4.2 Metadata quality

Different types of metadata have varying degrees of usefulness, accuracy, and clarity. The effectiveness of an information retrieval system heavily depends on the quality of its metadata. This module will explore essential criteria for good metadata, considering its application across various contexts and systems, which can influence how these criteria are applied​​.

RDA: Resource Description and Access Standards

Read

The Introduction to chapter 5, on page 93, in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

Previously, Hider discussed the purpose of various elements in resource description. This chapter examines what constitutes effective metadata, as not all metadata serves its functions equally well. Several aspects of metadata quality are crucial for specialists to consider, including both the elements and their values.

Hider revisits how some elements are more effective than others in meeting users’ information needs. The comprehensiveness of metadata, with more elements, better addresses these needs. Accuracy and clarity of values are also vital, as is consistency at both the element and value levels.

Studies by Park and Tosaka (2010) and Theimer (2013) highlight key evaluation criteria for metadata quality, such as accuracy, consistency, and completeness. While there is no consensus on the definition or measurement of ‘quality’, effective understanding and communication of metadata quality can greatly enhance user satisfaction.

 

Functionality

We know how metadata is intended to serve particular functions, including the FRBR functions—find, identify, select, obtain and navigate. The elements best suited to support these functions will depend on the user’s information needs and may be limited by the retrieval system in use.

Read

‘Functionality’ (pp. 94-95), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

The utility of various metadata elements depends on the information context, including users’ needs, behaviours, and the retrieval system used. As it is impractical to include every possible element in a description, it is essential to select those most supportive of functional requirements. Determining these elements can be challenging and often involves some guesswork, but a thorough study of the information context can provide valuable guidance.

Metadata specialists should understand the workings of their information organisation systems, including indexing and display rules. If an element is not indexed, it cannot aid in resource retrieval; if it is not displayed, it cannot help in resource selection. Ideally, specialists should influence what is indexed and displayed.

Understanding users is perhaps the most difficult aspect, as their needs are varied and often unclear even to themselves. Hoffman (2009) found that many library cataloguers lacked knowledge of their users and their needs, despite the principle of ‘user convenience’. Hoffman calls for more user-oriented research and evidence-based cataloguing standards. Similarly, Swan Hill (2008) argues for cataloguing policies and database quality discussions to be based on user research rather than assumptions or personal experiences.

User studies are increasingly prevalent in information organisation and often reveal not only users’ needs but also the impact of metadata on users. This includes how language in metadata can marginalise or exclude certain groups. This awareness has led to calls for updates to metadata vocabularies to reflect current social norms (Adler, 2017)​​.

 

Comprehensiveness

In an ideal scenario, we would describe all attributes of a resource. However, this is impractical as there can be countless attributes. Fortunately, many attributes are not crucial. For instance, most users do not care about the colour of a book’s cover. Nevertheless, the comprehensiveness of descriptions can greatly impact their effectiveness. There has been a longstanding debate in library cataloguing regarding what constitutes a ‘minimum’ description and the standard amount of detail that should be provided​.

Read

‘Comprehensiveness’ (p. 95), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

The comprehensiveness of resource descriptions differs from the sheer amount of metadata, as it considers both the number of elements and the extent of each value recorded. There is a clear trade-off between comprehensiveness and cost, with more detailed descriptions generally taking longer to create. Metadata creators often face the choice of describing more resources in less detail or fewer resources in more detail. The level of detail may depend on the material’s nature or the user’s needs. For instance, internet resources might be described with less detail to support discovery rather than selection, as users can inspect these resources directly.

University libraries might provide more detailed descriptions than school libraries. Library cataloguers often discuss ‘minimum’ and ‘full’ records, with most libraries having a mix of both. This approach allows libraries to balance the need for detailed information with the practicalities of time and resource constraints​​.

 

Accuracy and clarity

The quality of metadata involves not just the selection of elements but also the values within those elements. Ensuring accuracy in descriptions is crucial, as inaccuracies can lead to user frustration, missed resources, and a lack of trust in the system. Even experts can make mistakes, but it is vital to minimise these and correct any errors. A key attribute of successful metadata specialists is their meticulous attention to detail.

