Nobody Thinks It’s Easy


June 2005

Nobody Thinks It’s Easy

CIDMIconNewsletter JoAnn Hackos, CIDM Director

April 2005 saw the highly successful completion of the 7th annual Content Management Strategies conference in Annapolis, MD. With over 200 attendees, we exceeded our expectations and held the largest conference ever. The conference featured three tracks this year: Management, Technical, and DITA. Although track attendance was fairly well spread, the DITA track was most popular. Supported by IBM, the various presentations on DITA (Darwin Information Typing Architecture) demonstrated that the new topic-based authoring standard is garnering a great deal of attention in the technical information community.

In all three tracks, speakers provided vivid and interesting examples of their implementation of content management solutions. A wide variety of approaches demonstrate that there is no single solution. Some departments have developed their own systems; many are taking advantage of the vendor product offerings, as exemplified by the 15 exhibitors who gave their support to the conference.

Although I concentrated on the management track throughout the conference, one message was clear across the 50 plus presentations: “There’s nothing easy about structured authoring or content management.”

It takes time, resources, and lots of creative energy to implement a content management solution, not because tools are difficult to use but because achieving an acceptable return on investment and meeting business goals requires a significant amount of change. I’m pleased to report, in fact, that the vendors have been quite responsive to our usability requirements. User interfaces have improved greatly over the past several years. In fact, I could easily make the claim that most of the component-based content management systems are quite “user friendly” and can be implemented in a short time in most organizations.

Component-based content management continues to remain in the hands of a small number of vendors, although we continue to see new entries to the market. At the conference, the exhibit area featured

  • Astoria Software
  • XyEnterprise
  • Trisoft
  • Idiom
  • X-Hive
  • Vasont Systems
  • AuthorIT
  • EMC (formerly Documentum)
  • Mark Logic
  • SiberLogic
  • Percussion Software
  • Innodata Isogen
  • OmniLingua
  • Quadralay Corporation

In addition, Arbortext, developer of the XML editor, Epic, was a conference sponsor. Each of these vendors provide functionality that enables us to assemble documents, help systems, web sites, and other deliverables from a repository of topic-sized and smaller components. They support the key principles of modular reuse that is so essential to increasing efficiency and decreasing costs within our organizations. A number of the vendors also emphasize the importance of multiple language content management and support that functionality within their products.

However, as you are well aware, finding the right content management solution is a small part of the content management equation. That’s where the hard part comes in. Achieving our goals for efficient automation of product tasks and the reuse of content modules across multiple deliverables requires a new discipline in information development.

I believe it useful to examine four of the implications for this required discipline:

  • Structured content
  • Topic-based authoring
  • Minimalism
  • Collaboration

Structured Content

I find that very few organizations have thoroughly structured content throughout their information libraries. Individual documents may be more structured as a consequence of the work of a particular information developer but if you were to look across the library, you would find significant differences in the way in which information is organized.

Because of the focus on desktop publishing, legacy documents will have format tags that follow a department style standard. As long as the information developers have followed the standards with regard to format tags, the documents are consistently formatted. But formatting alone does not provide structured content. To produce structured content means that the information developers have presented content in a consistent manner. With structured content, we would encounter procedural, conceptual, and reference information addressed in the same way throughout the library. For example, conceptual information would be presented in the same way (definition of the concept itself, a set of examples, a set of non-examples, perhaps an analogy) in each case.

The lack of a rigorous structure for the actual content means that parts of the information cannot be combined into new assemblies without appearing to have been written by different authors. With a rigorous content structure, all the authors follow the same structure.

Topic-Based Authoring

More organizations, in my experience, have some semblance of a topic architecture in their legacy information. However, in most cases, the topic structure is not applied in a rigorous manner, which means that topics are embedded within other topics rather than treated independently.

In a solid topic-based authoring environment, topics are standalone pieces of information. Users can understand how to perform a task, understand a concept, acquire background knowledge of a topic, and find critical reference information that is not buried in the midst of other information types.

Moving to a topic-based architecture, accompanied by structured writing, is rarely a matter of converting legacy content. Most of the time, the content will need to be rewritten to follow the topic and structural rules.


Although minimalism is not strictly a requirement for component-based content management, minimalism presents an organization with a significant opportunity to improve quality while reducing volume. I strongly advise organizations to pursue a minimalist agenda at the same time they are restructuring content in a topic-based architecture.

All of our research in user information requirements, and certainly the outcome of user research by many of our member organizations, suggests that we produce far too much irrelevant content. See articles in this issue by Henry Korman, Susan Harkus, and Erling Nielsen et al. about the importance of ensuring that content is relevant to user needs.


Not only does successful content management require that legacy information be restructured and new information written in a new way, managers find that they must revise information-development processes to emphasize collaboration. For an organization to achieve aggressive reuse goals and automate the production of multiple deliverables from a topic library, you must collaborate on planning, designing, and developing the modular content. I believe that content planning can no longer take place at the book or help system level. Rather, content must be planned at the topic level so that a set of unique, interchangeable topics is constructed. For this to happen, information developers and architects must work in a highly collaborative environment. The days of isolated writers working on the mountain top are no longer appropriate.

It Takes Work to Succeed

It’s easy and convenient for a sales team to tell you that content management is easy. All you have to do is convert your content into XML or “chunk” it according to heading levels. It’s quick and you’ll be done in no time. In fact, automated conversion tools will be the answer to your dreams.

It was quite obvious from all of the presentations by managers, architects, and writers at Content Management Strategies that robust solutions that genuinely promise to save resources and costs in the future do not come easily. The presenters showed us that it takes time and hard work from team members to succeed.

As usual, however, the outcome of some pain is enormous gain. With minimalist, user-centered information, developed in a collaborative environment in which information developers create rigorously structured topics, you can not only show immediate cost reductions and efficiencies, but you have built a structure that will continue to produce financial and quality benefits. CIDMIconNewsletter