• Functional Requirements for Bibliographic Records •

A Conceptual Model

for the Bibliographic Universe

Barbara Tillett

Library of Congress

Cataloging Distribution Service

2

Background

From 1992-1995 the IFLA Study Group onFunctional Requirements for BibliographicRecords (FRBR) developed an entity-

relationship modelas a generalizedview of thebibliographicuniverse, intendedto be independentof any catalogingcode or implemen-tation. The FRBRreport1 itselfincludes a descrip-

tion of the conceptual model (the entities,relationships, and attributes or metadata aswe’d call them today), a proposed nationallevel bibliographic record for all types ofmaterials, and user tasks associated withthe bibliographic resources described incatalogs, bibliographies, and otherbibliographic tools.

IFLA continues to monitor the application ofFRBR and promotes its use and evolution.The IFLA Cataloguing Section’s WorkingGroup on FRBR, chaired by Patrick LeBœuf,has an active online discussion list and a Website at http://www.ifla.org/VII/s13/wgfrbr/wgfrbr.htm. The Web site includespresentations, training tools, a hotlinkedbibliography, and much more.

Terminology

FRBR offers us a fresh perspective on thestructure and relationships of bibliographicand authority records, and also a moreprecise vocabulary to help future catalogingrule makers and system designers inmeeting user needs. Before FRBR ourcataloging rules tended to be very unclearabout using the words “work,” “edition,”or “item.”2 Even in everyday language,we tend to say a “book” when we mayactually mean several things.

For example, when we say “book” todescribe a physical object that has paperpages and a binding and can sometimes beused to prop open a door or hold up a tableleg, FRBR calls this an “item.”

When we say “book” we also may mean a“publication” as when we go to a bookstoreto purchase a book. We may know itsISBN but the particular copy does notmatter as long as it’s in good conditionand not missing pages. FRBR calls thisa “manifestation.”

This brochure is available in PDF format on the Web at: http://www.loc.gov/cds/FRBR.html (Revised February 2004)

3

When we say “book” as in ‘who translatedthat book,’ we may have a particular text inmind and a specific language. FRBR callsthis an “expression.”

When we say “book” as in ‘who wrotethat book,’ we could mean a higher levelof abstraction, the conceptual content thatunderlies all of the linguistic versions, the storybeing told in the book, the ideas in a person’shead for the book. FRBR calls this a “work.”

Entities

The JSC is examining AACR2 to updatethe terminology to be clearer when we meanwork, expression, manifestation, and item,following these FRBR “Group 1” entities.

FRBR’s “Group 2” entities are personand corporate body that are relatedto “Group 1” entities through specificrelationships. These relationships reflectthe role of the person or corporate bodywith respect to the work, expression,manifestation, or item. FRBR’s modelshows us how important such role informa-tion is for performing user tasks and forassisting a user to navigate through thebibliographic universe. (Note: This universemay be limited to our local catalog or maybe the realm of global resources availablethrough the Web.)

The value of this ‘role’ information becomesvery apparent in light of FRBR. We needto regain the lost link of relator terms andcodes in our bibliographic records. It istime to re-examine a change in cataloging

practice that abandoned use of “relator”terms and codes to cut cataloging costs.In hindsight we can see that decision wasunfortunate for future users of our recordsand should be reversed to allow greaterflexibility in manipulating bibliographicdata and offering better information tousers as they navigate our catalogs.

FRBR “Group 3” entities are the subjectsof works. These can be concepts, objects,events, places, and any of the “Group 1”or “Group 2” entities. For example, you canhave a work about another work or a workabout a person or corporate body.

4

BibliographicRelationships

A lot of attention has been given tothe inherent relationships among theentities in the Group 1 hierarchy of work,expression, manifestation, and item.Additionally, there are many other richcontent relationships that enable collocationof related items and navigation throughthe sometimes complex network of thebibliographic universe.3

Content relationships can be viewed asa continuum from works/expressions/mani-festations/items. Moving left to right alongthis continuum we start with some originalwork and related works and expressions and

manifestations that can be considered“equivalent,” that is, they share the sameintellectual or artistic content as realizedthrough the same mode of expression.Next we come to works/expressions/mani-festations that are related through a“derivative” relationship. These comprisea range of new expressions, such astranslations, different performances, slightmodifications and editions that move alongthe continuum across a magic line wherethey become a new work yet still relatedto some original work. To the far righton this continuum we find ‘descriptive’relationships that involve new worksdescribing some original work. FRBRreminds us of the importance of theserelationships and keeps us focused on thoseof most importance to meeting user tasks.

Relationships in the Organization of Knowledge, edited by Carol A. Bean and Rebecca Green, 2001, p. 23, “Bibliographic Relationships” byBarbara B. Tillett, Figure 2, © 2001 Kluwer Academic Publishers Boston, with kind permission of Kluwer Academic Publishers.

