way we view cataloging operations today. Given the magnitude of these changes a reexamination of the Harvard CONSER program is appropriate.

    Harvard/CONSER Task Force

    In July, 1995, a task force was appointed and charged to review the organization and operation of the CONSER program at Harvard and to prepare a report to the University Council. The specific questions the task force was assigned were:

    I. Harvard's Investment in Serial Publications

    The Harvard libraries hold nearly 300,000 serials, of which more than a third are active subscriptions. For many disciplines, serials provide the most up-to-the-minute scholarship and form the indispensible core of published research. Serials account for at least 30% of the annual materials budget of most libraries, and represent an ongoing and escalating source of expenditures (increases of 20% or more in the cost of foreign periodicals are expected for FY97).

    At the same time Harvard's investment in serials is increasing, so too is its investment in abstracting and indexing tools to provide analytical access to serial articles, further testament to the importance of serial publications to scholarship and research. The quality of citations in these sources varies greatly, and full cataloging of serial titles in HOLLIS allows the user to locate items at Harvard more quickly and easily. This is increasingly important as more abstracting and indexing services offer convenient fee-for-service document delivery. Inadequate local access could result in the Harvard community paying for these works a second time.

    The appropriate level of description required to locate journal information in a collection of five hundred or five thousand titles is usually not enough to locate the same information in a large research collection. Records created by other libraries to meet local needs frequently do not meet the needs of our researchers. The CONSER program defines a cataloging standard which meets the needs of most research libraries.

    II. CONSER Cataloging

    Serial cataloging tends to be complex because serials are complex. (See sample "genealogy" of a serial in Appendix A.) CONSER cataloging is, in fact, more complete than non-CONSER cataloging; the question is whether the additional information provided adds enough value to justify the effort of producing it. (See examples demonstrating before and after CONSER treatment in Appendix B.)

    CONSER cataloging standards emphasize the creation of detailed descriptions and consistent access points, including

         - Subject access points
         - Corporate body access points
         - Publication history, including earlier and later titles
         - Dates of publication

    These characteristics are frequently missing from non-CONSER records, and are critical for the accurate identification of the title and for the ease and clarity with which the title can be located and used. Cataloging serials to CONSER standards maximizes the university's investment in these crucial materials.

    III. The CONSER Office

    The HUL CONSER Office today serves a number of functions for the HUL community:

    IV. Room for Improvement

    CONSER cataloging and the current model of CONSER processing at Harvard have served the libraries and library users well, but outside forces, particularly the changing technological environment, suggest ways the model could be improved to increase efficiency, cost effectiveness, and flexibility. The Task group focussed on the following areas:

         - Elimination of redundant effort
         - Cataloging at a level appropriate to the material
         - Increasing libraries' flexibility to optimize their own workflow 

    There is redundancy in the current environment because all records, whether cataloged by CONSER staff or by local library staff, must be processed through the CONSER Office. Those units which do their own cataloging will key a provisional record in HOLLIS, input or edit a cataloging record in either OCLC or RLIN, print it, annotate the printout, and send the printout to the CONSER Office, where CONSER staff will call up the record in OCLC, edit it to match the annotations on the printout, and download it to HOLLIS. This process was established initially to insure quality control and to enforce participation. HOLLIS was designed to support this model, and will discard any non-recon serial record which is not downloaded by the CONSER Office.

    The task group recommends the following actions:

    1. ULC should reaffirm the university's commitment to catalog all current serials within the CONSER program. Mandatory participation insures a consistent level of access to serial collections throughout the university. The shared effort -- one Harvard library able to piggyback on another's CONSER record with a high degree of confidence -- reduces the total effort the university expends on serial cataloging. CONSER cataloging is a significant and high profile contribution the Harvard libraries make to the national database. Mandatory participation allows for predictable budgeting and staffing for the CONSER Office.

    2. New technologies should be exploited to streamline workflow, including

      1. Libraries should be urged to implement support for multiple communications sessions, via McGill, OCLC Passport for Windows (available January 1996), and RLIN Terminal for Windows, and to use the cut-and-paste and macro capabilities of these programs to streamline workflow and eliminate redundant keying. OCLC's Passport for Windows includes macros which will automatically generate some of the authority records required by the CONSER program. The CONSER Office and the HAAC Subcommittee on Serials, Series, and Continuations should serve as a clearinghouse for information on using this technology.

