ALA/ALCTS/CCS Committee on Cataloging: Description and Access - Task Force on Rule 0.24: Interim Report on the Delsey Model (1/1999)

ALCTS - Association of Library Collections & Technical Services

CC:DA/TF/Rule 0.24/3

January 6, 1999

Committee on Cataloging: Description & Access

Task Force on Rule 0.24

To:      CC:DA

From:    Martha Yee

Subject: Interim report from the Task Force on Rule 0.24

Attached is an interim report from the Task Force, constituting our evaluation of Part I of Tom Delsey’s model of AACR2 (The Logical Structure of the Anglo-American Cataloguing Rules, Part I). Given our charge, we have tried to focus on content/carrier and edition/manifestation issues in the model. One thing we would like to hear from CC:DA as a whole is whether or not we have sufficiently focused on these issues, or whether you feel we have overstepped our charge in some areas. A related issue is whether CC:DA needs to figure out another mechanism for responding to the report as a whole, such as creation of another task force.

Our goals for the next six months consist of the following:

Evaluate Part II of Tom Delsey’s model of AACR2, focusing on content/carrier and edition/manifestation issues in the model.
Prepare our final recommendations concerning Rule 0.24 for CC:DA consideration at the Annual meeting and for submission to JSC for their fall meeting.

With regard to the latter, we can already see at least two functions that are currently carried out by Rule 0.24:

It provides guidance in the cataloging of items with multiple characteristics that must be cataloged using more than one chapter in AACR2R.
It provides guidance to the cataloger concerning when to make a new record, i.e. when an item should be treated as a new manifestation or a new edition of a work already represented in the catalog.

We intend to make recommendations concerning both functions of Rule 0.24, and would appreciate any input CC:DA would like to give us to aid us in this process. In particular, we would be interested in gauging where CC:DA members and liaisons currently stand on the question of whether a reorganization of Chapters 1-13 of AACR2R based on ISBD area would be likely to improve the way AACR2R carries out the first function above. Would such a reorganization make more efficient the cataloging of items with multiple characteristics such as a digitized map that is continuously updated (currently cataloged using Chapters 3, 9 and 12)?

Thank you for any guidance you can give us.

THE LOGICAL STRUCTURE OF THE
ANGLO-AMERICAN CATALOGUING RULES, PART 1

General Considerations

Evaluation of the Modelling Exercise

Most members of the Task Force have an overall positive response to the data model. One Task Force member writes, “Delsey’s AACR data modeling exercise is valuable because it gives us another opportunity to reassess AACR to make sure the rules are written as well and as clearly as they can be written.” Another writes, “My initial reaction [to the model] was an intuitive sense of recognition: oh, yes, that is how things work.” On the other hand, a number of Task Force members confessed that the model was difficult to understand. A model of descriptive cataloging that is not perfectly clear to working catalogers and cataloging administrators may be somewhat dangerous. Errors in logic may be allowed to slip by unnoticed. Also, one Task Force member points out that the model just about approaches the length of AACR2R itself. Such length, complexity and density is likely to prevent most commentators from having the time or the knowledge to be able to subject it to a thorough analysis for accuracy.

One Task Force member felt that the language of the model was needlessly arcane, and that better understanding of the model would be promoted by employing more familiar and more concise words, especially if they were more carefully defined.

Why This Model?

It would be helpful if this document contained 1) a description of the entity-relationship model and the object-oriented model (or at least a citation to a book or article that might provide an accurate description of these two models), together with 2) a statement addressing the question of why it is useful to use a data model developed to describe business databases to describe our data. Some thought the data model oversimplified, and overly rigid in places. One Task Force member pointed out that “we are, in fact, superimposing a model constructed in a different field of study (computing) onto cataloging. It’s not always going to be a good fit.”

The model as applied by Delsey to cataloging raised a number of questions about relational and object-oriented models in general:

Can one attribute apply to more than one entity? For example, the title on the item may apply to both a particular edition and a particular work (as when it actually represents the uniform title chosen for that work).
Does the relational model and/or the object-oriented model require that it be possible to create an exhaustive list of all possible values for any given domain? We collect many types of data concerning which one would not want even to attempt to create exhaustive lists of all possible values.
Can either model deal with many attributes that may be blank in any given instance? Not all items belong to a series. Not all items have an edition statement. etc.
Can either model deal with complexities such as the following: a corporate body responsible for release (i.e. a publisher or distributor) may sometimes also be considered to be a kind of author, as in the case of learned societies from which a work may emanate, or as in the case of original distributors of motion pictures, but usually is not (as in the case of a commercial publisher of a current trade publication). Publishers of rare books are often indexed and searched, but publishers of current trade publications usually are not. This kind of thing is left up to the judgement of catalogers. So is the publisher an attribute or an entity, or sometimes one or the other and sometimes not?
Is either model hospitable to variable length data, including data composed by the cataloger or transcribed from the item, as opposed to selected from an exhaustive list of all possible values?
Is either model hospitable to many complex relations, as when a particular document is one of many works by a particular author, a particular edition of a particular work, a member of a series, and a volume in a multi-volume set?
And the basic question—why do this? Why try to model our data after the data found in a business database? What advantages do we gain by this? What have we found out from having done this that we didn’t already know?

“Real World” vs. “Abstract”

It would help if the document indicated why it was felt necessary to distinguish between the two. Perhaps it is time for us to figure out why we want to use the term ‘abstract,’ and why we keep changing our minds about what is ‘abstract’ and what is ‘concrete.’ Before AACR1, the abstract thing was a ‘work,’ and the real thing was an ‘edition’ or an ‘issue.’ In AACR1, this was still applied to monographs, but for everything else, the abstract thing was a ‘work’ and the real thing was an ‘item.’ In AACR2 (before Delsey), the abstract thing was a ‘work’ and the real thing was an ‘item.’ FRBR calls work ‘abstract’ and item ‘concrete’, but fails to designate either expression or manifestation in this regard. Now Delsey suggests that ‘item’ is abstract and ‘edition’ is real (see below)! Where are we going with this? As noted above, the concepts ‘real world’ and ‘abstract’ seem to be very oddly assigned to entities in this model. For example, ‘Edition’ is considered to be a ‘real world entity,’ as denoted by a square box, as opposed to ‘Item,’ which is considered to be an ‘abstract concept,’ as denoted by a box with rounded corners. Some of us would have thought that just the opposite was true: the cataloger begins with a real thing, the item, and, based on clues provided within the item, makes a decision concerning whether it is likely to be a particular edition of a particular work, both more abstract concepts. Also, it reminds some of us of one of the first lessons we learned in cataloging—that our job is not to figure out what is really the case, e.g. who is really the author, or whether this text is really different from that one that purports to be a different edition, but rather to report on how the document represents itself (by means of transcription), and to correct that only when it is known to be erroneous. However, one Task Force member feels that there are cases in which catalogers do figure out ‘what is really the case’ and report it to users.