Descriptions must also be understandable to the intended users. Sometimes, cataloguers use abbreviations and jargon that are not widely known, which can be detrimental. Therefore, clarity is essential in all forms of communication, including metadata .

Quiz

The following are examples of abbreviations and jargon which were once routinely used by library cataloguers. Do you understand what is meant by all of them? You’ll find the answers later in this module!

• t.p. title page
• ill. illustrator
• ports.
• repr. reprint date?
• fl.
• prelim.

Read

‘Accuracy’ and ‘Clarity’ (pp. 95-97), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

Accuracy

The accuracy of metadata is crucial since resources are often found, identified, selected, obtained, and explored based on small pieces of information. Even a minor error can have significant consequences. Inaccuracies can occur in observing the attribute or recording the value. Metadata specialists, though experts, can make mistakes, particularly if unfamiliar with the resource. Their errors are more detrimental due to the high trust in their descriptions. Modern computer systems facilitate easier editing and error spotting, with editorial capacities extending to the public in a Web 2.0 environment. Certain inaccuracies, like misspellings, can be identified by computer systems. When an attribute of a resource changes without corresponding updates in the description, it creates inaccurate metadata. Updating descriptions incurs costs, so metadata specialists must decide if the change warrants attention.

Clarity

Accurate metadata is ineffective if unintelligible. Elements recorded in an unknown language or using unfamiliar terms and abbreviations can be problematic. Library cataloguers often use jargon and abbreviations only they understand. There is a growing awareness among metadata specialists to translate technical language for broader understanding. Even common language terms can vary in clarity and may need careful selection, especially in contexts like website menus. Language evolves over time, necessitating updates to terms to maintain clarity. Disambiguating metadata with clarifying terms is common, especially for homonyms. Controlled vocabularies ensure homonyms have only one meaning. Metadata should be succinct to present more information in limited space and improve search precision.

 

Consistency

Consistency is a crucial aspect of metadata quality that pertains to both elements and values. Standardisation is essential as it helps users become familiar with a system, thereby making it easier to use. Using standard values can also enhance the system’s effectiveness by increasing recall, a key measure of retrieval success. Recall refers to the proportion of relevant resources retrieved from the database.

For instance, consider the subject element where the term ‘Motor cars’ is consistently used to describe vehicles also known as cars or automobiles. If a searcher uses ‘motor cars’ as the query term, they would retrieve all relevant resources about motor cars. Without a standard value for this term, the searcher would need to guess and search all possible terms—automobiles, cars, autos, motors, etc. This could result in missing suitable terms, thus reducing recall.

Read

‘Consistency’ (pp. 97-98), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

Note that Hider points out that consistency might not always be considered beneficial.

Consistency

Using the same elements and values across descriptions enhances retrieval and ensures semantic interoperability within and across systems. Consistency helps users read and interpret descriptions, aiding all metadata functions (finding, identifying, selecting, obtaining, and exploring). Factors affecting indexing consistency include the number of terms assigned, controlled vocabulary vs. free text indexing, vocabulary size, subject matter characteristics, indexer factors, available tools, and item length. Studies on indexing often show low inter-indexer consistency, but consistent indexing generally indicates accuracy. However, variations in indexing can reflect different user perspectives, supporting the case for social tagging. Increasing consistency between indexer and searcher involves mutual learning. Standardisation is key for consistency, with standards for elements, values, formats, and transmission being crucial. Consistency remains a major concern for metadata specialists due to its impact on information organisation.

 

Authority Control

Indexers and cataloguers frequently use standardised or controlled vocabularies to enhance the effectiveness of retrieval systems, commonly controlling subject terms. However, they can also control names of authors, organisations, resource titles, and other elements if they are willing to invest the necessary time and effort.