      2. The CONSER Office should follow national developments currently underway to allow batch-loading of CONSER records and NACO authority records into OCLC. This would allow units to create new records directly in HOLLIS and ftp them to OCLC, saving several processing steps. (Maintenance of CONSER records will still have to be done in OCLC for the foreseeable future.)

      3. OIS should remove the HOLLIS restriction on non-CONSER records. This will allow units to download existing serial records in OCLC or RLIN instead of keying a provisional record. It will save time, keystrokes, and will put the best available record in the catalog in the interim until CONSER cataloging is complete. It will also allow units to use existing serial records in OCLC (or RLIN) to upgrade recon, non-CONSER records easily with any better cataloging available.

      Distributing the ability to download CONSER records and permitting non-CONSER records to be downloaded would promote faster throughput resulting in enhanced HOLLIS access in a shorter timeframe. Work flow would be streamlined by eliminating paper transactions required by the current system. Online editing eliminates duplication of clerical tasks by local libraries and CONSER Office.

    3. The CONSER Office and the libraries performing serial cataloging should review the new Core cataloging standard for serials, which defines a less-than-full cataloging record which nevertheless conforms to CONSER guidelines. The Core standard was designed to address the need to reduce unnecessary cataloging work while maintaining high standards of description and access. Libraries may find that Core cataloging meets their needs for selected materials.

    4. The CONSER Office should work with libraries to redesign workflows with an eye toward improving efficiency and effectiveness, streamlining processing, and eliminating duplicate work.

    5. A three-tiered system of HUL CONSER participation should be introduced, to allow libraries to choose the best "fit" for their environment:

      Level 1 Libraries: Independent CONSER units

      • Level 1 libraries would assume responsibility for all services currently performed for them by the CONSER Office. All serials cataloging would be performed locally including: creation of original CONSER records, authentication and maintenance of CONSER records, and enhancement of non-CONSER copy on OCLC. The library could download non-CONSER records and enhance them in HOLLIS. After CONSER treatment, those records would be uploaded to OCLC and RLIN.

      • Level 1 libraries would assume responsibility for all tasks currently performed by the HUL CONSER Office -- authentication, NACO input, etc. There would be some trade off in time saved vs. time used - the need to extensively edit paper records would be eliminated, but it would be replaced by the need to edit online. There would be improvement to the current system which requires editing both on paper locally and online at the CONSER Office.

        Duplicate clerical tasks would be eliminated and replaced with what have been CONSER Office provided services. Level 1 libraries would have more control over their workflow and over when their cataloging is completed.

      Level 2 Libraries: Decentralized copy cataloging

      • A level 2 library would copy catalog serial records from the utilties. Authenticated CONSER records would be added directly to HOLLIS with no additional processing required.

      • Creation of original records, enhancement of member contributed OCLC records and authentication and maintenance of CONSER records would continue to be done through the HUL CONSER Office. HOLLIS target records for copy cataloging could be downloaded or created with cut and paste macros, eliminating the need to key target records.

      • Target record numbers could then be emailed to CONSER Office where substandard records would be enhanced.

      Level 3 Libraries: All cataloging by the CONSER Office

      • Level 3 libraries would continue to depend upon CONSER Office for all serials cataloging as they do now.

      Implementation issues for the three-tier system:

      Training for Level 1 and Level 2 would be coordinated by CONSER Office. The CONSER Office would retain responsibility for review, revision, and certification of CONSER records produced by independent libraries.

      HAAC Standing Subcommittee on Serials, Series, and Continuations would examine all records produced during training period and would periodically examine sample records after certification of an independent library to insure that standards are maintained.

      As internal circumstances change, libraries would be able to move among the proposed levels.

      The CONSER Office uses the "HUL" logon in OCLC which allows permanent changes to be made to serial bibliographic records in the OCLC database. Also, serial records being claimed/updated on OCLC for HOLLIS must be targeted or reclaimed on OCLC using this logon. If a three tiered CONSER system is implemented, a decision must be made as to whether Level 1 and 2 libraries also use the HUL logon or whether their own logon codes should be enabled for CONSER.


