| |
|
|
ACRL/NEC Serials Interest Group ProgramsBringing Order Into Chaos: Issues and Applications in Cataloging Electronic Journals: A Report from the ACRL/NEC Serials Interest Group Fall 1999 ProgramPatricia A. Hatch Abstract: Presents a summary report of the Association of College and Research Libraries New England Chapter’s Serials Interest Group’s Fall 1999 program, which dealt with the theoretical and practical issues concerning the cataloging of electronic journals. The program’s speakers dealt with CONSER cataloging issues, developments with bibliographic utilities, cataloging workflows and the philosophical questions surrounding this topic. This author’s summary report focuses on the key points made by each of the speakers in their presentations. More than 100 technical services Librarians from around the New England area gathered at Harvard’s Gutman Education Center in Cambridge, Massachusetts for the ACRL New England Chapter Serials Interest Group’s Fall 1999 Program, which focused on the issues faced in cataloging electronic journal collections. Patricia Hatch, Librarian at the Institute for Business & Home Safety, and Chair of the ACRL New England Chapter Serials Interest Group, welcomed the participants and guests to the forum. The program emphasized that there are many philosophical and practical issues concerning the cataloging of electronic journals. Ruth Haas, Team Leader of the Serial Cataloging
Team in Widener Library and Head of Harvard CONSER UPDATE: Ruth Haas, David Van Hoy and Ann Sitkin Ruth began the presentation by focusing on two areas:
1. OPAC Access Aggregator Databases are those which hold many individual titles with an underlying database that defines the characteristics of all of the titles contained within it. An aggregator database can be one which contains a group of journals from one publisher (such as the Emerald Library from MCB Publishing) or a database can focus on a particular subject, such as law, but consist of materials published by a variety of vendors. An example of this type of aggregator is LEXIS/NEXIS. The contents of an aggregator database can be very stable or very erratic. Those that are erratic will constantly add or delete titles to their database, often without notice. Some start with the first volume published in a journal, such as JSTOR; others start only with the most recent, and retrospective holdings can vary greatly by publisher and/or title. In 1989, CONSER established a Task Group to investigate the best way to represent aggregator databases in public catalogs. The Group acknowledged early on that vendors would be brought in as partners in this process. Phase one of the Task Group's work focused on acquiring records to create record standards. Vendors were approached to contribute CONSER records which would meet the following criteria:
EBSCO Publishing was the first vendor to make available a set of these CONSER records which can be downloaded from their web site. EBSCO is a CONSER member. In addition, Proquest Direct also promised to provide records by the end of 1999. Other vendors would make records available through OCLC, but these records would not be available for purchase as record sets. Phase II of the Task Group's work focuses on more difficult questions. Acquiring the records and downloading them into a database was the easy part; maintaining these records is difficult. How do you deal with title changes, databases which add titles, delete titles, etc.? The answer that the Task Group has come up with to date: there is no one solution to these issues, and you will have to stay tuned. 2. Pattern/Holdings Information CONSER is creating a database where common information about patterns and holdings can be shared among libraries. These pattern holdings records are based on the US MARC Format 853 and 863 tags. Ruth displayed an 853/863 example: 853:00: |8 1 |a v. |b no. |u 12 |v r |i (year) |j (month) |w m |x 07 The 853 tag establishes the publishing pattern for the piece. Subfields a and b are the captions. Subfields i and j are placeholders for the year and month. The subfields u, v, w and x provide the pattern publishing information. The subfield u indicates that a volume of this piece consists of 12 issues. The subfield r indicates that when the volume changes, the numbering of the pieces restart. The subfield w indicates that the piece is published monthly, and the subfield x indicates what month the volume increments. In the example above, this means that the volume changes in the month of July. CONSER has established another Task Group to review three issues surrounding this topic: A. Internal Pattern Needs of Individual Systems This issue revolves around the fact that each integrated library system (ILS) has a slightly different system of interpreting its needs for holdings information. Even those are based on the US MARC format for holdings can differ. In this situation, a file of named generic pattern fields are available locally for check-in staff to set up the predictive check-in and holdings functions. A few vendors do not use USMARC for recording this information. Some use straight or "raw" USMARC. The Library of Congress, in preparing for implementing their new ILS, anticipates setting up several hundred of these named patterns. Addressing the sharing of patterns information among libraries with a common ILS vendor is only part of the problem. The vendors themselves need to be approached about implementing a more uniform way to access the handling of this data. We may find that it may even be necessary to undertake changes to the holdings format to meet our needs. B. Patterns Reflecting the Actual Piece Like bibliographic information, all libraries share the same publication patterns for titles held in common. This information could be communicated in the CONSER bibliographic record and adopted for use in the local system. The Task Group is setting up a trial initiative to begin adding and updating this information in CONSER records