Name authority files contain authority records that specify the preferred form of a name (authorised access point) and references from its variations (variant access points). There are three main types of name authority files:

• Personal names (names of people)
• Geographic names (names of places and geographic features)
• Corporate names (names of organisations or groups)

Leading library agencies, such as the National Library of Australia and the United States Library of Congress, maintain and publish their authority files for the benefit of other libraries and information agencies. The standardisation of metadata values is known as authority control in the library context.

Authority files do not directly help users find specific resources in a catalogue; instead, they maintain consistency in the catalogue record headings. An authority file is a collection of records containing preferred headings for names, series, subjects, and titles. These records provide the preferred heading and non-preferred alternatives, along with notes on their derivation. To locate specific resources, one must search the catalogue for catalogue records​​.

Read

‘Vocabulary and authority control’ (pp. 98-103), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

The Schools Catalogue Information Service (SCIS) uses controlled vocabularies, namely SCIS Subject Headings and the Schools Online Thesaurus. For example, SCIS Subject Headings uses the term ‘Motor cars’, with ‘Used For’ references from ‘Cars’ and ‘Automobiles’ and ‘Broader Term’ references such as ‘Motor vehicles’, ‘Roads’ and ‘Transport’.

Vocabulary and Authority Control

Achieving metadata consistency can be challenging but improves retrieval system effectiveness. A controlled vocabulary, a standardised language for indexing, ensures that each term has a single meaning. This consistency relieves users from the cognitive load of recalling synonyms. Controlled vocabularies also handle variant spellings and antonyms. They are structured with cross-references to aid navigation and accuracy in indexing and searching. Controlled vocabularies use qualifiers and scope notes for further clarity and conceptual demarcation. Consistency between indexers and searchers is enhanced through controlled vocabularies, though differing views on resource topics may still cause inconsistency. Authority control in library cataloguing involves creating authorised name and subject headings to maintain consistency, even as names change. This process, though time-consuming, supports bibliographic control, with external files aiding libraries without their own authority files. Authority control extends to names of people, organisations, titles, and series names, increasing consistency and supporting information retrieval.

 

Activity

Probably the most comprehensive set of authority files is maintained by the Library of Congress. Take some time to explore this site, particularly its Help on Searching Authorities.

Search the authority file for the subject, name and title authority headings for your favourite topic, author and book. Are they included in the authority files? Look at your results in ‘Labelled Display’.

 

Quality Assurance

It is evident that some metadata is superior to others, particularly in specific contexts. However, no agency, not even the Library of Congress, has the resources to create flawless metadata. Therefore, the challenge is to determine what constitutes metadata that is sufficiently good. Over the years, numerous guidelines and standards have been developed to address this, but there is no definitive answer. Each agency’s solution will vary based on its budget, clientele, systems, expertise, and other factors. Even with established policies, it is challenging to ascertain when benefits no longer justify the costs. Nonetheless, agencies can set standards and strive to adhere to them as much as possible.

Read

The remainder of Chapter 5 (pp. 103-109), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

Principles of Best Practice

Several aspects of metadata quality have been discussed, serving as criteria for evaluating information resource descriptions. Not all metadata is equal, and it can be challenging to determine the superiority of some resource descriptions over others. Good metadata ultimately supports effective information retrieval. Adherence to standards suggests professional involvement, but the standards must fit the information context. A skilled metadata specialist knows when to deviate from these standards and add necessary elements not included in templates. While strict adherence to standards can save time, it may also result in less relevant resources for users. Metadata creation involves judgement and developing the knowledge and skills to produce high-quality metadata. Specialists need to understand their users, collections, resources, and systems. They must also be detail-oriented, clear, succinct, systematic, and thorough, applying principles and standards.

