Toward a global access to bibliographic information:
Converging patterns, new paradigms
Zorana Ercegovac InfoEN & UCLA

†The author presented this paper as a keynote presentation at the COBISS/SICRIS User Conference in Maribor, Slovenia, 29 November 2000. The Conference was organized by IZUM, Institute of Information Science, Maribor, Slovenia.

Table of Contents

1. Introduction

2. Evolution of objectives of library catalogues

2.1 Cutter's Objectives (1876)

2.2 Paris Principles (1961)

2.3 OPACs on the Web and new opportunities

3. Factors that continue to affect how information is organized

4. Functional Requirements for Bibliographic Records (FRBR)

4.1 FRBR Objectives (1998)

4.2 Anatomy of the model

4.3 Areas for further study

5. Normalization at the international level

5.1 Authority records

5.2 Finding the common language

6. Convergence of multiple metadata today

Concluding remarks



1. Introduction

In order to set a framework for our discussion in the area of universal bibliographic control, I will be looking at the information organization models conceptually and historically. I want to start off by reviewing three sets of objectives of library catalogues and the means to achieve the stated objectives since the turn of the 20th century. Roughly, every fifty years, new conceptual frameworks led to cataloguing revisions, which, in turn marked distinct methods of organization and access to bibliographic information. In particular, this paper will discuss some of the converging efforts with regard to compatibility among existing metadata, normalization of headings, and sharing of cataloguing records at the international level. I will be also reviewing factors and the effect they have had on the cataloguing theory and practice. Examples are technological developments, economic pressures, heterogeneity of resources, including the exponential growth of digital resources, and heterogeneity of users, including remote users.

Over the past century, library catalogues have been guided by different requirements and new technologies; cataloguing theory and practice was re-examined and re-defined with regard to user tasks, access mechanisms, and design considerations. Library catalogues provided local access of their holdings to local readers during the early era of cataloguing (1876 to about 1950s). Next era of cataloguing, from early '60s to early '80s, saw rapid developments of pre-OPACs (online public access catalogues); used machine readable cataloguing to assist libraries and national bibliographic agencies to share bibliographic and authority records; extended cataloguing rules in order to identify and describe a variety of formats and media (e.g., serials, manuscripts, music, cartographic materials, sound recordings, graphic materials, computer files). We have the opportunity and responsibility to re-examine the position of today's OPACs with regard to a larger universe of networked bibliographic knowledge. Users are both regional customers and remote searchers; resources are heterogeneous, many of which virtual, dynamic, and often difficult to describe with current tools, structures, procedures, and conceptual frameworks. However, we use essentially the same rules (e.g., 1967[AACR], 1978[AACR], 1988[AACR2], 1998[AACR2R]) that have worked well with book-like resources. For example, "chief source of information" (per Rule 1.0A1) is well understood, title page is pretty much standardized, at least for books (after about 600 years of the printed book), with distinct places for titles and chiefly responsible creators,editions are well understood, and so forth. When these cataloguing rules are applied to new tasks, different resources, and new modes of accessing resources, they do not fare as well. Now is the time to start thinking about different ways of organizing, accessing, displaying, and relating information. I hope that this paper will generate more questions than answers with regard to the participation of Slovenian libraries in the universal bibliographic control.


2. Evolution of objectives of library catalogues

2.1 Cutter's Objectives: Early Anglo-American cataloguing tradition

In 1876, Charles Ammi Cutter stated the three types of objectives that a library catalog ought to achieve. These are the finding objective, the collocating objective, and the choice objective, as follows:

1. To enable a person to find a book of which either

A. the author

is known

B. the title

C. the subject

2. To show what the library has

D. by a given author

E. by a given subject

F. in a given kind of literature (e.g., 655 form/genre such as cartoons, correspondence, criticism and interpretation)

3. To assist in the choice of a book

G. as to its edition (bibliographically)

H. as to its character (literary or topical)

Cutter's person is the local user who uses his or her local library to find a known (printed) book (as well as other materials with similar characteristics) of which s/he knows author, title, or subject information. The second objective would help the person find a set of books by a given author (e.g., Mark Twain) on a given subject (e.g., children on the Mississippi River), or in the desired genre (e.g., literary criticism, biography). The choice objective would help the person differentiate among available expressions (e.g., translations, abbreviated editions, first editions, scholarly editions, editions with bibliographic footnotes, extensive annotations and references) of the same work. Cutter did not explicate the distinction between work (a prototype of different editions) and its expressions. An example may be a particular translation of Twain's Huckleberry Finn in Slovenian; or editions of Homer's Odyssey as the two titles illustrate:

The Odyssey of Homer. Translated [in English] with an introduction by Richmond Lattimore.

New York : Harper Perennial, 1967. --from UCLA's ORION2

Homerus - avtor. Sovre, Anton - prevajalec. Odiseja. v Ljubljani : Mladinska knjiga, 1964.

212 str., [6] str. pril. , 19 cm. (Knjižnica Sinjega galeba ; 98). --from COBISS online catalog

This was the era that founded the American Library Association; that witnessed devising of the Dewey Decimal Classification and Relative Index (by Melvil Dewey in 1876); of the Rules for a Printed Dictionary Catalog (by Charles A. Cutter in 1876); of the Library of Congress Subject Headings (1898); and the ALA's Catalog Rules: Author and Title Entries (1908). The same landscape saw books as the dominant bibliographic format. Libraries acquired and described their holdings independently of one another. There was little cooperation between and among libraries across the United States. Standardization however had just started to take shape through the Library of Congress card-distribution program (1901) in order to help libraries exchange cataloguing records; these early cooperative efforts would help reduce high costs in original cataloguing and control backlogs of unprocessed material. It was not until the Depression in 1930s that the notion of cooperative cataloguing was the compelling alternative to expensive local cataloguing. However, cooperative cataloguing and union catalogues required uniformity in headings and standardization, in general.

