CONSERline (ISSN 1072-611X) Newsletter of the CONSER Program - Published by the Library of Congress, Serial Record Division
No. 13, January 1999
Greetings in 1999. This issue of CONSERline is devoted primarily to electronic serials. After reading the individual contributions to the CONSER Annual Report and some very interesting comments regarding electronic serials, I invited CONSER institutions to share their experiences with us in more depth. I am very pleased to have reports from the national libraries of Canada and Australia, GPO, and MIT. I think you will agree that we have come a very long way in a very short time in our ability to provide access to electronic serials.
Nevertheless, a major challenge faces libraries trying to provide access to e-serials in aggregator databases. Ruth Haas reports on the work of a CONSER working group that is investigating ways in which libraries can cooperate with vendors in providing access for these very expensive resources.
The emergence of e-serials has also highlighted the need for good training. This issue begins with two reports on the new Serials Cataloging Cooperative Training Program. It is perhaps significant to note that the initial course on "basic serials" will include both print and electronic serials. As Wayne Jones reminds us, we can no longer afford to treat them as exotic. Be sure to note the information session to be held at ALA on Friday, January 29, 7:00-9:00 PM..
And of course, a major step still wanting is the revision of the Anglo-American Cataloguing Code to accommodate electronic serials. As this issue of CONSERline is 'going to press' the initial stage of the AACR review process is coming to completion. The four groups established early in 1998 have completed their deliberations and the final reports will be posted to the CONSER Web site. The CONSER AACR Review Task Force will meet for a full day at ALA to discuss the reports and help me with the preparation of the final report which will be sent to the Joint Steering Committee in early April.
-- Jean Hirons, CONSER Coordinator, Library of Congress
As I reported in July, the Serials Cataloging Cooperative Training Program aims to provide standardized training materials based on CONSER documentation and trained trainers in order to increase knowledge of serials cataloging principles. Much has happened since July and I'm happy to report that we are on schedule and that the course materials and trainers should be ready by the end of June. Here are the highlights of recent developments.
Course materials. The first course, Basic Serials Cataloging, is currently under development by Cameron Campbell at the University of Chicago. The course will cover both original and copy cataloging of print and electronic serials. Course materials will consist of a trainer's manual, a trainee's manual, a Power Point presentation diskette, and pre- and post-class exercises on the CONSER Web site. The course materials will be sold by LC's Cataloging Distribution Service and should be available in June.
Testing the course materials. I'm happy to report that Steve Savage at Wayne State University in Detroit is arranging for a test of the course that will be given in February and include librarians from the Detroit Area Library Network (DALNET). Steve Shadle (University of Washington) and I will give the two day session, which will be followed by a half-day focus group discussion on the course contents and presentation.
Trainers. A call for trainers was sent out in December and is open until February 15. The qualifications are a firm knowledge of serials cataloging and CONSER documentation plus experience with training. Applicants must also be available to give two sessions a year and supply names of references and a letter of support from their supervisor.
The trainer portion of SCCTP is being coordinated with the PCC Standing Committee on Training. SCCTP trainers will be considered as PCC trainers, may be asked to serve on the Standing Committee, and will help with the revision of course materials as needed.
Train-the-trainer sessions. There will be two train-the-trainer sessions held in June. The first, June 9-10, will precede the annual meeting of the North American Serials Interest Group (NASIG) at Carnegie Mellon University in Pittsburgh. The second, June 23-24, will be held in New Orleans just before ALA. I am very grateful to the NASIG Board for contributing funds that will be used to support both the train-the-trainer sessions and the test at Wayne State.
Going to ALA? Come to the SCCTP Information session, which will be held Friday evening, 7:00-9:00 in the Philadelphia Marriot, Salon K. All those interested in any aspect of the program are encouraged to attend: potential trainers and trainees, supervisors and administrators, and training providers. Note: this is not a training session! Bring your ideas and questions to share with us. SCCTP is continuing to evolve and you can help.
-- Jean Hirons
Greetings from Beverley Geer and Bea Caraway, both of Trinity University in San Antonio, Texas. When Jean Hirons mentioned the idea of a training program last year, we jumped at the opportunity to participate in its development and have enjoyed serving on the official Steering Committee. She has done an excellent job of keeping the serials cataloging community up-to-date on the program's progress, so if you are like us, you are eagerly looking forward to its start. We are both excited about the training, but for different reasons.