Quality Assurance

Given the impact of metadata quality on information retrieval, it is logical for agencies to monitor and evaluate resource descriptions and improve them if necessary. The professionalisation of metadata creation and management helps ensure quality. Many metadata specialists hold relevant qualifications, though these often relate to broader fields like librarianship or archival studies and do not guarantee high-quality metadata creation. Continuous professional development supports further skill development. Agencies can also assure quality by adopting best-practice standards, which do not always confirm quality.
Ultimately, the metadata itself must be examined and assessed. Many agencies establish quality control (QC) processes to check metadata when it is entered into the retrieval system. More independent evaluations, or audits, may also be conducted as part of a quality assurance (QA) process. QC and QA processes are particularly important when metadata creation is outsourced. Metadata evaluation can involve various criteria, often checked by computer programs for errors, but deeper evaluations are usually done by human experts. Agencies often create their own checklists or scorecards for evaluation. Assessment should focus on the cataloguing system as a whole rather than individual records, considering the specific context rather than a universal ‘gold standard’. Studies of metadata quality begin with understanding user needs. High-quality metadata is of little value if the system in which it operates is ineffective. Information retrieval system evaluations consider criteria like recall, precision, response time, usability, and user satisfaction, with metadata quality influencing these factors.

Cost–Benefit Analysis

High-quality metadata is often expensive and created by professionals earning professional wages. The more detailed the description, the longer it takes and the more it costs. Some agencies may find free metadata, such as downloadable catalogue records, but most high-quality metadata sources charge for their services. Metadata costs must be weighed against benefits, considering metadata specialists’ finite resources and time. While some agencies must comply with specific standards, they usually choose to provide high-quality metadata. The value of metadata often follows a ‘long tail’ distribution, with a few core elements used frequently and many minor elements used occasionally. Determining the point where costs outweigh value is challenging. Although financial costs can be estimated, monetising metadata value is difficult, with few attempts made. Cost-benefit analysis is increasingly necessary for tight funding environments. The business case for metadata must be made in both the public and private sectors. Under budgetary pressure, agencies may cut metadata expenditure, but metadata’s low profile does not equate to low importance. Metadata specialists must demonstrate the importance of resource description, arguing that some resources deserve detailed metadata. Implementing different levels of resource description within a single system is now common. Each agency must decide on the appropriate levels and resources for a detailed description. Shared costs across institutions can make professional descriptions more affordable, and the next chapter explores collaborative efforts in metadata creation​.

 

4.3 Metadata standards

Read

The Introduction to Chapter 7 on pp. 123-124, in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.).  Facet Publishing.

Here Hider outlines:

• the broad characteristics of metadata standards;
• gives the focus of this chapter, namely metadata standards relating to elements, format and transmission; and
• indicates that the examination of metadata standards in this chapter will relate to ‘key information domains’.

The previous chapters discussed the advantages of implementing various metadata standards. These standards enhance the consistency of resource descriptions and support the sharing and quality of metadata. A standard is more than just a guideline; it is a prescribed practice, not merely what is customary. Standards can be internal to an organisation or intended for a wider professional community, even at a national or international level. For it to be disseminated, a standard is documented, sometimes becoming a de facto standard when widely adopted by practitioners, even without formal recognition. Many metadata standards fall into this category, though others are developed through extensive consultation and formally approved by professional bodies or standards organisations. The International Organization for Standardization (ISO) describes standards as ‘documented agreements containing technical specifications or other precise criteria to be used consistently as rules, guidelines, or definitions of characteristics, to ensure that materials, products, processes and services are fit for their purpose’. Standards cover all aspects of metadata, including values, elements, format, and transmission, with some addressing multiple aspects.

Metadata standards can range from simple lists of elements to detailed documents with definitions, relationships, usage conditions, analysis recommendations, and options. While detailed standards promote consistency and quality, they can also be more challenging and costly to implement. Various factors influence the adoption of a metadata standard, including its purpose, resource characteristics, design, granularity, interoperability, support, growth potential, extensibility, reputation, ease of use, and existing expertise. Some standards are more suitable than others for specific contexts, often being developed for a particular professional community, such as librarians, archivists, or curators. There is often a balance between the need for consistency and best practices and the need to address local requirements and economic constraints. Chapter 7 explores key metadata standards across different information domains, beginning with those developed by the extensive community involved in the World Wide Web.

