Knowledge Organisation II: Cataloguing Canons and Principles, Centralised and Co-operative Cataloguing, CCC and AACR-II; ISBD, MARC 21, CCF, RDA, FRBR and BIBFRAME; ISO 2709, Z39.50 and Z39.71; and Dublin Core, METS, MODS and EAD
2. Library cataloguing codes: CCC and AACR-II, and the move to RDA
Ranganathan's Classified Catalogue Code (CCC), first published in 1934 (fifth edition 1964), prescribes a classified catalogue in two parts. The classified part arranges main entries by class number, so the catalogue mirrors the classified shelf and shows subjects in their filiatory sequence. The alphabetical part holds added entries and index entries: book index entries (under authors, collaborators, titles and series), class index entries derived by chain indexing (so that users who know a subject's name can find its class number), and cross-reference index entries. A classified catalogue collocates subjects systematically; a dictionary catalogue, following Cutter, files all entries (author, title, subject) in one alphabetical sequence.
The Anglo-American Cataloguing Rules appeared in 1967 in separate North American and British texts. The second edition, AACR2 (1978, edited by Michael Gorman and Paul Winkler), unified them; it was revised in 1988, 1998 and 2002, with updates to 2005. Part I covers description, built on ISBD with eight areas and levels of description; Part II covers the choice and form of headings, uniform titles and references, including the "rule of three" (with more than three authors, main entry goes under title). AACR2 has been succeeded by Resource Description and Access (RDA), published in June 2010 and implemented by the Library of Congress, the National Library of Medicine and the National Agricultural Library on 31 March 2013. RDA is based on the FRBR and FRAD models, is designed for digital and linked-data environments, and has since been realigned with the IFLA Library Reference Model in a restructured RDA Toolkit (2020).
| Point | AACR2 | RDA |
|---|---|---|
| Basis | Card-catalogue tradition, ISBD areas | FRBR and FRAD entities and user tasks, now the LRM |
| Type of material | General material designation, such as [electronic resource] | Content, media and carrier types (MARC fields 336, 337, 338) |
| Abbreviations | Many: p., ill., ed. | Few: "pages", "illustrations"; transcribe what you see |
| Many creators | Rule of three | Record as many as needed; only the first is required |
| Relationships | Implicit | Explicit relationship designators (author, illustrator, translator) |
3. Standards of bibliographic record formats and description: ISBD, MARC 21, CCF, RDA, FRBR and BIBFRAME
The International Standard Bibliographic Description (ISBD), developed by IFLA from 1971 (ISBD(M) for monographs) and issued as a consolidated edition in 2011, fixes the elements of a description, their order, and prescribed punctuation that lets records be understood across languages. The areas are: 0 content form and media type; 1 title and statement of responsibility; 2 edition; 3 material or type of resource specific; 4 publication, production, distribution; 5 material description; 6 series and multipart monographic resources; 7 notes; 8 resource identifier and terms of availability. Each area after the first is preceded by a full stop, space, dash, space, and within area 1 a space-slash-space introduces the statement of responsibility and a space-colon-space other title information.
| MARC 21 tag | Content |
|---|---|
| Leader (24 characters) | Record length, status, type and other processing information |
| 001, 005, 008 | Control number; date of latest transaction; fixed-length data elements |
| 020 | ISBN |
| 100 | Main entry, personal name |
| 245 | Title statement |
| 250 | Edition statement |
| 260 / 264 | Publication (264 is used under RDA) |
| 300 | Physical description |
| 490 | Series statement |
| 5XX | Notes (500 general note, 520 summary) |
| 650 | Subject added entry, topical term |
| 700 | Added entry, personal name |
| 856 | Electronic location and access (URL) |
MARC (machine-readable cataloguing) was developed at the Library of Congress in the 1960s under Henriette Avram (MARC II, 1968). MARC 21 (1999) harmonised USMARC and CAN/MARC and has formats for bibliographic, authority, holdings, classification and community information data. Each variable data field has a three-digit tag, two indicators and subfields marked by a delimiter and code ($a, $b, $c). The Common Communication Format (CCF), first published by UNESCO in 1984, was designed as a simpler common exchange format between the many MARC variants and abstracting services. The Functional Requirements for Bibliographic Records (FRBR, IFLA 1998) is a conceptual entity-relationship model: Group 1 entities are work, expression, manifestation and item; Group 2 are person and corporate body (family was added by FRAD, 2009); Group 3 are concept, object, event and place; and the user tasks are find, identify, select and obtain. The IFLA Library Reference Model (LRM, 2017) consolidated FRBR, FRAD and FRSAD and added the task explore. BIBFRAME (Bibliographic Framework), launched by the Library of Congress in 2011, with BIBFRAME 2.0 in 2016, is the linked-data model intended to replace MARC; its core classes are Work, Instance and Item, expressed in RDF.
