Metadata Basics

What are metadata?

The Oxford English Dictionary defines metadata as: “Data that describes and gives information about other data.” We create and use metadata every day through software, social media, streaming services and more – metadata are key ingredients in content systems because they help information seekers find resources through the exchange of data. Metadata also aid in information organization and identification, and provide important context for resources that are not clearly defined.

Metadata come in many forms. Webpages use metadata to link from one page to another; webpages also use metadata to make their content visible to search engines, this can be done locally by searching within a website, or on a larger scale, through a search engine like Google. For example, when you type keywords into Google, it searches for those word within websites’ metadata and pulls a list of results. Metadata can also be used by businesses to determine details about the products they carry both in-store and online, and they also maintain metadata about transactions to analyze sales and personalize customer experiences. For example, online shopping websites often offer product recommendations based on previous purchases – they use metadata to find similar items within their site. In these examples metadata can include: subject, number of visits or likes, cost, material, author, organization and many more.

Metadata are often divided into a few categories, but there are no set standards for developing and using metadata, so the use of these categories can often vary.

Types of metadata:

• Descriptive metadata: used for finding and understanding resources
  • Examples: Title, author, subject, genre
• Administrative metadata
  • Technical metadata: used to define technical details about a resource
    • Examples: File type, file size, creation date/time
  • Preservation metadata: used to maintain a resource for long-term use
    • Examples: Preservation event
  • Rights metadata: used to define the intellectual property rights and use information
    • Examples: Copyright status, license terms
• Structural metadata: used to describe how parts of a resource fit together
  • Examples: Sequence, place in hierarchy
• Markup languages: used to integrate metadata with the content’s other structural features
  • Examples: Paragraph, heading, list etc.

Creating Metadata

To create metadata, there are a few necessary tools to make it readable by both humans and machines:

• Schema – a list of defined elements or data points that can be used to gather information about a resource. These elements can include a title, a date, an author or creator, or a more general identifier.
• Standards – inform us how to populate each of the elements within the schema. There are three types of standards:

1. Content Standards – define each element of the schema and explain where each piece or type of information should go
2. Data Value Standards – lists of standardized subject terms, such as controlled vocabularies, discipline-specific vocabularies, or encoding and formatting standards
3. Data Structure Standards – explain how to encode and structure metadata to ensure it is machine readable

• Best Practices – ensure the selection and use of a specific metadata schema by determining which elements will make up a metadata profile for a specific project or collection. Best Practices should include a decisive selection of content, data value, and data standards structure.

High quality metadata should be accurate, complete, consistent, and offer interoperability.

Based on University of California Santa Cruz Univeriy Library’s Metadata Creation guide and Jenn Riley’s “Metadata: What is Metadata and What is it for?"