 

These domains and their related metadata are principally:

• Web publishing – HTML, XML and RDF
• Libraries – AACR, ISBD, RDA, FRBR/FRAD, LRM, MARC and Z39.50
• Digital Libraries – Dublin Core, MODS and METS, OAI-PMH and OpenURL
• Archives – ISAD(G) and EAD
• Museums – SPECTRUM, CIDOC and CRM

Metadata standards in libraries

The main standards used for descriptive cataloguing are:

• International Standard Bibliographic Description (ISBD) – the conceptual framework for AACR and still used in part in many libraries
• Anglo-American Cataloguing Rules (AACR) – now superseded but found in legacy data in library collections
• Resource Description and Access (RDA) – the current standard for descriptive cataloguing
• Functional Requirements for Bibliographic Records (FRBR) and Functional Requirements for Authority Data (FRAD) – the conceptual framework for RDA — now consolidated into the Library Reference Model (LRM).

Read

‘Libraries’ up to ‘AACR and ISBD’ (pp. 129-131), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

Libraries

The first guidelines for constructing library catalogues were introduced by the French revolutionary government in 1791. Although earlier catalogues and bibliographies sometimes included descriptions of their construction, these were more explanatory than prescriptive. Over time, conventions such as listing authors by surname and transcribing titles as they appeared on title pages became standard in English-speaking regions. However, the arrangement of catalogue entries and what information to include were still debated, particularly as large 19th-century libraries required consistent cataloguing rules due to their vast collections. The British Museum’s Department of Printed Books, later the British Library, spent years developing the ’91 rules’ under Sir Anthony Panizzi’s guidance. Although the accompanying catalogue was never completed, these rules had a lasting influence on cataloguing practices.

In 1876, Charles Cutter’s Rules for a Printed Dictionary Catalog became the foundation for the first cataloguing rules published by the American Library Association (ALA) in 1878, followed by the Library Association in the UK in 1881. These early rules, driven by an interest in cooperative cataloguing, covered headings, descriptions, and cross-references, providing detailed guidance for cataloguers. Cutter’s rules also included the creation of subject headings, although later rules generally did not address this, leading to the development of separate subject heading lists, such as the Library of Congress Subject Headings (LCSH).

Cutter’s objectives for a library catalogue were to help users find books by author, title, or subject, show the library’s holdings, and assist in selecting books by edition or literary quality. These objectives reflected the evolving role of catalogues as tools for intellectual access to resources, a concept that remains central to library cataloguing today. In 1908, the initial cataloguing rules from the ALA and the British Library Association were replaced by a new code for author and title entries. Although widely adopted, these rules proved insufficient as libraries encountered numerous exceptions, leading to a revised and expanded edition in 1941.

Seymour Lubetzky, tasked with reviewing the rules at the Library of Congress, argued for a set of guiding principles to streamline the rules. This led to a more concise code in 1949, covering descriptions but not headings. Lubetzky’s principles later formed the basis for the ‘Paris Principles’, established during an international cataloguing conference in Paris in 1961.

 

As this module focuses on metadata standards relating to elements, format and transmission, in this section we will examine the standards used for descriptive cataloguing, with a particular focus on Resource Description and Access (RDA), an international cataloguing standard implemented in many countries during 2013.

 

ISBD

The International Standard Bibliographic Description (ISBD), established by the International Federation of Library Associations and Institutions (IFLA) in 2011, provides a structured framework for creating bibliographic descriptions in a uniform, readable format. This standard enables libraries worldwide to share bibliographic records and ensures that these records are understandable, regardless of the language used. ISBD was foundational in the development of the Anglo-American Cataloguing Rules (AACR). Although the current cataloguing guidelines, Resource Description and Access (RDA), are built on a different framework, certain descriptive elements and punctuation from ISBD are still incorporated into the MARC coding system used in RDA-compliant cataloguing.

ISBD:

• Specifies the elements needed to describe and identify all types of material;
• Establishes an order for these elements; and
• Prescribes the punctuation to precede each element.

