Marguerite E. "Maggie" Horn Library Systems Implementation Specialist Office of Library & Information Services, System Administration (SUNY) [email protected]

So, the world is ready for the industrialization of technical services. Let's speed things up; let's take short cuts; it's just so many widgets; everyone uses google anyway ... who cares! The biggest new trend in technical services appears to be the lack of emphasis on technical services in the library (excuse me, "information science") schools and also in our own shops. Get everyone out of the back room and into the front room. The stuff in the back room will be taken care of by outsourcing or by electronic access. Materials and records for the same will magically fly in the door, organize themselves for the shelves, provide access in catalogs, and always be correct. I don't think so!

I am occasionally asked to lecture on serials cataloging to the one technical services class taught at our local information science program. The last time I lectured, the instructor stated, "I know you believe in the details of what you're presenting, but you also say 'no one pays attention.' So, why bother discussing the details?" My comeback is: "If you don't know the correct way to do things, you don't know when you are messing up." Throughout my thirty year career with six different institutions, I have participated in data migrations to seven different library systems. I am constantly amazed by the number of people who are in charge of the migrations who do not understand MARC; do not understand how their non–MARC data is presented in their catalogs; do not know what standards have been used at their sites; do not know that their cataloging practice may be non–standard. Many of these people were just handed the job from someone who had "set things up". They have no knowledge of their system. They have no knowledge of any other way to process materials. They have no knowledge of cataloging standards or rules. Basic underlying principles don't exist for them.

Their entire training appears to be on–the–job rote learning with no discussion of why we do what we do. Why is this entry point needed? Is it because that is correct cataloging? Is it because the legacy system couldn't do the correct manipulation of the data? Is it because a faculty member long ago wanted a particular added entry? Should this record be exchanged with others in a union catalog environment? If we are not educating staff to know why they are doing something, then we will not be able to share each other's records.

If you don't know the correct way to do things, you don't know when you are messing up

Trina Grover Ryerson University [email protected]

The changing world of cataloging Now that metadata is a hot topic in the information retrieval literature, becoming a cataloger is no longer a vocational blunder, because the definition of cataloging is in transition. On one hand, we see a trend toward industrializing the work in order to increase productivity and decrease costs. On the other hand, there is a new focus on the fundamental principles of bibliographic description and a strong desire to modify the theory and practice of cataloging work to better serve the information discovery and retrieval needs of our users.

Increasingly, librarians need improved technical computing skills to do their jobs. As the volume of electronic content skyrockets and libraries become progressively more dependent on networked information, our work involves both managing information content (the traditional domain of librarians) and managing the technologies used to communicate, access, or deliver information content (the traditional domain of computer scientists). Technology is having an impact on cataloging practice as well as theory. In practice, online tools and documentation have drastically reduced the amount of time it takes a cataloger to consult rules and standards. Online discussion lists, bibliographic utilities and the Internet have made it much easier to find out how other catalogers solved the problem that you currently face. The theoretical basis of cataloging is also undergoing review and change. The fundamental basis of bibliographic description (as per AACR2R) concentrated on the physical appearance of items. The proliferation of electronic resources encouraged cataloging– rule review and revision. Committees are currently investigating the feasibility of basing catalog records on the content of a resource rather than the physical container1. In a networked environment lacking volumes of printed data with easily identifiable title pages, this makes perfect sense. If fully implemented, FRBR2 would produce the biggest change cataloging has seen in the last century, and have a significant impact on the indexing and display of records in library catalogs. Aside from the exciting changes to our theory and practice, many catalogers today are performing different work than they did twenty years ago as well as participating in work outside of their technical services departments. Original cataloging is delegated to others while we search for faster and cheaper sources of copy, train staff on what's new in AACR and MARC21 (currently it is the introduction of a new class of materials called integrating resources), attend committee meetings, teach bibliographic instruction sessions, serve at the Reference desk, whatever. Within cataloging work, the movement toward outsourcing, including using purchased record sets to catalog large collections that are donated or purchased, and the implementation of technologies such as Z39.50, have also changed the expectations placed on cataloging staff and the skill sets required to do the job. Working with MARC Sharing cataloging copy through online networks revolutionized the way we get bibliographic information into library catalogs. In many libraries, copy cataloging is the only way it is done; there is neither time nor expertise to create original records. While no formal agreement is needed to use Z39.50 to download records from any OPAC that has a public Z39.50 server, this method still requires one to search for individual records. Now the next wave of copy– cataloging is here, in the form of batchloading. By batchloading, I mean that files of MARC records (either included for free or purchased with new acquisitions) are loaded into the catalog, eliminating the need to catalog titles individually The acquisition of an