Too Hierarchical with Too Much Emphasis on Part-Whole Relationships?

The model as a whole strikes most of us as being too hierarchical, with too much emphasis on part-whole relationships.

Hierarchy:

We think the following six facets cover the six most significant factors about any given item. According to this analysis, for any given item, you would pick one element from each facet. With those six facets designated, you should be able to decide how to handle any particular aspect of description:

Facet 1, Type of publication (use of the term ‘publication’ is unfortunate, however, as most unpublished manuscripts should probably be treated as monographs in this regard, and there may be a few unpublished serials; one Task Force member suggests that we try to come up with a better term than ‘type of publication,’ ‘type of stabilization,’ perhaps?):

Facet 2, Content (hardest to define, but we could start with the content categories already designated in AACR2): See the categories we suggested in our response to key issue 1:

Facet 3, Published vs. unpublished

Facet 4, Physical carrier:

Facet 5, Method of management:

Facet 6, Reproduction:

Concerning facet 6, some Task Force members support getting away from the requirement to determine simultaneity or succession of release in the case of reproductions, as such facts can be very difficult to determine, particularly in the case of electronic resources. Up until now, catalogers have been required to designate one item an ‘original’ and another a ‘reproduction’ in order to use any type of multiple version approach to cataloging.

Sometimes descriptive rules would have to be different based on the category in Facet 1 that an item fell into; sometimes rules would have to be different based on the category in Facet 2 that an item fell into; and so forth.

Most of us don’t quite see why these facets need to be given the different kinds of hierarchical relationships to each other that they are given in Figures 4-11, where type of publication and class of material are ‘belongs to’ categories, published vs. unpublished and library management vs. archival management (collection) require different figures to express the difference, physical carrier is a hierarchical part of document, and reproduction is seen as an attribute (‘Method of replication’) of Manufacture, and as a relationship type (Tab 27, page 10). It might work better to treat all 6 facets as equal in importance. If AACR2 were to be reorganized based on ISBD areas, according to Recommendation 1, each chapter (ISBD area) could begin with general rules, and then qualify them as necessary for variations in the 6 facets listed above, giving one or another facet precedence as judged optimum in any given area.

However, one Task Force member is not too concerned about the degree of hierarchy in the model. He writes, “I’m not sure I agree with your characterization of the relationship among entities as hierarchical. There is a logical hierarchy: DOCUMENT contains CONTENT set as INFIXION stored on PHYSICAL CARRIER housed in CONTAINER, but I’m not sure that I would imply too much here; I certainly would not imply priority."

Part-whole:

Part-whole is a relationship type AND a major part of the central hierarchical structure of the model, appearing in the form of Document part and Content part directly under Document at the center of the model. The emphasis seems particularly odd when you reflect on the many ways AACR2 allows part-whole relationships to be expressed:

Whole only described, with parts listed in a contents note.
One record for all parts, which are listed in the title field separated by semicolons (when the whole has no collective title).
One record for the whole and separate analytic records for the parts, each containing a series entry for the whole.
Separate records for the parts, with the whole represented only by a series entry.
One record for the whole with a contents note, and in addition, separate records for the parts with IN: notes describing the whole.
Separate records for the parts with WITH: notes listing all of the other parts making up the whole.
Whole only described with parts briefly mentioned in the physical description as accompanying material or even as “illustrative matter.”
Whole only described with parts briefly mentioned in a note, such as a bibliography note.
Multi-level description of whole and its parts.

Since AACR2 leaves libraries pretty much to their own devices to decide which of the above techniques to use, based on factors such as distinctiveness of titles, value of items to users of the library, need for intensive subject analysis, etc., and since any given item could be correctly described using any of several of the above techniques, it seems odd to make this particular relationship so central to the model.

We are confused here, too, about whether the model is really trying to describe AACR2 as it is. Document Part, for example, includes both an accompanying teacher’s guide, and an analyzable volume, yet under current AACR2 rules these two different items would be treated completely differently, as would the items listed in a contents note and the illustration statement in the physical description of a book (both included as part of Content Part).

Extent:

The document lists Duration as an attribute of Content, under Item, for Musical content and Cinemagraphic (shouldn’t this be ‘cinemaTOgraphic’?) content, yet as a data element under the attribute Extent of carrier under the entity Physical carrier for items that measure extent in number of pages, leaves, etc. Some of us don’t quite understand why this different treatment of extent was considered necessary. Also, we don’t think either treatment takes into account the high likelihood of a change in extent signalling an actual change in edition, proven by research on both monographs (Studies of Descriptive Cataloging, 1946) and moving image material (Martha Yee’s research). That fact means that extent has a much more important function than simply characterizing the content of an item.

One Task Force member argues that number of pages is really extent of carrier for a book, not extent of content; extent of content would have to be measured by means of counting the number of words. However, we have used page counts (as the only clue available) to signal probable content change for so many years, research (see above) has shown that it is a very reliable clue, and no one will ever be able to count words in all the millions of volumes in our research libraries, so another Task Force member hates to jettison this valuable clue—or at least consign it to ‘mere’ physical carrier status …

Reproduction:

Reproduction is seen as an attribute (‘Method of replication’) of Manufacture (TAB 20, p. 2), and as a relationship type (Tab 27, page 10). However, treating reproduction as such a minor aspect of the model does not seem to take into account the fact that two chapters of AACR2 (Chapter 9, Computer files, and Chapter 11, Microforms) are devoted to common methods of reproduction (digitization and microfilming). Is this really a data model of AACR2 as it is?

One Task Force member defends the model, though, saying “Two chapters in AACR2 may be devoted to common methods of reproduction, but I doubt that anybody would know that by reading the text of those chapters. The concept of reproduction (as the Mulver TF quickly figured out) is not really explicit in the rules. If you are seeking guidance about how these chapters should be applied to the description of reproductions, then I support your quest, but this isn’t part of Tom’s descriptive model (because it isn’t currently in the rules) nor is it one of the issues he chose to address regarding changing or extending the rules. Nice to know he left something for the rest of us to do.”

