This is a joint message/paper prepared by Jean Hirons and Regina Reynolds which presents a proposal to adopt a Modified Model C approach. This model has arisen from the problems encountered in the process of redefining the term "serial." We have concluded that Model B is not adequate as either a short or long-term approach. Additionally, Modified Model C demonstrates some "natural" break points which can help us determine where latest and successive entry conventions should be applied. This paper also proposes a revised plan of action for the rule revision process.
In the paper "Issues Related to Seriality," Crystal Graham and Jean Hirons recommended that Model B, which would redefine "serial" by removing the requirement for numbering and successive parts, be adopted as a short-term solution and that the library community work towards the better long-term solution of Model C, which would divide the bibliographic universe between static and ongoing publications. Participants at the conference agreed; however, there really wasn't very much discussion. There were two basic reasons why Crystal and Jean did not recommend an immediate adoption of Model C. First was the feeling that the cataloging community was not ready for such an approach. Second was the potential problem of implementing Model C within USMARC, the bibliographic utilities, and our local systems. We did not want to propose a solution that could not be implemented within a reasonable time frame. Recent discussions, however, have led us to believe that indeed the library community is ready and that there may be a way to implement Modified Model C within USMARC without making major changes. Jean recently discussed these ideas with Sally McCallum (Chief, Network Development and MARC Standards Office) and was encouraged to proceed.
Since ALA in January 1998, AACR Review Group 1 has been working on the redefinition of "serial" according to Model B. A "working draft" was sent to a large number of individuals for comment. A relatively small number of comments have been received, largely in favor of the definition, however, none were wildly enthusiastic. Those commenting noted the problems with loose-leafs, the potential difficulty of determining what is a serial and what is a monograph, and, in one case, a desire for a more rapid move to Model C. The basic problems with Model B are twofold: 1) by removing the requirement for numbering and successive parts, key differentiating characteristics for traditional print serials, it may be difficult to distinguish monographs from serials; and 2) by retaining the requirement for the intention to be continued indefinitely, we impose a distinction on publications such as loose-leafs and databases that is difficult to make and serves little purpose. Thus, we feel that Model B is not even a workable short-term solution. Furthermore, we don't believe that we have the luxury of implementing one solution with the idea of moving towards another, given that any change will be disruptive and will take time.
Loose-leafs are particularly problematic under Model B and Ann Sitkin (Harvard) has formed a small group of experts to review the ramifications of incorporating at least some loose-leafs within the Model B definition of serial. Loose-leaf catalogers cannot see any benefit in having to distinguish loose-leafs that are intended to be continued indefinitely from those that aren't. In fact, this is a category of material for which one cannot make that distinction, at least not easily, because publication patterns for loose-leafs differ from both monographs and serials. Thinking about the similar difficulties in determining intention to continue indefinitely for many electronic publications also makes us wish for a more rapid move to a different, more reasonable approach.
The particular problems mentioned above are perhaps only symptoms of an underlying problem with Model B--that simply expanding the current definition of "serial" and dealing only with entities that are intended to be continued indefinitely does not promote the concept of the "ongoing publication" that was addressed in the Hirons/Graham paper. It is only a partial solution. What Regina and I are proposing in this paper is a more encompassing approach that is more in keeping with the spirit of the JCS paper.
Embrace the concept of the "ongoing entity" as the overarching concept under which other categories of entities, such as "serial," "loose-leaf," "database," etc. will fall. In this process, also define "monographic entity." The realm of ongoing entities would encompass all bibliographic entities that are intended to be continued (for some period of time) and which are characterized as being either a succession of equal discrete parts, each with an associated title, or being a single entity to which updates and/or additions are integrated into a seamless whole, with one title associated at any given time with that whole entity. The fact that an entity is electronic and capable of revision does not alone make it ongoing.
"Ongoing entity" refers to the very basic nature of the work and its content. Within this broad category, entities are differentiated by their differing forms of issuance: successive discrete parts (serials, series, and multiparts); and integrating parts ( loose-leafs; databases; web sites). The distinction between Model C as given in the paper and this proposal, is that the intention to continue indefinitely is no longer a primary dividing point between categories of ongoing publications. However, we propose to retain the distinction for successively-issued entities, i.e., serials and multi-parts. Thus, we have a more flexible model that allows us to make distinctions where they are useful and not to force them when they are not useful.
We have excluded from the realm of "ongoing entities" entities that are supplemented or updated without the parts being integrated into the whole (e.g., monographs with annual supplements, monographs with pocket parts, etc.). While we recognize that these entities are not complete and require some of the control of other ongoing entities, we have excluded them because the supplements are subordinate to the base work and are not likely to impact on the description of that work as to its title, statement of responsibility and so forth.
Under this proposal, the category of "serial" remains largely unchanged--as the subset of successively-issued entities, generally bearing a designation, that are intended to continue indefinitely or for an indeterminate period of time. However, under this proposal serials fall under the larger umbrella of "ongoing entity." Thus, we can provide rules that recognize and accommodate an entity's ongoing nature and its potential for change that can be applied to all "ongoing entities." This will help resolve many of the current problems which are caused by entities having to be divided strictly into monographs and serials.
A copy of this paper and a diagram of the revised model appear here.
C. POTENTIAL IMPACT
AACR2: In the rules, we envision a single chapter, perhaps a revision of chapter 12, entitled "Ongoing Entities." The chapter would include a general section with rules and guidance pertaining to all ongoing entities and would be divided into sections treating the different categories of ongoing entities, successively-issued and integrating entities, perhaps following the hierarchical breakdown in the revised model. Where appropriate, identification of bibliographic elements would be prescribed rather than exact transcription, and different rules would apply, as appropriate, based on the characteristics exhibited by the different categories of ongoing entities. We could also envision including the rules for what constitutes a title change and conventions for handling changes in entry--now in Chapter 21-- in this revised chapter or in a separate section within Chapter 21.