aggregator package of online journals is often the impetus for batchloading. A consortial purchase (where libraries form a collective and buy products in bulk to save money and processing costs) may also result in hundreds or thousands of new titles acquired with one order. If cataloged one at a time, it would take months to finish a batch of 1000 ebooks, for example, or 2000 electronic journals. The challenge is then to catalog the collection in a timely manner while ensuring that quality does not suffer in the rush. Batchloading a set of records for the whole bunch (or part of it) is one way to accomplish this. The availability of record sets continues to increase. Sources include aggregators and publishers3 offer MARC records (and in some cases labeled and barcoded "shelf– ready" materials) for a fee, or free, with purchases; files available for purchase from vendors and utilities such as OCLC collections sets4 and MARCIVE; sets created by cooperative projects among like– minded libraries5 An online discussion list exists to share information about sources of free MARC records and files, http://groups.yahoo.com/group/Free_MARC_Records. There are many sources of records, and the quality of those records varies significantly. In order to minimize the introduction of errors into the catalog, it is important that catalogers take the time to scan the data in a batch file before loading it to determine what editing should be done, either before the file is loaded or afterwards. Editing files of MARC records requires some basic file conversion and manipulation skills in order to ensure that misleading or confusing information is not loaded into catalog, such as notes that indicate restrictions on access to a title. URLs created for one site may not work for another site, thus it is almost always necessary to edit the MARC record to match local system needs. Those with programming skills may choose to use MARC.pm, an open source PERL module for manipulating MARC data6. For those of us who are not programmers and have no one to delegate this type of work to, there is MARC editor software. MARC editor software is recommended over converting the file to ASCII and editing with a word processor, since a single incorrect byte in a file can have unpredictable results. One such editor is MarcEdit 7, a Windows– based application that is available free on the web. Developed by Terri Reese at Oregon State University, this software is easy to use as it is, although some might choose to integrate it with other software applications and programming/scripting languages like the Windows Scripting Host, PERL or C++. For catalogers with no programming expertise, I highly recommend it.

Finally, catalogers will require systems authority in their local ILS to set load parameters and ensure that records are saved and indexed properly. It is wise to test a few records in each file to determine the effects of this type of "cataloging" on indexes and displays. At the local level you might decide that item records are not necessary for ebooks since they do not circulate. Conversely, the omission of item or checkin records might unfavorably affect the OPAC display.

An annotated list of MARC Specialized Tools can be found at http://www.loc.gov/marc/marctools.html.

A niche specialization that will thrive The demand for high quality metadata will continue to increase. In order to keep up with the demand for metadata in our libraries, be it in the MARC format or something else, cataloging managers must strategically implement computer technology to solve problems. Catalogers can do this only if they have some computer competency in the first place. The skill-sets in cataloging are evolving and incorporating more Systems– like expertise, a trend that will continue as new technologies grow in importance. Loading record sets is an opportunity for us to keep up with the demand for MARC data in our libraries and enhance the information in our OPACs. Batchloading is an efficient approach to cataloging when it does not introduce inaccurate information into the catalog and mess up the indexes. With careful preparation and software like MarcEdit, the craft of cataloging and the industrialization of the work can live together in some kind of harmony. Footnotes 1A complete re– write of AACR2R rule 0.24 shifted the focus of bibliographic descriptions from the "physical form of the item in hand" (1988) to "bring[ing] out all aspects of the item being described, including its content, its carrier, its type of publication, its bibliographic relationships, and whether it is published or unpublished" (1998). 2 Functional Requirements for Bibliographic Records http://www.ifla.org/VII/s13/frbr/frbr.pdf ; see also OCLC research on FRBR http://www.oclc.org/research/projects/frbr/index.shtm 3 Sets can be acquired from netLibrary for ebooks, and from Proquest and Ebsco for electronic journals 4 http://www.stats.oclc.org/cgi-bin/db2www/wcs/wcs_cols.d2w/Electronic 5The National Library of Canada created sets of records for journals included in the Canadian National Site Licensing Project (CNSLP) and made them available via FTP to CNSLP members for free 6 http://marcpm.sourceforge.net See also Highsmith A. et. al, 2002. MARC it your way: MARC.pm in Information technology and libraries 21(1). 7 http://www.onid.orst.edu/~reeset/marcedit/html/index.html References Banerjee K. New technology and its impact on cataloging. http://home.earthlink.net/~banerjek/papers/ola2000.html Banerjee K. Taking advantage of outsourcing options: Using purchased record sets to maximize cataloging effectiveness. http://home.earthlink.net/~banerjek/papers/bulkrecords.html Danner R.A. Redefining a profession. http://www.law.duke.edu/fac/danner/callweb.htm Swan Hill J. and Intner S. Preparing for a cataloging career: from cataloging to knowledge management. http://www.ala.org/congress/hill-intner_print Back to Table of Contents