4. Standards for bibliographic information interchange: ISO 2709, Z39.50 and Z39.71
| Standard | What it does |
|---|---|
| ISO 2709 (Format for information exchange; first published 1973) | Defines the structure of an exchange record: a 24-character record label (leader), a directory of tags, lengths and starting positions, and variable fields separated by field separators, ending with a record terminator. MARC 21, UNIMARC and CCF records all use this structure. |
| ANSI/NISO Z39.50 (Information retrieval protocol; version 3, 1995; also ISO 23950) | A client-server protocol that lets one system search and retrieve records from another system's database without knowing its native search language; used for copy cataloguing and virtual union catalogues. SRU (Search/Retrieve via URL) is its web-based successor. |
| ANSI/NISO Z39.71 (Holdings Statements for Bibliographic Items) | Specifies how to display holdings statements, what volumes, issues and parts a library holds, consistently across single-part items, multi-part items and serials, so union lists and ILL systems can read them. |
5. Metadata standards: Dublin Core, MARC 21, METS, MODS and EAD
Metadata is structured information that describes, explains, locates or otherwise makes it easier to retrieve, use or manage a resource: "data about data". It is commonly divided into descriptive metadata (for discovery: title, creator, subject), structural metadata (how the parts of a compound object fit together: pages, chapters, files) and administrative metadata (for management: technical, rights and preservation metadata). The Dublin Core Metadata Element Set came out of a workshop at Dublin, Ohio, in 1995 hosted by OCLC; its fifteen elements are title, creator, subject, description, publisher, contributor, date, type, format, identifier, source, language, relation, coverage and rights. All are optional and repeatable; it is standardised as ISO 15836 and ANSI/NISO Z39.85, maintained by the Dublin Core Metadata Initiative, extended by the DCMI Metadata Terms, and required in unqualified form by the OAI-PMH harvesting protocol.
| Standard | Maintained by and year | Purpose |
|---|---|---|
| MARC 21 | Library of Congress and Library and Archives Canada, 1999 | Rich descriptive metadata for library catalogues; also expressible as MARCXML |
| METS (Metadata Encoding and Transmission Standard) | Library of Congress, 2001 | An XML wrapper that packages descriptive, administrative and structural metadata for a digital object; sections are the METS header, descriptive metadata, administrative metadata, file section, structural map (the only required section), structural links and behaviour |
| MODS (Metadata Object Description Schema) | Library of Congress, 2002 | An XML schema for descriptive metadata, richer than Dublin Core and simpler than MARC, using language-based tags such as titleInfo, name and originInfo; often used inside METS |
| EAD (Encoded Archival Description) | Library of Congress with the Society of American Archivists; version 1.0 in 1998, EAD3 in 2015 | Encodes archival finding aids hierarchically, from the whole collection down through series and files to items |
Key takeaways
- Cutter's objects (1876), the Paris Principles (1961) and IFLA's ICP (2009, revised 2016) define catalogue functions; Ranganathan's six canons are ascertainability, prepotence, individualisation, sought heading, context and consistency.
- Centralised cataloguing (LC cards, CIP, national bibliographies) catalogues once for all; co-operative cataloguing (OCLC WorldCat, IndCat) shares the work; CCC (1934) builds a two-part classified catalogue with chain-indexed class index entries.
- AACR2 (1978; revised to 2005) has been succeeded by RDA (published 2010, adopted by LC, NLM and NAL on 31 March 2013), which is based on FRBR and now aligned with the IFLA LRM (2017).
