Designing Your Information Architecture for Content Reuse: Five Best Practices

By Amber Swope, Principal Consultant, JustSystems - Reprinted with permission from DM Review

The increasing popularity of Darwin Information Typing Architecture (DITA) means that more users within an organization are looking to repurpose and reuse content across the enterprise. To realize the promise of reuse with DITA, you must optimize the mechanisms it supports and understand how to implement it. When considering the implementation of a reuse strategy, consider the following five best practices.

Design a map structure for single-source output. DITA enables the user to create multiple output documents from a single source. To take advantage of this support, organize the topics into reusable units using DITA map files. DITA map files are the XML manifest files that specify hierarchical and topic-to-topic relationships or links. For online output deliverables, such as HTML files or compiled HTML (.CHM) files, use a master map to organize the submaps. If this deliverable is to be displayed in a graphical user interface (GUI) with other deliverables, such as in an online help system “bookshelf,” use a container or wrapper topic to be the “book” that represents the deliverable on the bookshelf.
For printable output, such as portable document files (.PDF) files, consider using the Bookmap specialization included in DITA 1.1. You can create the bookmap to organize the content into the familiar book structure, including parts and chapters, and specify the metadata for the deliverable. If you have created reusable maps, you can use them as chapters or simply include them with other topics within chapters for the book. This allows you to take advantage of the additional bookmap functionality and still use the same maps for both the printable and online output deliverables.
Use metadata for conditional processing. By applying conditional values to content both within topics and in maps, you can generate output that contains only the specified content and links. The first step is identify the values on which to process, then to create a DITA processing (.ditaval) file that lists properties and all the possible values. By default, DITA includes properties for conditionally processing by audience, platform and product. You can easily add additional properties or customize the default properties with your own values. Next the authors must apply the appropriate values to elements within topics, topics within maps, or maps within maps. This means the author must know all of the appropriate values and when to apply them. The key to this exercise is not having too many values. If there are too many values or the value structure is too complex, meaning that multiple values need to be applied to the same elements or topics, authors will not apply the! values correctly. After the values are applied, you can generate the output using the appropriate value for each deliverable. If the values are not correctly or consistently applied or you do not specify the correct values during generation, you will not get the correct output. You need to have a rigorous verification regime to make sure that the deliverables contain the anticipated content.
Understand the dependencies for repurposing content. Enterprise content is often applicable to more than one purpose or deliverable in a company. The first challenge to repurposing content is identifying the content other authors want to access and reuse. This requires communication and coordination between all the stakeholders. Start small and work with those teams with whom you have the best communication and/or share the most content.

The next challenge is storing the content in a manner and location so that the other users can access it. This is particularly difficult when each department uses a different repository for content storage and will require the support of IT to address. In many cases, most of the users who want your content will not have access to the repositories in which you store it.

Lastly, you need to understand the way these authors want to consume the content. If they are not authoring in XML, they will need generated output. This means you need to know into which format you must generate the output and how often you need to regenerate it. If they are also authoring in XML, they will want to reuse the XML source. To avoid unintentionally changing reused content and unknowingly propagating changes throughout the enterprise, you must have a reuse strategy in place to know who is reusing or consuming what content.
Reuse controlled content using content references. Controlled content is content that cannot be changed without authorization. The first consideration is identifying who owns the content - who has authorization to make changes. The second consideration is training authors to reuse it. In order to reuse controlled content, create a master list of phrases or other content in one or more files, then reference the content into the applicable location in individual topics. For example, if your content contains regulated warnings, create a topic with the legally approved warning and reference or pull the content into each topic rather than rewriting the warning every time. This provides a single version of the truth and saves authors time. Another good example of controlled vocabulary is an acronym list that provides the full explanation of the acronym for authors to reference rather than retyping. The keys to successfully reusing controlled content are that the content must be easy to find and the authors must reference all instances of content. Authors will not reuse content that they cannot easily find, and if they do not insert the proper reference elements, the content cannot be updated globally.
Update dynamic content using content references. Dynamic content is content that can be updated at any time. Unlike controlled content, the ability to update dynamic content is not limited to a specific owner. This means that you must communicate when content is reused and when it is updated. You can do this either via technology or process. If you store the content in a content management system (CMS), the workflow support can manage this communication. Authors can easily see if someone else is reusing their content, and the CMS provides a mechanism to inform authors when content they are reusing has changed.

In the absence of technology, you must create a process to handle this communication. One strategy is to create topics of common information and store them in an area where all parties know that content is available for reuse. The rules for the common area are that no one updates the content without approval from all potential consumers and that all potential users are notified when a change occurs. This policy allows authors to have confidence that they can reuse the content at the element, topic, or map level without risk. In cases where there is a set of content that may appear en masse in multiple deliverables, create a map to organize the topic and store it in the common area. This will save authors time when it comes to creating maps and specifying links.

Enterprises that have followed the above best practice methods are getting the most benefits from using DITA.

About Amber Swope
Amber Swope is a principal consultant at JustSystems where she applies her information architecture and DITA experience to help clients address their content related business challenges. Swope is an experienced information architect with almost 20 years in the information development field. At IBM she led the first HTML to DITA migration project for the Rational division and implemented DITA in a production environment. She is also a member of the Oasis DITA Technical Committee and is participating on the multiple subcommittees. Amber has authored numerous papers and articles on information design, development, and architecture and has presented at many leading industry conferences. Contact Amber at .

Related article: To DITA Or Not To DITA?

More articles about Content Reuse : DITA : Information Architecture