Technical Services No Longer Daniel CannCasciato Head of Cataloging Central Washington University Library [email protected] Introduction The industrialization of a process need not only entail the utilization of better methods to attain the same goals and higher productivity. It is not limited in effect to doing more of the same with less, or fewer resources. Nor is it limited to doing more of the same with the same resources. The industrialization of a process is, or should be, an opportunity to increase production and to diversify and to grow. Specifically, the industrialization of the cataloging process (or technical services in general) in libraries should be increasingly used as a means of doing more qualitative activities with whatever resources are needed. The end results should be a greater success at achieving our goal of providing patrons access to the collection. A byproduct of this achievement could be a greater recognition within the profession of librarianship of the service we provide and a greater understanding of the substantive activities involved in that undertaking. A good place to start is by advocating that we no longer accept the umbrella descriptor of these activities as "technical" services. Of non–understandings The label of technical services is a barrier to better, or even a clear, understanding of cataloging processes. Some personal examples of this are the following. I was asked recently to prepare a presentation on the "technical activities" that take place in the catalog department. At the same time, other departments (and the only ones who were, in the end, allotted time for the actual presentation) were asked to discuss services to our patrons. At another time a few years ago, when a tour was being given to a group of students, the narrative that "this is where they slap the labels on the books" accompanied the brief stop in the catalog department. Neither of these situations demonstrates an understanding of the process of cataloging. Perhaps we are hampered by our own success in having industrialized our workflow by adapting to new technologies early, quickly, and comprehensively. The development of the bibliographic utilities and the cooperative programs in cataloging are amazing successes which could serve as models for other areas of librarianship. Instead, I believe they remain below the radar. I believe they do so in part because the opaque label of technical services is a hindrance to understanding. There are many technical activities in catalog departments, but so, too, are there such activities in each of the other functional areas of the library. Systems departments are highly technical by their very nature, and yet are not described as the place where you get toner for the printer. Departments involved in digitization projects are highly technical in processes, yet not referred to as the scanning department. Reference, more and more involved in instructing patrons in the use of electronic resources and programs, is not described as the place that gives people direction to the bathrooms and public telephones. It is undeniable that those mundane or prosaic activities take place in those respective departments. Yet cataloging alone has been frequently reduced in descriptions to some of its most routine activities. Another example of what this lack of understanding of cataloging produces can be seen in the tone and content of some fairly recent articles. These articles all have the common characteristics of containing significant fundamental errors or misunderstandings of the cataloging process and yet have also called for significant changes in the world of cataloging. They have appeared in mainstream library journals. Barbara Baruth (2000), William Y. Arms (2000), Fred Ayers (2001), and Roy Tennant (2002, 2003), non-catalogers all, have had their views prominently published1. In effect, they speak to the profession at large on the subject of cataloging. Corrections from others are not always so prominently published2. Thus, the errors that have been promulgated likely remain credible to the larger world of librarianship. Why? There are a number of contributing factors, beginning with our label. The details of cataloging are neither understood nor well represented in the world of librarianship. Worse, the contributions of cataloging and technical services departments are not understood. Technical services, Public services, or Patron services? In most libraries cataloging is considered one of the technical services. Yet the defining characteristic of technical services remains unclear. That is, is utilizing a computer and a shared resource a technical service only? If so, as I've demonstrated earlier, most other departments in the library should now be known as technical services. If providing substantive access to the collection, facilitating the use of materials, and guiding patrons to materials is a public service, then how can cataloging not be considered a public service? Not only does the old label and distinction no longer fit (if it ever did), it is potentially harmful in this age of continued budget cuts. Cataloging is not a technical service per se. That is, our goal is not that of a technical accomplishment. Our goal is the provision of a fundamental patron service. Cataloging must be labeled and promoted as such. If providing substantive access to the collection, facilitating the use of materials, and guiding patrons to materials is a public service, then how can cataloging not be considered a public service? Success by continuing to industrialize Oddly, I think our successes at industrializing cataloging workflows, which have in the past allowed others to misunderstand or ignore our contribution to patrons, can be used to help assure us a place in librarianship of the future. As we continue to automate or industrialize certain aspects of our workflows (subscription services for cataloging records, as an example, or development and implementation of macros for routine processes as another), we must also guarantee that we grow professionally. Industrializing might free staff time, but that does not mean our work is done nor that our staffing needs have automatically reduced. We must advocate for participation in professional development areas that increase patron services. Those areas can be wholesale additions of table of contents information, or enhanced participation in authority control projects (locally or nationally). The reclassification of materials or enhanced description and subject access are areas that need further development and research. As the industrializing processes continue, naturally, our continuing participation in patron services will grow as well. We must continue to promote to the profession at large not only our technical successes, but also our content specific achievements. We must emphasize the impact that our successes have on our goal3. Technical services has always been an opaque label. We must instill in the professional at large a better understanding our pertinent mission and our substantive accomplishments: not in the realm of technical services, but in that of patron services. Footnotes 1Baruth and Tennant both conflate a communications format (MARC) with a cataloging code (AACR). This is a fundamental error. Ayers profoundly errs by laying failures in software at the door of authority control processes. (To compound the problem, his article appears in a cataloging journal.) Arms, as Michael Gorman (2003) very soundly demonstrates, makes a number of flawed arguments throughout his piece and others that Gorman cites.. Tennant (2003) continues in this type of misunderstanding of catalogs and cataloging. Even John Berry (2002) gives unsound advice regarding how a complex world should be reduced to an American viewpoint. This is promoted as an improvement in the OPAC for searching the term "civil war.". 