BIBFRAME: Basics and Resources

BIBFRAME is a bibliographic framework for the description of physical and online objects to make them accessible on the Web by using a standard Linked Data model. It is a replacement for MARC.

Why the move to BIBFRAME?

The Machine Readable Cataloging (MARC) standard was developed over forty years years ago to make bibliographic records usable by computers.  MARC records were the basis for online public library catalog (OPAC) records to make items searchable in library catalogs. This sufficed until the 1990s and the rise of the Internet.

However, MARC was so entrenched that it took another decade before the Library of Congress made the first move towards converting MARC to a Web standard with MARCXML.  This schema essentially reformatted the MARC record fields and subfields into an XML schema but was not an entirely new model.  MARCXML wasn’t widely adopted.

So in 2011 the Library of Congress, along with the consulting company Zepheria, set out to create a new bibliographic framework called BIBFRAME to make library records conform to Web standards. BIBFRAME is a web-first Linked Data model intended to make library records accessible to the Web at large.

The Basics

BIBFRAMEThe BIBFRAME initiative was announced in an open letter “A Bibliographic Framework for the Digital Age” (dated October 31, 2011) by Deanna Marcum from the Library of Congress. In this letter it was recognized that the MARC record was outdated and a new format was needed for the Internet age.

So BIBFRAME was developed on the RDF model using Linked Data through the Bibliographic Framework initiative.  Its current draft specification is version 2.0.

The BIBFRAME Model consists of the following core classes:

  • Creative Work – a resource reflecting a conceptual essence of the cataloging item.
  • Instance – a resource reflecting an individual, material embodiment of the Work.
  • Authority – a resource reflecting key authority concepts that have defined relationships reflected in the Work and Instance. Examples of Authority Resources include People, Places, Topics, Organizations, etc.
  • Annotation – a resource that decorates other BIBFRAME resources with additional information. Examples of such annotations include Library Holdings information, cover art and reviews.

BIBFRAME Model

Within the Vocabulary of BIBFRAME there is a current total of 53 classes and subclasses falling under Resource such as Work, Instance, Authority, and Annotation.  A BIBFRAME Resource can have 289 current properties such as absorbedBy, classificationLcc, doi, format, relatedInstance, title, uri, to give just a few examples.

A good explanation of BIBFRAME was given in an ALCTS webinar titled “From MARC to BIBFRAME: An Introduction” by Victoria Mueller from Zepheira and Carolyn Hansen from the University of Cincinnati.

With millions of MARC records created over the last several decades, an obvious question arises: How will MARC records be converted into BIBFRAME?  The Library of Congress addressed this need in their paper “Bibliographic Framework as a Web of Data” (see Resources below):

A key part of supporting the BIBFRAME model is in providing tools and supporting services for helping migrate from MARC to a Linked Data environment. They should provide a means of navigating the output of a declarative BIBFRAME pipeline which takes existing MARC 21 data and translates this to the BIBFRAME model.

BIBFRAME.ORG offers a MARC to BIBFRAME Transformation Service to convert MARCXML files.  There is also an experimental open source marc2bibframe XQuery utility from the Library of Congress (and a Python version by Zepheira).

Resources

Here are some great online resources to learn about BIBFRAME:

  • BIBFRAME Editor – Open source editing software downloadable from Github.
  • BIBFRAME FAQ – Frequently asked questions and answers from the Library of Congress.
  • BIBFRAME Listserv – Bibliographic Framework Transition Initiative Forum.
  • BIBFRAME.ORG – Is an index site to BIBFRAME Initiative, Model & Vocabulary, and Implementation and Testing sites.
  • Bibliographic Framework Initiative – Library of Congress website with official BIBFRAME information, specifications, FAQ, tools, news, and more.
  • Zepheira – Linked Data and BIBFRAME training from the company which was consulted by the Library of Congress to develop the BIBFRAME specifications.

