Peter Meyer, Elkera

XML and its ancestor SGML have been in use for over two decades. SGML found initial support in the aerospace, military, and large commercial publishing industries. Slowly, XML has gained increasing use in a wider range of enterprises for technical documentation. This adoption has been given a recent push by widespread promotion and support for the DITA XML schema.

Many predictions have been made by consultants and software vendors that the use of structured content management is about to become more widespread, even mainstream in enterprise content management. Many of us involved with structured content management have fallen into that trap. The technology can be so powerful, and we wonder why the rest of the world just does not get it.

Despite the optimistic predictions and the enthusiasm of its proponents, structured XML (as opposed to the low semantic, presentational XML in word processing software) has not become a mainstream tool in enterprise content management. Why, after all those years, is it that such a powerful technology is not used more widely?

There may be many reasons but I am going to suggest just one critical factor: usability. To be truly successful, all good technologies have to become almost invisible to their users. The workings of the product are mostly invisible to users of motor vehicles, telecommunications equipment, and home appliances to pick the obvious ones. Unfortunately, this invisibility has not yet happened with structured XML authoring and publishing.

In this article, I focus on some of the issues involved with print publishing, the role of standards such as DITA, and the measures Elkera has taken to try to make XML print publishing applications more accessible to non-technical content managers.

The problem with XML publishing

XML documents must be transformed or styled into human readable formats in print and on the web. Every detail of presentation information for every output has to be defined in rules that can be applied during the transformation or styling process. Sometimes this process is easy enough and the output requirements stable so as to not be a problem.

In many cases, such as for technical documentation, it can be quite difficult to produce high quality print publications from XML. Particular problems include the following:

  1. Many XML schema are very complex and create a very large or even infinite number of element combinations. If the style rules don’t provide any or the desired formatting for a needed element context, the results will be unsatisfactory.
  2. It is difficult to define in advance the precise requirements for all desired publishing outputs. When moving from manual publishing systems, all existing example documents are likely to include inconsistencies and arbitrary formatting. Rarely does any one existing document show examples of all required contexts and the desired formatting. If the developer does not have a comprehensive statement of output requirements, he or she cannot create all the necessary style rules. Often adjustments will have to be made over an extended period as the application is bedded down.
  3. Transformation and styling languages such as XSLT and XSL-FO are very complex and have a steep learning curve for new developers. It is costly to train documentation managers to use such languages, even if they are inclined to do so. It is costly to call in specialist developers.
  4. Publishing requirements change to meet new customer requirements or to deal with changing corporate styles. There may be an ongoing need to have specialist style rules developers available to meet these needs.

The effect is that XML-based print publishing applications can be expensive to develop and maintain. Larger organizations can afford to provide developers with the necessary expertise. Smaller to medium organizations or even documentation departments in large organizations may not be prepared to incur those costs.

It can be difficult to persuade documentation managers that they should learn new, complex skills or surrender control. It is difficult to persuade the financial managers that they should incur the setup and maintenance costs of complex applications. If structured XML is to truly move into the mainstream, it must overcome these and other usability barriers.

The basis of solution

Sometimes it is necessary to create style rules from scratch for a new, complex schema. Creating style rules for new schema will likely remain difficult and is unlikely to become a mainstream activity.

There is a growing range of standard schema in particular industries that can take care of most user needs. Once a standard is developed, it is possible to create applications that support the standard. Thus, there are open source and commercial XSLT rules for XSL-FO publishing applications to support schema such as DocBook and DITA.

On their own, standards don’t solve the problem. In practice, it is always necessary to customize out-of-the-box style rules to meet specific application requirements. The need for customization raises most of the problems identified earlier. It may not be too hard to get close to the desired layout but it can be very difficult to produce one that is exactly as desired. Once developed, it can be difficult to adapt style rules as circumstances change.

The use of standard schema such as DITA is an essential foundation to a solution but it is not a complete solution. What is also needed is a simpler model for style rules development to reduce the skill level of the developer for the most common development and maintenance tasks.

Visual layout design in Elkera XML Print

Elkera XML Print ( is designed to address the limitations described earlier. It provides two ways to develop print publishing style rules. Developers can created coded applications using the styling language if they find this easiest to maintain. Alternatively, it is possible to create a set of style rules that use a Word document as a visual layout design template.

Once style rules are setup for a schema, most layout properties can be defined in the Word document using skills most document designers already possess. Designers can edit page layouts, headers, footers, paragraph and character styles without needing to know anything about the underlying style rules. Standard layouts for tables can be adjusted using simple variables files, rather than editing complex style rules. Only more complex changes may require some understanding of style rules.

Elkera XML Print includes comprehensive, foundation layout rules for DITA and supports DITA specialization. It thus combines the benefits of a standard schema and a visual development environment to reduce the need for specialist developer skills. For many organizations using DITA, Elkera XML Print can put documentation managers back in control of print layouts and reduce the cost of style rules development and maintenance.