Bea is the serials cataloger at Trinity. As is often the case when beginning a career in a relatively small institution, she learned by doing and by reading, without the benefit of training or revision from an experienced cataloger. In recent years, the CONSER Cataloging Manual and the CONSER Editing Guide have served as her cataloging bibles, and SERIALST as her support group. After several years, she still feels less than expert as a serials cataloger. Nevertheless, she is now called upon to train her paraprofessional assistant to do serials copy cataloging. In response to many a question, she must stop and look up the answer, or finally say, "I'm not sure, but I'll try to find out." There's nothing like teaching a subject to find out what you don't know about it! If the Serials Cataloging Cooperative Training Program were already underway, she believes her paraprofessional assistant would get much more skilled training from the program than he now is receiving from his supervisor, and in a shorter time. Furthermore, he would be joined in his education by others in similar situations and would begin to feel more a part of the serials cataloging community-which plays, as we all know, a vital role in our work lives. Bea hopes that when an opportunity to attend the training program finally presents itself, her assistant will be able to attend, even though she will already have introduced him to serials copy cataloging. (To be absolutely honest, she would like to attend a session herself-but don't tell anyone else that!)
Beverley is the head cataloger at Trinity University, but don't let her title fool you; she has plenty to learn too. Like Bea, she learned by doing and reading, but she also had the benefit of good on-the-job teachers. In the past few years, she has been called upon to teach others, both on the job and in workshops. She has learned that people come to workshops with different levels of knowledge (basic versus advanced), different expectations (needing to learn basic serials cataloging versus needing to know more about cataloging electronic serials), from different circumstances (some come from a large cataloging department and others are the only catalogers). She has been frustrated by her inability to tailor a workshop to reach each of those people, and there never seems to be enough time to do a truly good job. The Serials Cataloging Cooperative Training Program promises to solve these problems. Indeed, the Program states as two of its goals: a) to provide an ongoing program of serials cataloging instruction for participants with varying levels of skills and expertise; and b) to provide uniform core training materials that are adaptable to meet specific user needs. For Beverley this is a dream come true. Now if only she can pass the requirements to be a trainer! She cannot think of anything more fun (well, there are a few things, like dancing at Gruene Hall on a Saturday night).
In the end, however, we both appreciate the beauty of any program that promises to organize and standardize yet another aspect of our lives, catalogers that we are!
-- Bea Caraway, Serials Cataloger, and Beverley Geer, Head Cataloger, Trinity University
The National Library of Canada (NLC) began cataloguing electronic serials in 1994 with the start of an Electronic Publications Pilot Project to acquire, preserve and make accessible a selection of Canadian electronic publications. Although remote electronic Canadiana is not yet subject to legal deposit laws, NLC has a mandate to collect and preserve Canada's national heritage in all its forms. While NLC was gaining experience in selecting and archiving electronic publications, the staff in Bibliographic Access were faced with the task of providing appropriate bibliographic records for them. Over the last four years, we have encountered many forms in which electronic documents present themselves, and have gradually worked out some ways to deal with the bibliographic enigmas arising out of the creativity of the Internet. By January 1999, NLC had catalogued some 550 electronic serials (and 850 electronic monographs, for a total of about 1,400 titles).
The bibliographic records we create for remote electronic serials appear in the NLC's AMICUS database, as well as in the new CD ROM version of Canadiana, the national bibliography (and in the microfiche version). Brief bibliographic records for these e-publications appear in the Web version of the National Library's catalogue, ResAnet (external link), where a hotlink from the URL in field 856 of the bibliographic record brings the searcher directly to the electronic text itself archived on the National Library's server. The majority of the bibliographic records are forwarded to OCLC as CONSER records, and are also registered with the ISSN Network and appear on the ISSN Compact CD ROM.
What Are We Cataloguing?
The National Library has defined a remote access or networked publication as a digitally-encoded information resource made available to the public through a communication network. While we also acquire and catalogue Canadian CD ROMs, diskettes and other forms of electronic publications, this article focuses only on remote access serials such as those accessible through the World Wide Web.
NLC selectively acquires electronic monographs and serials published by Canadian publishers as well as the Canadian federal and provincial governments. Titles include those originally published on the Internet, as well as those which have been scanned or digitized from print versions. Special emphasis is placed on electronic publications of the Government of Canada. The National Library itself has also created digitized projects on the Web, and cataloguing has been provided for these titles once they have been archived as part of the National Library's electronic collection on the NL Website.
