Skip to Main Content

Tulane University Digital Library Metadata Guidelines

These guidelines were approved by the Digital Production Review Committee in September 2019.

Purpose of the Metadata Guidelines

The Tulane University Digital Library Metadata Guidelines provides specifications for creating metadata records for digital resources disseminated in the Tulane Digital Library and the Louisiana Digital Library (LDL). The LDL is a collaborative digital repository that supports Tulane’s digital collections as well as those of participating higher education and cultural heritage institutions in Louisiana. This document updates and expands upon the guidelines presented in the Provisional Guidelines for Descriptive Metadata and the Louisiana Digital Library Core Record Guidelines.

 

Application of these metadata best practices supports the following objectives:

  • Ensure quality control for metadata records
  • Improve discovery of resources
  • Increase interoperability across all collections created by the Tulane University
  • Increase interoperability with other digital libraries participating in the Open Archives Initiative
  • Inform users on the digital object structure and necessary viewers needed to view the digital resource
  • Assist with management and long‐term preservation of digital files

 

To further these goals, these guidelines attempt to set a standard for describing objects which complies with MODS, DPLA, and emerging standards for linked open data. This documentation includes general information for creating metadata content and matching it to particular metadata schemes that should allow content owners to create metadata that can be transformed and improved using automated processes. It also includes further information about which information in RDA, DACs, and other shared content guides informed our choices.

Metadata is generally defined as descriptive information of other resources. It supports the discovery, use, management, and preservation of these resources. There are three commonly accepted types of metadata:

  1. Discovery/Descriptive metadata: information used for the indexing, discovery and identification of a digital resource. Examples are resource title, creator of the resource, and subject(s) of the resource.

  2. Structural metadata: information used to explain the internal organization of the resource and/or its relationship to other data; useful in display and navigation of digital resources; may include information on what intermediary tools are required to view the content of a resource.

  3. Administrative metadata: information needed to manage the resource over time, including technical information such as the resolution of the image, file size, file format, hardware/software used to produce the digital resource, and when the object and related metadata were last updated.

The original Tulane Digital Library Metadata Guidelines draws heavily from the CCDL Guidelines for Metadata Application in the Claremont Colleges Digital Library Version 2 June 2007, the LOUISiana Digital Library Style Manual for Scanning and Cataloging Version 5 March 2005 and the Dublin Core Metadata Initiative web site.

 

The updated Tulane University Digital Library Metadata Guidelines draws heavily from the Tulane Digital Library Metadata Guidelines Version 1.0, Metadata Object Description Schema Version 3.7, the Provisional Guidelines for Descriptive Metadata Version 1, the Louisiana Digital Library Core Record Guidelines, the Louisiana Digital Library Metadata Guidelines, Describing Archives: A Content Standard (DACS) Second Edition 22 April 2016, Princeton University Library MODS Documentation, and DPLA’s Metadata Application Profile. Specific thanks to Cara Key, Metadata Librarian at Louisiana State University, author of many of the guidelines reference in this document and contributor to many of the transformation tools for metadata upon which the workflow described depends.

 

This version of the guidelines was drafted by Charlie Tillay.

Types of Metadata

MODS (Metadata Object Description Schema) is an XML schema and guidelines for encoding a resource description. It supports discovery and management of resources, and access to them, as well as exchange and management of encoded descriptions. MADS is an XML schema and guidelines for encoding an authority description. It supports control, normalization, and management of some types of data used in a resource description, as well as exchange and management of encoded descriptions. MODS and MADS can be used independently. To support those who implement them in an integrated manner, the definitions of elements and attributes in MODS and MADS are coordinated whenever appropriate, and there is an ability to reference a MADS record from a MODS record. They were developed by the MODS/MADS Editorial Committee and the Library of Congress. At this time, TUDL uses MODS for the metadata format in Fedora, but MADS is being considered for an authority format.

The shared goals of MODS are:

  1. Support localization and customization needs     
  2. Accommodate widely adopted descriptive practices     
  3. Maintain a relatively small number of elements and attributes to reduce training, application, and implementation costs     
  4. Support the communication of resource and authority descriptions     
  5. Support validation of the encoding     
  6. Allow use of MODS/MADS elements by other standards and in application profiles     
  7. Maintain continuity of structure and content     
  8. Maintain a single way to encode a piece of information
  9. Accommodate indexing of data in the description     
  10. Accommodate presentation of data in the description     
  11. Make element and attribute names as intelligible as possible to a general audience 12. Allow for extensibility to include data from richer element sets     
  12. Accommodate information about the metadata and record itself     
  13. Accommodate conversion to and from other commonly used resource and authority description encodings (such as Dublin Core, MARC, VRA Core)     
  14. Accommodate controlled vocabularies that are commonly used in resource and authority

Additional MODS goals are:

  1. Allow full description of whole-to-part and similar types of relationships
  2. Support encoding a description for any type of resource
  3. Support encoding the relationship of an agent to a resource

Additional MADS goals include:

  1. Support encoding an authority description for a name, title, name/title, topic, temporal entity, genre, geographic entity, hierarchical geographic entity, occupation

Dublin Core is an internationally recognized metadata standard comprised of fifteen elements used to describe a resource. The semantics of these elements have been established through consensus by an international, cross‐disciplinary group of professionals from the library, museum, publishing, computer science and text encoding communities and other related fields of scholarship. The Dublin Core Metadata Initiative (DCMI) Element Set has been approved by ANSI and assigned the number Z39.85.

The Louisiana Digital Library previously used the Dublin Core Metadata Element Set and qualifiers as defined by the Dublin Core Metadata Initiative as its metadata schema and converts its records to DC when appropriate, such as for collaborative projects that only utilize DC.

Characteristic of the Dublin Core element set are:

  • Simplicity of creation and maintenance allowing a non‐specialist to easily and efficiently create descriptive records for digital resources.
  • Commonly understood terminology and semantics that are universally understood and supported.
  • International in scope ensures that the standard will address the multicultural and multilingual nature of networked resources.
  • Extensibility by allowing additional elements to be added that make sense within a specific discipline. Additional elements can be linked to Dublin Core to meet the need for extensibility to aid in additional resource discovery and granularity needed for access.

According to LinkedData.org, linked data "refers to a set of best practices for publishing and connecting structured data on the Web. Key technologies that support Linked Data are URIs (a generic means to identify entities or concepts in the world), HTTP (a simple yet universal mechanism for retrieving resources, or descriptions of resources), and RDF (a generic graph-based data model with which to structure and link data that describes things in the world)."

Islandora currently allows a digital collection manager to create and edit RDF triples for data, however it is not "native linked data." Upcoming versions of Islandora will be truly based on RDF triples and native linked data. As those mappings are chosen, they will be added to this guide. Information about how the mappings from MODS to RDF triples is progressing is available on the Github page for the Islandora Metadata Interest Group, below.

Additional information will be provided below in the interim, including Tim Berners-Lee's TED Talk on Linked Data, OCLC's Linked Data for Libraries, LinkedData.org, and an interactive web primer on linked data and the semantic web.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.