2Rebecca Guenther�s substantive response to Tennant was published, of course, but it's in the letters section and not given near the prominence of the article on which she comments. 3For a heartening development in this area, see the program announcement for the Ohio Library Council's professional development program for 2003. The presenter is Janet Swann Hill who "believes that technical services people need to stop being apologetic and become more visible both in their libraries and in the field at large." http://www.olc.org/WorkshopDescription.asp?ID=218 (Viewed March 12, 2003.) References Arms W., 2000. Automated digital libraries. D-Lib magazine 6 : no. 7/8: [10 p.] Ayers F., 2001. Authority control simply does not work. Cataloging and classification quarterly, 32:2:49-59. Baruth B., 2000. Is your catalog big enough to handle the web? American libraries, v. 31:7: 56-60. Berry J. N., 2002. Keep it simple! Library journal Oct. 15, 2002: 8. Gorman M., 2003. The Enduring Library. American Library Association, Chicago. p. 28-32. Guenther R., 2003. MARC: Not dead yet. Library journal Jan. 15, 2003. Viewed online. Tennant R., 2002. MARC must die. Library journal, Oct. 15, 2002: 26-28. Tennant R., 2003. Library catalogs: the wrong solution. Library journal, Feb. 15, 2003: 28. Back to Table of Contents New Challenges for Cataloging Management Sharon Walbridge Assistant Director for User Support Services Washington State University [email protected] Cataloging departments today are likely leaner than in the past decade. This downsizing has come at a time when there are expanded expectations. Public services want all manner of material incorporated in the catalog. This includes table of contents, images, links to electronic resources, etc. In order to meet these expanded expectations, cataloging managers must be constantly alert to services that can provide some of the products that meet those expectations. There is little chance that the new demands will be accompanied by additional staff positions. Cataloging is now more than ever a hybrid – part productivity driven activity and part intellectual exercise, with emphasis on the latter in many libraries. Almost nothing is sacred in cataloging anymore. Priorities may change weekly. Messages from public services often do not carry a consistent message. "Yes, we want you to treat electronic resources as top priority, but we also expect you to continue to catalog incoming purchased material with the same speed and we also want you to reduce any and all backlogs." At some institutions, the technical services area is still treated as the slave with public services acting as the master. The realities of staffing, workload, shared cataloging challenges, keeping up with rule changes, planning and implementing new systems and equipment, etc. have little to do with expectations outside of technical services. So, how are we to cope? By questioning every old belief and procedure, and by taking advantage of what help is available. Two areas to look for help are suppliers of cataloging records and providers of cataloging related services. These need to be considered keeping in mind any possible problems/concerns with outsourcing. With many libraries still facing a lack of cataloging records for parts of their collections � gifts, old documents, microform collections, etc. – they cannot afford to look at the old model of perfect catalog record. It is difficult to respond positively to requests for adding table of contents, etc. to records when you are struggling to keep up with the new material coming in. We need to maximize use of existing records, even if they are not "perfect". If an existing record has 80% of what is needed, then you have saved a goodly amount of time in not having to create that 80%. Add whatever is needed, and get on to the next title... What can systems do for you that you do not have to do manually for yourself? Authority control and capturing statistics are two possibilities. Authority control can be a time-consuming process if done in the course of cataloging each title. Subscribing to automated authority control processing through any of the various organizations that offer such service makes sense. Utilizing the statistics gathering capabilities of local systems or bibliographic utilities also makes sense. Almost nothing is sacred in cataloging anymore. Keys to improving productivity lie in a number of areas including training, documentation, clear expectations, and collecting and interpreting of statistics. Giving each of these attention may be akin to "short term pain for long term gain" if the cataloging unit has not been diligent about them in the past. But, it is worth the effort as clarity in all of them will enable the department to move ahead. Having clear responsibilities for each of the areas will save time and effort and will improve communication. The many possible roles of MLS librarians in TSD include leaders, trainers, project managers, analysts, interpreters of rules. The division of responsibilities within the cataloging department, dividing activities between librarians and staff, can be a source of tension if expectations are not made clear. For staff who have been cataloging for decades, the introduction of new library school graduates or catalogers with little experience can lead to animosity. Cataloging managers must be aware of this potential for tension and work with both staff and librarians to ameliorate the situation. Making clear the responsibilities and contributions of each group will help. One