Cataloguing Standards and Tools
With a number of cataloguers in NLC involved in creating bibliographic records for electronic publications, there is an ongoing need to clarify and communicate new thinking on cataloguing practices for these publications.
The following standards and guidelines are applied in the course of cataloguing:
- Anglo-American Cataloguing Rules, 2nd revised edition, chapter 9;
- CONSER Editing Guide and CONSER Cataloging Manual;
- Cataloging Internet Resources : a Manual and Practical Guide, 2nd edition, edited by Nancy B.Olson, and published by OCLC in 1997;
- Canadian MARC Communication Format : Bibliographic Data (and soon its successor, MARC 21);
- Library of Congress Subject Headings, Canadian Subject Headings, Repertoire de vedettes-matiere;
- Library of Congress Classification, Dewey Decimal Classification, 21st edition.
Because NLC publishes the national bibliography for Canada, we create a separate bibliographic record for an electronic publication which has also appeared in another form (in other words, we do not add the electronic version to a record created for a print version). However, we create a single record for an e-publication which has been archived in more than one electronic format, for example, in both PDF and HTML. Bilingual English/French publications are catalogued in both English and French, following the National Library's policy for all publications. If an electronic serial has been scanned from a printed version, there is a good chance that the printed version has already been catalogued. If this is the case, we may be able to copy the record with minor changes; if not, original cataloguing is needed.
The processing of electronic publications has been mainstreamed, and these publications are assigned a cataloguing priority and level of cataloguing using the same criteria as are applied to all other publications. These criteria generally give priority to current Canadiana, especially in subject areas of special emphasis for the National Library such as Canadian literature, history, music and library science. Electronic serials are catalogued in the Serials Team in NLC's Government Publications and Serials Cataloguing Division.
Why Are e-Serials Dear to NLC Cataloguers Hearts?
Our cataloguers have found, as have many others, that current definitions of serials and monographs in AACR2R do not easily accommodate the changing nature of electronic publications and the emergence of hybrid ongoing publications. The National Library, as a member of one of the AACR author bodies (the Canadian Committee on Cataloguing) and as a long-time CONSER member, is participating in current efforts to propose changes to AACR for the treatment of serials and seriality.
The volatile nature of the Internet makes it easy for publishers to revisit a publication and make changes after it has gone public. We have found that a description of an electronic serial based on the earliest issue may not provide a sufficiently accurate description for later issues. There may also be a number of titles embedded in one publication, making the chief source of information difficult to identify. The look and feel of electronic serials can vary greatly over time, again making it difficult to identify the chief source of information, or what the title may be. Because Internet resources are so dynamic, we provide an indication of when the item was viewed, taking the form of a 500 note. "Title from title screen (viewed on [date])." The publisher's URL is also subject to change without notice. At present, National Library cataloguers do not update the publisher's URL in bibliographic records, but the URL for the National Library's archived copy is kept up-to date (and is in fact required to create a hotlink from the bibliographic record in ResAnet to the full archived electronic text).
Sometimes it is even hard to know where the publication begins, for example in cases where the top level Web page or "burst" page points the user to English and French versions of the publication. In this case, NLC has created two separate bibliographic records, one in English using the title display screen of the English text as the chief source, and one in French, using the French title display screen. Another challenge of bilingual publications is the varying presentation of the same document in different electronic formats. For example, a serial can appear in a parallel bilingual or tete-beche format in the PDF version, but have separate language editions in the HTML version. We have resolved many of these cases by preparing separate bibliographic records for each language edition, but each case must be examined on its own merits. Similarly, a publisher may use a different title for the different versions of an electronic serial, such as PDF and HTML.
What Lies Ahead?
NLC cataloguers are resolving problems on a case by case basis, in consultation with other cataloguers and bibliographic standards specialists. In a rapidly changing environment, those involved in the cataloguing of electronic serials work hard to remain abreast of changes in technology, cataloguing principles and electronic publishing practices.
According to NLC cataloguer Sharon Reeves:
"While cataloguing remote access electronic publications can have its frustrations for cataloguers such as myself, it also provides interesting challenges. To develop cataloguing solutions which allow us to describe a publication accurately while following the spirit of international cataloguing standards, thought and consultation is necessary. It is a continuing challenge to keep up with the pace of change, but one at least feels at the forefront of a new field of cataloguing. It is undeniable that electronic publishing will continue to grow in the millennium, that is until the next publishing format is developed. Whatever that may be, we'll be ready."