Since TAB 20, p. 2 is bearing the weight of reproduction in this analysis, shouldn’t it include microfilming, digitization, film to tape transfer, sound disc to tape transfer, etc.?

Type of Publication:

Type of Publication seems to tease out one aspect potentially present in all items (monograph and serial). However, it is confusing that in Figure 4 ‘Document’ “belongs to” ‘Type of Publication,’ but then ‘Serial publication’ shows up as a sub-type of ‘Document’ (Tab 7, page 48), causing one to wonder whether this data model is clean and logical yet. Is this revealing of a failure in logic in AACR2, or a failure in logic in the model itself? Two Task Force members are fairly sure this is due to AACR2’s failure in logic. One of them writes, “I think what Tom is doing with the TYPE OF PUBLICATION entity (as opposed to the subtypes of DOCUMENT) is to point out that CHIEF SOURCE is determined by two things: CLASS OF MATERIALS and TYPE OF PUBLICATION. Rather than adding Chapter 12 to Table 1 where it would have nothing in common with the other chapters, he chose to treat TYPE OF PUBLICATION separately (which logically it is). Note also that TYPE OF PUBLICATION has a particular role within the model; we aren’t filling in fixed field codes that identify some feature of the item; we are selecting a chief source (which is sort of a fundamental point of departure for applying the rules); all other consequences of TYPE OF PUBLICATION are not relevant at this point in the model. As to the use of ‘publication’ as a DOCUMENT sub-type, Tom explicitly recognizes that this is a problem; it should be "serial document" but that’s not the terminology the rules use—this is a further implied restriction in the definition of serial that is probably not warranted; I believe that there are manuscript serials that I would consider unpublished. Again Tom recognizes this problem in his discussion of Issue #3.”

Infixion:

Despite the fact that this concept does not appear anywhere in AACR2, this is a most intriguing one, and one it is good to get out on the table. It doesn’t seem to be very cleanly applied yet, though. For example, sound vs. silent for film refers to a difference in content. Either a film is made as a silent film, in which case it was not intended to have sound (with the possible exception of a music track), or it was made as a sound film, in which case a silent copy would be defective. One Task Force member suggests, though, that “Yes the sound and color is inherent in the CONTENT, but I don’t think Tom is using CONTENT in that way. I suspect that he would say that the sound that is inherent in the CONTENT is only present because it is realized through some kind of encoding and then stored on a carrier. He is making logical distinctions, not ‘real’ distinctions: you can’t have CONTENT without the INFIXION (or the CARRIER, for that matter), but you can try to separate the concepts (?results of authorship vs. results of encoding vs. results of manufacture? or something like that).”

There is some strange categorization going on here, with sound designated an attribute of sub-type film/video, but color of films and video designated as an attribute of sub-type graphic content (and one might ask where ‘graphic content’ is referred to in AACR2 as it is, too; is this really a model of AACR2 as it stands now?).

And color and infixion in general seem to have been ignored for primarily textual materials. What about color illustrations in a book or on microfilm? What about infixion (formatting?) as applied to text? Text is listed under ‘INCLUDES’ but not under ‘ATTRIBUTES’. In the digital environment, at any rate, this might be a useful concept to deal with ASCII vs. Word 6.0 versions of text or different font sizes, and even in the print environment, it might be a way to deal with braille, vs. large-print vs. regular text editions.

Also, the relationship of infixion to reproduction needs to be examined, as it seems to be a concept that should be applied whenever an item is reproduced in a slightly different format, e.g. when a film is reproduced with a different aspect ratio, or a color film is reproduced in b&w for television broadcast, or perhaps when a book is reproduced on microfilm (ink on paper infixion becomes photographic infixion?)?

p. 22 seems to equate to infixion the stages of film preprint leading up to the print, yet TAB 22, page 15 classifies video reproduction/copying stages under COPY. This inconsistency should be cleared up, and in addition, the stages of still photography and sound recording reproduction should be included.

One Task Force member suggests that the logic behind infixion could be cleaned up a bit if it were compared to the “OCLC When to Input a New Record” guidelines that describe widespread AACR2 cataloging practice.

An electronic resources expert comments that “With regard to ‘Infixion,’ the model encompasses the attributes of ‘sound,’ ‘sectoring,’ and ‘recording density,’ which are related only to local/direct access computer files (as distinct from remote access files).” How is it proposed to apply the concept of ‘infixion’ to remote access electronic resources?

Publishing:

Figure 4 indicates that a corporate body can be responsible for release (i.e. can be a publisher), yet TAB 15 does not mention the publication/distribution area rules and TAB 7 p. 13 lists a publisher as an attribute of Document, rather than an entity in its own right (as Figure 4 would seem to indicate). This makes one wonder whether this model can deal with the complexities of noncommercial publication which is sometimes seen as a kind of corporate authorship vs. commercial publication which is generally not vs. motion picture distribution which is seen as a kind of authorship for the primary distributor but not necessarily for subsequent distributors, etc.

Why is release an entity (TAB 21) when publication is an attribute of Document (TAB 7, p. 28-36); compare, for example, TAB 21, page 3 with TAB 7, page 30. Is TAB 21 meant to cover original release of the first edition only?

More Specific Comments:

TAB 1, ITEM: Shouldn’t published document and published document part include series statement in the attribute list? Two editions of a work can differ in whether or not they have a particular series statement. One Task Force member also thought extent belonged in this list (see comments above).

TAB 7, DOCUMENT: p. 12, Note 1, the difference in rules for statement of responsibility for a sound recording vs. the statement of responsibility for a motion picture or videorecording do NOT relate to the “physical form of the document” but to the content! Musical composition and performance are just two of the many subsidiary functions that go into creating cinematographic content (and other functions are considered more primary, such as direction and screenwriting), while musical composition and performance are primary for the creation of purely musical performance on sound recording.

TAB 9, CONTENT: Where is sound content that is not literary or musical? Bird song? Thunderstorms?

TAB 16, PRODUCTION: p. 3: query concerning place of production for motion pictures. Just as a point of information (as it does not apply to AACR2, just to a manual designed to accompany it), rules for archival moving image materials already place the country of production agency in Area 3 (257 field in USMARC). Archival moving image catalogers also make a note (518 in USMARC) to indicate the location at which a film was shot, if applicable. — p. 7: How does recording system differ from infixion?

