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

Glossary · 1 · Documentation models

Structured authoring

Also known as: Structured writing, Structured content authoring, Structured documentation

German: strukturiertes Schreiben

In technical communication, structured authoring is a way of writing in which content is created as discrete information units that follow a formally defined content model — a schema, document type definition, or specification — rather than as free-form pages shaped by a word processor. Meaning is marked up explicitly (a warning is tagged as a warning, a step as a step), while typography, pagination, and layout are applied later by a publishing process. Because the markup is validated against the model, the same source can be checked automatically, reused across deliverables, and rendered to print, HTML, or machine-readable formats. Structured authoring is one of the six documentation models described in the encyclopedia subject area Technical documentation models and is usually combined with the others rather than used alone.

  • structured authoring
  • content model
  • XML
  • single-sourcing
  • documentation models

In one sentence

Structured authoring: writing to a validated content model in which meaning is marked up and layout is applied at publishing time.

Example

A machine builder authors maintenance procedures as validated XML data modules, so the same source produces the printed manual, the HTML help, and the content delivered to a service portal.

Structured authoring grew out of the generalized markup idea: describe what a piece of content is, not how it should look, and let separate processing decide the appearance. That idea was standardized as SGML in ISO 8879:1986, which remains published and was last reviewed and confirmed in 2020; XML and HTML both derive from it. In documentation practice, structured authoring today is implemented with XML-based content models such as DITA, DocBook, and the aerospace and defense specification S1000D, and increasingly also with lighter-weight formats that are constrained by templates and validation rules.

The defining feature is not the file format but the constraint: authors write into an agreed schema, and the result is checked by schema validation before it is published. Writing to a content model does not by itself make documentation correct, complete, or compliant with any regulation — it makes the structure of the content predictable and testable.

How it applies

  • Content model first. Before writing, the team agrees on an information model: which information types exist, which elements are allowed where, which metadata is mandatory. This is the artifact that governs authoring, not the style guide alone.
  • Semantics instead of formatting. Authors tag safety notices, prerequisites, steps, parameters, and identifiers as such. Downstream, a transformation decides whether a warning appears as a boxed panel in the PDF or as a callout in the help system.
  • Validation as a gate. Schema validation catches missing mandatory elements and wrong nesting; rule-based checking (for example with Schematron) can enforce editorial policies that a schema cannot express, such as "every procedure needs a tools list."
  • Reuse and variants. Stable units plus references and keys allow one source to serve several machine variants, options packages, or markets; this is the basis for single-sourcing and multichannel publishing.
  • Machinery and software documentation. For machinery, structured sources make it practical to assemble operating and maintenance information per configuration and to feed documentation into standardized delivery formats; for software, the same approach supports API reference generated from source plus hand-written conceptual content in one validated model.
  • Tooling and roles. Structured authoring usually implies a structured editor, a repository or component content management system, and a publishing pipeline. It also shifts work: information architects maintain the model, authors work within it, and release engineers own the transformations.
  • Cost of entry. Modeling, migration of legacy documents, and author training are the real effort. Projects commonly start with a narrow model and widen it, rather than adopting a full industry specification on day one.

Structured authoring vs. topic-based authoring

The two are often used interchangeably but answer different questions. Topic-based authoring is about granularity and independence: content is written as self-contained topics that make sense on their own. Structured authoring is about constraint and validation: content follows a formally defined model, whatever its size. A topic can be written in an unstructured word processor, and a structured source can contain a long, book-like hierarchy. In most real implementations the models are layered — DITA, for instance, supplies a structured model whose units are topics — which is why the encyclopedia treats them as separate models that combine.

External references