2.2 Paris Principles and IFLA's Initiatives

It was not until the International Conference on Cataloguing Principles (ICCP), and Statement of Principles that were adopted at the Conference, that brought cataloguing theory and practice at another conceptual level (IFLA, 1971). In 1943 Lubetzky was initially hired for only several months to study underpinnings of the (1949) Code at the Library of Congress. He would resign his post in 1962 to become a faculty member at the newly created School of Library Studies at UCLA. Lubetzky's investigation (1960) at the Library of Congress in the 40s and 50s prepared a necessary groundwork that culminated at the first major IFLA Initiative held in Paris. The adopted principles at the Paris Conference (1961) focused on the descriptive cataloguing, applied to the choice and form of headings and entry words, differentiated between the finding and the collocation functions of the catalog, and eliminated the choice objective (Cutter's third objective). In particular, the functions stated that the catalogue should be an efficient instrument for ascertaining:

1. whether the library contains a particular book specified by

(a) its author and title, or

(b) if the author is not named in the book, its title alone, or

(c) if author and title are inappropriate or insufficient for identification, a suitable substitute for the title; and

2. (a) which works by a particular author and

(b) which editions of a particular work are in the library.

According to ICCP, the term "author" applies equally to personal authors and corporate bodies under whose names entries may be made (IFLA, 1971, p. 6). A work was defined as the intellectual creation of an author (a family of documents with the same intellectual or artistic content). A book, was defined as a particular object that embodies the work (e.g., different editions by size, special editions with bibliographic footnotes, references and illustrative material, but not the medium; for example, various renditions of Julius Caesar in books, film adaptations and television series are considered as separate works).

The Paris Principles of 1961 were manifested in the first edition of AACR (1967). The rules also considered conformity with the ISBD(M), and were open to include non-book materials.

This was an era that preceded the development of MARC at the Library of Congress. The user is still operating at the local level; s/he comes to the library to identify, find, select, and borrow library material. Besides books, other non-book materials became widely represented in libraries. With early technological advances and economic pressures to share cataloguing records in order to control costs in cataloguing and reduce backlogs of unprocessed material, there was a growing interest in cooperation and standardization worldwide (there are many other reasons that support cooperative efforts among libraries, such as inter-library loan, collection development, verification, and preservation). These developments served as the backdrop of the International Meeting of Cataloguing Experts held in Copenhagen in 1969. The Meeting established international standards for the form and content of bibliographic descriptions. The first of the standards developed under that resolution, the International Standard Bibliographic Description for Monographic Publications, was published in 1971. Both the Paris Principles and the ISBDs have served as the foundation for a variety of national and international cataloguing codes (see Table 1).


ISBD publications

ISBD (M) for monographic publications

Prelim. Version, 1971; 1st std ed. 1974; rev. 1978; rev. 1987

ISBD (S) for serials

1974; 1st std ed. 1977; rev. 1988

ISBD (G) Gen std bibl description

1977; rev annotated ed. 1992

ISBD (NBM) for non-book materials

1977; rev ed. 1987

ISBD (CM) for cartographic materials

ISBD (CM) for cartographic materials

ISBD (A) for older monographic public

1980; 2nd rev ed. 1991

ISBD (PM) for printed music

1980; 2nd rev. ed. 1991

ISBD (CF) for computer files

1990; rev. as ISBD (ER) below

ISBD (ER) for electronic records

1st ed. 1997

Table 1: ISBD family of cataloguing rules for different formats (

Other contributions of the Meeting constituted a Working Group to complete the Annotated Edition of the Paris Statement of Principles (IFLA, 1971; chaired by Eva Verona); and another Working Group on an International Standard Bibliographic Description (chaired by A. J. Wells).

About the same time, information science researchers for the first time set out to evaluate effectiveness of information retrieval systems by creating a precision-recall measure (Cleverdon). Precision was defined as retrieval of only those records (documents) that are relevant; in library catalogues, it is achieved partly through authority control. Recall was defined as retrieval of all relevant documents in a given database; it is achieved through cross references. For the first time, library catalogues (and other retrieval systems) could be evaluated in terms of how well they retrieved potentially relevant documents in response to user queries.

2.3 OPACs on the Web and new opportunities

Syndetic structure of library catalogues is of utmost importance now when very-large-databases are growing exponentially; they are being measured in petabits. In large databases, the user will often retrieve a portion of relevant publications without knowing what and how much has been missed. Library catalogues, through their authority work, ensure that the user will retrieve only relevant documents from a given database with certain confidence.

Many OPACs are now available on the Web, alongside other online databases. As Table 1 shows, one of the latest additions among ISBDs is the format for cataloguing electronic resources (e.g., electronic texts, computer files, multimedia, and online services). According to Cataloging Internet Resources Using MARC 21 and AACR2 (Olson), bibliographic records code electronic location and access in MARC 856 field (e.g., 856 4# $u This means that with the 856 $u tag in the MARC format, the user can click on the URL address from bibliographic records and be connected to full documents and other Web resources. (The electronic location and access field is part of the MARC format only and, as of now, is not prescribed in AACR2R). This capability extends traditional boundaries of OPACs, which up until now, could not help a searcher to retrieve full-text papers. Today's OPACs on the Web compete with commercial online services, portals, and search engines. However, unlike America OnLine, Yahoo, and other robot-generated indexes, library catalogues bring to the Web their tradition in the authority control.

3. Factors that continue to affect information organization

Since the '60s, the dominant factors contributing to the change in bibliographic control continue to be of technological changes and economic pressures.

Technological changes were especially visible in library automated systems for the creation and processing of bibliographic data (e.g., starting with turnkey systems for specific library modules, to integrated library systems, and currently, to information gateways on the Internet). In major ways, technology has been the dominant factor in influencing creation of very large databases, such as OCLC, that contain records contributed and used by member libraries that participate in shared cataloguing programs. Technological advances have influenced other sectors and industries outside the library world. Examples are supermarkets that have replaced neighborhood stores (e.g., Safeway, Costco); drugstores that have replaced local pharmacies (apothecaries); super-bookstores (e.g., Borders, Barnes and Noble, that continue to replace single-owner bookshops; similar examples include commercial outlets (e.g., blockbuster), consolidation of chain department stores, the food industry, and self-service banking and gas stations. Technology is changing the way libraries operate at all levels: organization and bibliographic control. Economies-of-scale makes mass-distributed information cheap, fast and available, but at what price? At what common denominator should the quality of bibliographic records be accepted and promoted? Related factors are:

1. Growth of published output coupled with high costs of original cataloguing, and backlogs of uncatalogued material influenced two interrelated directions: interest in shared cataloguing in order to reduce cataloguing costs by minimizing duplicate cataloguing effort; and "minimal level" cataloguing in order to keep pace with the continued growth of publishing output. I will revisit these issues later, under normalization of headings that are acceptable by cooperating cataloguing agencies and their users.