thing is clear and that is the amount of time spent in actual cataloging has diminished considerably in many libraries. For more on this issue, see Rosann Bazirjian's recent article on the Role of Library faculty in a Team Environment (2003). A study of librarians in ARL libraries done several years ago by Stanley Wilder of Louisiana State University (1999) predicted that libraries, especially technical services units, and most especially cataloging departments would be facing serious staffing shortages in the next decade. As we see recruitment pools shrinking dramatically Wilder's predictions are becoming fact. Those library school graduate students who might have previously been interested in cataloging are perhaps now more likely to go to commercial database builders where they can do similar work and be better paid. So, it is likely that catalog librarians will become more and more scarce and the need to be very clear as to their role will become more and more necessary. The trend toward staff doing cataloging and librarians being leaders, problem solvers, analysts, advisors, trainers, etc. will continue and likely escalate. In addition to the potential change in emphasis regarding the roles of MLS catalogers, there is also a factor that effects staff. As the ranks are diminished in budget reductions, the need for lower level staff is reduced. The premium is on higher level staff who can be more flexible in terms of their work. Expectations are raised as staff are dealing with more complex materials and projects that call for independent judgement. The days of lower level staff and students doing the more straightforward work are rapidly coming to an end in some libraries. Where Promptcat can be used or records being supplied by book vendors, the need for such low level staff and students is reduced or eliminated. Staff who can deal with assigning LC call numbers and subject headings and can do original cataloging of subject matter such as literature may do the bulk of the work, freeing up MLS catalogers to focus on more difficult original cataloging, planning for projects, training, advising the staff, etc. Cataloging has changed from a fairly isolated endeavor to a highly collaborative one Cataloging has changed from a fairly isolated endeavor to a highly collaborative one. There is more group problem solving. As new cataloging challenges come along more quickly, the need for consultation and information sharing is heightened. An example is the rapid emergence of electronic links present in monographic materials. Just as libraries were getting an effective handle on electronic journals, catalogers began to notice links in monographs. There are many permutations from links to full text to links to publisher web pages that are largely advertising devices. Important links include those to materials that were previously found in monographs – bibliographies, indexes, and illustrations. At this point, maintenance issues for such links are still largely unknown. Quality and quantity are not mutually exclusive. They are partners, or two halves of a whole. In a sense, quality problems are more visible in the online environment than they were in the card catalog. What could not be found in the card catalog seemed to stay buried. Now people seem more inclined to report problems. However, the days of duplicative review of cataloging have gone away in many libraries. Acceptance of LC and PCC copy has grown as the pressure to get more done in less time by fewer people has grown. Can you afford to make mistakes in access points? No. Can you proofread every element in an existing record? Probably not. Cataloging managers must find ways to ensure quality while maintaining or improving productivity. How have libraries dealt with the influx of electronic resources? For years prior to this new form of information, many libraries added few new serials titles. Canceling of serials was more the norm. Not only are we met with serials by the hundreds and thousands, but the linking of these electronic titles in the online catalog is desired yesterday. Changes in provider often mean re–touching records again and again to ensure effective links. It is ironic that technical services areas are frequently perceived as conservative and traditional. Given that they were often the first areas within the library to automate (courtesy of OCLC) and that they are frequently the area where staffing is reduced in budget reductions (to preserve public service), technical services have had to change in order to survive. Library directors who think about it and are fair, acknowledge that this is the case, often saying that no area in the library has changed as much as technical services. Many libraries have reduced staff in acquisitions, cataloging and serials by aggressively seeking increased automation, adjusting workflow, taking advantage of commercial services including obtaining bibliographic records with books either through vendors such as Blackwell North America or through services such as OCLC's PromptCat. Conclusion At its base, cataloging provides access. It is the foundation upon which the library is built. The knowledge of record structure represented by a clear understanding of MARC is a valuable skill. If a library takes good advantage of its catalogers, it can and should involve them in all manner of database creation. Who better to break down information into a format that can be tagged and organized for ease of entry and retrieval. Who better understands the need for consistency and standards in building databases? Who better understands the online catalog and how it is indexed? Public services and systems folks would be well advised to use the skills of technical service folks. There are things that can be done to help others in the Library better understand the realities of the cataloging environment. One is the concept promoted by Anne Lipow – MARC for Public Services. Such sessions will help reference librarians understand the basis for indexing in the online public catalog. Local systems all have individual idiosyncrasies, but understanding MARC can help public services know why something retrieves like it does. Another strategy that can help technical and public service librarians better understand their shared environment is usability testing. Understanding how the catalog is used can help in cataloging decisions. How important are contents notes, secondary access points, etc.? Usability