BIBFRAME has been a buzzword in the library industry since the Library of Congress launched it in 2011. So, by now, you may know that it’s the descriptive data model that will eventually replace MARC. But you may not understand exactly how BIBFRAME works.
As you probably guessed, BIBFRAME uses a different classification system than MARC. It uses three core classes for bibliographic description :
- Work. The creative, intellectual content being cataloged (includes information like authors, languages and what it’s about).
- Instance. The physical embodiment of the work (includes information like publisher, place and date of publication and format).
- Item. The actual copy of an instance being cataloged (includes information like location, shelf mark and barcode).
Beyond these core classes, BIBFRAME has subclasses that flesh out relationships between resources further, like :
- People, organizations, etc., linked with a work or instance because they were the author, editor, artist, photographer, composer, illustrator, etc.
- Subjects. Concepts related to a work. A work can be tied to more than one concept (i.e. about more than one subject). Potential concepts could be topics, places, events, works, instances, items, agents, etc.
- Things that happened (and were recorded) that may be the content of a work.
But as helpful as it is to understand the core classes and subclasses that go into a BIBFRAME record, the most exciting and important BIBFRAME element is Linked Data.
Linked Data: The Backbone of BIBFRAME
Linked Data is data that’s linked to other data using a set of best practices for publishing and connecting data on the Semantic Web . The Semantic Web is an extension of the traditional web you’re familiar with. Except, instead of being a web of documents designed for humans to read and use, it’s a web of data designed for humans and computers to read and use .
If you’re struggling to wrap your mind around Linked Data and the Semantic Web, here’s an example:
When you search for something on Google, like “how to build a birdhouse,” you get highly relevant results at the top of the search page. You get a carousel of videos that teach you how to build a birdhouse and a list of related questions (and answers) about building a birdhouse, like “What is the best wood for a birdhouse?” How does Google know this information is relevant to your search?
Because whoever created the information followed best practices for publishing and connecting data. That means their data is linked to other data on the Semantic Web and can be read and understood by Google’s computers. What are the best practices for creating Linked Data?
To qualify as Linked Data, data must use three key technologies :
- URIs (Uniform Resource Identifiers) that identify things on the web. Every book, person, etc. on the web has its own URI.
- HTTP (Hyper Text Transfer Protocol) is connected to URIs to retrieve data. Basically, if you want to look up a URI, you use HTTP.
- RDF (Resource Description Framework) is the data model to describe and link data. When someone looks up a URI, RDF is used to provide useful information.
By including these three key technologies in BIBFRAME records, library data suddenly gets looped into the web of Linked Data on the Semantic Web. It’s connected to all sorts of new, helpful information outside of the library.
For example, with BIBFRAME, if someone searches the term “Little Women” on Google because they’re interested in the new movie, their search results would also show them that their local library owns the 1994 movie and they can rent it today. People who visit the library in person will get richer search results when they search the library catalog too, because library catalogs can pull relevant information from online.
So, by connecting library data to data on the web, BIBFRAME allows current library users to make useful connections between resources they would have missed otherwise and attracts new library users. Put simply?
BIBFRAME brings library data into the digital age by making it more useful, connected and visible than ever before.