2. Libraries have traditionally collected book-like publications. Early cataloguing codes prescribed description and identification of books. Emergence of non-book formats (e.g., printed music, serials, cartographic materials) has expanded the original scope of cataloguing codes. Besides print material, there has been an increasing need to adapt cataloguing codes and practices to accommodate cataloguing of electronic resources (ER). I will revisit this point under converging metadata. (Ercegovac, 1999).

3. Due to economic pressures worldwide, many libraries have shifted their collection development focus from "ownership" to providing "access" to global collections. This change in "ownership" expands traditional objectives of library catalogues which need to provide seamless access to physically owned documents within the library's walls as well as to virtual collections that are distributed across the world.

The changing environment that I just described led to a series of conferences, seminars and workshops. These were sponsored and organized jointly by international and national bodies including IFLA, ALA, OCLC, the Library of Congress and many other organizations. One of them was the 1990 Stockholm Seminar on Bibliographic Records, sponsored by the IFLA Universal Bibliographic Control and International MARC (UBCIM) Programme and the IFLA Division of Bibliographic Control (Svenonius). It was recognized that continuing pressure to do "minimal level" or "less-than-full" cataloguing required a careful re-examination of the relationship between individual data elements in the record and the needs of the user. It was also recognized that shared cataloguing programs required an agreed standard for a "core" level record.

One of the resolutions adopted at the Stockholm Seminar called for the commissioning of a study to define the Functional Requirements for Bibliographic Records (FRBR). "The purpose of this study is to delineate in clearly defined terms the functions performed by the bibliographic record with respect to various media, various applications, and various user needs. The study is to cover the full range of functions for the bibliographic record in its widest sense--i.e., a record that encompasses not only descriptive elements, but access points (e.g., names, titles, subjects), other 'organizing' elements (classifications, etc.), and annotations." (FRBR, p. 2)

4. Functional Requirements for Bibliographic Records (FRBR)

IFLA Study Group on FRBR has used a conceptual model (Entity-Relationship model) for the changing environment that was just described. The model was first described by Chen in 1976. Accordingly, an "entity is a 'thing' which can be distinctly identified. A specific person, company, or event is an example of entity. A relationship is an association among entities. For instance, 'father-son' is a relationship between two 'person' entities." (Chen, p.10) This model sought to answer the question of how should the data in a bibliographic record be organized and structured for intellectual and physical access at the international level? What are the functional requirements for a bibliographic record that serves different user needs?

4.1 FRBR Objectives

The functional requirements for bibliographic records are defined in relation to the four tasks of the user as follows:

1. to find entities that correspond to the user's stated search criteria (i.e., to locate either a single entity (originally known as the finding objective) or a set of entities (the collocating objective) in a file or database as a result of a search using an attribute or relationship of the entity.

2. to identify an entity (i.e., to confirm that the document described in a record corresponds to the document sought, or to distinguish between two texts or recordings that have the same title); identify a work; and an expression of a work.

3. to select an entity that is appropriate to the user's needs (i.e., to select a text in a language the user understands, or to choose a version of a computer program that is compatible with the hardware and operating system available to the user); select a work, an expression, or a manifestation.

4. to acquire or obtain access to the entity described (e.g., to acquire an entity through purchase, loan, etc. or to access an entity electronically through an online connection to a remote computer); obtain a manifestation of a work (FRBR, pp. 8-9).

Once again, the objectives of library catalogues were modified to include the variety of materials, as well as the full range of physical media described in bibliographic records (FRBR, pp. 7-8). However, the finding and the collocating objectives collapsed into a single objective.

The model developed in the study is comprehensive in scope but not exhaustive in terms of the entities, attributes, and relationships that it defines. The study is comprehensive in terms of the range of materials, media, and formats that are covered (e.g., textual, cartographic, audio-visual, graphic, and three-dimensional materials; to paper, film, magnetic tape, and optical media; and to acoustic, electric, digital, and optical recording modes). The basic elements of the model are entities, attributes, and relationships. The three groups of entities, are:

4.2 Anatomy of the Model

Group 1 ENTITY consists of the following four levels (FRBR, Chapter 3, pp. 12-29):

Work is defined as a distinct intellectual or artistic creation (e.g., Homer's Odyssey; La Chanson de Roland; Shakespeare's Hamlet). At this level, we do not think of any particular realization or manifestation of Homer's or Shakespeare's work. The work exists as an abstraction that is recognized by its creator and title, or simply by title, like in the case of the oldest extant epic poem in French, La Chanson de Roland. The Study recognizes that different cultures or national groups may have different criteria for determining of what constitutes a new work (FRBR, p. 16). Revisions, editions, abridgements or enlargements

Expression is defined as the intellectual or artistic realization of a work (e.g., translations, abridgments or enlargements, revisions and modifications, arrangements, adaptations of the same work, regardless of typeface, font size, page layout; however, the intellectual or artistic content is same). Translation to another language is considered a different expression of the same work. Example to the Slovenian language:

Homerus - avtor. Sovre, Anton - prevajalec. Odiseja.

A particular (say, first) edition is also considered a different expression of the same work. However, all editions, translations, and other expressions of the same work belong to the same bibliographic family, Homer's Odyssey. Another example:


Liszt's Hungarian Rapsody


the composer's score


edition by Arcadi Volodes

Manifestation is defined as the physical embodiment of an expression of a work (includes all the physical objects that bear the same intellectual or artistic content and physical form). Examples: the book, The Odyssey of Homer, that is published by HarperCollins Publishers; another by Mladinska knjiga in 1964:


Homerus - avtor. Sovre, Anton - prevajalec. Odiseja.

v Ljubljani : Mladinska knjiga, 1964. 212 str., [6] str. pril. , 19 cm. (Knjižnica Sinjega galeba ; 98).