Here are some popular papers about BIBFRAME:

BIBFRAME AV Assessment: Technical, Structural, and Preservation Metadata by Bertram Lyons and Kara Van Malssen

BIBFRAME AV Assessment: Technical, Structural, and Preservation MetadataThis paper, undertaken on behalf of the Library of Congress, investigates how metadata of audiovisual material can be best handled using BIBFRAME.  Sections cover preservation, structural, and technical metadata.   Appendices give examples of video, audio, and film examples.  Published in 2015, revised January 4, 2016.

Access the full-text paper (PDF).

Bibliographic Framework as a Web of Data: Linked Data Model and Supporting Services by the Library of Congress

Bibliographic Framework as a Web of Data: Linked Data Model and Supporting ServicesStraight from the source, this paper is an early look into the BIBFRAME model to introduce the subject and begin discussion.  It covers the four classes (Creative Work, Instance, Authority, and Annotation) and vocabulary.  The papers covers related initiatives such as OCLC’s WorldCat, Schema.org, RDA, and FRBR.  Published in 2012.

Access the full-text paper (PDF).

The Relationship between BIBFRAME and OCLC’s Linked-Data Model of Bibliographic Description: A Working Paper by Carol Jean Godby

The Relationship between BIBFRAME and OCLC’s Linked-Data Model of Bibliographic Description: A Working Paper“This document describes a proposed alignment between BIBFRAME and a model being explored by OCLC with extensions proposed by the Schema Bib Extend project, a W3C-sponsored community group tasked with enhancing Schema.org to the description of library resources.”  The paper also covers FRBR and gives examples of BIBFRAME in Turtle and RDF/XML syntax.  Published in 2013.

Access the full-text paper (PDF).

The Virtual International Authority File (VIAF)

VIAF Authority Record for Harper Lee

The authority file goes global.

It was probably inevitable that there would be an attempt to create a global authority file for library bibliographic data.  Any worldwide authority file would need the support of major national libraries and library organizations.  We now have the first viable candidate.

VIAFThe Virtual International Authority File (VIAF) is not a new service, but one that has become more prominent in recent times due to the direction of OCLC (starting in 2012) and the rise of linked data. It began as a joint project of the Library of Congress and the German National Library, then the National Library of France and OCLC, but now has a long list of contributors.

From OCLC, the VIAF service has these characteristics:

  • A collaborative effort between national libraries and organizations contributing name authority files, furthering access to information
  • All authority data for a given entity is linked together into a “super” authority record
  • A convenient way for the library community and other agencies to repurpose bibliographic data produced by libraries serving different language communities

The VIAF combines the national authority files—which contain data such as names of authors and companies, conferences, places, and subject headings in their native languages—into a single international name authority service.

There are several benefits of a global name authority file (or service). Obviously, there is the fundamental benefit of providing uniform names for indexing and searching.  With a global authority file, the connecting application, such as a discovery service, can retrieve and display names based on the language of the application’s user interface—and can switch on-the-fly.  It enables the generation of “see” and “see also” links across languages.  The VIAF allows developers to search and access the authority data using an API.

As mentioned at the beginning, the VIAF is one source which can be accessed via a linked data URI.  Here’s an example:

creator: {
@id: “http://id.loc.gov/authorities/names/n78089036”,
label: “Wodehouse, P. G. (Pelham Grenville), 1881-1975”,
sameAs: “http://viaf.org/viaf/sourceID/LC|n78089036”
}

The above operation translates a Library of Congress control number for an author to the related VIAF record.

You can search the VIAF at viaf.org.

View a typical authority record at viaf.org/viaf/46734193.

VIAF authority Record Map for P. G. Wodehouse

Linked Data: Basics and Resources

Linked Data is a set of practices which involves the publishing, sharing, and connecting of related data across the Web in a structured format, preferably using an open access license.

Why use Linked Data?

Using Linked Data is a way to make online library resources—even those items having metadata and located in online library catalogs—accessible to the Web at large by publishing the data in a standard, openly-accessible way.