TAB 19, EQUIPMENT: One could argue that many of the GMDs and SMDs signal equipment needs. For example, you can’t play a sound CD on the same machine as an audiocassette. — p. 3: Panavision (a motion picture wide screen process) is 1) “infixion” (aspect ratio, tab 11, p. 16), 2) “recording system” (tab 16, p. 7) and 3) “equipment”, since it has a projection requirement. Is it worth making all of these distinctions concerning the functions of this one data element?

TAB 32, CONTENT TO CONTENT RELATIONSHIPS: Equating the translation relationship with the adaptation and sequel relationships without distinction here seems to imply that the entity CONTENT includes both work and edition, although this is nowhere explicitly stated. Since rules in Part II of AACR2 clearly distinguish between translations (two editions of the same work) and adaptations or sequels (two related works), and no rules in Part I contradict this, some of us think this is an area where the model fails to reflect AACR2 as it is.

Response to Key Issue 1

1. Does the concept of class of materials as currently reflected in the code serve as a viable basis for an extended structure accommodating new forms of digital materials?

[TF] There are four positions held by members of the Task Force:

The concept of class of materials as currently reflected in the code could work if it were cleaned up so that it was based on content only, not carrier. (1 person)
The code works as is and should not be changed. (1 person)
The code should be reorganized by ISBD area. (6 persons)
The code does not work well for items with multiple characteristics, but some other solution besides organization based on content or ISBD area should be devised. (3 persons)

Task Force members and consultants heard from:

Note that two members of the Task Force have not yet cast their votes.

If the concept of class of materials is modified, some members of the Task Force would like to see a less hierarchical approach than is taken in the data model. Instead, they would like to see a more facetted approach that can accommodate an item that participates in three or more different classes of material, without giving priority to any one class. According to this analysis, for any given item, you would pick one element from each facet. With those facets designated, you should be able to decide how to handle any particular aspect of description.

The following facets are suggested:

Facet 1, Type of publication (use of the term ‘publication’ is unfortunate, however, as most unpublished manuscripts should probably be treated as monographs in this regard, and there may be a few unpublished serials; one Task Force member suggests that we try to come up with a better term than ‘type of publication,’ ‘type of stabilization,’ perhaps?):

Facet 2, Content (See the categories suggested below in 1.a.)

Facet 3, Published vs. unpublished

Facet 4, Physical carrier:

Facet 5, Method of management:

Facet 6, Reproduction:

See the comments in the overall evaluation of the model concerning the desirability of getting away from having to determine whether two items were simulataneously or successively released (i.e., whether one is the “original”).

We don’t quite see why these facets need to be given the different kinds of hierarchical relationships to each other that they are given in Figures 4-11, where type of publication and class of material are ‘belongs to’ categories, published vs. unpublished and library management vs. archival management (collection) require different figures to express the difference, physical carrier is a hierarchical part of document, and reproduction is seen as an attribute (‘Method of replication’) of Manufacture, and as a relationship type (Tab 27, page 10). It might work better to treat all 6 facets as equal in importance. If AACR2 were to be reorganized based on ISBD areas, according to Recommendation 1, each chapter (ISBD area) could begin with general rules, and then qualify them as necessary for variations in the 6 facets listed above, giving one or another facet precedence as judged optimum in any given area.

[1.a] Are the classes that group materials on the basis of the form of the physical carrier—sound recordings, motion pictures, videorecordings, computer files, and microforms—conceived in sufficiently precise terms to enable an unequivocal determination as to the class to which a new form of physical carrier would belong (e.g., as digital technologies continue to evolve, will it be possible to make clear distinctions between a sound disc and a computer disc);

[TF] Most members of the Task Force agree that there are problems with physical carrier classes, although the problem may not be a precision failure, but rather a cross-classification failure.

[TF] Examples of problem materials supplied by one Task Force member include:

[TF] A music expert on the Task Force likes Delsey’s analysis of the complexities of class, and says “This is clearly where we have run into problems with music materials—where we have one set of rules for representing the graphic representation, another set for the sound representation, and another for the visual (which includes sound). All this is further complicated with a multimedia representation of a musical work.”

[TF] One member of the Task Force argues that the analysis concerning ‘form of physical carrier’ in question 1.a is inaccurate, and that Table 1 is misconceived. Instead, she argues that the chapters in AACR2 that are based on form of physical carrier are only chapters 2 (books), 9 (computer files) and 11 (microforms); consider, for example, the fact that a book can consist wholly of graphic content, wholly of cartographic content (as in the case of atlases), or wholly of musical notation, in addition to consisting wholly of alphanumeric content, or consisting of some mixture of the above. In contrast, motion pictures and videorecordings can contain only ‘cinematographic’ or ‘moving image’ content, and that is intellectual and artistic content, not mere ‘carrier’. In addition, the reason for distinguishing between ‘cinematographic’ and ‘moving image’ content in Table 1 is unclear; they are the same thing. Likewise, a sound recording can contain only sound; it cannot contain graphic or cinematographic content or computer data; and sound is its ‘intellectual and artistic content,’ not merely a carrier. This analysis betrays a distinct bias against sound and moving image as intellectual and artistic content! A better, more logical, and cleaner analysis of the ‘classes of materials’ might simply distinguish ‘form of content’ from ‘form of physical carrier’, dropping the category of ‘form of expression’ altogether, and listing the following forms of content:

[TF] It is also imperative to include a ‘mixed’ category, as ‘mixed’ has always been with us (e.g. the old matter of deciding among illustrated text (text predominant) vs. graphic content with text (graphic content predominant) vs. a work made up of both text and graphic content inextricably intertwined) and always will be (e.g. many kinds of multimedia, as well as Delsey’s example of a work consisting of musical notation that can be “played” as musical sound; our computer files expert describes web sites as “a mix of data and programs, which may include text and images along with sound and animation”). It would be wise to include as well some explicit and principled instruction to guide catalogers in determining when one type of content is predominant and when they should fall back on defining something as mixed content.

[TF] Table 1 is difficult to understand; if 'x' indicates exclusion, what does a blank box indicate? What does a ‘greyed in’ inclusion indicate?

[1.b] Are the rules set out for each class of materials sufficiently comprehensive to accommodate the various types of content and forms of intellectual expression that might be recorded on the physical carriers included in that class (e.g., do the rules for computer files adequately cover digital forms of text, musical notation, sound and video);

[TF] No (with one possible dissension).