Another manifestation would be the one produced on CD-ROM. A manifestation is a subset of a larger intellectual or artistic product. At the same time, a manifestation is exemplified by an item.

Item is defined as a singular concrete unit. The same item may have lost certain pages, or may have been autographed, or otherwise changed at different locations. For example, the above item in the Slovenian may have certain physical features that are specific to a particular copy.

Group 2 ENTITY includes Persons and Corporate bodies. These entities are responsible for the intellectual or artistic creation of Group 1 entities. These may also serve as the subjects of works.

Group 3 ENTITY is taken to represent an additional set of entities that serve as the subjects of works: Concept, Object, Event, Place. These entities are typically subjects of works.

The attributes are the means by which users formulate queries and interpret responses when searching about a particular entity (FRBR, Chapter 4, pp. 30-55). The Study defines attributes of a work (Section 4.2), of an expression (Section 4.3), of a manifestation (Section 4.4), and of an item (Section 4.5). Similarly, attributes of a person and of a corporate body (Sections 4.6 through 4.7), as well as of a concept, object, an event, and place are defined (Sections 4.8 through 4.11).

The function of relationships is to link works, expressions, manifestations, and items, and to facilitate construction of search queries to local and remote collections of objects. Basic bibliographic relationships, as defined by Tillett (1991) include:

Horizontal relationships for equivalent or semi-equivalent expressions, include editions, translations, revisions, abridgements, and arrangements. See and see also references, as well as notes are used to signal "editions," "versions," translations," "based on…", and so forth, between and among works, expressions, manifestations, and items. Vertical or hierarchical relationships include part<->whole relationships between different works, expressions, manifestations, and items. Chronological or sequential relationships include relationships between numbered series; these are linked with notes such as "continued by…", and so forth.

FRBR study examines relationships in the context of the entries defined for the model. The relationships may be between one work and another, between one expression and another, between a manifestation and an item, etc. (FRBR, p. 56).

For the purposes of the FRBR study, the users of bibliographic records are seen to encompass a broad spectrum, including library users and staff, publishers, distributors, retailers, and the providers as well as users of information services outside traditional library settings. The study also takes into account the wide range of applications in which bibliographic records are used: in the context of collection development, purchasing, cataloguing, inventory management, circulation and interlibrary loan, and preservation, as well as for reference and information retrieval.

The main sources used included the International Standard Bibliographic Descriptions (ISBDs), the Guidelines for Authority and Reference Entries (GARE), the Guidelines for Subject Authority and Reference Entries (GSARE), the UNIMARC Manual, and the AITF Categories for the Description of Works of Art.

4.3 Areas for Further Study

 The FRBR model is intended to provide a base for common understanding and further work. The model could be extended, as another study, to cover the additional data that are normally recorded in authority records, such as subject authorities, thesauri, and classifications, and relationships between those entities (FRBR, pp. 5-6).

The model is independent of any particular cataloguing code. National bibliographic agencies might find it useful to adapt the model to apply within their particular cultural context and bibliographic tradition. Further study could be done on the practical implications of restructuring MARC record formats to reflect richer hierarchical and reciprocal relationships outlined in the model.

5. Normalization at the international level

5.1 Authority records

AACR2, Part 2 details how to construct the authorized forms of names used as access points (per Part II: Headings, Uniform Titles, and References that pertain to choice of access points, Chapter 21; and headings for persons (ch. 22), geographic names (ch. 23), corporate bodies (ch. 24), uniform titles (ch. 25), and references (ch. 26). The traditionally promoted unity to a single, local library catalogue contradicts with the principle of trying to provide cataloguing records that serve the needs of the remote users across the globe (IFLA, UBCIM). In this international context where global databases are growing exponentially, authority control for names of persons, corporate bodies, series, places, and subjects becomes even more critical. IFLA's UBCIM Working Group on Minimal Level Authority Records (MLAR) and International Standard Authority Data Number (ISADN) (chaired by Tillett, 1998 in IFLA UBCIM) noted that requiring everyone to use the same form for headings globally was not practical. The Working Group has allowed the preservation of national differences in authorized forms for headings to be used in library catalogues and national bibliographies that best meet the language and needs of local users. It should be noted that Section 7.1 on Choice of Uniform Heading in the Annotated Edition (IFLA, 1971, p.23) reads:

When editions have appeared in several languages, preference should be given to a heading based on editions in the original language; but if this language is not formally used in the catalogue, the heading may be derived from editions and references in one of the languages normally used there.

At the international level, the crossover might be accomplished by first finding the mandatory minimal set of data elements that should be provided in all authority records (e.g., record ID, entity category, ISADN, language, script, nationality, authorized headings, variant forms of the authorized headings [see], related authorized heading [see also], as well as historical and biographical information); by linking authority records for the same entity through existing record numbers; and by providing switching for displays of authorized headings on an international scale. IFLA's MLAR envisioned ISADN as a standard number for an authority entry to be present in all variant records to serve as the ID. The record ID # from one country's authority record could be added to the authority record of another country to provide the link between the authority records for the same entity. To this end, Project AUTHOR (with members from the national libraries of Belgium, France, Portugal, Spain, and UK) mapped national records in various formats into UNIMARC using a program called USEMARCON. Each of the partners created a sample set of authority records to test. With the advent of Internet technologies, some of the linkages might become realized in a desired form of compatibility to participant national libraries. The results of this transnational application of national authority files are reported by Sonia Zillhardt and Francoise Bourdon (1998).

It is worth noting that over thirty years ago, Kaltwasser of Bayerische Staatsbibliothek in Munich, presented some of the problems with the use of Library of Congress cards in German libraries arising from different rules for headings. At the International Meeting of Cataloguing Experts in Copenhagen (IFLA, 1970), within the Shared Cataloguing Program, Kaltwasser proposed that problems of matching books and Library of Congress cards would be solved by the use of the International Standard Book Number (ISBN).

