Metadata - Types and Conceptual Models

Types of Metadata What Is Metadata? Metadata is information that describes other data. It helps us understand, organize, find, and use resources—whether those resources are digital documents, images, databases, or even physical objects like library materials. Think of metadata as the "label" or "catalog card" that tells you what something is, where it is, and how to use it. The image above shows a traditional card catalog—a classic example of metadata in action. Each drawer contains cards (metadata) that describe the books in a library. The cards contain information like the book's title, author, and subject—information that helps you discover and identify the resource. Understanding the Major Types of Metadata There are many ways to classify metadata depending on your needs. Rather than thinking of these as mutually exclusive categories, understand that a single resource often has multiple types of metadata attached to it. Here are the main types you need to know: Descriptive Metadata Descriptive metadata supplies the information needed to discover and identify a resource. It answers questions like "What is this?" and "How do I find it?" Common examples include: Title of the work Author or creator Abstract or summary Keywords or subject terms Publication date This is the type of metadata you interact with most when searching libraries, databases, or the internet. When you search for a book by title or author, you're using descriptive metadata. Structural Metadata Structural metadata describes how different parts of a resource relate to and are organized within each other. It tells you how the pieces fit together. For example: How individual pages are grouped into chapters in a book How tracks are organized into albums in a music collection How images are sequenced in a photograph collection How chapters relate to sections in a document Structural metadata is essential for complex resources where the organization of components matters. Without structural metadata, a digital book would just be a collection of loose pages with no clear order. Administrative Metadata Administrative metadata contains practical information needed to manage and maintain a resource over time. It helps answer "How do I control and preserve this?" Key elements include: Resource type (is it a book, image, video, dataset?) File format and size (technical details about storage) Creation and modification dates Access permissions and restrictions (who can view or use this?) Rights and copyright information Version information This type of metadata is crucial for libraries, archives, and organizations that need to track their digital collections and control who can access them. Technical Metadata Technical metadata (also called internal metadata) describes the technical details and structure of data elements. It's the "behind-the-scenes" information that systems need to process and display resources correctly. Examples include: File type (.pdf, .jpg, .mp3, etc.) Data formats and encoding standards Image resolution and color depth Audio specifications (bitrate, sample rate) Software and hardware requirements Technical metadata ensures that a resource can be properly opened, displayed, and used by different systems and applications. Business Metadata Business metadata translates technical details into information that business users can understand and act on. It bridges the gap between technical systems and business operations. This includes: Data definitions (what does this data element represent in business terms?) Business requirements and rules Process flows and workflows Terminology and business glossary Key metrics and performance indicators Timelines and deadlines For example, while technical metadata might describe a data field as "VARCHAR(50)", business metadata explains that this field contains "Customer Name" and defines the business rules for what makes a valid customer name. Process Metadata Process metadata documents how data is created, collected, processed, transformed, and used throughout business operations. It answers the question "How was this data produced?" This includes: Which processes created or modified the data When transformations occurred What calculations or rules were applied Which systems or people handled the data What quality checks were performed Process metadata is critical for understanding data lineage—the complete history of where data came from and what happened to it along the way. <extrainfo> Reference Metadata Reference metadata provides information about the contents and quality of statistical data, such as accuracy measures, data sources, and confidence levels. This is less commonly discussed than other types. Statistical Metadata Statistical metadata (also called process data) describes the processes and methods used to collect, process, or produce statistical data. It's specialized for statistical datasets and quality assessment. Legal Metadata Legal metadata provides information about copyright, ownership, licensing, and legal rights associated with a resource. It includes: Creator and copyright holder information Public licensing terms (like Creative Commons licenses) Legal restrictions on use Accessibility Metadata Accessibility metadata describes how a resource can be adapted to meet the needs of different users, supporting universal access. Examples include: Whether the content is suitable for users with dyslexia or other learning differences Available formats (braille, large print, audio versions) Captions and transcripts for video content Reading level and complexity information </extrainfo> Metadata Classification Systems While all these types of metadata exist, experts have developed frameworks to organize them into logical groups. Two major classification systems are particularly important: The NISO Framework: Three Core Types The National Information Standards Organization (NISO) identifies three foundational types of metadata that apply to most resources: Descriptive Metadata — enables discovery and identification (title, author, keywords) Structural Metadata — describes how resource components are organized Administrative Metadata — supports management and preservation This is a simplified framework that groups all metadata into these three categories. Think of this as the most fundamental way to think about metadata. The Kimball Framework: Three Functional Categories Data warehouse expert Ralph Kimball proposes a different classification based on the function metadata serves: Technical Metadata — describes the technical implementation and structure (file formats, database schemas, system specifications) Business Metadata — translates technical details into business language and context (definitions, rules, metrics) Process Metadata — documents how data flows through systems and transformations (data lineage, processes, workflows) The key difference: Kimball's framework focuses on how metadata is used rather than what type of information it provides. It's particularly useful for understanding data in business and analytical contexts. How Do These Systems Relate? These two classification systems answer different questions: NISO answers: "What are the main categories of metadata?" (Focuses on nature/purpose) Kimball answers: "What functions does metadata serve?" (Focuses on use/audience) Both frameworks are valid and useful. For example, technical metadata fits into NISO's "administrative" category, but serves Kimball's "technical" function. Business metadata might be part of NISO's "descriptive" or "administrative" metadata, but serves Kimball's "business" function. Key Takeaways When studying metadata, remember: Metadata describes other data and comes in many types depending on what you're trying to accomplish A single resource typically has multiple types of metadata working together (descriptive for discovery, structural for organization, administrative for management) The NISO three-type system (descriptive, structural, administrative) is the most widely recognized framework for classification The Kimball three-type system (technical, business, process) emphasizes the functional role metadata plays in organizations Different types serve different audiences — some metadata is for humans finding resources, some for systems processing data, and some for business users understanding context