Switching to XML-based content management can save big money in the long run by promoting content reuse and multichannel delivery. It can also be an expensive, time-consuming challenge akin to building a house from a pile of lumber. Using the Darwin Information Typing Architecture (DITA) makes the project more like building with prefabricated modular components.
This analogy, offered by Frances Gambino, director of documentation services at software vendor Information Builders, is typical of the enthusiasm greeting DITA, an open XML-based architecture for planning, authoring, managing and publishing technical content in print and on the Web. Developed by IBM and approved as a standard in June by OASIS, DITA promises huge savings for organizations that publish user manuals, maintenance documentation, catalogs and other technical content.
Information Builders adopted DITA as its corporate standard more than a year ago. "Many of our software products share components, so we might have the same chapter in three different manuals," says Gambino. "DITA takes a topic-based approach, so we can share chapters and subsections [across] manuals."
DITA was developed to promote multiformat reuse, particularly for help menus and knowledge bases organized by topic. DITA has built-in support for industry- and company-specific "specialization" (or customization), so it's easy to bring new content into existing DITA-based document type definitions (DTDs).
"Companies often spend tens if not hundreds of thousands of dollars developing custom DTDs, yet [those models] often turn out to be inflexible and costly to maintain," says consultant JoAnn Hackos of Comtech Services. "DITA will reduce these costs because it provides a ready-made set of DTDs that offers all the advantages of a standard."
In a merger scenario, for example, two companies embracing DITA would have a much easier time merging information by exploiting the standard's hierarchies and specialization features. And for manufacturers purchasing subcomponents from myriad suppliers, standardizing on DITA would enable content from those suppliers to be ingested, rebranded and delivered without the cut-and-paste tedium that's common today.
The OASIS standards initiative had support from technical content publishers including BMC, IBM, Intel, Nokia, Oracle, Sun and the U.S. Department of Defense, as well as from technology vendors ArborText and Idiom Technologies. The buzz on DITA is spreading, however, so vendors are scrambling to support it.
Information Builders implemented DITA using software from Idiom that also integrates with the vendor's content translation and localization system. "Once a DTD is developed for English, there's little additional work to move it into localization," says Gambino, adding that the move has saved "years of work" over the DocBook-based approach previously considered.
DITA won't displace standards such as ACORD in insurance, ATA in the aerospace industry, SPL in the pharmaceutical industry or other industry-specific DTDs. And early studies show that "at least half of all DITA implements involve specialization of the standard," says content management expert Ann Rockley of The Rockley Group. Nonetheless, DITA is designed to be extensible, and the out-of-the-box DTDs are just the boost many organizations need to get up to speed on structured content and reuse.
— Doug Henschen
|[ FAST LANE ]|
Data aggregation partner to speed up Cognos reporting. HyperRoll has entered into a partnership that will bring Cognos the same type of data aggregation technology that Hyperion now uses in its Essbase OLAP product. Cognos expects HyperRoll to help its reports process in one tenth to one hundredth of the time they currently require.