An example of normalizing names has been achieved in Getty's Universal List of Artists Names® (ULAN) ( It links artists' names to authorized name headings internationally. It takes the user-centered approach. It recognizes the multiplicity of preferences that a single name may have in different countries, languages and time periods. For example, an 18th century German composer may have different names in French, English, Italian, Dutch, Spanish, Russian, and Japanese. Specifically, an Italian 16th century painter, engraver, printmaker, and draughtsman, who was born in Dalmatia (ca. 1510-1563), Schiavone, Andrea is also known in the literature under the names of Andrea Esclabon, Andrea Meldolla, Andrea Schiavone veneziano, Andre Schiavon, And. Schiavone, A Schiavone, Medulic, Andrija, Meldola, Andrea, Meldolla, Meldolla, Andrea, etc. (See Figure 1 for the full array of synonymous names).

ULAN database links each name to birth and death dates, to variations in spelling, geographic locations, type of artist's work, and relation that an artist has to other people (e.g., poets, composers, playwrights, and political activists). Compare the method of name authority work with the traditional setting in OPACs: Franz Liszt, 1811-1886.

Similarly, one can provide multiple geographic terms to refer to the same place (e.g., Getty's Thesaurus of Geographic Names™, TGN). As Figure 2 illustrates, TGN uses multiple synonymous place names for Ljubljana as follows: Lyublyana, Lubiana, Laibach, Luvigana, Emona (ancient), Aemona (ancient Roman). The work in this area is experimental, fragmented and increasingly commercialized.

Standardization is hampered at the international level by numerous factors, including:

1. Language between and among cooperating libraries (languages, dialects; alphabets; transliteration schema; abbreviation rules; spelling variants).

2. Local Cataloguing traditions may not be compatible across different sets of national cataloguing rules. For example, Anglo-American tradition as represented in AACR and German's Regeln für die Alphabetische Katalogisierung, RAK, have different treatments with regard to what constitutes a corporate name. What level of detail is acceptable across different cataloguing traditions? Verona, Kaltwasser, Lewis, and Pierrot examined and considered over twenty cataloguing codes and practices in various countries (IFLA, 1971, x-xii).

3. Machine Readable Cataloguing formats (e.g., MARC 21, Universal Machine Readable Cataloguing or UNIMARC, and INTERMARC, BLMARC) and coding schema (e.g., XML, SGML, TEI, EAD) that define individual fields and their relationships also differ.

4. Vocabularies and classifications may be culturally biased (e.g., AAT, UMLS, DDC, UDC, LCSH, ULAS, TGN, Ei Thesaurus).

5. Specific search, display, management, and print features of local automation system differ (e.g., GEAC, DRA, VOYAGER/Endeavor, and numerous in-house produced systems).

5.2 Finding the common language

The Library of Congress has cooperated with the British Library, and the National Library of Canada. Other efforts are noted with the Program for Cooperative Cataloging (PCC) and others. The Deutsche Bibliothek is comparing rules with AACR2R and making some proposals to change some rules to achieve higher level of compatibility with AACR. Project REUSE+ studied possible improvements for the exchange of multipart records between USMARC and German bibliographic data and within the FRBR framework (Eversberg). Similar rule comparisons are going on in other countries. AACR2 has been most recently translated to Czech, Catalan, Greek, and Russian. Other countries are considering translation to their own languages.

Difficult questions before national bibliographic agencies that are interested in sharing records internationally include:

1. What elements are necessary (sufficient) in a bibliographic record at a minimal level description?

2. What elements are necessary in authority records (IFLA UBCIM Working Group on "Minimal Level Authority Records" and International Standard Authority Data Number, ISADN's "Mandatory Data Elements for Shared Authority Records" (1998), as well as IFLA UBCIM Working Group, current progress on the functional requirements and numbering for authority records, FRANAR)? How to pull up existing authority records for the same entity when forms may vary from file to file? It was proposed to use the existing authority record control numbers to provide a link, and to make display easier (Delsey; Tillet, 1997). See also examples in Figures 1 and 2 from Getty's authority lists, ULAN and TGN.

3. Which cataloguing rules are currently being used among COBISS member libraries? Is the level of compatibility among cataloguing rules examined (Društvo Bibliotekara Hrvatske, 1970; AACR2R (1998))? Which parts of codes should be tested?

4. How many languages are currently being used by cooperating agencies? Which alphabets are currently used by cooperating agencies?

5. Which cooperative programs for cataloguing bibliographic records are currently being used? How successful have they been? (e.g., OCLC, CORC)

6. How is integration currently achieved across different materials? At the level of a catalogue, are bibliographic records representing different materials being integrated in the library catalogue? At the level of a code, are rules for different materials treated uniformly?

Ultimately, regional consortia will make decisions based on their own needs and capabilities but also noting that certain principles will remain: cooperation at the international scale, different modalities of standardization (the ones that recognize the identity and richness of participating units), flexibility, scalability, and open architecture protocols (i.e., Z39.50).

6. Convergence of multiple metadata today

Diversity of library materials by type and medium is now further complicated by the growing amount of different electronic resources that are available on the Web. Traditionally, libraries were often divided by their respective collections by type and medium: there was a map library; music library; film archives; newspaper library; special collections library; rare book room; special languages library, and so forth. Everything else was a general library that collected, organized, preserved, and disseminated book-like documents for their library patrons. In other words, the format and medium often provided an organizing principle for libraries. However, library catalogue would describe and identify these diverse library collections by applying cataloguing rules across different formats and media as uniformly as possible. On the other hand, people from non-library traditions (e.g., geographers and cartographers, museum experts, etc.) devised independently their own rules for description of their respective collections; these rules reflected different seeking approaches of their respective users. In other words, people who seek maps and geospatial entities typically search by the area and scale rather than by author and title (Ercegovac, 1998). Maps differ from books, and users who search for maps use different attributes than those who search for books. In addition, many non-book entities lack such bibliographic markers as title pages, authors, or specific titles.

This is furthermore complicated with two additional factors: volume of electronic sources that needs to be catalogued; and remote users who navigate through virtual collections for different purposes, at different levels of expertise, and with varied expectations and capabilities. Even at the level of a single OPAC, numerous user studies have demonstrated that users have many serious difficulties when they search large databases in order to find what they want rapidly and effectively. Clearly, we need a new conceptual framework, such as FRBR, to construct libraries as gateways to global repositories that are accessible to everyone, anyplace, and anywhere, provided they have access to the Internet. This may be a starting point to guide the next Code (2008?).