[1.c] If certain classes are defined on the basis of the type of content or the intellectual form in which the content is expressed, how is a determination to be made as to which criterion takes precedence in classing new forms of material (e.g., will digitally encoded musical notation that can be “played back” in the form of sound be classed as music or as a sound recording);

[TF] One Task Force member points out that we can’t answer this question until we develop the list. However, others point out that the problem of how to determine precedence in cases in which content is mixed, or cases in which rules based on content conflict with rules based on carrier or type of publication, is not going to go away no matter how the rules are arranged. If the option described in Recommendation 1 below is taken (using ISBD areas as the primary organizing element), it will still be necessary to provide catalogers with principles for according precedence when rules conflict in this manner.

[TF] The Task Force did a quick study of how often rules in any given ISBD area differ based on content, carrier or type of publication, and found that the potential for the need to set up tables of precedence in a reorganization might be rather high (although this was a very quick and dirty study). See below:

Content chapters:

Chapter 3, Cartographic materials
Chapter 5, Music
Chapter 6, Sound recordings (for new works, e.g. ethnographic recordings, bird songs)
Chapter 7, Motion pictures and videorecordings (for new works and adaptations of existing works)
Chapter 8, Graphic materials
Chapter 9, Computer files (for programs and new works)
Chapter 10, Three-dimensional artefacts and realia

Carrier chapters:

Chapter 2, Books pamphlets and printed sheets
Chapter 6, Sound recordings (for musical performances and talking books)
Chapter 7, Motion pictures and videorecordings (for musical performances only)
Chapter 9, Computer files (for digitized copies and versions of non-computer files)
Chapter 11, Microforms

Type of publication chapters:

Chapter 12, Serials

Other:

Chapter 4, Manuscripts (including manuscript collections) (published vs. unpublished; method of management)
Chapter 13, Analysis (part-whole relationships)

Where the Rules Differ from Chapter to Chapter:

CHIEF SOURCE

Content (3,5,6,7,8,10)
Carrier (2,7,8,9,11)
Type of publication (12)
Other (Early printed monographs (2.13),4,13)

AREA 1

Content (3,5,6,7,8,9,10)
Carrier (7,9)
Type of publication (12)
Other (Early printed monographs (2.14),4)

AREA 2

Content (9)
Type of publication (12)
Other (Early printed monographs (2.15))

AREA 3

Content (3,6,9)*
Carrier (9)
Type of publication (12)

*Note that archival moving image materials (Chapter 7) also use Area 3 for country of production.

AREA 4

Content (5,6,8,10)
Carrier (6)
Type of publication (12)
Other (Early printed monographs (2.16), 4, early printed music (5.4B2))

AREA 5

Content (3,5,6,7,8,9,10)
Carrier (2,6,7,9,11)
Type of publication (12)
Other (Early printed monographs (2.17),4,13)

AREA 6

Type of publication (12)
Other (4)

AREA 7

Content (3,5,7,8,9,10)
Carrier (6,7,9,11)
Type of publication (12)
Other (Early printed monographs (2.18),4,13)

AREA 8

Other (4)

[1.d] If an item falls within more than one class, how is an order of precedence for applying specific rules that differ for each of the relevant classes to be determined?

[TF] See comment above.

Recommendation 1: Use the model developed for this study to assess options for restructuring Part I of the code to facilitate the integration of rules for new forms of expression and new media. One option for consideration would be to use the ISBD(G) areas of description as the primary organizing element for the overall structure of Part I.

[TF] To some on the Task Force, the option seems a good suggestion, and could help a great deal in clearing up areas of illogicality in AACR2. It could make it possible to allow content, carrier, and type of publication to drive the rules as necessary. As one Task Force member points out, “If the rules are no longer made up of a few categories for which complete sets of rules are provided, then we can have as many categories (‘classes’?) as we need to specify the scope of special rules.” This could make the rules easier to apply to items with multiple characteristics (such as digitized manuscript maps or serial digitized sound recordings) and help to ensure better consistency in practice for clear-cut cases. No rules can avoid having to appeal to cataloger judgment in cases that are not clear-cut, of course, but the more the principles behind the rules are incorporated into the rules themselves, the more likely it is that cataloger judgement will be based on a sound application of those principles. Reorganization could make the code shorter. Also, one Task Force member who works for a bibiographic utility points out that “‘Extensibility’ is a key factor in the viability of the cataloging rules. A finite list of classes of materials has been a barrier to the ease with which new forms can be accommodated.”

[TF] Some point out that the cross-classification in chapters 1-13 of AACR2 has actually been causing cataloging problems for years, prior to the advent of electronic resources, as in the case of serial nonbook materials and microfilm maps.

[TF] All agree that such a massive reorganization of the rules would take a great deal of effort and expense. The decision as to whether that effort and expense are worth it will have to hinge to some degree on a guess as to the future importance to libraries of electronic documents and other such documents with multiple characteristics. If in the near future all materials heretofore collected by libraries (books, journals, music, maps, videos, sound recordings, etc.) will be distributed in electronic form, many of them in continuously updatable form, a reorganization of the rules could potentially make the cataloging of these materials more efficient and therefore less costly. If, on the other hand, the bulk of materials collected by libraries continues to consist of single-characteristic materials that fall neatly into just one of the existing 13 chapters of AACR2R, reorganization of the rules could make routine cataloging more difficult. Currently, there is some evidence that libraries are willing to leave the provision of access to most electronic materials in the hands of commercial entities that are similar to the abstracting and indexing services we have relied on for years for access to journal articles. The question is whether this is a permanent situation, or merely temporary pending the solution of the copyright and other problems that currently prevent widespread distribution of monographs, music, films, etc. in digital form.

[TF] At least one member of the Task Force does not think that reorganizing the rules will solve the major problems connected with the multiple versions issue, and believes we should be “focussing on revising the cardinal principle to permit the creation of clusters of smaller records in bibliographic families where commonly-held data is recorded once. Likewise we need to eliminate the focus on physicality.”

Response to Key Issue 2

2. Does the physicality inherent in the concept of DOCUMENT constrain the logical development of the code to accommodate the cataloguing of electronic resources?

Some of us think ‘document’ sounds too much like text to be used for audiovisual materials, music and computer programs. We also don’t think the term ‘document’ necessarily implies physicality.

[2.a] The DOCUMENT entity would have to be defined in terms that are not necessarily linked to the notion of physicality;

[TF] Okay. We have to confess, though, we don’t see why this is “not a trivial matter.” We have never necessarily tied our decision about what to catalog to the way a publisher has physically published an item. If it is important to the cataloging agency to catalog a content part (to use the terms of the model), i.e., to define it as the item cataloged, the cataloging agency has done so.