Canadian Information by Subject
In addition to creating standard bibliographic records for electronic publications, the National Library has joined the quest for other, non-traditional means of providing access to the burgeoning volume of electronic documents on the Internet. The Library's Canadian Information by Subject (CIBS) (external link) Web service is one such means of providing direct access to Web sites on Canadian topics, by adapting a traditional tool (Dewey Decimal Classification, 21st ed.) for use on the Web to provide a structured, hierarchical subject overview of Websites, while at the same time indexing the sites using bilingual, controlled-vocabulary subject terms. CIBS now indexes some 2,800 sites of Canadian interest, and through hotlinks brings the searcher directly to the site he is seeking. Links are checked automatically, and new sites are added on an ongoing basis. This form of metadata has been found to be highly effective as a resource discovery tool, and has consistently registered as one of the most visited Web pages in the National Library's Website.
The National Library has participated in a number of experiments with other forms of metadata (of which bibliographic information can be considered one form). The National Library was one of several Canadian federal government departments creating GILS (Government Information Locator Service) records for National Library publications and services as part of a GILS Pilot Project to provide Canadian citizens with improved access to government information. As well, some of the digitization projects undertaken by the Library (for example, the Canadian Massey Commission on National Development in the Arts, Letters and Sciences) included the creation of metadata records using GILS.NLC is also interested in the potential use of the Dublin Core.
The following resources are available on the Web:
- The National Library of Canada's collection of electronic publications (external link).
- The National Library's Web homepage (external link), and
- information on how to purchase Canadiana (external link), the National Bibliography on CD ROM.
-- Liz McKeen, Director, Bibliographic Access, National Library of Canada
In June 1996 the National Library of Australia established the PANDORA (Preserving and Accessing Networked Documentary Resources of Australia) Project (external link). Since then the Project has reached many milestones. It has:
- established an archive of selected Australian Internet publications, including serial and non-serial titles;
- developed policies and procedures for the archiving of such publications, including the provision of access to them; and
- developed a proposal for a national approach to the long-term preservation of Australian Internet publications.
Since the Project's inception, the Australian Electronic Unit has been responsible for the selection, archiving, cataloguing and maintenance of titles in the archive. The Unit began with three members of staff and has since grown to five. Although it was initially envisaged that the processes of acquiring, cataloguing and archiving Internet publications would be integrated into mainstream technical service workflows, this has not occurred. As the work is quite different to that carried out by staff dealing with print and other physical format materials, we have found it beneficial to operate with a separate unit to perform all the tasks required when dealing with online publications. Perhaps better technology and simpler processes will alter this in the future.
Procedures for cataloguing electronic publications were developed using AACR2, CONSER Cataloging Manual, OCLC's Cataloguing Internet Resources: a Manual and Practical Guide and LC's Guidelines for the use of field 856. We settled on a simple call number for all online publications, both serial and non-serial. It is "Internet"! The holdings statement for serial titles is: "See Internet for available issues". This statement provides flexibility and requires no updating when issues may disappear in this dynamic environment.
For the first six months of the project, to test the resources required for cataloguing online publications, we catalogued a number of titles, some of which we did not intend to archive. However, we found this practice to be too resource intensive due to the maintenance required to correct out-of-date URLs and discontinued the practice. We made the policy decision to catalogue only those Internet titles to be included in the PANDORA Archive. Since one of our chief selection guidelines is that we do not archive a title that has a print equivalent, (we prefer to preserve the print version), we now rarely use the one record approach.
We have continued to update our cataloguing procedures in line with changes that have occurred over the past two years, for example, the introduction of the "viewed on date" in conjunction with the note describing the source from which the title was taken. Staff at the National Library of Australia have been contributing to discussions in the AACR review and we are awaiting the final outcomes of the review to further clarify cataloguing policy for electronic publications.
Archiving Internet titles continues to provide us with major technological problems; selecting and cataloguing are the easy tasks. Prior to archiving online publications, permission to archive is sought from the publisher, producer or author as there is currently no Legal Deposit provision nationally in Australia relating to electronic publications. A Bill that will include electronic publications under Legal Deposit is expected soon.
