Glossary Updates12 new terms added to the glossaries · October 2, 2026, 22:44 CEST
AI TechDocKnowledge

Wiki · iiRDS core concepts · Information units

iiRDS Topic: the self-contained unit of topic-based documentation

The Topic is the iiRDS information unit for content that covers a single subject and can be understood without further context: one task, one concept, one reference table. It is the natural target when topic-based content from DITA or a component content management system is delivered to portals, service apps or retrieval systems. This guide explains how iiRDS defines a Topic, how it differs from a Document and a Fragment, how the kind of topic is expressed with a topic type instead of a subclass, and which metadata makes a topic findable in context. It also shows a Turtle example with a rendition, a topic type, a lifecycle phase and a component, and explains how topics are arranged into documents through directory nodes rather than direct links.

Valid for iiRDS 1.3Last reviewed Level: Introductory

  • Technical documentation
  • Machinery
  • MedTech
  • AI retrieval

Definition

In iiRDS (intelligent information Request and Delivery Standard), a Topic is an information unit that deals with a single subject and can be understood on its own — for example one task, one concept or one reference table.

The specification defines a topic as an information unit that covers a single subject and describes it as self-contained: the reader understands it without any additional information. Examples in the specification are a task about turning on a mobile phone and a list of all user-interface elements of a program. That test separates the Topic from the Fragment, which needs surrounding context.

iiRDS does not define subclasses such as task topic or concept topic. The kind of a topic is metadata: iirds:has-topic-type points to a standard topic type such as iirds:GenericTask, iirds:GenericConcept or iirds:GenericReference. Safety content is marked with an information subject, not a topic type.

Like every information unit, a topic has an IRI, carries its metadata directly and delivers its content through one or more renditions.

In one sentence

An iiRDS Topic is a subclass of InformationUnit for one self-contained subject, typed as iirds:Topic and described by metadata so it can be delivered on its own.

Example

A machine builder publishes the service content for a palletizer from its component content management system. The procedure “Replacing the air filter” becomes one iiRDS Topic with an HTML rendition for the service portal and a PDF rendition for print. It is typed with iirds:GenericTask, related to the lifecycle phase iirds:Maintenance and to the component “air filter”. When a technician scans the machine and selects the filter, the portal shows exactly this topic. The warning at the start of the procedure stays inside the topic, so it is never shown without the steps it belongs to.

How it applies

Technical documentation

The Topic is the delivery counterpart of a module in topic-based authoring. Each topic is written, reviewed and translated once and can then appear in several deliverables. In iiRDS, it carries its own metadata, so a portal can find it by product, phase or topic type without knowing which manual it came from.

Migration

Topic-based sources such as DITA map to iiRDS Topics directly after publishing. Legacy PDF manuals can be made granular without rewriting: a chapter addressed with a RangeSelector can be typed as its own Topic, as long as it is self-contained. Content that needs context is better modeled as a Fragment.

MedTech

Procedures for cleaning, setup or troubleshooting of a device can be delivered as individual topics to the point of use. Language versions and revisions of the same topic link to one information object, so a portal can show the current revision in the reader’s language. Delivery granularity does not change which content must be approved.

AI and retrieval

A self-contained topic is a strong retrieval chunk: it can be answered from correctly without its surroundings. Topic type and lifecycle phase work as filters, for example task content for “how do I” questions. The topic IRI gives an assistant a stable source to cite.

Topic vs. Fragment vs. Document

Unit Scope Understood alone Typical example
Topic One subject Yes Task “Replacing the air filter”
Fragment Part of a subject No, needs context A single hazard statement or table row
Document Many subjects for one purpose Yes, read as a whole Operating instructions

When in doubt, ask whether a reader could act correctly on the unit without seeing anything else. If yes, it is a Topic.

In RDF

A task topic with metadata and two renditions. Shown in Turtle for readability — inside a package, metadata.rdf is written in RDF/XML.

@prefix iirds: <http://iirds.tekom.de/iirds#> .

<https://example.com/iirds/components/air-filter> a iirds:Component .

<https://example.com/iirds/replace-air-filter> a iirds:Topic ;
    iirds:title "Replacing the air filter"@en ;
    iirds:language "en" ;
    iirds:has-topic-type iirds:GenericTask ;
    iirds:relates-to-product-lifecycle-phase iirds:Maintenance ;
    iirds:relates-to-component <https://example.com/iirds/components/air-filter> ;
    iirds:is-version-of <https://example.com/iirds/objects/replace-air-filter> ;
    iirds:has-rendition [
        a iirds:Rendition ;
        iirds:format "text/html" ;
        iirds:source "content/replace-air-filter.html"
    ] , [
        a iirds:Rendition ;
        iirds:format "application/pdf" ;
        iirds:source "content/replace-air-filter.pdf"
    ] .

In topic-based authoring and modular architectures

In topic-based authoring, a DITA topic or a CCMS module is the unit of writing and reuse; the iiRDS Topic is the same unit at delivery time. DITA information types map onto iiRDS topic types: a DITA task becomes a topic with iirds:GenericTask, a concept one with iirds:GenericConcept. The map that orders topics is not carried inside the topics. iiRDS models order separately with directory nodes, so one topic can appear in several documents without duplication. In modular delivery architectures, this separation lets a portal assemble topics dynamically by metadata, while the directory structure preserves the author’s intended reading order where it matters.

Concept cluster

How this concept connects to the other iiRDS core concepts.

Also searched as

  • iiRDS Topic
  • iirds:Topic
  • iiRDS topic type
  • topic in iiRDS
  • iiRDS topic-based content
  • iiRDS Topic vs Fragment
  • topic
  • topic module

Questions this article answers:

  • what is a topic in iiRDS
  • difference between iiRDS topic and fragment
  • how to map DITA topics to iiRDS
  • iiRDS topic type task concept reference

Explore further

Frequently asked questions

What is a topic in iiRDS?
A Topic is an information unit that covers a single subject and can be understood on its own, such as one task or one concept. It is typed as iirds:Topic and carries its own metadata.
What is the difference between an iiRDS Topic and a Fragment?
A Topic is self-contained. A Fragment, such as a single warning or one step, is only understood together with additional context.
How do I say that a topic is a task or a concept?
Use the relation iirds:has-topic-type with a standard topic type such as iirds:GenericTask or iirds:GenericConcept. iiRDS does not define subclasses of Topic for this.
Is safety a topic type in iiRDS?
No. Safety content is marked with an information subject such as iirds:SafetyInstruction, so a topic can be a task and a safety instruction at the same time.
Does a DITA topic automatically become an iiRDS Topic?
Not automatically. A DITA topic becomes an iiRDS Topic when it is published, typically to HTML, and described in the package metadata with an IRI, a rendition and metadata.
How are topics linked to the document they belong to?
Through directory nodes. A table of contents is a linked list of iirds:DirectoryNode instances that point to the document and to its topics; topics do not reference each other directly.

Conclusion

The iiRDS Topic turns the authoring principle of one subject per module into a delivery unit that systems can filter and assemble. Keep topics genuinely self-contained, express their kind with a topic type and their subject with information subjects, and leave the order of topics to directory structures.

By knowledge.aitechdoc.world · Published September 25, 2026 · Last reviewed · Next review due

The wiki articles are written to stay valid: each states the version of the standard it was checked against and the date of its last review, and is reviewed again at least once a year or when a new version of the standard appears. Definitions follow the cited specification; examples and the sections on how a concept applies are editorial commentary by AI TechDoc Blog.