Properly described and published, online library resources could be harvested by search engines and linked to from other online resources.  In order to do this, Linked Data must be structured in a standard way for the Web.

On the cataloging and description side, Linked Data from online resources such as the Virtual International Authority File (VIAF) can serve as global authority records.

The Basics

Linked Data mugTim Berners-Lee defined his four principles of Linked Data in 2009:

  1. Use URIs as names for things
  2. Use HTTP URIs so that people can look up those names
  3. When someone looks up a URI, provide useful information, using the standards (RDF*, SPARQL)
  4. Include links to other URIs so that they can discover more things.

Linked Data has two components: URI and RDF.

URI (Uniform Resource Identifier) is a location as name of a digital resource on the Web, typically given in the form of a Uniform Resource Locator (URL) that also gives the protocol (ex. HTTP) of accessing it.

RDF (Resource Description Framework) is a standard structured file (often in XML) which describes a digital resource and contains URIs.

For a Linked Data statement, you need three components: subject, predicate, and object.  These together are referred to as a triple.

<subject> <predicate> <object>

Subject – what’s being described and linked, a URI or blank node.
Predicate – describes the connection, always a URI.
Object – the resource being linked to, a URI, literal, or blank node.

<Atlas Shrugged> <was created by> <Ayn Rand>

One framework for moving library metadata records into Linked Data format is BIBFRAME, the replacement for MARC being developed by the Library of Congress and others.  We will continue to see new tools created for the conversion of library metadata formats (such as Dublin Core) into Linked Data formats (such as RDF/XML and Turtle).

Finally, as more library records are converted and made available as Linked Data, we will see library vendors start to utilize Linked Data in their products.  For example, Ex Libris recently announced that the company has launched a program to “harness linked data technology in its resource management and discovery solutions” and published the paper Putting Linked Data at the Service of Libraries.

Resources

Here are some great online resources to learn about Linked Data:

Articles and books about Linked Data and libraries:

Common Ground: Exploring Compatibilities Between the Linked Data Models of the Library of Congress and OCLC by Carol Jean Godby and Ray Denenberg

Common Ground: Exploring Compatibilities Between the Linked Data Models of the Library of Congress and OCLC“Jointly released by OCLC and the Library of Congress, this white paper compares and contrasts the compatible linked data initiatives at both institutions. It is an executive summary of a more detailed technical analysis that will be released later this year.”  Published in 2015.

Access the abstract and full-text article (PDF).

Library Linked Data in the Cloud: OCLC’s Experiments with New Models of Resource Description by Carol Jean Godby, Shenghui Wang, and Jeffrey K. Mixter

Library Linked Data in the Cloud: OCLC's Experiments with New Models of Resource DescriptionPart of the “Synthesis Lectures on the Semantic Web: Theory and Technology” series.  OCLC looks at Linked Data in cataloging and its products.  Topics include authority records including VIAF, cataloging and FRBR, text mining, and the library Linked Data cloud.  Published in 2015.

Watch the companion video (59:50).

View details and find a place to buy or borrow at Google Books.

Library Linked Data: Research and Adoption by Erik T. Mitchell

Library Linked Data: Research and AdoptionThis title is actually an issue of Library Technology Reports from ALA Tech Source.  The report covers basic metadata in libraries and museums, Linked Open Data (LOD), case studies including BIBFRAME, and “issues, opportunities, and trends” in metadata. Published in 2013.

View details and find a place to buy or borrow at Google Books.

Linked Data for Libraries, Archives and Museums: How to Clean, Link and Publish your Metadata by Seth van Hooland and Ruben Verborgh

Linked Data for Libraries, Archives and Museums: How to Clean, Link and Publish your MetadataIn this book published by the American Library Association, the authors cover Linked Data and focus on the metadata standards supporting it. They present key concepts of metadata including metadata modelling, cleaning, reconciling, enriching, and publishing. Published in 2014.

View details and find a place to buy or borrow at Google Books.