Family of Works

5

Whole/part and part to part relationshipsare also in FRBR. When we providebibliographic control for electronic digitalresources, we find these whole/part andpart to part relationships especially relevant.For example, a Web site may be viewedas the “whole” and the components as its“parts,” or we may view the whole digitizedresource and its components as the partsthat will need to be tracked throughtechnical metadata for storing anddisplaying that digital information.

The part to part relationships include‘sequential’ and ‘accompanying’ or‘companion’ relationships. Companionrelationships can be either dependent orindependent, which will influence howmany bibliographic records we wouldmake for the related works and theirmanifestations.

In fact the number of records we make is adecision made up front by the catalogerbased on local policies reflecting local userneeds. We may choose to catalog at variouslevels: the collection of works (FRBR callsthis an aggregation), an individual work, ora component of a work. At the collectionlevel we may include a description of allthe parts and should provide access to eachcomponent. At the component level weshould provide a link to relate to the larger“whole.” FRBR reminds us that theserelationships are important factors forfulfilling user tasks regardless of what wechoose to view as the “whole.”

User Tasks

So what are these FRBR user tasks? Briefly,they are find, identify, select, and obtain.‘Find’ involves meeting a user’s searchcriteria through an attribute or a relationshipof an entity. This can be seen to combineboth the traditional “find” and “collocate”objectives of a catalog. ‘Identify’ enables auser to confirm they have found what theylooked for, distinguishing among similarresources. ‘Select’ involves meeting a user’srequirements with respect to content, physicalformat, etc. or to reject an entity that doesn’tmeet the user’s needs. ‘Obtain’ enables auser to acquire an entity through purchase,loan, etc., or electronic remote access.

Additionally, FRBR recognizes theimportance of being able to navigate, andwe could add other tasks relevant to specificusers, such as tasks for rights managementor preservation communities. These usertasks reinforce the traditional objectives ofa catalog, as described by Cutter in 1876 toenable a user to find and to collocate works.4

Whole-Part Relationships

Relationships in the Organization of Knowledge, edited by Carol A. Bean andRebecca Green, 2001, p. 26, “Bibliographic Relationships” by Barbara B. Tillett,Figure 3, © 2001 Kluwer Academic Publishers Boston, with kind permission ofKluwer Academic Publishers.

Shape = container / carrierColor = content

WORKS WITHIN WORKSComponents /Aggregates

6

Impact onCataloging Rules

Today FRBR provides an opportunity to re-examine our cataloging rules and principles.The Joint Steering Committee for Revisionof the Anglo-American Cataloguing Rules( JSC) is using FRBR not only to updateterminology, but also to re-examine andhopefully improve the traditional linkingdevices of uniform titles in light of FRBR.Perhaps an expression-level citation orwork-level citation can provide an improvedreincarnation of traditional uniform titlesthat would offer more collocation anddifferentiation capabilities than currentuniform titles. Other professional organiza-tions such as IAML, IFLA, ALA, etc. willbe engaged in this re-examination.

The JSC is also exploring the FRBR modesof expression and some of the attributes ofmanifestations to revisit GMDs (generalmaterial designators). Online system displaycapabilities (such as icons used in somesystems today) might now be evaluated asan additional means for conveying informa-tion about the mode of expression and thetype of carrier or container available,replacing GMDs as we know them todaywith a device that better meets user needs.

FRBR is reaching even beyond AACR.IFLA’s first International Meeting of Expertson an International Cataloguing Code(IME ICC), July 28-30, 2003 in Frankfurt,Germany, will also provide an opportunityfor re-examining the 1961 Paris Principlesin light of FRBR and today’s onlineenvironment.

Impact onBibliographic Structures

OCLC’s initial research of FRBR withrespect to its more than 40 million recorddatabase WorldCat has shown that over80% of these records reflect a singlemanifestation per work.5 We could interpretthis to mean that we could let our localsystems automatically create authorityrecords for us based on the headings weconstruct according to cataloging rules whenwe get the first work of a creator. We wouldonly need to do the more extensive work forthe less than 20% of items, once we got thesecond or third manifestation (as suggestedby Jennifer Younger several years ago6).More interestingly we could provide subjectheadings and classification to the authorityrecord for the work—do it once there, ratherthan redundantly for each bibliographicrecord for each manifestation. Our biblio-graphic records today typically reflectparticular manifestations.

Similarly, we could link the authority recordsfor persons and corporate bodies with therelated “work” authority records whenthere is a “creator” relationship to the workauthority record. Authority records for“expressions” could also be linked to theperson or corporate body authority recordsin a “realized by” relationship. Theserelationships could be used for the systemto establish the citation form for the workand associated expressions that can then beused as the linking device for collocationand navigation. New models of bibliographicstructures could evolve to better meet userneeds.

