Information Storage and Retrieval: Metadata and Descriptive Records

Lesson overview

Teach metadata as structured description that makes resources findable, manageable and trustworthy, using examples such as Dublin Core, provenance and controlled vocabularies.

CourseInformation Storage and Retrieval

Topic strandMetadata and semantics

Assessment styleWorked scenario, applied task and digital exam practice

EvidenceDefinitions, representation, method, result and interpretation

Starter

Write down the user need, the data being stored or searched, and the decision the system has to support. Then predict which representation, index, query method or governance control will matter most in this lesson.

Learning objectives

Define metadata and distinguish descriptive, administrative, structural and technical metadata.
Use Dublin Core-style fields to describe a resource.
Explain how controlled vocabulary and provenance improve retrieval quality.

Learning outcomes

Students can define metadata and distinguish descriptive, administrative, structural and technical metadata.
Students can use Dublin Core-style fields to describe a resource.
Students can explain how controlled vocabulary and provenance improve retrieval quality.

Key vocabulary, acronyms and terminology

metadata: Data that describes, manages or gives context to another resource.
Dublin Core: A widely used set of descriptive metadata elements such as title, creator and subject.
controlled vocabulary: An agreed list of terms used consistently for description.
provenance: Information about the origin, custody or history of a resource.
structural metadata: Metadata describing how parts of a resource relate to one another.
authority control: Consistent naming of people, organisations or subjects.

Detailed teaching notes

Core concept

Metadata is a retrieval tool, not clerical decoration. It gives resources stable descriptions that can be searched, filtered, cited, governed and linked across systems.

Representation choice

A descriptive record might include title, creator, subject, date, format, identifier and rights. Administrative metadata covers ownership and permissions; technical metadata covers file type, size or checksum; structural metadata connects chapters, pages or media segments.

Method and reasoning

Choose metadata fields by asking how the resource must be found, interpreted and managed. Controlled vocabularies reduce synonym noise, while provenance helps users judge trust and context.

Risk and limitation

Poor metadata can make a high-quality collection effectively invisible. Inconsistent names, missing dates and uncontrolled subjects create fragmented search results and weak analytics.

Degree-level deep dive

Metadata creates retrievability

A resource without metadata may exist in storage but remain practically invisible. Descriptive, administrative, technical and structural metadata give search systems the fields needed for discovery, filtering, trust and management.

Controlled vocabulary is a quality tool

Controlled terms reduce fragmentation such as 'AI', 'artificial intelligence' and 'machine intelligence' being treated as unrelated labels. Authority control performs the same role for people, organisations and places.

What excellent work shows

A strong answer chooses metadata fields from user needs, not from habit. It should include title, creator and subject where useful, but also provenance, rights, format, identifiers and controlled values when those affect retrieval or governance.

Concrete example to study

Dublin Core-style image record

title: Railway workers outside depot
creator: Unknown photographer
date: 1952
subject: rail transport; labour history
format: image/jpeg
rights: educational reuse permitted
provenance: donated by local history society

Reasoning

Title and subject support keyword search.
Format and rights support filtering and governance.
Provenance helps users judge trust, origin and interpretation of the image.

Reveal takeaway

Good metadata turns an otherwise anonymous file into a retrievable, interpretable resource because it supplies searchable fields, governance evidence and context for judging meaning.

Worked example

Scenario

A scanned historical map needs a searchable record.

Worked solution

Create descriptive metadata: title, creator, date, subject and geographic coverage.
Add technical metadata: image format, resolution and checksum.
Add administrative metadata: rights holder and permitted reuse.
Use a controlled vocabulary for place and topic terms so browsing and filtering are consistent.

Reveal model result

The metadata record turns a file into a retrievable resource. Without it, users may need to inspect every image manually.

\text{Define the data, choose the representation, apply the method, interpret the result.}

Applied retrieval task

Design metadata for a small image collection.

Deliverables

Create a Dublin Core-style record for one image.
Choose two controlled vocabulary fields.
Add one provenance field and explain why it matters.

Success checks

The record distinguishes descriptive from administrative metadata.
Controlled values are realistic and reusable.

Common misconception

Students often treat metadata as optional because it is not the primary content. In retrieval systems, metadata is often the most reliable bridge between user language and stored resources.

Quick checks

1. In this lesson, why does metadata matter?

It is only a presentational label and has no effect on retrieval quality.It replaces the need to understand user tasks or data representation.Data that describes, manages or gives context to another resource.

2. Which answer best shows degree-level understanding of metadata and descriptive records?

Define Dublin Core, apply it to the scenario, then explain the trade-off or risk.Name as many technologies as possible without explaining the scenario.Give a final recommendation without showing how the data or query was represented.

Digital exam practice

Example exam task

A publishing archive has thousands of scanned covers, but users can only search by filename and results are poor. In your answer, define the relevant objects or data structures, use course-specific vocabulary, show the method rather than only the result, and finish with a decision about the storage or retrieval system.

Notation and technical toolkit

dc:titleDublin Core title field Use to identify a descriptive metadata element.

subject = controlled termnormalised subject description Use to avoid uncontrolled variants.

provenancesource and history of a record Use when trust and auditability matter.

What a good answer is expected to show

A strong answer for Metadata and Descriptive Records the answer should propose metadata fields, explain Dublin Core-style description, include provenance and show how controlled vocabulary improves retrieval. It should connect the formal or technical representation to the user's information need instead of listing terms without using them.

How to solve it

Identify the retrieval failures caused by filename-only search.
Add descriptive, technical and administrative metadata fields.
Use controlled vocabulary for subject or genre.
Explain how provenance supports trust and interpretation.

Model answer

Reveal model answer

Filename-only search is weak because filenames rarely contain reliable title, creator, subject, date or rights information. I would create metadata records using Dublin Core-style fields such as title, creator, date, subject, format, identifier and rights. Controlled vocabulary terms for genre and topic would reduce inconsistent labelling, while provenance would record where each cover came from and how it was digitised. These fields support search, faceted browsing and governance.

Practise next

Write a metadata record for a podcast using at least six Dublin Core-style fields.
Explain why controlled vocabulary can improve recall and precision at the same time.

Self-marking criteria

Defines the scenario objects, data forms or system components before solving.
Uses the lesson vocabulary accurately and in context.
Shows a clear method with enough working for a marker to follow.
Connects the result back to retrieval, storage, analytics or governance.
States a limitation, trade-off or quality risk rather than presenting the answer as absolute.

Extension

Find a public archive record online and identify descriptive, technical, administrative and structural metadata.

Study route

Save a short worked answer from this lesson using this pattern: define the need or data, choose the representation, show the method, state the result and interpret the implication for the system.

Next lesson: Relevance, Ranking and Retrieval Models.