One of the early solutions to providing rapid access to electronic resources (ER) has been to use Dublin Core as a simple content description model (Weibel). OCLC's Cooperative Online Resource Catalog (CORC) has been designed to link Dublin Core metadata and MARC 21 records with XML and other common structures, thus allowing opportunities for sharing records to an international pool. Many crosswalks have been written between USMARC and Dublin Core, GILS, FGDC, and SGML. As Table 2 shows, there are several concerns, especially in the area of interoperability between rich and controlled codes that are applied by trained cataloguers, and other metadata that facilitate rapid cataloguing of ER at a minimal level. One such concern is the practice of putting author statements (uncontrolled names in DC and GILS) into 1XX and 7XX authority controlled fields.

The following is a crosswalk between the 15 elements in the Dublin Core, MARC 21, and Government Information Locator Service (GILS).


Data Elements

Dublin Core (DC)

MARC 21 format



Name given to the resource

245$a title st/title proper


Contributor, also see Creator

An entity (persons, organizations) primarily responsible for making (contributions to) the content of the resource

1xx$a pn, cn, conf
700$a added entry, pn
710$a added entry, cn
711$a added entry, conf
720$a added entry, uncontrolled name

Contributor (Verfasser)
Originator (Urheber)

per Prussian Instruction

Subject & keywords


653$a (it, uncontrolled)
651$a (subj added, geo)
600, 610, 611$a (conf)
LCSH (650$a)
LCC (050$a)
DDC (082$a)

Uncontrolled term


Abstract of the resource

520$a summary, note



The entity responsible for making the resource available in its present form (e.g., university)

260$b (publication, dist)


Other contributors

Other than the Creator

700$a added entry, pn
710$a added entry, cn
720$a added entry, uncontrolled name


Date (associated with an event in the life cycle of the resource)

The date the resource was made available

260$c (date of publication, distrib., etc)

Date of publication

Resource type

web page, novel, TR

655$a (it, genre/form)



Ascii file, gif, jpeg

856$q (electronic location and access)

Available linkage type

Resource ID

Text string, number that uniquely identifies the resource (ISBN, URL)

856$u (URL)
856$b (access number)
ISBN (020$a)
ISSN (022$a)

Available linkage


The work, print or ER, from which this resource is derived

786$n (data source entry/note)

Sources of data


Should coincide with the Z39.53 3 char code

546$a (language note)
041$a (language code)
Z39.53 (041$a) lan code

Z39.53 lan of the resource


Relationship to other resources (e.g., chapters in a book, items in a collection)

787$n (nonspecific rel.)
IsPartOf: 773$n
HasPart: 774$n
IsBasedOn: 786$n
IsReferencedBy: 510$a
IsPartSeries: 490$a
Preced version: 780$t
Succeed version: 785$t

Cross reference relationship


The extent or scope of the content of the resource

500$a (general note)
255$c (stat/coordinates)
513$b (typ/report+time)

Supplemental data, e.g., spatial (coordinates), temporal (time period)

Rights Management

A link, URL, to a copyright notice

540$a terms governing use + reproduction note

Use constraints

Table 2: DC/MARC21 /GILS Crosswalk (November 1999). For details, see


Cataloguing Internet resources raises many important questions:

1. Is MARC 21/AACR2R appropriate for Internet resources?

2. How to incorporate electronic resources? (add a new chapter; incorporate into chapter 9 (computer files)); these are some of the open questions currently being debated.

3. How to deal with resources that have many versions (e.g., ASCII, Word, html, PDF, SGML)?

4. Is Rule 0.24 applicable for ER? It instructs cataloguers to base the description of the piece in hand (the "starting point for description is the physical form of the item in hand…"). (AACR2R, 1998)

5. How to deal with those resources that change location? An interim working solution to this problem has been to create a PURL ( that does not change and link each PURL with a corresponding URL that can change.

6. Who is qualified to catalogue Internet resources? (set of skills necessary to catalogue Internet resources)? Namely, will the people who are not professional cataloguers and librarians become involved in describing and organizing voluminous, heterogeneous, electronic resources? The new breed of cataloguers can now choose between different levels of detail to describe their collections (e.g., the Dublin Core; DC in Extensible Markup Language (XML); and Resource Description Framework (RDF), to mention just a few).

7. Are participating libraries sufficiently motivated to participate in cataloguing Internet resources, and at what level (e.g., Dublin Core, Cataloging Internet Resources using MARC 21 and AACR2R, 1997)?

8. Will electronic resources fall back to the current practice of Web online services that typically have no control over names and topics?

Questions with respect to standardization and integration between electronic and the traditional library materials have not been resolved. Some of these issues are presented in papers at: http://www.nlc-bnc/ca/jsc/confpap.htm.

Are we starting to see convergence of coexisting (and more compatible) metadata that had developed independently among different traditions to organize and access their respective objects? (Examples: scholars have developed and used the Text Encoding Initiative (TEI), under the auspices of the Center for Electronic Text in the Humanities ( for scholarly resources (Sperberg-McQueen and Burnard). The TEI guidelines give recommendations on what features of the text to encode and how to encode them. Designed for multipurpose use, the TEI header provides metadata needed by librarians who will catalogue the text, scholars who will use the text, and software programs that will operate on the text. Several groups within the library community are now examining the possibility of using TEI header information as a source of data in the automated creation of MARC21 records for electronic documents. Computer programs have been produced that convert TEI-tagged bibliographic headers into MARC21 formatted records. Librarians have developed the AACR2R/MARC21-based cataloguing for library resources. The geographers, cartographers, and the geospatial community has separately developed their own system for describing and accessing geospatial objects (Federal Geographic Data Committee, FGDC, metadata). Archivists use their own metadata (Encoded Archival Description, EAD) to describe archival inventories and registries (Ruth). Museum and visual resources are described with several metadata (e.g., Categories for the Description of Works of Art, CDWA; the Museum Educational Site Licensing project, MESL; the Visual Resources Association, VRA). See Special Topic Issue on Integrating Multiple Overlapping Metadata Standards (Ercegovac, 1999a; 1999b).