on OCLC. We need a better understanding of workflows in the individual libraries to pinpoint instances in which this information is used by what staff and how it can be efficiently added to, or extracted from, the bibliographic record and how it can be kept up to date. OCLC has authorized CONSER to use a new field, the 891, to communicate such information among the libraries. At Harvard, there are about 150,000 holdings records with 853 fields in them. The University is now in the process of reviewing these to identify which could appropriately be dumped into the CONSER records to seed this project. In the near future, CONSER will be sending out a request on their listserv for participants to be a part of this trial project. C. Retrospective Holdings Information The usefulness of retrospective holdings information at the national level has been questioned by members of the Task Group. Isn't this, by definition, local information? However, as publication patterns change or are updated on the CONSER record, what should happen to this once valid information? Should we just delete the information? We need to look very carefully at libraries' needs now and in the future before we make the decision not to retain it. But then again, if the decision is to retain it, how should it be done? More questions than answers. The original Task Group is being reorganized to be able to more specifically address these issues. The hope is that solutions can be found so staff at individual institutions are not recreating, one at a time, the same information over and over. This is certainly the basic premise of the CONSER program, which has proved a valuable and successful approach to serial data. In addition to following the work of the Task Group, interested individuals should also watch for minutes of the At-Large Meetings that take place at ALA and ALA Midwinter because there are always reports by these task groups, as well as minutes from the Operating Committee. David's talk focused on two areas:
In other words, are the computer file characteristics or the serial file characteristics going to be given in the 008 field of the bibliographic record and which is going to be given in the 006 field of the bibliographic record.
1. Choosing the OCLC Workform If you think back to the time of pre-format integration, CONSER was cataloging computer file serials on serial workforms. Other people were sometimes putting them on the computer file workform, so you sometimes had duplicates in the database. Then, format integration happened and there was a change over to the computer file format. Everything got merged. In those cases where there were duplicates, OCLC merged the duplicates. There was actually a mass changeover by OCLC to the computer file format. Fairly recently, there was a re-definition of type code "m" by MARBI. This re-definition of this type code meant that, for the most part, textual serials now belong back on the serial workform. But not all computer file serials belong on that workform. Software, numeric data, for example, still belong on the computer file workform. So, this time around, there is not going to be any mass change by OCLC. The CONSER libraries are, for the most part when they encounter these records, changing them, when necessary, to the right workform, the serial workform. Non-CONSER libraries can also do these changes for their own use if they want and are encouraged to report these changes to OCLC. But, for your own use, you can make this change. In his documentation provided to workshop participants, David provided examples of those records that should be on the computer file workform and those that should be on the serial workform. 2. Using One Record or Two The CONSER Cataloging Manual module defines two ways in which an online serial can be described: by creating separate records or by noting the existence of the online version on the record of the original print version of the title. Since the development of the single CONSER record standard, it has been widely applied by many libraries. At the time we developed it, knowledge of electronic serial versions was too limited to warrant defining categories for which the standard would be applicable. In recent years, however, a number of institutions have given careful thought to the issue and developed institutional policies. If you look at the URL of the CONSER Working Group, Single or Separate Records, you will find a list of links to libraries that have policies regarding this issue. With the increased knowledge of the types of electronic versions and the usefulness and drawbacks of the single record approach, it now seems time for CONSER to move beyond the experimental nature standard and define more clearly the situations in which it is best applied. The main result of this Working Group's deliberations are these "rules of thumb" that have been developed. There are two instances where the single record approach has been recommended:
There is one instance where a separate record approach has been recommended:
The most common example of this is newspapers; newspaper web sites seem to have so much more information online than in its print counterpart. These rules of thumb are also available in the new module 31 of the CONSER Cataloging Manual which can be found at the CONSER Web Site. Why Rules of Thumb instead of Just Rules? The main reason is because the CONSER libraries do not agree. There's a significant divergence of opinion as to how to deal with this situation. On the one hand you have libraries that are not doing the single record approach at all. On the other hand, you have libraries that are even going beyond these rules of thumb, such as GPO, and are "piggy backing" everything. For the GPO, it is primarily a staffing issue. You may have seen a posting on AUTOCAT from Adam Schiff from the University
of Washington discussing the use of a single record for monographs. The LC
draft guidelines, in short, permit it. Draft interim, subject to change.
