glossaryPage.heroH1
glossaryPage.heroSubtitle
glossaryPage.definitionTitle
A MARC record is a machine-readable bibliographic record that describes a library item — book, journal, ebook, AV material — in a standardised structured format. MARC (Machine-Readable Cataloging) was developed by Henry Avram at the Library of Congress in the 1960s; MARC 21 is the dominant variant in North America, UK, and Australia, while UNIMARC and IFLA-specified variants serve parts of Europe and Asia.
glossaryPage.howItWorksTitle
A MARC record is a sequence of fields, each identified by a three-digit tag, holding subfields delimited by the dollar sign ($a, $b, $c, etc.). Field 245 holds the title statement (245$a is the main title, $b is the subtitle, $c is the statement of responsibility). Field 100 holds the main author (personal name); 110 holds corporate body; 130 holds uniform title. Field 020 holds the ISBN; 022 the ISSN; 050 the Library of Congress call number; 082 the Dewey Decimal call number. Subject headings live in 6XX fields (650 for topical, 651 for geographic, 600 for personal-name subjects). The leader (24 characters at the start) declares record type, bibliographic level, and encoding level. Library systems import MARC records via Z39.50 from Library of Congress, OCLC WorldCat, or British Library; the OPAC then indexes the relevant fields for patron search.
glossaryPage.whySchoolsTitle
Libraries use MARC because it is the lingua franca that lets every cataloging system speak to every other. A cataloger at a small school library can pull a fully-cataloged MARC record for "To Kill a Mockingbird" from the Library of Congress in seconds, drop it into the local OPAC, and have the title searchable that afternoon — no re-typing, no missing subject headings, no inconsistent author-name formatting. According to IFLA (International Federation of Library Associations), 95%+ of the world's academic and public libraries store catalogs in MARC or a MARC-derived format. Even institutions experimenting with newer formats like BIBFRAME or RDA-native records typically export to MARC for interchange. For acquisition workflows, vendors (Ingram, Baker & Taylor, Follett) supply MARC records with new-book shipments — the library scans the barcode and the catalog entry appears.
glossaryPage.keyFeaturesTitle
- Standardised structured format — fields and subfields with three-digit tags
- MARC 21 (North America, UK, Australia) and UNIMARC (Europe, Asia) variants
- Z39.50 / SRU protocols for record exchange between libraries
- Authority records (MARC 21 Authority) for consistent author and subject naming
- Subject heading vocabularies — LCSH, Sears, MeSH integrate as subfields
- Maps to MARCXML for web/API exchange and BIBFRAME for linked-data future
glossaryPage.faqTitle
Is MARC obsolete? Should we be using BIBFRAME instead?
MARC remains the dominant working format. The Library of Congress launched BIBFRAME in 2011 as a linked-data successor; uptake has been gradual. As of 2024-2025, most library management systems continue to read and write MARC 21 as the primary format, with BIBFRAME experimentation in research libraries and the LoC. For school, college, and most university libraries, MARC is what vendors ship, what cataloging consortia exchange, and what library systems index. BIBFRAME interest exists; MARC remains the operational standard.
What are the most-used MARC fields a librarian should know?
Common bibliographic fields: 020 (ISBN), 100 (personal author), 110 (corporate author), 245 (title statement), 250 (edition), 260/264 (publication info), 300 (physical description), 490 (series statement), 500 (general note), 520 (summary/abstract), 650 (topical subject heading), 700 (additional author), 856 (electronic location/URL). Plus the leader and 008 (fixed-length data elements). LCSH (Library of Congress Subject Headings) supplies the controlled vocabulary for 650.
How does a library import MARC records?
Via Z39.50 query (real-time pull from Library of Congress, OCLC WorldCat, British Library, or peer libraries), via MARCXML or MARC binary file upload (vendor-supplied records for new-book shipments), or via SRU (Search/Retrieve via URL — the modern HTTP-based Z39.50 alternative). Most library systems support all three. A cataloger typically queries by ISBN; the system returns a matching MARC record; the cataloger adds local fields (call number, location, accession number, copy number) before saving.
What is the difference between MARC 21 and UNIMARC?
MARC 21 emerged from the merger of US MARC and Canadian MARC in 1999 and is used in North America, UK, Australia, and many Anglophone library systems. UNIMARC was developed by IFLA in 1977 to support multilingual cataloging and is used in France, Portugal, Italy, Russia, Greece, and parts of Africa and Latin America. The two are conceptually similar but use different field numbers (UNIMARC 200 = MARC 21 245, etc.). Conversion tools (MarcEdit, USEMARCON) translate between them.