[2.b] The attributes of the entity DOCUMENT would have to be reviewed and extended as necessary to include attributes unique to networked electronic resources;

[TF] Fine. We don’t think anyone could have any objection to this, although there might not be agreement on which attributes are “unique to networked electronic resources;” we would hope a principled approach would be taken such that any attributes that were defined could be extended to new kinds of materials in the future (or even old kinds that have been inadequately described up until now).

[2.c] The definition of DOCUMENT PART would have to be reworked so that it is not necessarily linked to the notion of physical separability;

[TF] We don’t quite follow here. Doesn’t CONTENT PART deal adequately with physically inseparable content that we nevertheless want to catalog on a separate bibliographic record? Perhaps ‘physical separability’ is being taken too literally here, and we just have to make sure we retain our current freedom to define the object of a cataloging record according to the needs of the cataloging agency when complex part-whole relationships exist …

[TF] A music cataloging expert on the Task Force comments, “I can apply Delsey’s analysis to a textual work, but have difficulty extending it to non-print works.”

[2.d] The “stored on” relationship between INFIXION and PHYSICAL CARRIER would have to be reviewed to ensure that they could be applied, as appropriate, to non-physical (i.e., intellectual) elements of a DOCUMENT.

[TF] Much of our concern with physical carriers has to do with availability of a desired manifestation of an edition of a work. Availability for ILL may be affected by whether it is a book or a microfilm. If it is on microfilm, you are going to have to have special equipment available to view it. These types of concerns evaporate concerning electronic documents. Note though that something like information about physical carrier has to be known in order to access an electronic document: one has to know that it is digital and that it is on the Internet in order to know to look for it there rather than on the shelves of a library. In other respects, some of us would support dispensing with physical carrier as an aspect of the description of an electronic document. However, for identifying and describing monographic editions*, it would still be useful to gather information about extent of content (not extent of physical carrier); we would hope that methods to measure and record extent of content of static/monographic editions (not integrating/continuously updated works) would develop as the field of electronic documents becomes more mature.

*For serials, it would still be useful to gather information about coverage (where dates or volume numbering exist).

[2.e] The criteria for determining CHIEF SOURCE OF INFORMATION would have to be reviewed to ensure that they could be applied, as appropriate, to non-physical (i.e., intellectual) elements of a DOCUMENT.

[TF] Okay. It would help, however, to have a clearer statement of what is meant by this. Does it merely mean that the rules should sanctify transcription from a home page or initial screen of an electronic resource? Or does it imply that the cataloger might be asked to supply a title or other information that is not presented anywhere in the item cataloged, or supersede information presented in the item cataloged with supplied information?

Recommendation 2: Use the model developed for this study as the basis for examining the feasibility of modifying the internal logic of the code to accommodate documents that are defined in non-physical terms. Consultation should be undertaken with experts in the area of electronic document architecture.

[TF] We support this in principle, but would not like to see the rules become very rigid in regard to defining ‘documents,’ ‘document parts’ and ‘content parts’ suitable for being the object of a bibliographic record, or in defining their ‘boundaries.’ We have always relied on cataloger judgement in this regard, and should continue to do so. Too many factors need to be taken into account: the distinctiveness of titles, the need for subject analysis, the needs of the cataloging institution, etc.

[TF] Again, there is some questioning of the value of conducting a massive reorganization of the rules in order to accommodate the cataloging of electronic resources. The decision as to whether that effort and expense are worth it will have to hinge to some degree on a guess as to the future importance to libraries of electronic documents and other such documents with multiple characteristics. If in the near future all materials heretofore collected by libraries (books, journals, music, maps, videos, sound recordings, etc.) will be distributed in electronic form, many of them in continuously updatable form, a reorganization of the rules could potentially make the cataloging of these materials more efficient and therefore less costly. If, on the other hand, the bulk of materials collected by libraries continues to consist of single-characteristic materials that fall neatly into just one of the existing 13 chapters of AACR2R, reorganization of the rules could make routine cataloging more difficult. Currently, there is some evidence that libraries are willing to leave the provision of access to most electronic materials in the hands of commercial entities that are similar to the abstracting and indexing services we have relied on for years for access to journal articles. The question is whether this is a permanent situation, or merely temporary pending the solution of the copyright and other problems that currently prevent widespread distribution of monographs, music, films, etc. in digital form …

[TF] All are agreed that the internal logic of the code should not be modified in such a way as to erode the ability to describe the physicality of traditional items that still make up the bulk of most libraries’ collections.

Response to Key Issue 3

3. Is the division of the universe of objects described into two categories—published and unpublished—adequate to accommodate the description of digital objects disseminated on-line?

[TF] We would suggest it might be useful to ask this question a different way: Does the concept of ‘unpublished’ have any meaning in a world in which anything can be digitized and put on the Internet? Why do we make the distinction now? In effect, treating something as unpublished simply means not including a publisher and publication date in the description. There may also be a presumption that an unpublished item or collection is unique, and therefore, the cataloger need not feel so constrained to follow standard rules of description since the record will never be used for copy cataloging? Is it not true that our main interest in whether or not something is ‘manufactured,’ ‘released,’ or exists in ‘copies,’ is in whether or not multiple copies of the same edition of the same work exist, and therefore whether we need to build a description that will be usable by many different institutions to describe their copies? There certainly is a value in including a publisher statement on a published item in terms of indicating whether or not the information contained has been refereed or validated in some other way by a reputable organization, and this value does not go away in the online environment; it is still very valuable to know who put particular information on the Web, and therefore how reliable it is likely to be. A corollary question might be: When something currently unpublished (e.g. a manuscript) is digitized and put on the Web, should the digitized copy still be described as unpublished?

[TF] Most Task Force members agree that putting something on the Web should be held to constitute publication for cataloging purposes. One Task Force member indicates “it is a slippery slope trying to differentiate electronic copies which we ‘own’ by means of receiving an FTP file, those for which we license [access] to a virtual copy, and those freely available on mirror sites.” One Task Force member wonders, though, whether we can define manufacture, release, copy, etc. in a way that makes sense for both print and digital. A serials expert on the Task Force asserts that the current concepts of manufacture, release and copy, impression, issue and edition do not work for serials now.

