Once more, but with meaning – how will the Semantic Web affect technical documentation and technical authors?

Web technologies expert, John Fintan Galvin, is claiming 2010 will be the year of the Semantic Web, when semantic technologies really take off. If that is the case, how could it be used by technical communicators to deliver better user assistance?

The Semantic Web is all about the automation of connections between “resources” in a context-sensitive way. These connections can be made between anything defined as a resource, e.g. topics in a Help file, chapters in a user guide, other content, data, people, systems etc.

In the past, we’ve talked about how it will allow for more intelligent searching. 

However, the Semantic Web also offers technical authors the opportunity to automate the republishing of content (e.g. extracts of a user guide) across the Web – in all the places where this content could add real value to users.

Let’s look at some possible examples:

1. When someone poses a question in a Newsgroup or user forum concerning “adding a new user”, a summary of the relevant information from the user guide appears automatically in the right hand column of the Web page.
2. When users from ABC Inc. view the online Help for Oracle 11, they see links to the topics that have been judged the most useful by other ABC Inc. staff.
3. Users see Beginner, Intermediate or Advanced user information, depending on their individual status or preferences.
4. When someone raises a Support ticket, a summary of the relevant information from the user guide and/or the latest status on any similar support calls, appears automatically next to the form.

In such an environment, technical authors would need to do much more statistical analysis of what users want and how they behave – both modelling scenarios and analysing behaviours. This means looking at your chunks of user information (resources) and asking:

• What could we connect our content (resources) to?
• To whom could we connect it?

It also means that the content will need to be stored in a way that makes it possible to categorise the content semantically and to republish the chunks of information in a semantic format.