7

Systems Designand Applications

FRBR promises to have a profoundinfluence on future systems design. Vendorsand bibliographic utilities, like VTLS,OCLC, and RLG have already embracedthe FRBR conceptual model in designingtheir future systems. These and othervendors are engaged in discussion ofFRBR through the JSC’s Format VariationsWorking Group, led by Jennifer Bowen.Although somewhat slow to catch on in theUnited States, FRBR has been fundamentalto recent system designs in Australia andEurope for many years.7

Conclusions

FRBR’s terminology, relationships, anduser tasks are already assisting us to reviewour traditions in cataloging in light oftoday’s digital environment. This workwithin IFLA has spread worldwide andprovides a conceptual model to guide usfor many years to come. IFLA togetherwith other interested parties will continueto encourage the application of this model tofacilitate international standardization andreduce costs for cataloging on a global scale.

Dr. Barbara B. Tillett— Chief —

Cataloging Policy & Support Office

Library of Congress�Dr. Tillett is also the current

Chair of the IFLA Cataloguing Section,a member of the Joint Steering Committee

for Revision of the Anglo-American CataloguingRules, and was a consultant to the IFLA

Study Group on the Functional Requirements forBibliographic Records along with Tom Delsey,

Elaine Svenonius, and Beth Dulabahn.She is currently working with others on theextension of this model to authority controlwith chair, Glenn Patton, and the IFLA

Working Group on Functional Requirementsand Numbering for Authority Records

with consultant, Tom Delsey.

Notes1. Functional Requirements for Bibliographic Records, Final Report / IFLA Study Group on the Functional

Requirements for Bibliographic Records. – München : K.G. Saur, 1998. (UBCIM Publications, New Series ;v. 19) Also available as http://www.ifla.org/VII/s13/frbr/frbr.htm orhttp://www.ifla.org/VII/s13/frbr/frbr.pdf

2. Thanks to Patrick LeBœuf for the book analogy.

3. Tillett, Barbara B., “Bibliographic Relationships.” In: Relationships in the Organization of Knowledge,edited by Carol A. Bean and Rebecca Green. – Dordrecht : Kluwer Academic Publishers, 2001, p. 19-35.

4. Cutter, Charles A. Rules for a Printed Dictionary Catalogue. – Washington, D.C. : Government PrintingOffice, 1876, p. 10.

5. Hickey, Thomas. FRBR algorithms & tools [online]. [Dublin, Ohio]: [OCLC], June 20, 2002. Availablefrom http://staff.oclc.org/~hickey/presentations/frbrAlgorithms20020620 files/frame.htm

6. Younger, Jennifer. “Resources description in the digital age,” Library Trends, v. 45 (Winter 1997), p. 462-487.

7. Examples of recent applications include AustLit, Denmark’s VisualCat, OCLC’s Fiction Finder, the futuredesign for OCLC’s WorldCat, RLG’s Web union catalog plans, the prototype for VTLS’s Virtua system,and the underlying concepts of <indecs>, the ABC model in Project harmony (US, IK, and Australia), andIndiana University’s Variations2 digital music project.

References

Hickey, Thomas; & Vizine-Goetz, Diane. Implementing FRBR on large databases [online]. [Dublin, Ohio]:[OCLC], 2002 [cited 31 December 2002]. Available from http://staff.oclc.org/~vizine/CNI/OCLCFRBR files/frame.htm

Hickey, Thomas B., O’Neill, Edward T., & Toves, Jenny. Experiments with the IFLA Functional Requirementsfor Bibliographic Records (FRBR). In: D-Lib Magazine [online], Sept. 2002, v. 8, no. 9. Available fromhttp://www.dlib.org/dlib/september02/hickey/09hickey.html (ISSN 1082-9873)

O’Neill, Edward. FRBR: application of the entity-relationship model to Humphry Clinker: ALCTS/CCS/Cataloging and Classification Research Discussion Group, Saturday, June 15, 2002…Atlanta… [online].[Buffalo, NY: Judith Hopkins] June 2002? cited 27 August 2002]. Available fromhttp://www.acsu.buffalo.edu/~ulcjh/FRBRoneill.html

O’Neill, Edward. FRBR (Functional Requirements for Bibliographic Records): application of the entity-relationship model to Humphry Clinker. In: Library Resources and Technical Services (2002) v. 46, no. 4,p.150-159. (ISSN 0024-2527)

OCLC. OCLC research activities and IFLA’s Functional requirements for bibliographic records [online]. Dublin,Ohio: OCLC, cop. 2002 [cited 16 July 2002]. Available fromhttp://www.oclc.org/research/projects/frbr/index.shtm With links to OCLC’s four projects: Case study:the FRBRization of Humphry Clinker http://www.oclc.org/research/projects/frbr/clinker/index.shtm,Extending the case of Clinker http://www.oclc.org/research/projects/frbr/2 works.shtm, Algorithmdevelopment and testing http://www.oclc.org/research/projects/frbr/3 algorithm.shtm, Fiction Finderhttp://www.oclc.org/research/projects/frbr/4 fictfinder.shtm [cited 31 December 2002].

The text of this brochure was originally published in Technicalities v. 25, no. 5 (Sept./Oct. 2003).