In Your Hands:
Multi-Platform Dynamic Publishing

If you were to poke around on the website that delivers Contelligencegroup.com, you might be surprised at what you would not find there. There are no HTML files or static web pages. In fact, there really is no concept of a webpage in the convention sense of the word. There is no duplication of content, and no custom code to maintain.

This site is an example of Dynamic Publishing. What you see is determined by rules we have established to deliver content that we expect to be most relevant to your personal profile. Where does that profile come from? While the approaches to determining user profiles are nearly limitless we are not trying to pry into your browsing behaviors. If you came here from a LinkedIn Group the site notices which group it was and assumes that visitors from the XML in Practice Group might be looking for something different than visitors from the Social Media and Emergency Management Group. If you followed a Twitter link, the site personalizes the experience based on the hashtag used in the tweet. If neither of these applies, or if you just want to see what happens, look for a widget on this page that will allow you to change the personalization as if you arrived here from different origins.

We took this approach to the site for two reasons. First, Content is a broad and complex subject with business and technical issues that vary by industry and functional area within an organization. It can be overwhelming even to people who are expert in one area since they may still have little exposure to other areas. Personalization improves the site experience, a top priority for anyone who puts significant effort into designing a website.

The second reason is that there are very few examples of rules-based Dynamic Publishing on the Web. Dynamic Publishing itself is not rare—there are many examples across most industries and public sectors. However, real-time Dynamic Publishing of personalized web content is extremely rare, even though most organizations have wanted it for years and the technology has been available for more than a decade.

There are a lot of moving parts in a website launch and project teams have only recently started to appreciate the central role of content and the need for content strategy. It will be a while before those teams understand what the Content Strategists mean by Dynamic Publishing. This site uses international standards to deliver Dynamic Publishing with minimal effort—something we hope will inspire both project teams and their organizations to expect more from Web Content Management.

Contelligence.com is built around the OASIS Darwin Information Typing Architecture (DITA), an XML standard that goes beyond tag definitions to suggest both architectural approaches to information management and best practices for content creation, storage, and delivery. There are three key structures within the DITA standard that are important to the design and operation of this site:

1. A topic is the standard unit of content creation. Topics can be thought of as roughly analogous to sections in a document or webpage, however an unlimited number of topic types can be designed that conform to a specific content architecture. Even though an author may write several topics to cover a subject (think, several sections), the content will be most flexible and most reusable if each topic is written so that it is complete and contextually valid by itself. There are many examples of this approach in good writing and it is primarily a matter of style.
2. A map is an ordered list of individual topics that a content creator or curator intends to deliver together. Maps are hierarchical structures that can include topics, other maps, and metadata that can be passed to a processor.
3. A DITAVAL file encapsulates the rules concerning the actions should be taken based values derived from the content metadata or from the consumer’s interaction with the site.

All of the content presented on contelligencegroup.com is assembled based on DITA maps that indicate the DITA topics that are the potential content sources and then further refined based on a DITAVAL file. Topics have been authored in any number of browser interfaces. The most common is a WYSIWYG rich text editor that hides the existence of XML from the content creator. The topics themselves are static until checked out and revised by an author following basic standards for content management.

The content that is delivered on a specific page, the site navigation that appears on a page, and in some cases, the site structure itself is determined with DITA maps and DITAVAL files. These controlling files may have been authored previously or may be dynamically generated based on values collected at runtime.

There are many competing approaches to website development. When personalization is a priority, there is a tendency to invest extensive resources in custom development of web forms and database models for each content type. While the API environments of most Web Content Management products encourage this approach, there is an alternative. DITA’s implementation of XML can be integrated into almost any object model provided by the underlying WCMS, and when accomplished this will allow personalization to be based on the information model, as opposed to custom coding.