testing can shed real light on these issues. Cataloging is a hybrid activity, part productivity and part intellectual process. There is still much enjoyment and challenge to be found. Added to that is the opportunity for more collaboration than was true in the past – collaboration with public services to improve access, collaboration with academic departments in database creation, collaboration with staff to improve productivity. We hold the keys to convincing our colleagues of the value of what we bring to the library table. If we can do a good job at that, maybe then we can help with recruiting of new folks to the library profession (even to technical services) rather than seeing new library and information school graduates go to Microsoft. References Bazirjian, Rosann. "Role of Library Faculty in a Team Environment." Library Administration & Management vol. 17, no. 1 (Winter 2003) p. 33-38. Wilder, Stanley. The Age Demographics of Academic Librarians. New York: Haworth Information Press, 1999. Back to Table of Contents The Art and Science of Cataloging Felicia Uhden Manager, Bibliographic Services Seattle Public Library [email protected] Time: the present. One human being is seated before a computer terminal with an object in hand. A database search reveals no record of the object, so the human creates one. Cataloging is an amazing feat. It requires knowledge of the rules and practice of description, the policies and practice of subject analysis and classification, and the intricacies of the MARC format. The modern catalog librarian relies on knowledge of the mechanics, policies, and retrieval capabilities of both a bibliographic utility and a local online catalog system. Much of this knowledge is so ingrained that the cataloger makes use of it without conscious thought. At the same time, the cataloger thinks about the record in the context of the systems where it will reside. Is the cataloging treatment chosen consistent with records for similar items in the utility and in the local catalog? Most importantly, the cataloger considers the user and the uses that are likely to made of the record being created. It is almost as if she peers through the computer screen at the user who will be searching the system. This user may be a patron with extremely specific information about the item sought, a researcher who is only interested in certain categories of things, or a librarian trying to locate a work using imperfect clues and third–hand descriptions that have been uncovered during the reference interview. Just as we inspect records created by generations of librarians who preceded us, some of our records will exist far into the future. Is this act of creation, communion, and time travel that we term cataloging an art or a science? Should machines undertake all or part of this work? If so, which parts? At the dawn of the 21st century, does cataloging, as we know it, have a future? To ponder the answers to these questions, consider the tasks that make up the creation of a catalog record. First, the object is examined to determine its title, creator, maker, and date of publication. These facts are relatively easy to gather from the object itself. One goal of description is to provide a surrogate that will let a catalog user identify the object. Today, searchers are using copies of covers to identify desired objects. Why don't we pursue the use of scanning and character recognition capabilities to provide perfectly accurate copies of the chief source of information, publication information, and tables of contents in our records? Instead, we continue to create surrogates in a labor–intensive and error–prone fashion that can still leave doubt in the user's mind about what the record represents. Description proceeds according to well–established and well–known rules. Most of these elements would be difficult or impossible for a machine to complete. Creating some elements, such as the extent of the item, does not demand a high level of intellectual activity or judgment. Information for other elements, such as series information or bibliographic history, can come from sources outside of the piece. Deciding what notes and which creators' names to include requires judgment about the usefulness of the information. Discovering the nature or scope of a work and making a coherent assertion about it requires human judgment; the information can be critical to the user's decision about the work's value. Machines are better at providing some elements. A link to an electronic copy of an index is more valuable to the user than a statement about the existence of an index. At the dawn of the 21st century, does cataloging, as we know it, have a future? Rather tediously, the description is marked up for computer systems to interpret, display, and manipulate the contents. Systems provide some assistance with these tasks: supplying default codes for some elements, validating some relationships among data, and verifying some numeric content. Machines can and should proofread and correct some elements of records. However, humans still expend too much energy on coding that is created for the use of machines. We act as servants to the machine rather than the reverse. Given the prescribed phrasing and limited vocabulary of many descriptive elements, why doesn't the machine parse through the text and perform tagging and coding for some statements? Next, the human proceeds to more difficult, and sometimes nearly impossible, questions about the work. Does it have relationships to other works that could be important to users? What is the item about? Once these questions have been answered to the cataloger's satisfaction, he must refer to authority files to verify the form of the names. The cataloger must apply judgment in choosing the correct form for the proper entity. Among the hundreds of John Smiths who have created intellectual works, which John Smith created the work in hand? The cataloger assigns subject headings from a particular list of terms. If Library of Congress Subject Headings are used, then the cataloger consults policies about the kinds and numbers of headings to be assigned, looks up practices for the subdivision of terms, and checks the vocabulary list. A classification number is assigned to the work as well. The cataloger determines the primary subject of the work and considers the treatment of the subject to ascertain the discipline of the work as a starting point for choosing