[3.a] The DOCUMENT entity would have to be defined in terms that would encompass digital objects that effectively lack a physical dimension (see the discussion of the “physicality” issue, above);

[TF] Fine. Most members of the Task Force view this as something of a non-issue.

[3.b] The concept of MANUFACTURE would have to be reexamined to determine its applicability in a networked environment;

[TF] See above.

[3.c] The concept of RELEASE would have to be modified to extend the notion of “making copies available to the public” to modes of dissemination in which the “copies” made available are not necessarily physical in nature;

[TF] Fine. Television archives have already done this for off-the-air recordings of news and public affairs programs.

[3.d] The concept of COPY and the specifications for the scope of the entity would have to be reviewed with respect to the inclusion of on-line displays of a digital object, document source, printouts of screen displays, etc.;

[TF] Yes. And perhaps this could be done in a principled way so as to include other similar situations, such as the negatives and masterpositives and sound tracks in various formats that are created in the course of making prints of a motion picture.

[3.e] The attributes of the entity COPY would have to be reviewed and extended as necessary to include attributes unique to “copies” of digital objects;

[TF] Yes. Note that there isn’t much now in AACR2 that deals with copies (i.e. holdings?), other than in rules for rare books. To suggest an expansion in this area is perhaps to suggest that AACR2 should contain rules for the data contained in the USMARC holdings format just as it now contains rules for the data contained in the bibliographic and authorities formats. Perhaps it should, but that would represent a huge change in scope. A serials expert on the Task Force is emphatic that AACR2 should NOT deal with the description of holdings.

[TF] If we can discuss the possibility of there being copies of something that “lacks a physical dimension,” perhaps we could at the same time discuss the possibility of there being copies that differ only physically, as in the case of a microform copy of a book or journal, or a digitized copy of a text.

[3.f] The concepts of IMPRESSION, ISSUE, and EDITION would have to be reworked to accommodate groupings of copies that are the product of on-line transmission;

[TF] Yes. These concepts have been out of date for a long time anyway, since the end of widespread use of printing presses, and with the explosion of methods for reproduction of all kinds, not just digitization. The code should have a statement of principle that addresses the need to identify and describe (demonstrating both sameness and differences) items that contain the same work, but differ from each other in ways that are significant to users, such as being in different languages, having different editors, having revised text, being illustrated or not, having the same text, but different titles on the title page, etc.

[TF] We suggest that the following types of grouping need to be directly addressed by the code:

Variant content editions

Editions by appendage (called expressions in FRBR) (same content stream, but with commentary, biographical/critical material, etc. added)

Title editions (not dealt with by FRBR) (same content stream, but with a change in title or statement of responsibility that is likely to influence the way users cite and seek the work of which it is a manifestation of an expression)

Identical content editions (called manifestations in FRBR) (same content stream but with a change in either physical format or distribution or packaging information that does not influence citation behavior)

[TF] Some on the Task Force think that the code is currently ambiguous on the question of whether the cataloger’s job is to simply report what the item says with regard to edition, or whether the cataloger’s job is to determine the true facts in the case (concerning whether or not there are actual changes in content between two items that purport to be either different editions or the same edition). Others can live with the fact that the cataloger’s job is to record all available evidence as to edition (and that includes not just edition statements, but all elements of the bibliographic description), placing a heavy reliance on how an item represents itself, but that the cataloger cannot always determine whether two items really have the same content stream or not.

[3.g] The concept of TYPE OF PUBLICATION would have to be reviewed to ensure that it adequately accommodates the range of “publication” possible through the on-line dissemination of a digital object (see the discussion of the “seriality” issue below).

[TF] Yes.

Recommendation 3: Using the model developed for this study as a frame of reference, examine the issues raised with respect to the notion of “publication” in a networked context in consultation with experts in the area of electronic documents.

[TF] We would hope that these issues would be examined in consultation with many other kinds of experts as well.

Response to Key Issue 4

4. Can the notion of “seriality” as reflected in the code be extended to accommodate electronic forms of “publication” or dissemination of documents “intended to be continued indefinitely”?

[TF] Yes, but we would not want to see it extended to cover all electronic documents on the grounds that all can potentially change over time. At that point, it becomes a meaningless and confusing distinction. We wonder if the rise of the electronic environment is not forcing us to recognize something that has always been true, that is that all works can potentially change over time. Works of nonfiction can change in the form of revised editions. The texts of works of belles lettres can be tinkered with over time by various textual editors. If a work is seen as an abstract concept made concrete by its various editions, in a sense the work changes every time a new edition is published. In the electronic environment, there is the potential for all types of work to change over time by means of overlay, rather than by means of the succession of discrete publications. This is certainly a situation that we must deal with, but we are not sure calling everything a serial is the solution. That is not in fact what CONSER serials experts suggest should be done. The serials expert on the Task Force agrees that a digital map, digitized books, tables of contents, static documents like committee minutes and reports are clearly monographic in nature. Even given that, there is a considerable amount of as yet undefined territory between the electronic resources that are clearly monographic in nature and those that are clearly serial in nature.

[TF] Nevertheless, there are a number of works that may be conceived of as serials by users, but have become ‘integrating entities’ on the Web (to use CONSER terminology). These include:

Electronic equivalents of print serials, which are often presented in database format, or with a current issue and an archive of previously-published articles.
Reference works that were previously published annually in paper (or in loose-leaf format) that are now updated continually on the Web so that they never go out of date.
Web sites such as CNN Interactive that mimic newspaper web sites.

[TF] We would like to suggest that we go back and ask ourselves why we made the distinction between monograph and serial in the first place. It appears to some of us that one of the major reasons has to do with the way serials have been published in hard copy, and the implications of that for the organization of technical services departments in libraries. A library must make sure that it receives each successive issue that it has paid for, and it must check it in in order to report to reference librarians and users on exactly which issues have arrived and which haven’t. The serials expert on the Task Force agrees that it is probably not contemplated that we will continue to check in electronic journals in this fashion, as it would be rather difficult to carry out such procedures on a continuously updated database or “integrating entity.”