LC has so far limited this approach to collections, photographs, maps, etc.
They have not yet applied it to typical monographs yet. Their guidelines
are slightly stricter than CONSER on this issue. There truly must be a 1:1
correspondence between the print and online versions of this. Something
reproduced in .pdf format, for example. OCLC has also proposed the same
thing. Ann began her discussion by noting that AACR2 does not contain any kind of rules for dealing with electronic materials, and this could be one of the reasons why so many of us are floundering with this issue. Ann summarized a paper written by Jean Hirons and others revising AACR2 to accommodate seriality. This paper was written in response to a request from the Joint Steering Committee at the Toronto Conference in 1997, which took a close look at electronic resources and seriality and asked Jean to prepare this paper. The goals of the revision are:
What's missing in AACR2?
The recommendations can be characterized into the following:
1. General Recommendations These include :to expand the introduction of the AACR2 code to include guidance on what constitutes a serial and when to make a new record. To reorganize Part I, description by ISBD area. To include rules for changes in each area of the description. Add a third part to AACR2, dealing with relationships to other works and manifestations. Add some new definitions to AACR2. The major points of the recommendations are to define two types of publications. The bibliographic world would be divided into these categories: finite and continuing.
To add definitions to the glossary. The new definition of "serial" reads: a bibliographic resource issued in a succession of discrete parts usually bearing numeric or chronological designations that have no predetermined conclusion. What's different in this definition is the word "usually." New definitions for the glossary have also been recommended that are not currently included in AACR2. 2. Recommendation 2 Looseleaf publications have been working with guidelines published in 1989 from the Library of Congress: Adele Hallam's Cataloging Rules for the Description of Looseleaf Publications. It was thought that since those were so successful, maybe we should apply those to other integrating resources such as electronic journals. So the recommendation here is to look at Adele Hallam's guidelines and see what would be appropriate to add to AACR2. This does not mean to incorporate them as they are but just to incorporate what would be appropriate. Secondly, is to eliminate the concept of "chief source of information". The purpose of this recommendation is to take a more holistic approach in recognition that continuing resources cannot be adequately described from just one issue. The gist of this recommendation is as follows: For all continuing resources, record the title and statement of responsibility and edition statement from the latest piece in hand. These can be derived from different sources of information for title, statement of responsibility and edition statement, for different types of resources. This recommendation is necessary because there are differences between printed serials and looseleaf publications and CD-ROMs and so forth that affect the preferred source of information. So they are recommending that for an electronic journal, the entire resource is the preferred source; for remote access databases and web sites, the title screen, home page or main menu would be the preferred source. For direct access electronic resources, eye-readable information on the disk label would be the preferred source. And for looseleaf publications, the title page or title page substitute would be the preferred source. Inaccuracies: they recommend correcting obvious typographical errors, rather than transcribing these. For web resources, introductory words such as "Welcome to.." or "Disney presents..." or those type of introductory words if the cataloger deem them unimportant, do not include in the title transcription. Do not record other title information for continuing resources. For serials, this is not anything new. For other types of continuing resources, this may be a problem. Looseleaf publications, for example, often have the substance of what they are about in other title information. Recommendation 2.8: If the first issue lacks a designation, supply one in brackets or such as volume 1 or its equivalently or alternatively supply a chronological designation if considered important. This recommendation is necessary because the current definition of serial requires a designation or chronology. But if the new definition of serial (see above) which says "usually" has a designation, is adopted then this recommendation may not be necessary. If the numbering of a serial begins again without a caption, apply new series in brackets or its equivalent. This eliminates the need to make a new record, and an undercurrent agenda of all these changes are trying to cut down on the number of new records which have to be