There was a significant alignment between ISBD and the Anglo-American Cataloguing Rules, as ISBD principles were integrated into AACR. While RDA is based on the Statement of International Cataloguing Principles, its guidelines for recording data are separate from its guidelines for display.

AACR

The Anglo-American Cataloguing Rules (AACR), including its second edition (AACR2), served as the primary cataloguing standard for English-speaking librarians until the introduction of Resource Description and Access (RDA) in 2013.

AACR adopted the ISBD model by creating chapters for 12 different formats of material commonly found in libraries, such as books, manuscripts, microforms, music, and realia. Each chapter addressed the ISBD areas of description in sequence and provided specific rules for cataloguing information in each area. An introductory chapter provided general rules applicable to all material formats, also organised according to the ISBD areas of description. Additionally, AACR included a section on constructing name authority records and instructions on the use of ‘main entries’ and ‘added entries,’ which are similar to RDA’s ‘authorised access points.’

 

Read

‘AACR and ISBD’ (pp. 131-134), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

AACR

Lubetzky’s work led to the creation of the Anglo-American Cataloguing Rules (AACR) in 1967, a new cataloguing code covering both headings and descriptions, including non-book materials. Published in American and British versions by the respective Library Associations, AACR was widely adopted by libraries across the English-speaking world, particularly by the Library of Congress (LC). However, it was less commonly used in non-English speaking countries, where local cataloguing codes, based on regional publishing and bibliographic practices, were more prevalent. For instance, the German and Austrian libraries followed the Instruktionen für die Alphabetischen Kataloge der Preußischen Bibliotheken (Prussian Instructions) and its successor, the Regeln für Alphabetische Katalogisierung (RAK).

In the early 1970s, the Anglo-American library community sought to align AACR with the newly established International Standard Bibliographic Description (ISBD) and to unify the American and British texts. In 1974, a Joint Steering Committee (JSC), including representatives from the British Library, the Canadian Library Association, the LC, the American Library Association (ALA), and the Library Association (LA), was formed to revise AACR. This resulted in the second edition, AACR2, published in 1978, which was subsequently adopted by major libraries in the English-speaking world. AACR2 underwent several revisions in 1988, 1998, 2002, and 2005 to address new types of resources and the challenges posed by the web. By its 2005 revision, AACR2 spanned 750 pages, although a simplified version, Concise AACR2, was also available. While AACR2 has largely been replaced by Resource Description and Access (RDA), many of its rules have been incorporated into the new standard.

ISBD

As cataloguing communities worldwide continued to develop their own rules, the Paris Principles demonstrated that common ground was possible. In 1971, the first International Standard Bibliographic Description (ISBD) was published, initially for monographs. Over the next four decades, additional ISBDs for different types of materials were released, culminating in a consolidated edition in 2011, which superseded all previous versions, including ISBD (G), which covered general library materials.

ISBD prescribes the elements cataloguers should include in their descriptions of library resources and how these descriptions should be presented. The elements are organised in a specific order, with designated punctuation to separate them, although these punctuation marks can be replaced by standard terms if needed. The advantage of this punctuation is its language neutrality, a feature particularly beneficial in the era of card catalogues, where space was limited.

The 2011 consolidated edition of ISBD is designed to be compatible with web standards, including those supporting the Semantic Web. It includes an ISBD namespace, a defined element set, and corresponding URIs for use in digital environments. Although ISBD is independent of any specific cataloguing code, it serves as a foundation for them, enhancing interoperability. ISBD has maintained a close relationship with Anglo-American codes, and its current element set aligns with those in Resource Description and Access (RDA) and the Functional Requirements for Bibliographic Records (FRBR), focusing on the description of manifestations.

ISBD covers only core elements, allowing libraries to supplement their descriptions with additional data, including linked data. However, specific types of resources may require more detailed treatment in certain contexts. The ISBD’s structure, with its arrangement into ‘areas of description’ and their respective elements, provides context that enriches the meaning of each element. However, some meaning may be lost when these elements are expressed individually, as in RDF triples used in linked data environments. Despite these challenges, the consolidated ISBD is praised as a reliable and trusted ‘quality brand’ for librarians globally.