Archiving is carried out using Harvest software, which has been specifically modified by IT staff at the National Library of Australia for the task. This software was not originally designed for such a task and has many limitations. For example, it is not capable of gathering titles that have been developed using database and Java technologies. To complement the limitations of Harvest staff began to investigate other software packages designed specifically for gathering web sites. Currently, a package called WebZip is being used and is frequently successful in gathering publications where Harvest has been unsuccessful. However, archiving is a most time consuming task. Each title that is gathered must be compared with the original publication, as one of our aims is to present an archived version that generally retains the look and feel of the original. Exact reproduction is not always possible, however.
Although technological problems in archiving are significant, the archive has grown steadily over the past year, especially the last six months. We feel that this is a major achievement given the capabilities of the gathering software lag well behind those of the software being used to develop web sites. Many more titles have been selected for preservation but as yet have not been archived. This is of major concern to us as a number of them may disappear before the technology is developed to allow us the opportunity to archive them. One of the greatest challenges being faced by national libraries, which have the responsibility for preserving their country's published documentary heritage, is ensuring long-term access to online publications. Through its Digital Services Project, the National Library of Australia is addressing the need for more sophisticated, automated technology and a large scale, robust archive management system to manage the gathering and archiving of online titles (among other functions). More information on the DSP (external link) is available on the Web.
-- Anne Daniels, Senior Librarian, Australian Electronic Unit, National Library of Australia
Serial catalogers within the Cataloging Branch, Library Programs Service (LPS) of the United States Government Printing Office (GPO) first began recording Uniform Resource Locator (URL) information in records for physical format serials in 1994. In the course of reviewing serial issues, catalogers noticed the presence of URL data in an increasing number of paper issues. Initially, URL data were recorded in the 530 note field. Later, in consultation with Jean Hirons, our practices changed to conform to CONSER guidelines for use of the 856 field.
The use of existing serial records for recording these URLs, as distinct from producing a record for an electronic version of a physical format serial, was consistent with GPO's long established policy of using a single record to represent both paper and microfiche versions of serials and monographs. This practice was expanded to include use of an existing paper/microfiche record to record Internet related access prior to the formal issuance of the CONSER Editing Guide's "single record" option.
CONSER's policy allows institutions to use existing records for physical formats to provide hot linked access to electronic versions and authorizes CONSER members to code such "single" serial records for CONSER. In a technical sense, many of GPO's records that have been updated with URL data do not represent "cataloging the Internet", but the recording of Internet related data in existing physical format records. CONSER's "single record" option offers GPO an important means of providing timely access to works made available via the Internet.
Electronic Serials, Cataloging, and the Federal Depository Library Program (FDLP)
Understandably, most CONSER institutions provide CONSER with cataloging records that reflect local institutional acquisitions and collection management decisions. However, GPO's Cataloging and Indexing Program, as required by 44 U.S.C. 1710-1711 has a broader mandate to identify, catalog, and index works, published in all media, by U.S. Government agencies. With the advent of "GPO Access" legislation (P.L. 103-40), which expanded the scope of the program to include access to electronic information through cataloging and locator services, the tasks of providing Internet related access multiplied.
Generally, GPO's selection of serials, monographs, and maps, reflects the scope of the FDLP rather than "local" collection policies. Works in the "program" are "official" government publications of public interest or educational value, that are not restricted for reasons of national security.
This broad mandate to catalog and index works published by more than 250 U.S. Government agencies creates a challenging dynamic that requires innovative practices that help assure timely cataloging of all works in all media. More than fifty percent of all works processed or cataloged by GPO's Cataloging Branch are serials. Approximately 80% of our cataloging is original.
Physical format works are distributed to more than 1,350 selective and regional depository libraries throughout the United States. Although depository materials are maintained in separate collections within most of these depository libraries, the records for such materials are integrated within the Online Public Access Catalogs (OPACS) of most member libraries. To be integrated, our records must and do conform to CONSER and other national level standards.
Complex Manifestations of Serials Remains the Norm
Although a trend towards "Internet only" publishing seems to be increasing among Federal agencies, a review of GPO's CONSER related records indicates that the majority of works published via the Internet continue to reflect a physical format manifestation rather than Internet only availability. Given our need to keep cataloging as current as possible, to meet the needs of FDLP and other libraries, continued use of CONSER's "single record" approach remains a critical component of GPO's efforts.
While the "single record" approach is useful for many institutions, the continued mixture of multiple versions, as distinct from single Internet only versions, continues to add to the complexity of decision-making faced by serial catalogers. From a workload perspective, producing an original cataloging record for a single Internet only serial could be less time consuming than verifying the suitability of a physical format record for recording various electronic versions of physical works.