- ISBD has areas 0 to 8 with prescribed punctuation; MARC 21 (1999) uses tags such as 020, 100, 245, 260/264, 300, 650 and 856; CCF (UNESCO, 1984); FRBR (1998) has work, expression, manifestation and item; BIBFRAME (2011; 2.0 in 2016) uses Work, Instance and Item in RDF.
- ISO 2709 packages records (24-character leader, directory, fields); Z39.50 is a search-and-retrieve protocol (successor SRU); Z39.71 standardises holdings statements; Dublin Core has 15 elements (ISO 15836); METS wraps, MODS describes (the syllabus's "MODES"), EAD encodes finding aids.
Practice questions (10)
Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.
How many elements are there in the simple Dublin Core Metadata Element Set? Type the number.
Numerical answer — type the value.
Show answer
Answer: 15
The Dublin Core set has fifteen elements: title, creator, subject, description, publisher, contributor, date, type, format, identifier, source, language, relation, coverage and rights, all optional and repeatable.In MARC 21, which field holds the title statement?
Show answer
Answer: D — 245
Field 245 is the title statement, with $a title proper, $b remainder of title and $c statement of responsibility. 100 is the personal-name main entry, 650 a topical subject and 856 the electronic location.In the FRBR model, a particular English translation of Premchand's Godaan, considered as a text, is an example of:
Show answer
Answer: A — An expression
A translation is a distinct realisation of the work in a language, which FRBR calls an expression. The work is the abstract creation, a manifestation is a particular edition or printing, and an item is a single copy.Which of the following are among Ranganathan's canons of cataloguing? Select all that apply.
Show answer
Answer: B — Canon of sought heading; C — Canon of prepotence
Sought heading and prepotence are cataloguing canons, with ascertainability, individualisation, context and consistency. Exhaustiveness (of arrays) and modulation (of chains) are classification canons of the idea plane.Assertion (A): Major US national libraries still catalogue with AACR2 rather than RDA. Reason (R): The Library of Congress, the National Library of Medicine and the National Agricultural Library implemented RDA on 31 March 2013. Choose the correct option.
Show answer
Answer: D — A is false, but R is true
A is false and R is true. RDA, published in 2010, succeeded AACR2, and the Library of Congress, the National Library of Medicine and the National Agricultural Library adopted it together on 31 March 2013, so they no longer catalogue with AACR2.Which standard governs the display of holdings statements for bibliographic items?
Show answer
Answer: B — ANSI/NISO Z39.71
Z39.71 specifies holdings statements. Z39.50 is the information retrieval protocol, ISO 2709 the exchange record structure, and ISO 15836 the Dublin Core standard.Which statements about metadata standards are correct? Select all that apply.
Show answer
Answer: B — MODS uses language-based tags and is simpler than MARC; C — EAD is designed to encode archival finding aids; D — The structural map is the only mandatory section of a METS document
METS requires only the structural map, EAD encodes finding aids, and MODS uses words such as titleInfo instead of numeric tags. Dublin Core elements are all optional and repeatable, so the last statement is false.How many characters long is the leader (record label) of a MARC 21 or ISO 2709 record? Type the number.
Numerical answer — type the value.
Show answer
Answer: 24
The leader is fixed at 24 characters and carries the record length, status, type of record, bibliographic level and the base address of data; the directory and variable fields follow it.BIBFRAME, the Library of Congress's linked-data replacement for MARC, is built around which core classes?
Show answer
Answer: A — Work, Instance and Item
BIBFRAME 2.0 models Work, Instance and Item, collapsing FRBR's work and expression into Work. Work-Expression-Manifestation-Item is FRBR, title-creator-subject are Dublin Core elements, and leader-directory-fields is the ISO 2709 record structure.In Ranganathan's Classified Catalogue Code, class index entries in the alphabetical part are derived by:
Show answer
Answer: C — Chain indexing
CCC derives class index entries by chain indexing, which Ranganathan devised to lead from subject names to class numbers. PRECIS was Derek Austin's system for the British National Bibliography, KWIC is Luhn's keyword-in-context index, and citation indexing is Garfield's.