 

 

AACR and SCIS standards

From its inception, SCIS used AACR as its standard for descriptive cataloguing. SCIS made clarifications, amendments and additions to AACR to better meet the needs of Australian and New Zealand school libraries and recorded these in its SCIS Standards documentation. However, the descriptive cataloguing in SCIS catalogue records created from 1 July 2013 is based primarily on RDA.

 

RDA

Resource Description and Access (RDA) is a cataloguing standard designed specifically for the digital world, and as such represents a change in direction from the earlier, print-dominated AACR2. With RDA the focus has changed, from preparing a catalogue record for the item in hand, to describing the resource in its bibliographic context—that is, building the set of identifying characteristics and relationships that are important to meet the FRBR user tasks, and also relating that resource to others in its bibliographic ‘family’.

RDA provides instructions on what information to record (the ‘content’ of resources), not how to record it (the ‘encoding’ of resources). RDA translates the ideas set out in the FRBR and FRAD conceptual models into a set of instructions on how to record information about bibliographic resources. RDA describes these instructions as ‘guidelines, not rules’—a completely opposite approach to its predecessor AACR2, which contained inflexible rules that governed every aspect of bibliographic description, even the punctuation used.

Read

‘RDA and FRBR/FRAD’ (pp. 134-143), in Hider, P. (2018). Information resource description: Creating and managing metadata (2nd ed.). Facet Publishing.

RDA and FRBR

The Anglo-American Cataloguing Rules, Second Edition (AACR2) was developed during the era of card catalogues, well before the rise of the internet. Even in its later revisions, AACR2 continued to use terms like ‘headings’ and ‘main entries’, which are less relevant in today’s online environment. Although its alignment with the International Standard Bibliographic Description (ISBD) was advantageous in the 1970s, by the 2000s, before ISBD was consolidated, this compatibility had become a limitation. Modern online catalogues tend to modify the ISBD format to improve readability, using labels and omitting punctuation. As a result, instead of releasing another edition of AACR, a complete overhaul was undertaken to create a new standard, Resource Description and Access (RDA). Released in 2010 by the Joint Steering Committee for Development of RDA (now the RDA Steering Committee), RDA was designed to produce resource descriptions that are both computer- and human-friendly.

Unlike AACR, RDA does not prescribe a specific format like ISBD but focuses on content—specifically, the elements and their values. It accommodates traditional ISBD-format descriptions while allowing for other formats more suited to the digital age. RDA’s schema includes more elements than AACR2, partly because RDA defines elements more narrowly and does not limit itself to ISBD elements. It also covers authority records more comprehensively, identifying hundreds of possible elements for both bibliographic and authority records, though only a core subset is mandatory.

Although RDA extends beyond AACR2 in coverage, many of its rules are based on the older code, meaning records created under RDA may look similar to those created under AACR2. However, there are differences: RDA replaces many specific material rules with more generic ones and has attempted to be more linguistically neutral, choosing the most appropriate name form based on the catalogue’s language rather than defaulting to English. Nevertheless, this move has led to the abandonment of certain Latin terms, like “et al.,” which could hinder international cataloguing exchanges.

Perhaps the most significant difference between AACR2 and RDA is that RDA is built on a conceptual model, the Functional Requirements for Bibliographic Records (FRBR). While AACR2 was grounded in principles, RDA is explicitly structured around FRBR and its related models, such as the Functional Requirements for Authority Data (FRAD) and the Functional Requirements for Subject Authority Data (FRSAD). These models conceptualise the catalogue as a network of entities (like works and authors) with attributes and relationships that support user tasks. Although FRBR has been influential, it has also faced criticism for its complexity and for being more suited to computers than to human users.