the classification. The goal of classification is to organize a collection, so the resulting classification number is checked against the context of the collection. Machines are working on the classification of electronic content, but print, video, and sound content remain extremely difficult for a machine to approach. Some subject terms map rather easily to classification; many do not. Electronic classification schedules linked to subject terms in existing bibliographic records provide a wonderful tool for classifiers. They have not yet replaced them. Finally, the cataloger is forced to complete a number of fields that exist solely for computer reading and manipulation based on the intellectual content of the record. Although the coding provides a way for machines to quickly sort and limit retrieval based on many characteristics–– e.g. format, intellectual level, geographic area, musical composition type, presence of indexes, bibliographies and illustrations––our computer systems rarely, if ever, make use of this information. Instead, systems perform these functions by searching human–readable fields. The codes used in the machine–readable fields are based on existing information in the record. Often this information uses terms from a limited vocabulary and appears in a prescribed location. Why can't the machine use the rules for coding and vocabulary lists to populate the machine–readable fields based on the text that has already been entered? The task of record creation is complete. Because the record exists for many years, the task of record maintenance has just begun. Machine processes have been very successful at altering records to comply with changes in coding practices. They enjoy mixed success in dealing with changes to the text of headings under authority control. The programmers of some local systems were better at exploiting the capabilities of authority records and may have possessed a better understanding of what these records contain than programmers of other systems. Some local systems have extremely flexible methods for making global changes to bibliographic records in response to changed headings. Other systems force staff to use such onerous processes to open and edit records that one might conclude the system designer had an expectation that records rarely, if ever, need to change. Bibliographic system search and retrieval capabilities have not developed as rapidly as Web–based search systems. With bibliographic record content so prescribed, our retrieval systems should be using information about where the information is found in the record as well as how many times the term occurs to provide meaningful ranking of results to users. Subject headings use controlled vocabulary lists; electronic records exist that could be used to match natural language keywords against desired or related terms from this controlled vocabulary. In many systems, the use of cross–references in authority records is limited to searches in the index itself and sometimes only used if records in the system contain text that matches the heading exactly. If the catalog contains only a subdivided term and the only existing authority record is for the term without subdivision, no cross–references will be available to the searcher. Catalogers, with their deep knowledge of both content and placement of information in bibliographic records, need to advocate for a stronger voice in choosing, configuring, and developing systems. The user is expected to locate relevant items by wisely choosing the proper search terms and selecting the best index from a multitude of indexes. Catalogers, with their deep knowledge of both content and placement of information in bibliographic records, need to advocate for a stronger voice in choosing, configuring, and developing systems. Our goal should be search and retrieval systems that apply this kind of knowledge in a way that is transparent to users. Time: the future. One human being is seated before a computer terminal with a work in mind. A database search using a few words in a single catalog search window reveals, near the top of the list, a record of the desired work with a picture of the book's cover, a replica of the title page, complete tables of content, and a shelf location listed. Is this the result of art, science, or magic? The author gratefully acknowledges the contributions of her colleagues at Seattle Public Library who constantly stimulate and inform her thinking about the art and science of cataloging. Back to Table of Contents Eli's Coming (and he has some questions) Fred R. Reenstjerna Cataloging Librarian Douglas County Library System [email protected] Introduction Eli Whitney should be revered as the greatest 19th–century American figure in library technical services. Not Dewey, not Cutter – just that longsuffering Connecticut Yankee, Eli Whitney. Whitney's real claim to our recognition was not his celebrated cotton gin, an invention that changed the South but brought him virtually nothing because of patent infringements. Whitney's genius was that he perfected the "American factory system," which is synonymous with the assembly line and interchangeable parts. In 1800, rifles were the pinnacle of American craftsmanship. Each rifle was a work of art by an individual gunsmith; every part was made by hand. It's hard to imagine, but every rifle in the American army 200 years ago was a unique example of craftsmanship. Sure, all rifles of a type might conform to specifications, but the notion that a part from one rifle could be taken off and put onto another rifle was as foreign a concept as letting women vote. Whitney's bold idea was to make machines that would make rifle parts. This was the origin of the American machine–tool industry, and the beginning of the modern factory. Whitney designed a saw that cut the wooden gunstocks, and other machines that could be guided to cut metal parts of rifles, so that someone else could assemble the parts into a completed firearm. Eli Whitney envisioned meta–machines – machines that made other machines – almost 2 centuries before the Dublin Core Metadata Initiative defined data about data. The Process... All too many libraries' Technical Services departments function as though Eli Whitney had never existed. Rather than design processes (it is called Processing, remember?) to handle a continuous flow of information, some librarians treat each new item as though it were the first book they had ever