Key Aspects of Metadata Modelling for Organising and Managing Content

Understanding metadata is crucial for those who plan and manage content. Since we are always making, sharing, and using information, it's important that we have ways to sort and control this huge amount of data. The workshop on Metadata and Taxonomy with Madi Weland Solomon was designed to provide an in-depth look at the key facets of metadata modeling.

Metadata: Beyond Just Data #

But, what exactly is metadata? At its core, it's "data about data." Whether it's describing an asset's form and content or serving as a narrative, metadata can take on various roles. For instance:

• Descriptor: Highlighting specific attributes of the data.
• Perspective: Offering a unique viewpoint on the information.
• Narrative: Telling a story about the data.
• Frame: Setting boundaries or context for the data.

Descriptor: This metadata identifies key data attributes, such as a digital photo's date, time, camera model, resolution, and location (if GPS data is available), aiding in organization and identification in databases or catalogs.

Perspective: This metadata provides a unique viewpoint or interpretation, such as the genre or style of a music track (e.g., jazz, classical, rock), helping users filter and understand content through these musical categories.

Narrative: Narrative metadata offers context and background, like a historical document's significance, related events, and impact, enriching understanding beyond textual content.

Frame: Frame metadata defines the scope or limits of data, such as a research paper's field (e.g., molecular biology, astrophysics), methodology (experimental, observational, theoretical), and geographical scope, contextualizing research within scientific inquiry.

Classifying or grouping metadata can be likened to the task of sorting M&Ms by attributes such as color, size, or taste, like we did it in the workshop. This simple analogy serves as a foundation for understanding how, in more complex scenarios, these groupings contribute to forming detailed data models encompassing various elements like assets, user profiles, and industry-specific information. In the business context, adept use of metadata can streamline several processes, including automated content identification, efficient workflow management, and adherence to compliance regulations.

Metadata is multifaceted. Its types include:

• Descriptive Metadata: Helps in resource discovery and identification.
• Structural Metadata: Indicates the composition of compound objects, considering aspects like file format and size.
• Administrative Metadata: Assists in asset management, including rights and preservation metadata.

Descriptive Metadata: Used for discovery and identification, it includes titles, authors, abstracts, keywords, and subjects. Example: In a library database, a book's descriptive metadata covers its title, author, publication date, genre, and summary.

Structural Metadata: Describes how parts of a compound object are organized, like the arrangement of pages and chapters in a book. Example: In a digital magazine, it details the organization of articles, images, and ads.

Administrative Metadata: Helps manage resources, detailing creation time, file type, and access permissions. It has two sub-types:

• Rights Management Metadata: Relates to intellectual property and usage rights, like copyright and usage restrictions for an online photograph.
• Preservation Metadata: Aids in long-term digital resource management, recording file format, access tools, preservation history, and integrity checks.

Additionally, metadata can be categorized as Asset, Subject, Use, and Relational metadata, focusing on various attributes from ownership to relationships and rights.

The Evolution of Data Representation #

The digital landscape has undergone a significant transformation, moving from isolated documents to a sophisticated network of interconnected data. Central to this shift have been organizations like the World Wide Web Consortium (W3C) and initiatives like schema.org, which have played pivotal roles in guiding this evolution. Tools such as "Trooly" highlight the capabilities of modern metadata, demonstrating its power by rapidly scanning billions of pages to evaluate their trustworthiness using a range of criteria. This evolution showcases the increasing complexity and connectivity of our digital world.

The Fluidity of Language #

As language continually evolves, adding roughly 800 new words to the English language annually, it underscores the need for a dynamic and adaptable metadata governance system. Such a system is essential to keep pace with the changing linguistic landscape, incorporating new terminologies, slang, and synonyms. This adaptability ensures that metadata remains relevant and effective in reflecting current language use and trends.

Integration and the Role of AI #

The aim of integration and interoperability is to ensure the seamless interaction of different systems. Semantic interoperability, which ensures that the information exchanged is understood in a meaningful way, stands out here.

However, the real turning point is the emergence of machine learning and AI. While data models guide the machines, humans play a central role in providing context. This interaction paves the way for machine understanding, which is fuelled by conceptual relationships and neural networks. Neuroscience further underpins this by emphasising that mental models or 'frames' help to rationalise human thinking. Unlike machines, humans have a natural aptitude for framing as they rely on experience, simulation and constraints.

The Imperative of Interoperability #

Metadata plays a critical role in the digital environment, acting as a guide to help users locate the information they need. By assigning relevant metadata to digital objects, it creates a network that links related content seamlessly. This connectivity enhances the user experience, improves research efficiency, and sharpens the accuracy of content recommendations. The cornerstone of this seamless interaction is shared metadata, promoting interoperability across a range of platforms.

Unraveling Interoperability Barriers

But what if it's hard to find what you're looking for online? Even when you find it, why are there sometimes problems using it? This happens often because the metadata isn't consistent, the way data is organized doesn't match up, or sometimes, the right metadata isn't there at all.

Metadata Capital: Amplifying ROI through Interoperability

Content, like articles or videos, costs money to make. But its real worth comes from how often it's found and used. Every time content is reused, it lowers the cost of making new stuff and increases its value. It's a simple idea: the easier it is to find and use content (thanks to good metadata), the more it can be reused, and the more valuable it becomes.

The Power of Semantic Integrations #

When we delve into semantic metadata, structured around universally recognized standards, we unlock a world of contextual depth. Such metadata can interlink with a variety of datasets, ranging from the huge repositories of Wikipedia and Geonames to specialized government statistics and industry standards.

How Metadata Works in Real Life

Think about the story of a painting called "Allegory of Fortune." Someone found this painting at a flea market and bought it cheaply. Its real worth was hidden until specialists at Christie's, an auction house, realized it was important. This painting, made by an artist named Dosso Dossi, shows us that sometimes, luck plays a big part in finding valuable things.

Making active metadata usable

Take for instance, Whisk/Samsung Food - where metadata translates to tangible components like ingredients, chemical compounds, nutritional content, and even factors like perishability and product availability. This granularity of data streamlines processes and decision-making.

Deciphering FEEDS

The way we use Feeds has changed a lot. It's not just for updating content, but also for keeping people interested over time. Big companies like Netflix use 'Scale Feed' to share their huge range of content with people all around the world. Then there's 'Social Feed', which tries to keep users coming back by focusing on content that's about people interacting and joining in. And 'Identity Feed' is about giving people a special selection of content that helps create a community feeling and makes them more loyal to a brand.

Conclusion #

Metadata plays an important role in the organisation and enhancement of digital content and forms the backbone of content management and usability. It comes in different forms - descriptive, structural and administrative - and is central to connecting disparate systems and managing rich digital content. The integration of metadata with AI and machine learning is transforming data processing and interpretation, making it an essential element of our digital interactions. As the digital landscape and language evolve, metadata systems must adapt to remain relevant and effective. More than just a technical tool, metadata adds depth and context to digital information and is proving indispensable when it comes to uncovering hidden art assets or refining global content feeds. Understanding and using metadata effectively is key to navigating and maximising the potential of the digital world.

Where to go from here #

https://www.contentstrategy.at/lecture-reports/how-to-create-a-metadata-model-a-guide?x-craft-preview=zz81MQpddQ&token=Y7VbTFl4miSfj8Od6oORLvapKzGrPkJn

 

References #

Baca, M., & Getty Research Institute (Hrsg.). (2016). Introduction to metadata (Third edition). The Getty Research Institute.

Diamond, D. (2016). Metadata for content management: Designing taxonomy, metadata, policy and workflow to make digital content systems better for users (CreateSpace edition (v1.1)). CreateSpace.