The more recent IFLA Library Reference Model (IFLA-LRM), which aims to consolidate FRBR, FRAD, and FRSAD, introduces new entities like ‘nomen’ for naming concepts and reconfigures others, such as ‘personal agents’ and ‘collective agents’. While this model attempts to improve data accuracy for the linked data environment, its complexity may make it challenging for users and less reusable for Semantic Web applications.

RDA’s structure, which is divided into ten sections and 37 chapters, mirrors the FRBR framework. It covers the description of various entities and their relationships, but it is designed as an online tool (RDA Toolkit) with extensive hyperlinks, making it complex and time-consuming to learn. Libraries may adapt RDA to meet local needs, as seen with the Library of Congress’s Descriptive Cataloging Manual, which details how RDA is applied.

RDA also aligns with the Statement of International Cataloguing Principles by IFLA, aiming to facilitate international sharing of bibliographic and authority data. However, the transition to RDA has been gradual and costly, with some questioning whether it was worth the effort. Despite its potential for a radically different approach, most RDA-based records still resemble those created with AACR2, primarily because of the continued use of the MARC format. Significant changes, such as cataloguing as linked data, would require a new encoding format, but such a shift is yet to occur. The benefits of such a transformation remain uncertain, with ongoing debate about the necessity and value of moving away from MARC and towards RDA’s vision.

 

 

Although built on the foundations of AACR2, RDA is based on the theoretical framework of two conceptual models: Functional Requirements for Bibliographic Records (FRBR) and Functional Requirements for Authority Data (FRAD). RDA focuses attention on the tasks the user carries out in the process of resource discovery: find, identify, select, obtain, and explore. Thus the organisation and structure of RDA is radically different from that of AACR2. RDA is designed to be independent of ISBD, but is flexible enough to allow for the incorporation of ISBD elements such as punctuation and order of descriptive elements if required by information agencies. This flexibility is very important for maintaining compatibility between existing records and new records created under RDA. RDA also incorporates the concept of core elements, with the inclusion of other specific elements being left to the discretion of the information agency or the individual creating the data. (RDA 0.6)

 

FR Family

I have learned that Resource Description and Access (RDA) incorporates concepts from the Functional Requirements for Bibliographic Data (FRBR) and Functional Requirements for Authority Data (FRAD) models. These, along with the Functional Requirements for Subject Authority Data (FRSAD), have been unified into a single conceptual model known as the IFLA Library Reference Model (IFLA LRM). Published by IFLA in 2017, LRM was developed to address inconsistencies among the three separate FR models. LRM, like FRBR, is a high-level conceptual reference model based on an entity-relationship framework. While RDA currently uses the three original FR models, the updated RDA Toolkit is aligned with the new LRM model. However, in the ETL505 studies, the focus will remain on the RDA Toolkit based on the FRBR model.

FRAD and LRM User Tasks

The user tasks in FRAD are similar to those in FRBR but focus on the functions of authority records rather than bibliographic records. Name authority records represent the agents associated with information resources, such as authors and publishers—referred to in FRBR/FRAD as Group Two entities (agents, including persons, families, and corporate bodies). While authority records are not frequently used by end-users, they are essential tools for cataloguers to standardise the names of authors and other agents.

LRM focuses on the user tasks involved in retrieving and accessing resources from library catalogues and bibliographic databases. These tasks include find, identify, select, obtain, and a newly added task, explore. The explore task enables users to follow links within a displayed record, connecting them to other related resources. This feature enhances the user’s ability to navigate and explore the collection through the catalogue.

Table 4.1. FRBR, FRAD and LRM user tasks

FRBR user tasks	FRAD user tasks	LRM user tasks
find (a resource meeting the user’s search criteria)	find (information about an entity)	find (a resource meeting the user’s search criteria)
identify (distinguish between similar resources)	identify (distinguish between similar entities)	identify (distinguish between similar resources)
select (a resource appropriate to the user’s needs)	contextualise (clarify the relationship of the entity to others)	select (a resource appropriate to the user’s needs)
obtain (access the resource)	justify (a task for cataloguers more than end-users)	obtain (access the resource)
		explore (discover resources using relationships between them)