encountered. They turn Technical Services into a pre–industrial cottage industry, where no part is interchangeable with any other and each item is handcrafted. I have never worked a Reference Desk where customers have come up and commended the library on the elegance of its cataloging. I have, however, fielded queries or complaints from people who could not find their desired materials using the library's catalog. Here are some concepts to think about when reviewing your own Technical Services workflow: ONE: Why do you make your customers wait for a book because it's not on file in Dublin, Ohio? That means, why don't you do original cataloging immediately if you cannot find a book in OCLC's database? Any OCLC member can contribute bibliographic records at the most rudimentary level (meaning it can be upgraded by another member library), and your library gets a credit against its charges for every record contributed. The practice of dumbing–down Technical Services to make every MLS librarian a copy cataloger, as many libraries do, is a scandalous waste of resources. Why do you employ MLS catalogers, anyway? Why wait for an outsourced bibliographic record to appear in your utility? More importantly, with all the pious platitudes about serving local information needs, why does any library's management tolerate delaying access to information because local staff lacks cataloging skills? There's no magic in the drinking water in Dublin, Ohio, that turns people into Super Librarians. The OCLC librarians and the LC librarians and the librarians with elbow patches cataloging at the ARL libraries all earned an MLS, same as we did. We all took Cataloging classes, all learned about the MARC record (hopefully), and all know how to read the standard documentation of our profession. I know I do. If you're not up on the latest AACR2R revision, buy a copy and read it. No book budget for professional development? Read OCLC's Bibliographic Formats and Standards online (at http://www.oclc.org/bibformats/en/about/index.shtm), no subscription necessary, to get your bearings concerning MARC fields with which you (or your automation vendor, to judge by examples I�ve seen) are unfamiliar. The practice of dumbing–down Technical Services to make every MLS librarian a copy cataloger, as many libraries do, is a scandalous waste of resources. Why do you employ MLS catalogers, anyway? Why wait for an outsourced bibliographic record to appear in your utility? More importantly, with all the pious platitudes about serving local information needs, why does any library's management tolerate delaying access to information because local staff lacks cataloging skills? TWO: Why do people in the original information management profession refuse to educate themselves in current information management trends and technologies? Example: The refusal of so many Technical Services librarians to abandon an obsolete telecommunications technology such as OCLC's Passport™, when the lower–cost alternative CatME™ is available, makes me despair for the profession. Why does any holder of a graduate degree in information management –and that's what the MLS really is– refuse to learn new, more cost–effective, ways to manage information? You think it's not just some Luddite resistance to change? Then why are Passport commands executed by pressing the F11 key, when every other software from Excel™ to Tomb Raider™ uses "Enter"? Could it be that the F11 key is closest to the location of the red "Send" button on the old dedicated OCLC terminals of 20 years ago, and somebody couldn't make the warp shift to pressing a different button? In the 1970's, Gerald Shields wrote that we librarians were at a crossroads. We could embrace the emerging information technology and lead society, Shields wrote, or we could go on "playing dress–up in the attic of civilization." Far too many librarians seem to have decided that their calling is indeed to play dress–up, as if our entire profession were nothing but an audition for The Music Man's Marian Paroo role. Come on, librarians, we are the watchers on the walls of civilization, the keepers of the flickering flame that is the long saga of human struggle and creativity. Can't we at least keep up with the societies we allege to serve, making information available to people in a timely and technologically germane manner? THREE: Who exactly is minding the store, or at least the cash register? Why doesn't anyone hold to account fiscally irresponsible practices such as choosing to pay continuous telecommunications charges instead of switching to a batch mode, or wasting professional salaries on copy cataloging instead of contributing original work? Where is the leadership – in the profession, in the institutions, or in the governing bodies? I know how cost–effective my cataloging practices are, because I document them and monitor them. This doesn't require a math degree, but it does require sufficient commitment to the profession to document practices, to be aware of national standards, and to analyze workflow to select the most cost–effective means to deliver materials to customers. But that's (Gasp!) business, and we're librarians, not tradespeople. Cut the Victorian posturing, and explain why people who entrust us with their money (taxes, endowments, whatever) do not deserve a rational explanation of the resources purchased with that money. Conclusion Maybe there's an ugly reason that governing bodies pay so little attention to library fiscal management. Maybe we have become so utterly irrelevant to them, we don't even show up on their political radar. How about that?! Schools are in the thick of the fiscal fray, scrambling with police and fire and other needed services for ever–decreasing financial support. Maybe we have managed to make libraries reflect too many of our professional colleagues – distant, dismissive, and not really in the slightest way relevant to any aspect of human existence – such that nobody even knows we're there. We librarians are in the attic playing dress–up, while the adults are down in the front parlor, deciding the fate of the world. When Eli Whitney comes to your Technical Services shop, imploring you to examine your ways as Marley's ghost implored Scrooge, what will you tell Mr. Whitney? Will you tell him that, 2 centuries later, your library has failed to apply his system of organization to delivering informational materials to your customers? Why? Back to Table of Contents