created. Recommendations 2.10 and 2.11 deal with place of publication and the publisher. When it changes, add the latest and put the earliest or earlier in the statement of publication preceded by the word "originally." This emphasizes the concept of identification rather than transcription. But it raises a big issue of record maintenance. Every time a place of publication changes: think about that in serials. At ALA's Annual Meeting this summer, there was some discussion of this at MARBI and they looked at the possibility of having two 260s: one with the earliest place of publication and one with the current place and differentiating between the two with indicators. Recommendation 2.12: When the first or last date is not in hand, omit the date from the description. This is current practice, but they are adding, as an option: supply a probable date in brackets. Recommendation 2.13: Introduce the use of angle brackets into AACR2. It is now currently only an LC/CONSER practice to add known information in notes in brackets, also in the 260. The recommendation is to incorporate this practice into AACR2. Recommendation 2.14: Make notes on the title borne by the item or by which it is popularly known. The emphasis here is on popular titles, not necessarily anywhere on the piece. Recommendation 2.15: Replace "description based on" by "earliest issue consulted" and "latest issue consulted". And for integrating resources, provide "date viewed" so there is some sort of pinpointing what you looked at when you described this piece. 3. Recommendation 3 Deals with major/minor changes. One of their goals is to include, in the introduction, when to make new records. You do this by adopting a list of major changes (which equals, "I make a new record") and minor changes (which equals "I don't need to make a new record.") The recommendation is to adopt ISBDS and ISSN language on major and minor changes and to increase the types of minor changes in order to reduce the number of new records needed. The changes cover titles, corporate bodies and issuing bodies, and edition statement. A few examples of minor changes to titles that are being recommended include changes occurring after the first three words of a title and do not change the meaning of a title is not a title change. Current practice dictates that a title change occurs after the first five words of a title has changed. Changes in the representation of a word or words, i.e. changing from the singular or plural, or plural to singular, from abbreviated to spelled out. So therefore if you had a title called Fisheries Bulletin, and it changes to Fishery Bulletin, you do not need to make a new record. Addition, deletion or rearrangement of the name of the issuing body anywhere in the title that is not a name change, but just a change in representation, is now considered a minor change. The addition or deletion of words indicating type of resource, for example, "magazine", "bulletin", "journal" are all considered minor title changes. Minor changes to corporate body name include: the addition or deletion of articles, prepositions and conjunctions, changes in spelling and punctuation, changes in inflection of a word (singular to plural, plural to singular) and changes to the order of elements in the name or rearrangement of the name. The last part of the report deals with future considerations and the recommendation is to add a third part to AACR2 dealing with relationships to other works and manifestations. Part of that recommendation is to drop the use of uniform titles in many cases, not to be used any more for translations, and only to be used in successive entry titles. And also to give more prominence to links and linking entries in the rules (which are not properly dealt with currently in AACR2 and they wish to incorporate more of the MARC practice.) This report has gone to the Joint Steering Committee which issues and revised the AACR2 rules. They are currently meeting in Australia discussing this report and a number of libraries have submitted comments to this report and we are hoping to hear something. The results of the JSC's deliberations are slow in coming out, but if you are interested, you should watch for reports that come out of the JSC web site (found at the ALCTS Section at ALA, or through the National Library of Canada). OCLC and Electronic Serials: Research, Collaboration and Communication: Diane Baden Diane began her presentation by providing an overview of three electronic cataloging projects at OCLC: Intercat, NetFirst and CORC. These three programs represent a progression from what started as a purely manual, human intellectual process to more and more automated ways of cataloging Internet resources. 