      The use of HOLLIS as our mandatory shared database is the only bond among the university's libraries. The cataloging labor of the first library to create an original CONSER record alleviates the burden for other libraries cataloging the same title. The second and subsequent libraries need only add their holdings information to a shared bibliographic record.

      Harvard has long been considered an outsider in the development and implementation of national cooperative projects from which we benefit. Our leadership and participation in the CONSER program has been cited as our major contribution nationally.

      The task force is convinced that CONSER cataloging is an appropriate and valuable way to insure that serial titles throughout the university receive the level of access they deserve and users require. It is the procedures rather than the policy which need alteration.

      Past decisions to accept less than standard cataloging in exchange for short-term efficiencies have resulted in limited accessibility or costly clean up projects. In a world of remote-access to multiple online catalogs, we should create an environment in which a researcher can find information at Harvard at least as easily as he can at another institution. Given the size and complexity of Harvard's collections, that means we must do more, not less.

      We have learned rather painfully that it costs less to "do it right" up front. Time saved by using less than standard cataloging is far exceeded on the user end when lack of detail in the bibliographic record impedes the work of researchers and the public services staff they consult for assistance.

      CONSER Task Force

      Ruth Haas, CONSER Office
      Pat Johnson, Baker Library
      Janet Rutan, Countway Library
      Yitzhak Teutsch, Law School Library
      Judy Warnement, Botany Libraries
      Robin Wendler, OIS
      Jane Ouderkirk, Widener Library, Chair