At the one end of the spectrum, we have seen highly structured and semantically rich formats, such as EAD, the TEI header, FGDC, and MARC 21/AACR2R (1998), that give rich description and identification of resources at high cost. Somewhere in the middle is the Dublin Core 15-element data set that is intended to facilitate discovery of electronic resources across disciplinary boundaries and information agencies at a far lower cost (although we really do not know how much are we saving in a long run?). At the other extreme of this spectrum are browsing lists and robot-generated indexes that are often easy to use but lack authority lists, selectivity, and detailed description of resources.

In this landscape of multiplicity of existing metadata, the following are open questions:

1. How can the needs of different communities for different kinds of metadata be accommodated at a conceptual level, implementation level, and at what cost?

2. How can we ensure that resources, once described, can be located throughout their lives?

It does not seem likely that a consensus will emerge out of all these independently developed efforts; what is more likely to happen is at a level of some common understanding of a core set of descriptive data (e.g., from the Dublin Core at the minimum to FRBR on the other side of the spectrum). By adopting such a common understanding in principle, each national library or community of users, can continue to develop methods compatible with its own users' needs, cataloguing traditions, language characteristics, and so forth, while taking advantage of data and systems created by other groups.

Little or no empirical evidence exists to demonstrate the applicability of various metadata to provide recall and precision of the type that we have had in the library catalogues. How can we use less to do more for all users across the international repositories? Will less coupled with coding schema such as SGML (XML, etc.) be adequate in the context of universal bibliographic access and control for remote collections and remote users?

Concluding Remarks

This paper has reviewed different conceptual models for bibliographic control and access. We began by looking at Cutter's three sets of objectives for (local) library catalogues around the turn of the 20th century. Landmark outcomes of the International Conference on Cataloguing Principles (ICCP) were discussed. Various factors, such as advances in technology, economic pressures, and growth of publications including electronic resources, have influenced all created a growing interest in minimal level cataloguing as well as in shared cataloguing in order to reduce costs. It was time to again re-examine the question of how should the data in a bibliographic record be organized and structured for intellectual and physical access at the international level.

Under the headings of normalization with regard to the uniform headings (in Section 5), and under the heading of convergence of multiple metadata (in Section 6), this paper discussed current efforts among national libraries and bibliographic agencies to cooperate. The cooperation is seeing at the level of using authority records that would allow the preservation of national differences in authorized forms for headings that would be used in library catalogues and national bibliographies that best meet the language and needs of local users. Projects AUTHOR (with five European national libraries) and ULAN (by Getty Information Institute) were explained and used to illustrate this type of direction. Numerous issues were discussed that hamper standardization internationally.

Another level of cooperation is at the level of cataloguing codes and metadata. This paper maintains that studies should be established to examine level of compatibility between cataloguing rules. It is difficult to see how various proposed crosswalks can be practiced when many metadata (e.g., Dublin Core, GILS) are essentially minimal level cataloguing, do not require authority work, and assume very different philosophical underpinnings.


I wish to thank Dorothy McGarry of UCLA, for her willingness to read an entire copy of this paper in its draft form. Special thanks must go to Branko Zebec and the IZUM management for inviting me to participate in the COBISS User Conference 2000 in Maribor, Slovenia, and inviting me to publish my presentation paper in COBISS Obvestila.


Figure 1: Personal names normalized: ULAN Name Record


[16642] searched as Medulic Andrija

Schiavone, Andrea [GC,PR,VP]

(Italian artist, op.1527/30-1563) [WC]

(Italian painter, 1527/30-1563) [GC]

(Italian painter, ca.1510-1563) [PR]

(Italian painter, engraver, and draughtsman, ca. 1500-1563) [VP]

(Italian painter, printmaker, ca.1510-1563) [BA]

Andrea Esclabon [PR]

Andrea Meldolla [PR]

Andrea Schiavone [PR]

Andrea Schiavone veneziano [PR]

Andrea Schiavoni [PR]

AndrŽ Schiavon [PR]

And. Schiavone [PR]

A Schiavone [PR]

A. Schiavone [PR]

A. Schiavoni [PR]

A. Shiavoni [PR]

Medulic, Andrija [VP]

Meldola, Andrea [GC]

Meldolla [PR]

Meldolla, Andrea [PR]

Schavione [PR]

Schiavone [PR]

Schiavone (Andrea Meldolla) [BA, PR, WC]

Schiavoni [PR]

Schivne [PR]

Schivoni [PR]

Shiavone, Andrea [PR]



*BŽnŽzit; *Bolaffi, v.4, p.332; *Encic. italiana, v.3,

p.208; *Encyc. world art; Fiocco, G., ARTE VENETA, IV

(1950), p.33 ff.; George Goldner; Getty Photo Study Coll.

(Ptgs.); Grande enciclopedia Vallardi, v.14, p.264; *Libr.

of Congr. Name Auth. File, NAFL8084600; *Petit Larousse;

Richardson, ANDREA SCHIAVONE (1980), p.9 ff.; *RILA/BHA;

*Thieme-Becker, v.24, p.357; Uvodic, A. Andrija

Medulic...1934; *Witt checklist


*See full citation in ULAN bibliography.



Figure 2: Geographic place names normalized (from TGN)



Ljubljana (inhabited place)

Lat: 46 04 NLong: 014 30 E (represented in degrees minutes direction)

Lat: 46.067 Long: 14.500(represented in decimal degress & fractions of degress)

Note - Was strategic Roman city on road to Pannonia; ruined by Huns (451) & Magyars (900s); ruled by dukes of Carniola 12th cen.; under Otakar II of Bohemia 1270; went to Hapsburgs 1277; captured by French 1809, seat of Illyrian Provinces until 1813.


Hierarchical Position:




Ljubljana (C,V)

Lyublyana (C,V)

Lubiana (C,V)

Laibach (C,O)

Luvigana (H,V)..rebuilt by Slavs after destruction by barbarians in 5th cen.