1. Intercat OCLC considered Intercat to be a research project and it was basically a test: could we apply the rules of AACR2 to Internet resources? And so libraries were invited to participate to select resources based on their own collection development criteria and to attempt to catalog them using AACR2. Intercat existed in a web environment, via OCLC's home page, and then those records were also loaded into Worldcat. One of things that helped everyone cope with this project was a manual, now in its second edition by Nancy Olson, Cataloging Internet Resources: A Manual and Practical Guide. There were about 92,000 records generated from this project. Since the records have been integrated into the main OCLC database, Intercat really doesn't exist anymore. One of the things that came out of the Intercat project was a collaboration among all the MARC groups to develop a field to park the URL of Internet resources and the 856 field tag was born. 2. NetFirst NetFirst is still an active project. NetFirst was an attempt to apply a mix of automated and human resources toward web resources that were identified as being suitable for library use. These sites were meant to be selected based on some collection development criteria. There are appropriately 120,000 records in this database. This was the first attempt to have some automated selecting of resources, and then a human element applied. One of the interesting things about NetFirst and a lot of what OCLC has learned from NetFirst that has been applied to projects that came after was that after resources were selected, they were assigned descriptors, mostly Library of Congress subject headings, abstracts were written and Dewey Decimal classification numbers were assigned. This, at the time, was the human piece. OCLC is in the process of getting this part of it automated as well. There was an attempt to organize the information based on traditional library principles and subject hierarchies in order to make the materials retrievable. This was a clear reaction to the web search engine phenomena which generated thousands of hits in response to a search query. There is some automated URL and content verification, that is, if the web sites change in a substantial way the program fetches them back for human re-editing of the records. NetFirst is a database that is available through the FirstSearch system. It used to be part of PRISM (the former cataloging module), but it has been taken out of this module and placed into FirstSearch. 3. Cooperative Online Resource Catalog (CORC) CORC is the future of the OCLC system as Diane sees it, moving towards a web interface. CORC is an attempt to further automate the process of both harvesting (that is locating and bringing together web resources) and then describing, cataloging or whatever term you want to use. It contains tools for building, and then maintaining, pathfinders. Pathfinders are those web pages that your institutions put up with valuable resources, arranged by subject. The feeling here was that every library was doing this on their own, selecting, gathering, presenting. What if we shared a database of these subject guides to web information and made it available collectively. So the program collects sites, creates metadata, and then humans step in and complete the process. It's based on lots of new standards, and it integrates them with existing ones, so, for example, you can create a Dublin Core record in CORC and then convert it into a MARC record and load it into your system if you desire. And there is authority control in CORC, which is the first attempt to bring authority control over a part of the web. There are authority records and authority maintenance available on this project. So this is a database, a database created in CORC, both catalog records of sites and pathfinders, and then you also have these automated tools to help you catalog digital resources on your own and, if necessary, import them onto your local system. At the last "Big Heads" meeting at ALA, there was a roundtable discussion about how libraries were using CORC. There was a wide variety of uses, everything from cataloging free electronic journals (i.e., you didn't subscribe to them so you didn't want to give them full cataloging, but you wanted to point your users to them), to the creation of metadata standards for your own digital scanning projects. This began as a research project buy by the summer of 2000 this will become a for-sale OCLC service. The pricing is being determined now and the marketing strategy is being developed. It will probably be a subscription service. Diane provided some screen shots about the service to workshop participants. 4. Other OCLC Research Projects
This is an attempt to control the problem with migrating URLs. The PURL is an intermediate kind of database where you put the PURL address in your record and the PURL server adjusts the address of the URL if it changes to a new address. This is all part of the development toward URNs (Uniform Resource Names).
Dublin Core is an international project to define data elements for web resources in a way that will cross disciplines, that will encourage the creators of web documents to create their own metadata or descriptive information and then use that as the basis of a cataloging record. CORC uses Dublin Core.