{TF] One of the concerns of serials catalogers is that a work heretofore defined as a serial has “turned into” a continuously updated database (“integrating entity”) on the Internet, and if a user does an OPAC search and limits it to serials, they will see the record for the paper serial only unless the database is also defined as a serial. Even before the advent of the electronic serial, it was always dangerous to limit any OPAC search to serials, since many citation titles that look like serial titles are to titles that have not in fact been treated as serials in the library in which the search is being done. Nevertheless, the serials expert on the Task Force points out that the ability to limit OPAC searches to serials is very popular and widely used by users, and some even prefer to set up the limit in such a way that it applies to their entire searching session. She also points out that serial records exist separately in split catalogs like RLIN, Melvyl and LC; many serial union lists and the CONSER database have no monographic equivalent; and a user looking in one of those split catalogs or lists for the New York Times is unlikely to turn to the “books” catalog to search for an electronic version. Nevertheless, while we all agree that a person looking at a record for a work that is issued on paper as a serial should always be informed of the existence of an electronic version of that work if there is one, we are not all sure that it is necessary to treat the latter item as a serial in order to get this result.

[4.a] The concept of the DOCUMENT entity would have to be extended to encompass digital objects that effectively lack a physical dimension (see the discussion of the “physicality” issue above);

[TF] Okay.

[4.b] The concepts of MANUFACTURE, RELEASE, COPY, IMPRESSION, ISSUE, and EDITION would have to be reworked to accommodate the electronic transmission of a digital object within the notion of “publication” (see the discussion of the “publication” issue, above);

[TF] Okay.

[4.c] The definition of serial as a TYPE OF PUBLICATION would have to be reviewed to determine whether the criterion relating to the issue of successive parts implies the distribution of separate “physical” parts, and if so, whether the criterion could be modified;

[TF] Okay.

[4.d] Examining the implications of treating a “serial” issued as a continuously updated database of articles, etc. not as a set of document parts but as a single document intended to be continued indefinitely;

[TF] See comments above.

[4.e] Reworking the concept of TYPE OF PUBLICATION to reflect alternatives to the current binary division of the universe of publications into monographic and serial publications.

[TF] Yes, this would be better.

Recommendation 4: Continue the examination of the “seriality” issue initiated as a follow-up to the Conference on the Principles and Future Development of AACR, using the frame of reference set out in the model developed for this study as a tool to assist in the analysis of the issues.

[TF] Well, we can certainly support the continued examination of the “seriality” issue! The serials expert on the Task Force points out that un-numbered series and loose-leaf publications, even when not electronic, are not yet adequately covered in the rules, and integrating online resources certainly are not.

Response to Key Issue 5

5. What are the implications of applying the logic of the code to documents in which the intellectual or artistic content is not permanently “fixed” within a physical object?

[TF] Some of us feel that the degree of volatility shown by electronic documents presents an irreducible problem for cataloging that cannot be solved by the code, but must just be accommodated as best we can. To some extent, it is not a new problem. It is really the essential problem posed to the cataloger of a serial, and it is encountered in other areas of cataloging as well, e.g. the cataloging of texts subject to frequent revision.

Recommendation 5: Review the conventions and rules for reflecting change in the attributes of the item described, as currently established, to determine their applicability to changes in the attributes of digital objects, and extend them as necessary to accommodate a broader range of variables.

[TF] Some of us would recommend that this review not be limited to digital objects, but be extended to cover change in the attributes of both the publication, distribution area and the physical description area for all materials, to see if this cannot lead to a better solution to the “multiple versions” problem, or, as one of us would prefer to call it, the “near-equivalent copies” problem … Some of us are convinced that there are changes in these types of attributes (publication/distribution, and physical description attributes) that are too trivial to create a new edition of a work, and instead should be held to create a variant copy … On the other hand, members of the Task Force agree that we should not recommend solutions that cannot be implemented in USMARC, and at least one member of the Task Force feels that AACR2 will never be able to solve the multiple versions problem, since it is actually an exchange problem.

[TF] The serials expert on the Task Force suggests that the code needs to specify less detailed descriptions in order to accommodate documents in which the intellectual and artistic content is not permanently “fixed.” However, several other Task Force members feel that any change that is made to the code to accommodate volatile content or identification should not be allowed to obscure the identification and differentiation functions of bibliographic description; it is still important to ensure that one cataloger will be able to tell that her item is the same as that described by another cataloger in a pre-existing record, and to ensure that users are able to tell which records represent the same edition of the same work. If it is proposed to dispense with the description of certain attributes, or to ignore change in certain attributes, it is crucial to know which attributes these are. Some attributes, such as title and statement of responsibility, may have such an impact on users’ citation practice that change in them should not be ignored. Other attributes, such as physical format or distribution or packaging information may be less critical; however, it would be better to describe such variation at a lower level in the record than to simply ignore the variation altogether.

[TF] By the way, it is interesting that Table 2 carries the notation ‘[unspecified]’ so frequently under single-part monograph. In practice, users of AACR2 read rule 0.24, as well as the entire array of descriptive elements detailed in chapters 1-13, to mean that in all the instances of change listed in Table 2 one should ‘describe as a separate edition.’ Perhaps Table 2 should be modified accordingly, since it is meant to be a model of AACR2 as it is?

[TF] One Task Force member who works for a bibliographic utility writes, “What are the implications of describing a document whose content could change in ways that are not easily detectable, as well as [what are] the implications of determining whether the item you are cataloging is the same as the one that another library described yesterday (last week, this morning, etc.) in a shared database? … The intention of [OCLC’s ‘When to Input a New Record’ guidelines] has always been to help catalogers make pragmatic decisions and to avoid a lot of needless ‘churning’ about whether a difference is significant. They are inherently based on what can be observed about [a] physical item and what can be inferred from a bibliographic record.” Several Task Force members agree that incorporation into AACR2 of something like OCLC’s ‘When to Input a New Record’ guidelines or LC’s rule interpretation for 1.0, general rules (which, be it noted, conflicts with the OCLC guidelines in a number of significant ways) could help catalogers decide quickly in the majority of cases, leaving just the odd case in which “the variables have not all been predefined or cannot be identified in situ,” as one Task Force member puts it. The fact that the OCLC guidelines and the LCRI conflict on such an important aspect of descriptive cataloging points to the need for a clearer statement in the code itself concerning when to make a new record.

[TF] See our comments under key issue 3, question 3.f for our discussion of the concept of edition and the object of a cataloging record.

Committee on Cataloging: Description & Access

Task Force on Rule 0.24

THE LOGICAL STRUCTURE OF THE ANGLO-AMERICAN CATALOGUING RULES, PART 1

General Considerations

Response to Key Issue 1

Response to Key Issue 2

Response to Key Issue 3

Response to Key Issue 4

Response to Key Issue 5

THE LOGICAL STRUCTURE OF THE
ANGLO-AMERICAN CATALOGUING RULES, PART 1