Structured Authoring
DITA, Structured Authoring & XML Publishing Standard
Overview
What is DITA?
DITA (Darwin Information Typing Architecture) is an XML data model for authoring and publishing. DITA is built around the idea of topics: discrete pieces of information that can be combined and recombined into different documents.
Where to get the document
Key Features
- Topic-based authoring: content as individual topics, not monolithic documents
- Content reuse: topics reused across multiple publications
- Conditional publishing: different versions for different audiences
- Maps and bookmaps: define structure and organization
- Specialization: extend DITA to meet specific industry needs
Also written as: Darwin Information Typing Architecture · OASIS DITA · DITA XML · DITA 1.3
Who DITA Applies To
DITA is an open standard maintained by the OASIS consortium and is not tied to any industry or regulator. Nobody will require it of you contractually; it is chosen. That makes the evaluation different from a military or aviation specification. The question is not what compliance demands, but whether topic-based authoring and its reuse model fit the content you actually have.
DITA Versions, and Where 2.0 Stands
IBM donated DITA to OASIS in 2004, and DITA 1.0 became an OASIS Standard in 2005. Versions 1.1 (2007) and 1.2 (2010) followed, then DITA 1.3 in December 2015, published in Base, Technical Content, and All-Inclusive editions and corrected by Errata 01 (2016) and Errata 02 (2018). DITA 1.3 with Errata 02 is the current approved standard.
DITA 2.0 is still in development. The technical committee has released three pre-release betas (June 2021, October 2021, and July 2026) and has not approved it. It removes and renames enough markup that it is not backward compatible, and the DITA Open Toolkit supports it only as a preview.
Topics, Maps, and Reuse
Content is written as typed topics. The technical content package defines concept (what is it), task (how to do it, in general and strict forms), reference (facts for quick lookup), troubleshooting (a condition followed by cause and remedy pairs, added in 1.3), and glossentry (one sense of a glossary term).
Maps organize topics into publications and set the relationships among them; a bookmap is the map specialization for traditional books. Reuse runs through keys (names bound to resources in a map, referenced with @keyref), content references (@conref, and the indirect @conkeyref), relationship tables, and conditional processing, where attributes such as @audience, @product, and @platform are filtered or flagged at publish time by a DITAVAL profile.
Specialization and Publishing
Specialization lets an organization create new element types that remain interchangeable with conforming DITA tools, because every specialized element can be generalized back to its base type. It is how industries and companies adapt DITA without forking it, and it is the Darwin in the name.
The DITA Open Toolkit is the open-source publishing engine for DITA, licensed under Apache 2.0. Its built-in outputs include PDF (through XSL-FO), HTML5, XHTML, Eclipse help, HTML Help, and Markdown. Commercial authoring tools and component content management systems add editing, workflow, and reuse management on top.
DITA Revisions & Designations
Four names turn up together and mean different things:
- DITA 1.3
- The current approved OASIS Standard (December 2015), with Errata 02 (June 2018).
- DITA 2.0
- The next version, in draft: Beta 03 was released in July 2026, and it has not been approved.
- Lightweight DITA
- A simplified form described in an OASIS committee note (2018), not an approved specification.
- DITA-OT
- The DITA Open Toolkit: open-source publishing software, not part of the specification itself.
Where DITA Programs Go Wrong
- Adopting topics without adopting an information model. Unmodelled topics are just short documents, and reuse never materializes.
- Specializing early. Specialization is powerful and raises the cost of every future toolchain upgrade; most teams need it later than they think and some never do.
- Reusing by copy because keys and content references were never set up, which reintroduces exactly the divergence DITA exists to prevent.
- Choosing DITA for an aerospace program that will later be contractually required to deliver S1000D, and discovering the conversion cost at that point.
How It Relates to Other Standards
- DocBook
- The other established open documentation vocabulary. DocBook is book-shaped and rich out of the box; DITA is topic-shaped and built for reuse and specialization.
- S1000D
- The comparison that matters in aerospace and defense. DITA is lighter to adopt but has no equivalent of the CSDB or of S1000D's applicability machinery.
- W3C / XML
- The underlying layer. DITA is an XML vocabulary, so the transformation and validation tooling is the W3C stack.
DITA Questions
What does DITA stand for?
DITA stands for Darwin Information Typing Architecture. The full form is rarely used in practice and the acronym has effectively become the name, but it is worth unpacking: Darwin refers to the specialization mechanism, which lets element types evolve from existing ones, and information typing refers to the idea that content has distinct types such as task, concept and reference rather than being undifferentiated prose.
What is DITA?
An XML architecture for authoring content as topics (discrete, self-contained pieces of information), which are then assembled into publications using maps. Its distinguishing features are reuse across publications, conditional processing for different audiences and outputs, and specialization, which lets an organization extend the vocabulary without leaving the standard.
What is the current version of DITA?
DITA 1.3, approved as an OASIS Standard in December 2015, with Errata 02 (June 2018). DITA 2.0 is in draft: its third beta was released in July 2026, and it has not been approved.
What are the DITA topic types?
The technical content package defines concept, task, reference, troubleshooting (added in DITA 1.3), and glossentry. Concept answers what something is, task how to do something, and reference gives facts for quick lookup; troubleshooting pairs a condition with causes and remedies, and glossentry defines a single sense of a term.
What is DITA-OT?
The DITA Open Toolkit: the open-source publishing engine for DITA content, licensed under Apache 2.0. It transforms DITA into outputs including PDF (via XSL-FO), HTML5, and Markdown, and it supports DITA 2.0 only as a preview.
DITA or S1000D?
If a contract specifies one, that is your answer. Where the choice is genuinely open, it usually turns on whether you need what S1000D adds (a managed source database, lifecycle handling, and applicability across configurations) or whether that machinery is overhead for content that does not vary by tail number. DITA is materially quicker to adopt; S1000D carries more of the hard problems for you.
Is DITA suitable for regulated content?
Yes, and it is used for it widely. What DITA does not give you is the aerospace and defense lifecycle model, so where a program needs configuration-driven applicability and a controlled source database, those have to be built around it rather than assumed.
Where to Get DITA
DITA is an open OASIS standard: the DITA 1.3 specification is free to read and use, and the open-source DITA Open Toolkit (Apache 2.0) publishes from it.
Where DITA Applies
All standardsTechnical publications servicesTalk to us about DITA
Get In Touch
Let's Talk
Planning, building, or migrating DITA-compliant technical content? We can help.