Kilroy is a database that can be used for study purposes by the OCLC Office for Research. They are trying to collect data on the kinds of information that are found in Internet resources, how end users interact with that data, how often do the links to a resource not work, how much metadata is really out there, and how much are people using Dublin Core and so on. This is a foundation of research that allows them to develop products like CORC that can interact with the Internet with some predictability. Scorpion is the program underlying all this automated subject retrieval and assignment, which is being used by CORC and other OCLC programs. 5. Products that are in development
This is based on the criteria used to create the Major Microform Set, but for electronic resources. OCLC is now seeking participants to assist in cataloging sets of electronic resources. Catalogers are currently creating bibliographic records for this collection that can not only be purchased, but will allow libraries to subscribe to the updates for this material.
This program currently allows libraries to ask OCLC to send them notification when specific bibliographic fields have been updated. For example, a library can request notification of all CIP records that have been upgraded to full cataloging, or for a list of all records that have Table of Contents information added to it. An option that is being discussed is to add notification when all 856 fields have changed in a record. 6. Documentation Cataloging Electronic Resources: OCLC-MARC Coding Guidelines Cataloging Internet Resources: A Manual and Practical Guide, 2nd edition 7. Consortia and Administrative Issues The three issues covered here are cataloging, interlibrary loan and union listing.
What do you catalog? Do you only catalog those items that are specifically subscribed to, or do you catalog those resources that you point to on the Internet? Do you catalog using a single or separate record? How do you catalog aggregator and overlapping databases? Do you catalog a separate database or do you integrate it somehow? In shared databases, coordination of these issues are extremely important so that the user of a consortium system does not become confused. Other cataloging issues concern license restrictions: what happens when categories of your users do not have access? How do you maintain URLs in your bibliographic records?
License restrictions on databases are common. Do you want to integrate these with databases that can have interlibrary loan requests? What about those systems that allow ILL but only with a printed version (not an electronic one)? Changing coverage and content can also be a problem with ILL. Single vs. separate records also have implications for ILL.
OCLC is actively working on guidelines about this. How do you union list electronic resources? What is the definition of an electronic resource as a "serial". How do you handle licensing issues--if you cannot share it, how can you include in an Union List. Changing content and coverage is a major issue for Union listing, as is having information on single or separate records. NELINET's Union Listing group has come up with guidelines that are available on the NELINET web site. Diane stressed that NELINET is available to help libraries that use OCLC here in New England. Please contact their office if you need any assistance. Practical Issues and Obstacles in Cataloging Ejournals: Jill Thomas Jill structured her talk around the Boston College experience: BC's profile, the collection policy, the selection process, and the cataloging policies. 1. BC Profile Student population of 9,000 undergraduates and 2,000 graduates. Patron visits of over 815,000 last year. Over 3 million photocopy pages in that time as well. Book collection of 1.7 million volumes, approximately 20,000 serial titles, and 3 million microform and archival titles. The Acquisitions Budget for 1999/2000 is $5 million (the book budget), and the electronic materials will comprise 18% of this budget, or approximately $950,000. 2. Collection Policy An electronic journal is defined at BC as any serial publication that is available over the World Wide Web or through an Internet technology. The Access Policy is that the OPAC will serve as the official source for the electronic journal collection at Boston College. 3. Selection Policy The task of selecting electronic journals at Boston College is the work of the subject bibliographers, and there is selection criteria. The journal should enhance the existing print material, the resource should be easy to access, fair use access should be accommodated, a user-friendly interface should be part of the journal, archive facility should be available for back issues, and the resource should be stable and reliable. The Library will negotiate the best licensing agreement possible. As for duplication, when the library acquires an electronic resource, it will not currently mean that the print version of it will be withdrawn or no longer subscribed to. Issues