USMARC: We propose that the serials bibliographic level "s" be redefined to be used for all ongoing entities. We also suggest renaming the "serial type" (008/21) to "ongoing type" and expansion of the codes to identify entities such as loose-leafs, databases, and web sites. What this would mean is that within one 008 we could include all language materials that are considered ongoing entities, according to the defined term. We believe that the types of data coded in the current serials 008--publication status, frequency, and serial type--are also applicable to other types of ongoing entities and would be more useful than many of those in the books 008.
However, while we believe in somewhat of a purist approach to the rules, we could see making compromises when it comes to implementing this proposal in USMARC and our catalogs. For instance, we might want to keep language-based multi-parts in the Books 008. This is an issue for later consideration but we do not foresee that making some limited exceptions would seriously compromise the benefits of this proposal.
SYSTEMS: This approach could be problematic for searching and retrieval in some systems and is something we need to think about carefully. How many systems divide files? How would this impact on search qualifiers? Would it make sense to patrons? Perhaps the primary problem is that "true" serials would only be identifiable through several codes contained at a lower level within the record. What are the potential ramifications of this change?
RECORD MAINTENANCE: One of the big advantages of the proposed approach in USMARC is that it would group together those records that are most in need of change. This might make it easier for utilities, such as OCLC, and cooperative programs, such as CONSER and BIBCO, to focus on record maintenance and how it can best be achieved.
PROGRAMS: ISSN, CONSER, etc.: This proposal will pose some practical considerations for both ISSN and CONSER. If the definition of serial is retained within the broader concept of "ongoing entities," will serial programs also want to broaden their coverage? Would CONSER want to catalog databases, for instance, and might the ISSN Network want to assign ISSN to such entities? What are the characteristics of the various types of ongoing entities that might qualify them or disqualify them from ISSN? This proposal does not cover who should catalog what; that is an issued to be addressed later; however, we don't foresee major changes. By dividing ongoing entities into various categories, this model offers the possibility that groups can choose which types they will cover or not.
D. LATEST AND SUCCESSIVE ENTRY AND THE WORK
In our discussions and in reading the comments from AACR Review Group 1 concerning latest and successive entry, we found that we could justify single record (latest) and multiple record (successive) approaches based on a combination of the form of issuance and the perceived number of works.
For updating entities that are continually integrated into a single part (or parts in the case of loose-leaf services), it is easy to see that latest entry is a better fit. We also recommend that multi-parts continue to be handled using a single record because they are planned as a single work, even if that work takes some time to be completed. Serials, on the other hand, are issued over time and as they evolve may change their titles and scope and can become quite different works. Thus, we see these as a succession of evolving "sequential" works, or perhaps segments in a "mega-work." We agree with most of those in AACR Review Group 1 who argue that multiple records for successively-issued entities are easier to understand and better accommodate these publications. We hasten to say, however, that our current rules for title changes call for successive records for changes that do not in any way signal a new work (e.g., Atlantic and Atlantic monthly). And we recognize the need to work on this to reduce the number of meaningless changes, as best we can. There is no perfect solution; but improvements can be made and we are confidant that Group 3 will identify them!
Electronic journals are problematic because they take on different forms of issuance and are losing some serial characteristics, such as numbering and publication in issues. However, we would argue that these still be treated as "serials" because they are ultimately comprised of articles, which constitute distinct additive parts. We need to find the best way to describe them while recognizing their intrinsic nature.
E. WHAT DOES THIS MEAN? WHERE DO WE GO FROM HERE?
Assuming that most of you think this is a good idea and that the JSC is in support of this proposal, how does this change the current charges? Here are our thoughts.
The four groups should remain as they are. They each have distinct issues to address and should concentrate on identifying and resolving those issues. However, it is currently difficult to imagine the feasibility of keeping to the current proposed time frames and the idea of suggesting rule revision proposals over a period of time. We propose that we leave the actual drafting of rule revision proposals to several members of the CONSER task force, which would base their proposals on the recommendations from the groups. Our goal would be a single proposal to JCS, consisting of a revised chapter 12 and accompanying definitions. This process should result in a more unified and well thought out set of rule revision proposals. This new time frame will also allow for more international discussion and "buy-in," also essential to the ultimate adoption of this proposal. While the groups are working, we can test out the proposed concept on CC:DA and MARBI and ask for their assistance with the problem areas.
A suggested time frame: Groups 2A, 2B, and 3 should identify the specific issues they plan to address by ALA in June. They can begin discussing those issues at ALA and should have recommendations to the CONSER AACR Review Task Force by ALA Midwinter 1999. The Task force will then establish a working group to draft the proposals with a goal of completion by ALA in June 1999. This is ambitious, however, we think it can be achieved.
Group 1 will need to revisit definitions. Some of the terms we might need to define are: bibliographic entity (already in draft form), monographic entity, ongoing entity, form of issuance, sequentially-issued entity, serial entity, periodical, multi-part entity, and integrating entity
F. BENEFITS OF THE PROPOSAL
- Promotes the spirit of the paper and the need to integrate provisions for handling ongoing entities more broadly into the code
- Keeps works and their manifestations (e.g., print and online versions) broadly grouped together
- Does not force meaningless and difficult distinctions (e.g., intention to continue indefinitely for loose-leafs and Web sites)
- Collects the rules for entities that exhibit change in order to recognize similarities, while allowing for differences
- Allows us to include rules for integrating entities in one chapter rather than multiple chapters
- Could group together records in USMARC (under code "s") that are most in need of maintenance
- Provides an approach that we hope will cause minimal disruption to systems
- Is flexible enough to accommodate new forms of issuance and types of entities.