              TITLE: Forschungen zur brandenburgischen und preussischen Geschichte. 
                       Neue folge der "Mrkischen forschungen" des Vereins fur      
                       geschichte der mark Brandenburg.                             
          PUB. INFO: Leipzig, Duncker und Humblot.                                  
              NOTES: Vol. 45-54 include: bibliographie zur Gesschichte der Provinz  
                       Brandenburg und der Stadt Berlin, 1932-1943.                 
            INDEXES: Indexes: vol. 1-10 in vol. 10; vol. 11-30 in vol. 30.; vol. 31 
                       -40 in vol. 40.; vol. 41-50 in vol. 50.                      
              TITLE: Forschungen zur brandenburgischen und preussischen Geschichte : 
                        Neue Folge der "Mrkischen Forschungen" des Vereins fr      
                        Geschichte der Mark Brandenburg.                             
          PUB. INFO: 1. Bd. (1888)-Bd. 55, 1. Hlfte ([1943]); N. F., 1. Bd., Heft 1 
                      Leipzig : Duncker und Humblot, 1888-                           
        DESCRIPTION: v. : ill. ; 22-24 cm.                                           
          FREQUENCY: Semiannual                                                      
      LINKING NOTES: Continues: Mrkische Forschungen                                
              NOTES: Vols. 1-9, 11-   each in 2 parts separately paged; v.1-6, 15-   
                       have, besides the general t.-p. for each volume, also a       
                       special t.-p. for each part.                                  
                     Vols. for 1991-  issued by: Preussische Historische                   
                Kommission, Berlin.                                          
                     Vols. for 1888-1943 also available on microfilm from the Library of Congress
                Photoduplication Service.                
                     "Sitzungsberichte" of the society included in each volume, vols. for 1888-1943.              
             INDEXES: Bd. 1-10 in Bd. 10. (includes index to the earlier title); Bd. 
                        11-30 in Bd. 30.; Bd. 31-40 in Bd. 40.; Bd. 41-50 in Bd. 50. 
                ISSN: 0934-1234                                                      
            SUBJECTS: *S1 Prussia (Germany)--History--Periodicals.                   
                      *S2 Brandenburg (Germany)--History--Periodicals.               
             AUTHORS: *A1 Verein fr Geschichte der Mark Brandenburg.                
                      *A2 Preussische Historische Kommission.                        
              TITLE: Archives of otology.                                          
          PUB. INFO: New York.                                                     
        DESCRIPTION: v. ill.                                                       
          FREQUENCY: Bimonthly                                                     
      LINKING NOTES: Continues: Archives of opthamology and otology, ISSN 0092-5624
              NOTES: Also published in German as: Zeitschrift fr ohrenheilkunde   
              TITLE: Archives of otology.                                           
          PUB. INFO: v. 8-37; 1879-1908.                                            
                     New York, G.P. Putnam's Sons.                                  
        DESCRIPTION: 30 v. ill. 25 cm.                                              
          FREQUENCY: Quarterly, 1879-1897.  Bimonthly, 1898-1908                    
      LINKING NOTES: Continues in part: Archives of ophthalmology and otology, ISSN 
                     Other editions available: Zeitschrift fr Ohrenheilkunde       
            INDEXES: Authors and subjects: v. 22-28, in v. 28 (pp. 505-527).        
                     Authors and subjects: v. 29-35, in v. 35 (pp. 600-625).        
               ISSN: 0893-1380 = Archives of otology                                
           SUBJECTS: *S1 Otology--Periodicals.                                      
       ABBREV TITLE: Arch. otol.                                                    
              TITLE: Medieval Encounters : Jewish, Christian, and Muslim Culture in
                       Confluence and Dialogue.                                   
          PUB. INFO: Leiden; New York : E.J. Brill, 1995-.                        
        DESCRIPTION: v. ; 24 cm.                                                  
           SUBJECTS: *S1 Middle Ages--History--Periodicals.                       
                     *S2 Christianity--Relations--Islam--Middle Ages, 600-1500.   
                     *S3 Christianity--Relations--Judaism--Middle Ages, 600-1500. 
              TITLE: Medieval encounters : Jewish, Christian and Muslim culture in 
                       confluence and dialogue.                                    
          PUB. INFO: Vol. 1, no. 1 (June 1995)-                                    
                     Leiden ; New York : E.J. Brill, c1995-                        
        DESCRIPTION: v. ; 24 cm.                                                   
          FREQUENCY: 3 times a year                                                
              NOTES: Language note: Chiefly English, with some French and German.  
                       Summaries in English.                                       
                     Title from cover.                                             
               ISSN: 1380-7854                                                     
           SUBJECTS: *S1 Civilization, Medieval--Periodicals.                      
                     *S2 Christianity and other religions--Periodicals.            
                     *S3 Islam--Relations--Periodicals.                            
                     *S4 Judaism--Relations--Periodicals.                          
              TITLE: Punto cardinal revista De Accion Poetica   
          PUB. INFO: Miami, Fla.                                
           SUBJECTS: *S1 Latin america                          
              TITLE: Punto cardinal : revista de accin potica.                    
          PUB. INFO: Miami, Fla. : [s.n.],                                          
        DESCRIPTION: v. : ill. ; 22 x 28 cm.                                        
              NOTES: Description based on: 2 (jul. de 1967); title from cover.      
           SUBJECTS: *S1 Spanish American poetry--20th century--Periodicals.        
                     *S2 Spanish poetry--20th century--Periodicals.                 


      the process of reviewing records for content to ensure conformance to CONSER practice and conventions; and the addition of elements to indicate the degree of authoritativeness of data in the record.
      initially an acronym for CONversion of SERials, now for Cooperative Online Serials Program.
      CONSER database
      The set of serial records input/created or otherwise introduced to the OCLC database that are authenticated by CONSER participants. Although some or all CONSER records reside in the local databases of CONSER institutions, maintenance is performed on OCLC, making that the authoritative set of CONSER records.
      CORE Records
      bibliographic records containing only those elements deemed essential and mandatory by a national task force in conjunction with the Library of Congress.
      a terminal emulation software package which provides microcomputer access to HOLLIS.
      NACO (Name Authority Cooperative Project)
      similar in structure to the CONSER Program, NACO participants contribute name authority records to the national authority file.
      OCLC member contributed records
      serial records contributed to OCLC by OCLC member institutions which are not CONSER participants. Record quality and usefulness varies.
      provisional HOLLIS record
      a bibliographic record containing "order level" data with minimal descriptive information. Provisional records are used as markers for incoming catalog records with standard access points included.
      target record
      a provisional record containing only enough information to serve as a place marker for an incoming full record. Also called stub records.