The selection process feeds into their order process. Their acquisitions requests originate with faculty and graduate students, subject bibliographers, or other library staff. However, all requests are funneled into the ordering process through the appropriate subject bibliographer. The ordering process is handled in this manner: the subject bibliographer identifies a resource, contacts the electronic journal publisher or aggregator and arranges a trial of the product, obtains its cost and licensing information. Then the bibliographer emails the Head of Collection Development with the trial information, (i.e. dates of the trial, the URL and the logon procedures). And then, if the trial is successful, the subject bibliographer will fill out an electronic journal order form (Jill showed an example of such a form to the workshop participants). At the bottom of the form, the cataloging department is the last department the form is given to check off when the resource is cataloged. After it is cataloged, a copy of the form is returned to the Head of Collection Development and the bibliographer to let them know that the resource has been cataloged. The original requester (staff member, faculty member, student, etc.) is also notified. With these selection issues, some issues do come up. First of all, sheparding these materials through this multi-step process has not always been easy. This is a relatively new process for BC. Settling on an access policy (such as are we going to go through the web, are we going to use WebSPIRS, what is the aggregate, etc.) and honoring an acquisitions request while maintaining adherence to the collection development policy of the school generally has been hard. All subject bibliographers are tied directly to the academic departments. If that department has a very vocal faculty member who wants a resource cataloged, it is difficult to explain to that faculty member why it does not fit in with the collection development profile. 4. Cataloging Policies BC has adopted the separate record approach: two records, one for the print version and one for the electronic version. Also, in time to have a full record in the database, for all journals and all of their databases, what they have in their database is purchased journals through JSTOR or their MUSE project. Hopefully, they can expand that to include all the electronic journals that BC owns. It was one of the reasons why Jill was hired by BC six months ago. For now they also keep a running list of all of their electronic journals and they hope, as these are integrated into the OPAC, they will not have to keep this ongoing list. They hope to have a short update list of what's new and what's available, and have the reference librarians remind patrons that they should check the OPAC for a complete listing of all electronic resources (in keeping with the Library's access policy). Jill showed an example of an aggregator record from the BC database. While some subject coverage is shown, it is very difficult to show this kind of arrangement for multidisciplinary databases. Cataloging issues include:
Jill showed the group of an example of an electronic record that has some interesting information: The 362 field says "begins publication in July 1995". The 500 note says that the "description is based on a printout of April 1" The 515 notes says "includes issues published in print in September 1994." This is all on one record. Most of these notes do not show on the short display. What will the cataloger do if this changes? Things to think about:
Administrative, Organizational or Social Aspects of Integrating E-Resources Into Library Collections, Library Services and Library Processes--Matthew Beacom NOTE: Matthew's presentation can be found on his personal web site at Yale. Matthew informed the group that he wanted to look at the big picture issues with a focus on how we work together as an organization. A reflective report by a cataloger. Matthew began his discussion focused on three general concepts:
1. Integration of library and publishing Similar to what happened when manual card catalogs were transformed into integrated library systems. Now all departments were using a common database, and there was a chance for everyone to communicate all at the same time. It brought people together in ways that were unexpected. Now it seems that, with the advent of networked information and electronic publishing on the web, or on the Internet, we are seeing an even larger field being brought together. The whole communications cycle from the author to the reader has been, in some respects, squished together in ways that are sometimes predictable and sometimes unexpected. We are finding web sites now when authors can self-publish on the web and are finding huge audiences for their work. The author has skipped both the publisher and the library all together and went directly to the reader. So this raises a lot of questions for both publishers and for librarians about the role their institution plays in this new environment. What does a publisher do now and what does a library do now? 2. Collaboration Mainstreamed We've heard a lot about collaboration within an institution. Collaboration between a selector and a cataloger and public service librarians. We are going to see even more of this. It has become impossible for us to look at our jobs and say "my job is to catalog and my job stops there." This is really changing due to networked information. There is also a collaboration just within your own library but there are new opportunities to collaborate with faculty on the campus about materials. We can also collaborate with authors and publishers and distributors. Librarians and vendors are working closer together and that this will become more common in the future. 3. Follow the Reader Does anyone knows where we're headed in terms of where networked information is going to lead us to the reader? No one knows: not the libraries and not the vendors. The jury is really out on whether or not Internet vendors such as Amazon.com are going to survive if it doesn't show a profit some day. The best guide to follow is the individual path of the reader. The reader is interested in saving time and finding things in the quickest and most efficient way. This is a good thing and we should encourage it. Patrons would much rather have the quickest information than to actually search for the best information, in a database, see if the library owns it, ask for an ILL of it if the library does not own it, because they want the simple information now. Readers are looking for simple indexes and simple records. We have a problem in being able to do this: we can either provide simple indexes and complex records or complex indexes and simple records. We need to find a way to make things simpler and more efficient. These three concepts led Matthew to these four questions:
