Darwin Information Typing Architecture (DITA)

By Carl Haupt and Michael Nelson

The Darwin Information Typing Architecture (DITA) represents as major a breakthrough in online technical communication technologies as previously accomplished by the eXtensible Markup Language (XML). XML succeeded in separating a documentís content from its logical structure. The DITA goes a step further and separates a documentís logical structure from its taxonomy. This represents a quantum leap forward in the ability to construct online compound documents on-the-fly based on topical classification and cataloging.

Prior to the DITA, organizations were forced to utilize standardized library cataloging and classification methodologies such as MAchine Readable Cataloging (MARC), Anglo-American Cataloging Rules, Second Edition (AACR2), Dublin Core, and the Library of Congress Classification system (LCC). With the introduction of the DITA, organizations are freed from generic classification systems and can now create custom taxonomies which are coded into DITA-compliant XML-compatible Document Type Definitions (DTDs) or schemas. The DITA is an extensible framework for creating and managing custom taxonomies.

The end result of the ability to create custom taxonomies and apply them in a single-source environment is that document elements stored in an XML data repository can be searched, selected, revised, merged, tracked, and published, based on a topical classification system (taxonomy) unique to an organization. No longer are XML-based compound document systems restricted by the hierarchical tree structure.

Ann Rockley, a prominent consultant in the field of single sourcing, says, "DITA is not really a directly implementable model; instead, it defines how to construct your model." (Managing Enterprise Content: A Unified Content Strategy). "Organizations construct these models by coding taxonomies into DITA-compliant DTDs. These DTDs are then imported into an XML data repository, where reusable document elements are managed and stored." (2003, p.179)

The DITA also incorporates the concept of information typing:

Information typing is the practice of identifying types of topics that contain distinct kinds of information, such as concepts, tasks, and reference information. Topics that answer different kinds of questions can be categorized as different information types. The base topic types provided by the DITA ( a generic topic, plus concept, task, and reference ) provide a usable starter set that can be adopted for immediate authoring.
(Hackos & Priesley, 2005, p.8 )

Finally, the DITA Language Specification divides document elements into 18 distinct types, each with comprehensive descriptive abilities:

These element types include Topic, Concept, Reference, Task, Body, Table, Typographic, Programming, Software, User Interface, Utilities, Prolog, Related Links, Specialization, Map, Map Group, Commonly Referenced Descriptions, and Miscellaneous.
(Priesley & Hackos, 2005, p.1.0-5.0, 7.0-19.0)

This combination of abilities, providing extensibility to the architecture, is what gives the DITA its ability to adapt to custom publishing environments. As a meta-language, the DITA can morph into whatever shape needed for a particular purpose. The Organization for the Advancement of Structured Information Standards is currently working on version 2.0 of the specification, which promises to provide even greater flexibility in information management for those organizations adopting the DITA standard

The DITA is still in its infancy. Very few commercial implementations of the DITA currently exist; consequently, the DITA developers must create their DITA environments from scratch. Additionally, software tools supporting the DITA are scarce. Fortunately, software vendors have been quick to recognize the DITAís potential and a substantial number of the DITA-compliant programming tools are currently under development. Once those programming tools reach the developer base, the quantity and variety of the DITA implementations is expected to rise dramatically across multiple vertical markets.




Priestley M., Hackos, J. (2005). OASIS DITA Architectural Specification. 1(9).
         Organization for the Advancement of Structured Information Standards. P. N.A

Rockley, A. (2003). Managing Enterprise Content: A Unified Content Strategy.
        Indianapolis: New Riders.