How Best to Provide Access to Remotely Published Electronic Serials?
Another matter that complicates the decision-making associated with cataloging Internet related serials concerns the choice of an appropriate address for electronic access. Ideally, addresses, either via URLs (Uniform Resource Locators) in records or via PURLs (Persistent Uniform Resource Locators) in PURLs servers should provide users with direct access to works that are clearly reflected in bibliographic records. Unless works are archived and/or are available at a stable address, our practice within GPO is to take users to a clickable index or search window. Experience suggests that either form of intermediate access is likely to be more stable than are addresses that take users to specific issues. Following CONSER policies, we take users to the run of issues or to addresses that hold the promise of taking users to the run.
Providing Access via PURLs
At the present time, approximately 8,000 Monthly Catalog records are estimated to contain URL or PURL data. Most of these records remain available for hot linked access.
An important stage in the evolution of providing cataloging and locator services for electronic resources has been the large-scale application of OCLC's PURLs software. When properly maintained, PURLs provide a seamless redirect from a PURL in a bibliographic record to the most recent active URL at agency websites. Use of PURLs means that electronic access is maintained by changing addresses in servers in preference to repeatedly changing URLs in bibliographic records. To be suitable for large-scale applications, OCLC personnel agreed to implement new features associated with automatic links checking of PURLs and with development of reports modules. As with earlier releases, OCLC's most recent public release of PURLs software is freely available for downloading and use.
What's in the Future?
As GPO's Cataloging and Locator Services evolve in the context of providing access to an FDLP electronic collection, serials cataloging may be affected in at least two ways. First, the use of electronic archives may offer opportunities to catalog and provide appropriate access to an entire run of serial issues. If this objective is achieved, we expect to be able to catalog works that we are presently unable to represent as remotely accessible electronic serials. Second, if depository distribution of multiple physical formats is eliminated in favor of providing access to electronic equivalents, we anticipate that an increasing number of serials and non-serial works will be represented by "electronic only" records. Much remains unclear as to future developments. What is clear is that continued change and the need to adapt to change will continue for many years.
For more information, consult: Managing the FDLP Electronic Collection: A Policy and Planning Document, Library Programs Service, U.S. Government Printing Office, October 1, 1998. GPO cataloging policies for collection level records, computer files, and linking fields are available on the Web.
-- Thomas A. Downing, Chief, Cataloging Branch, United States Government Printing Office
To date, MIT has cataloged or provided access to about 500 electronic serial titles. I and the other serials catalogers want to share with you some observations made, problems encountered, and practices that we've adopted along the way.
- Our default cataloging method is the much-loved single record method. It saves us time and our public services staff and patrons prefer to see one record in the catalog rather than separate records for all the versions. We have, in fact, expanded our use of it so that we now "piggyback" the e-serial cataloging information not only quite properly on records for the original (generally, the print or CD-ROM versions) but also on those for the microform reproduction as well (the latter locally only, of course: we don't do that in the CONSER or OCLC record).
- The line between what is a serial and what is a monograph on the Web is a little blurry. The current AACR2 definitions simply are not adequate. When a publication which is an obvious serial in its tangible incarnation goes online, suddenly it's a database, or a serial with combined issues, or maybe it's just a monograph?
- It is beneficial to incorporate e-serial cataloging into our normal and well-established workflows. E-serials are no longer "exotic." MIT selects an e-serial or a package of e-serials, our digital resources librarian lets Serials Cataloging know that it is accessible and ready for cataloging, and we catalog it. We've also found that it's been beneficial to document our cataloging policies and procedures (including local ones) so that once we decide what to do we don't have to agonize when similar situations pop up later. This has proved useful because we don't have e-serials to catalog every day. We may receive access to a few titles, or a couple of packages, all in one week, and then some weeks may pass before we see any more. (Our documentation can be found on the Web.) (external link)
- Some e-serials we've cataloged have been well-designed and a breeze to catalog, but others are atrociously designed and have taken us a lot of time, effort, and, yes, frustration to catalog. On one the title of the electronic version was abbreviated and so poorly displayed and we decided against providing access to the title on the record for the print.
