Ancient Places in Today's Library: Pleiades URIs and MARC

By Gabriel McKee

In September, the ISAW Library submitted a proposal to the Library of Congress to add the Pleiades gazetteer to its list of authorized sources for subject heading terms. That same month the proposal was accepted, and Pleiades was entered into the official list and assigned an identifying code. With this code, place names from Pleiades can now be entered into library catalog records.

Though this may seem like a somewhat arcane bit of technical news, it’s actually a big step forward for both Pleiades and the role of libraries in the Linked Open Data movement. The ability to use Pleiades names in subject headings is useful for keyword searching, as it allows us to provide access to both the ancient and modern names of some locations. Under the cataloging rules used by American academic libraries, inhabited places are cataloged using their modern names. For instance, the latest ISAW publication, Graffiti from the Basilica in the Agora of Smyrna, is assigned the LC geographic heading İzmir (Turkey), the modern name of the city. Since Pleiades is now a recognized source of authoritative name data, we can now add to this book's record a geographic heading for the city’s Pleaides heading, which records not only one ancient name, but three: Naulochon/Smyrna/Palaia Smyrna. 

But additional name access is not all that this change allows.

recent change to MARC (MAchine-Readable Cataloging)—the standard format in which library catalog records are coded—allows for the entry of uniform resource identifiers (URIs) in subject headings. URIs—unique character strings used to identify a resource or thing—are one of the foundational principles behind Linked Data. Though you may not have heard of URIs, you probably use them every day—web URLs are a form of URI, and due to their utility and ubiquity most URIs are now structured in HTTP format and point to an online location. Though the names of places in Pleiades are useful, it is the unchanging URIs that Pleiades associates with those names and places that truly distinguish it as a 21st-century linked data resource. By encoding the URI for Smyrna ( in the metadata for a resource about that place, we create a connection between the resource, the conceptual place, and other resources that also connect to it. The Pleiades page in turn contains references to additional resources about the place in multiple periods. The metadata model known as the Resource Description Framework (RDF) describes individual information resources—from books to websites to physical artifacts—in three-part units of information (subject : predicate : object), called triples, that connect URIs to one another semantically, representing each resource as a part of web of interconnected information. Each portion of a triple is represented by a URI. For example:


RDF uses the relationships, represented by the links used above, to link resources to each other. A linked data library catalog or other database would use these links to draw connections between related resources. In the example above, a user would be able to easily navigate from Graffiti from the Basilica to Letter to the Philippians, other works by Polycarp, and other works concerning Smyrna. This would enhance the user's ability to discover information, and could highlight unexpected connections between different resources.

The use of URIs in library cataloging is relatively new, but has the power to transform the usefulness of cataloging and cataloging metadata. The Library of Congress is currently at work on the BIBFRAME (Bibliographic Framework) Initiative, an entirely new framework for resource cataloging that is intended to replace MARC. Though it is not likely to be implemented on a large scale for several years, BIBFRAME is built entirely on linked data principles, and will rely on URIs for connecting users to information. In preparation for this, the controlled vocabularies used for subjects and names are beginning to shift to a URI-based model.

The ISAW Library is ready to be an active agent in that conceptual shift. Beginning this semester, we will be adding Pleiades headings and URIs to many of our records for new materials. We are already beginning to think about different uses for this metadata, including the creation of browsable maps of our collection and the automatic updating of Pleiades pages with information about new resources that link to them. And we will also work to expand and enhance Pleiades itself, creating new Pleiades IDs for places represented in our collection but not yet in the gazetteer, particularly in Central Asia and Ancient China.

For more information about this project, please email or .