Emona (H,V)............................. ancient

Aemona (H,V) ancient Roman


Place Types:

inhabited place (C).....................founded by Augustus in 34 BC

city (C)

national capital (C) of Yugoslavian Slovenia from 1946, of

independent nation of Slovenia from 1991

transportation center (C)

industrial center (C)

educational center (C)



Aemona...........GRI Photo Study, Authority File (1998), 9525 [VP]


Emona......[GCPS]Encyclopedia Britannica (1988), VII, 420-421 [VP]

GRI Photo Study, Authority File (1998), 9525 [VP]

Laibach..........Encyclopedia Britannica (1988), VII, 420-421 [VP]

Webster's Geographical Dictionary (1984) [GCPS]

Ljubljana............... Times Atlas of the World (1994), 114 [VP]


Canby, Historic Places (1984), I, 526 [VP]

Columbia Lippincott Gazetteer (1961) [BHA]

Webster's Geographical Dictionary (1984) [BHA]

Encyclopedia Britannica (1988), VII, 420-421 [VP]

Webster's Geographical Dictionary (1984) [GCPS]

Lubiana..................Van Marle, Pittura Italiana (1932) [GCPS]

Encyclopedia Britannica (1988), VII, 420-421 [VP]

Luvigana.........Encyclopedia Britannica (1988), VII, 420-421 [VP]

Lyublyana..........Webster's Geographical Dictionary (1984) [GCPS]


American Library Association, Australian Committee on Cataloguing, British Library, Canadian Committee on Cataloguing, The Library Association, and The Library of Congress. (1998). Anglo-American Cataloguing Rules, 2nd ed. rev., edited by Michael Gorman and Paul W. Winkler. Chicago: American Library Association.

Chen, Peter Pin-Shan. (March 1976). "The Entity-Relationship Model: Toward a Unified View of Data," ACM Transactions on Database Systems, 1(1): 9-36.

Cleverdon, Cyril W. (1962). Report on the Testing and Analysis of an Investigation into the Comparative Efficiency of Indexing Systems. Cranfield, England: College of Aeronautics, ASLIB Cranfield Research Project.

Cutter, Charles A. (1876). Rules for a Printed Dictionary Catalog. U.S. Bureau of Education, Special Report on Public Libraries, Part II. Washington, D.C.: U.S. Government Printing Office.

Delsey, Tom. (1989). "Authority Control in an International Context." Cataloging & Classification Quarterly, 9(3):13-28.

Društvo Bibliotekara Hrvatske. (1970). Pravilnik i prirucnik za izradbu abecednih kataloga. Eva Verona. Prvi dio: Odrednice i redalice. Zagreb.

Ercegovac, Zorana, guest editor. (1999a). Special Topic Issue: Integrating Multiple Overlapping Metadata Standards. 50(13), November issue of the Journal of the American Society for Information Science.

Ercegovac, Zorana, guest editor. (1999b). "Introduction" to the Special Topic Issue: Integrating Multiple Overlapping Metadata Standards. 50(13): 1165-1168.

Ercegovac, Zorana. (1998) "Minimal level cataloging: What does it mean for maps in the contexts of card catalogs, online catalogs, and digital libraries?" Journal of the American Society for Information Science, 49 (8): 706-719.

Eversberg, Bernhard (1998) REUSE+ The part à whole relationship in German and American Cataloging Data.

Functional Requirements for Bibliographic Records: Final Report. IFLA Study Group on the Functional Requirements for Bibliographic Records. (1998). International Federation of Library Associations and Institutions. IFLA Universal Bibliographic Control and International MARC Programme, Deutsche Bibliothek, Frankfurt am Main. Approved by the Standing Committee of the IFLA Section on Cataloguing. UBCIM Publications, New Series Vol. 19. Munich: K. G. Saur.

IFLA Communications. (1970). Report of the International Meeting of Cataloguing Experts, Copenhagen, 1969. Libri, 20(1): 105-132.

International Federation of Library Associations and Institutions. (1962). Report: Proceedings of the International Conference on Cataloguing Principles, Paris, 9th-18th October, 1961. London: IFLA.

International Federation of Library Associations and Institutions. (1971). Statement of Principles adopted at the International Conference on Cataloguing Principles. Paris, October, 1961. Annotated edition with commentary and examples by Eva Verona; assisted by Franz Georg Kaltwasser, P. R. Lewis, and Roger Pierrot. London : IFLA Committee on Cataloguing.

IFLA. Universal Bibliographic Control and International MARC Core Programme (UBCIM). (1998). Mandatory data elements for internationally shared resource authority records. Report of the IFLA UBCIM Working Group on minimal level authority records and ISADN. Chaired by Barbara B. Tillett. (

Lubetzky, Seymour. (1960). Code of Cataloging Rules: Author and Title Entry: An Unfinished Draft. Chicago: American Library Association.

Olson, Nancy B. (ed.). (1997). Cataloging Internet resources: A manual and practical guide. 2nd ed. Dublin: OCLC Institute, OCLC. Available at:

Ruth, Janice E. (1997). "Encoded Archival Description: A Structural Overview," American Archivist, 60 (Summer 1997). Tag library available from Society of American Archivists (

Sperberg-McQueen, C. M. and Lou Burnard. (1994). Guidelines for Electronic Text Encoding and Interchange. Oxford: Text Encoding Initiative. Available at:

Svenonius, Elaine. (1992). "Bibliographic entities and their uses." In IFLA UBCIM Programme and the IFLA Division of Bibliographic Control: Seminar on Bibliographic Records Proceedings of the Seminar held in Stockholm, August 15-16, 1990. Edited by Ross Bourne, 3-18. Munich: Saur.

Tillett, Barbara B. (1991). "A taxonomy of bibliographic relationships." Library Resources & Technical Services, 35(2): 150-158.

Tillett, Barbara B. (1997). "International shared resource file: From authority control to access control." In: Authority Control in the Twenty-first Century: An Invitational Conference, 3/31-4/1.

Tillet, Barbara B. (2000). "Authority control on the Web: Initial draft for the Library of Congress Bicentennial Cataloging Conference," November 2000. Available at:

Weibel, Stuart. (1997). "The Dublin Core: A simple content description model for electronic resources." Bulletin of the American Society for Information Science, 24(1):9-11. The Dublin Core home page:

Zillhardt, Sonia and Françoise Bourdon. (1998). AUTHOR Project: Transnational Application of national Name Authority Files, Library Project PROLIB/COBRA-AUTHOR 10174, Final Report. Paris: Bibliothèque nationale de France.