1. Who does the work? When he began to think about this issue, the cataloging of ejournals, Matthew thought of who was going to do the work. This boils down to three key roles: leader, worker and part of the team. These roles are not exclusive (sometimes you must be a leader, a team player or a worker). These teams can be formal or informal teams. 2. What do we do? You need a "vision thing." You have to look at the larger role and not just the issue at hand. You need to look not just at your job, but the job of others, such as public service librarians. You need to look at your work in terms of what it provides to students and faculty. We need to keep in mind that we are managing collections for their benefit. Again, the integration of tasks is essential here. The way that Matthew views electronic resources is the same way he views the book collection: as a collection of materials that are integrated. We want to lead our patrons to these resources and how we present that information we need to ignore the format and bring them tools that can help them find this. The OPAC works fairly well here. Archiving and back issues are important issues. Help users now. The rate of change is also part of the process: Yale did not have a Web OPAC until today. They had Proquest one day and EBSCOHost the next. Creating policies and trying to enforce them is very difficult. Picking projects and what you choose to pursue is also a difficult process. In terms of jobs, we need to create, adapt and promote jobs that allow us to do the work and to present what we have created to our users. We need not only catalogers, but selectors and public service librarians who can advocate for these materials. Sometimes, some librarians take on these jobs without these "official" designations. There needs to be a more "group" environment within the libraries. Matthew showed some examples of such teams at Yale. These teams came up with excellent ideas and tangible achievements. Matthew also showed an example of an informal group that had been developed at Yale without any type of formal committee designation. 3. How We Do Our Work Tools: to help teams come together, tools must be developed. Shopping list types of web pages for example, that provides resources to individuals. We also need policies to help us do our work. National standards such as the CONSER Cataloging Manual and local policies for individual institutions. An example of Yale cataloging categories include providing cataloging for most electronic resources, catalog those that the selector requests be cataloged, and not catalog, for the most part, personal web pages. We have to be thoughtful about what projects we choose to look at. Obviously, electronic journals must be controlled. There are so many tremendous opportunities to digitize many collections. What projects to take on must be thought our carefully, with workflows, budgets, and resources correctly assigned to make these projects work. 4. Why do We Do Our Work? Why are we trying to bring control to the electronic resources? If there is a fundamental change in the delivery of information from the author to the reader directly, what is our role. Matthew sees two responses to this: the librarians' niche and the value of the library. The librarians niche is our relationship to our users. We have a professional responsibility to provide access to these resources to our users. This is what defines our profession. Most libraries are not a free-standing institution, they are a part of a larger whole. Amazon.com does not owe anything to the Yale student. The library and the University does, however. We restrict access to non-Yale affiliates and try to develop collection policies that reflect the students adequately. The library's value is that we build collections (which, in a networked environment, is really the heart of the matter). Selection can help build the appropriate resources. Bigger is not better. More is not better. If you search the web and get 3 million hits, how does the user find what he needs. Like going to a Barnes and Noble to search for children's books and not finding them there, and then going to a small children's bookstore and finding all of the books you want, it is not about how much you have, but how those resources match to the user's needs. This is where the library's value comes into place. Libraries can also describe that material for access. Whether it is Library of Congress subject headings or metadata tags, this has not changed. Conclusion Patty Hatch thanked the participants for their presentations. Many librarians who attended the forum felt that this is an issue that is constantly changing and catalogers will need frequent updates to try to stay on track with new developments. Many hoped that this would become an annual event on behalf of the Serials Interest Group. The members of the ACRL New England Chapter Serials Interest Group are: Patricia Hatch, Librarian, Institute for Business & Home Safety (Chair), Rebecca Breedlove, Head of Serials, University of Massachusetts, Boston; and Joseph A. Gabriel, Director of Technical Services, Gutman Library, Graduate School of Education, Harvard University. |
ACRL/NEC Home Page | SIG Home Page | About the Serials Interest Group| Planning Committee | Programs | Join the SIG
Confirm Your Address | Annual Reports | Feedback Form | Other Serials Resources