- The truism about the Internet is, well, true: Web sites are highly volatile. The arrangement of titles on a site may change. The site itself may move and therefore all those URLs we have diligently recorded in the catalog records will need to be changed. Even the titles of serials on a site may change or fluctuate or be retrospectively converted to the latest title. Keeping our holdings up to date is a real challenge. We are sometimes uncertain about publishers' archiving practices. How long will they retain the earliest issues that were there the day we cataloged the serial? Another problem is that we have no efficient and effective way to know when publishers add new titles on their sites. Very few publishers notify subscribers (JSTOR is the only one that comes to mind) and of course the result over time will be that there are titles to which we have access but for which there are no records in the catalog because we were never explicitly informed that we DO have access. One possible solution is to regularly revisit sites to see if titles have been added but this is time-consuming. Another method would be to take advantage of a service that would alert us when a site is updated. One site we are currently exploring and informally testing is called Mind-it. We would love to know what other CONSER libraries are doing in this regard.
- We have also been experimenting with a shareware link-checking program called LinkLint. We ran it for the first time in Dec. 1998, and hope to run it monthly in order to do a broad check on which URLs in our catalog records are not working. We have found that it does not produce the prettiest and most user-friendly of reports, but it does seem to do the basic checking we want it to.
- In the big debate of whether e-serials should be cataloged at all -- that is, whether records for e-serials should be created and included in the catalog -- we've come down on the side of including them in the catalog for now. There may come a time in the next months or years when meta-tags and other methods of providing non-catalog access to these resources will make traditional cataloging of them seem rather quaint. But for now we feel that it would be a great disservice to users to force them to rely solely on Web lists and search engines. Access to e-serials at MIT is now provided two ways: via the catalog and via lists of titles on the Libraries' website (including a list of e-journals (external link), and a list of other e-resources (external link), some of which are monographs).
- Cataloging electronic serials has caused us to re-examine and challenge some of our basic principles about bibliographic access. And that, we believe, is a good thing!
-- Wayne Jones, Serials Cataloging, MIT
The issue of providing access to full text journal titles that have been bundled together into a large single database was brought up during the spring CONSER Operations meeting.
These aggregator services are expensive resources, often containing hundreds of full text journal titles. Library patrons, as well as internal library units, such as ILL and reference departments, need a single source, preferably the local OPAC, to search for all serial titles which a library owns or has access to.
While access to these titles is made difficult by the sheer number of titles available from any single service, the problem is compounded by the maintenance issues as titles are added or dropped by the vendor with ease. Insuring the accuracy of information provided to the user requires constant attention from library staff.
At the spring meeting, a small Group, which includes Jeanne Baker (Maryland), Karen Calhoun (Cornell), John Riemer (Georgia), and Ruth Haas (Harvard), was set up to investigate the issues and interest among CONSER participants for possible solutions.
The Group initially focused its attention on surveying CONSER institutions about their needs, what means they are presently using to provide access, if access is not being provided, why not, and, in an ideal world, what means would libraries find best for providing access. Responses to the survey pointed to the fact that, faced with shrinking budgets, cooperative solutions, both among libraries and with vendors, need to be explored further.
Over the fall, a survey of the whole library community was undertaken by John Riemer. This survey seeks information on types of access being used, access methods libraries would most like to have available, and, the levels of willingness to participate in cooperative solutions to the problem. Results of this survey and possible further actions will be presented by John and Karen at the ALA Mid-Winter Cataloging Management Discussion Group (Saturday, January 30, 2:00-4:00).
With these surveys as a base, the Group will be assessing the structure, or structures, for bibliographic information which would be most useful within the library community and exploring partnerships among libraries and vendors which will most efficiently provide the data.
-- Ruth Haas, Chief Serial Cataloger, Harvard University
CONSERline (ISSN 1072-611X) is published at least semiannually by the Library of Congress, Serial Record Division. CONSERline is a cooperative effort with contributions from program members consisting of news of the CONSER Program and information of interest to the serials cataloging community.
For comments or suggestions, contact the editor, Jean Hirons, Library of Congress, Serial Record Division, Washington, DC, 20540-4160; [email protected] (email); 202-707-5947 (voice); 202-707-1778 (fax), or: Bill Anderson, CONSER Specialist, Library of Congress, Serial Record Division, Washington, DC, 20540-4160; [email protected] (email); 202-707-5185 (voice).
CONSERline is available in electronic form only and is free of charge. Back issues are available from http://www.loc.gov/aba/pcc/conser/conserline/
All materials in the newsletter are in the public domain and may be reproduced, reprinted, and/or redistributed as desired. Citation to the source is requested.