How to Create a Topic Map?

Topic maps are a new ISO standard for describing knowledge structures and associating them with information resources. As such they constitute an enabling technology for knowledge management. Dubbed "the GPS of the information universe", topic maps are also destined to provide powerful new ways of navigating large and interconnected corpora.

While it is possible to represent immensely complex structures using topic maps, the basic concepts of the model -- Topics, Associations, and Occurrences (TAO) -- are easily grasped. This paper provides a non-technical introduction to these and other concepts (the IFS and BUTS of topic maps), relating them to things that are familiar to all of us from the realms of publishing and information management, and attempting to convey some idea of the uses to which topic maps will be put in the future.[1]

Note

The original version of this paper was published in June 2000 and thus predates the development of XTM (XML Topic Maps). The purpose of XTM was to adapt the topic map standard (ISO 13250) for use with XML and the Web. XTM provides an alternative, XML-based syntax for expressing topic maps and also clarifies certain concepts, especially those relating to subject identity.

Since this paper deliberately avoids syntactical issues, the fact that there are now two standard interchange syntaxes for topic maps (HyTM and XTM) is not a problem. However, doing justice to the clarification of concepts would require a substantial reworking of the paper. While this updated version includes certain minor amendments to reflect changes due to XTM, readers are advised to consult [XTM 1.0] to get the full picture, paying particular attention to the notion of subject indicators, and the distinction between addressable and non-addressable subjects.

If the concepts described in this paper turn you on, look for pointers to further reading at [Ontopia 2002]. To see topic maps in action, try out the online demo of Ontopia's Omnigator, a generic topic map browser. If you find that interesting, download the free version of the Omnigator and try it with your own topic maps.

1. Introduction

Someone once said that "a book without an index is like a country without a map".

However interesting and worthwhile the experience of driving from A to B without a map might be in its own right, there can be no doubt that when the goal is to arrive at one's destination as quickly as possible (or at least without undue delay), some kind of a map is indispensable.

Similarly, if you are looking for a particular piece of information in a book (as opposed to enjoying the experience of reading it from cover to cover), a good index is an immense asset. The traditional back-of-book index can be likened to a carefully researched and hand-crafted map, and the task of the indexer, as Larry Bonura puts it [Bonura 1994], "to chart[ing] the topics of the document and [presenting] a concise and accurate map for readers."

In Troilus and Cressida Shakespeare used a different metaphor:

And in such indexes (although small pricks

To their subsequent volumes) there is seen

The baby figure of the giant mass

Of things to come at large

but also here there is the same sense of the index replicating, in miniature, the structures of its subject, in order to provide a more manageable view of the whole.

Perhaps it isn't surprising that Shakespeare chose not to use the map metaphor. After all, the art of cartography was still in its infancy in his time ... and so too were communications. Today the situation is quite different, the sheer speed of modern communications makes accurate and advanced mapping techniques of major importance. One answer to this problem in the realm of transportation is the GPS (Global Positioning System) . The answer in the realm of publishing and information management is the new international standard, Topic Maps [ISO 13250].

Up until now there has been no equivalent of the traditional back-of-book index in the world of electronic information. True enough, people have marked up keywords in their word processing documents and used these to generate indexes "automatically", but the resulting indexes have remained firmly within the paradigm of single documents destined to be published on paper. The world of electronic information is quite different, as the World Wide Web has taught us. Here the distinction between individual documents vanishes and the requirement is for indexes to span multiple documents, and in some cases, to cover vast pools of information, which in turn calls for the ability to merge indexes and to create user-defined views of information. In this situation, old-fashioned indexing techniques are pitifully inadequate.

The problem has been recognized for several decades in the realm of document processing, but the methodology used to address it -- full text indexing -- has only solved part of the problem, as anyone who has used search engines on the internet knows only too well.

The main problem with full text indexes is their lack of discrimination. They index everything: Imagine creating a traditional back-of-book index by taking every single word in the book, removing a couple of hundred of the most obviously useless ones, and then including every single usage of those that remain. Even with some intelligence to allow for inflected forms the result would be of no practical use whatsoever. Mechanical indexing cannot cope with the fact that the same subject may be referred to by multiple names (the "synonym problem"), nor that the same name may refer to multiple subjects (the "homonym problem"). And yet this is basically how a web search engine works (no wonder you always get thousands of irrelevant hits and still manage to miss the thing you are looking for!).

That is why new methodologies are called for. Topic maps provide an approach that marries the best of several worlds, including those of traditional indexing, library science and knowledge representation, with advanced techniques of linking and addressing. It is our firm conviction that they will become as indispensable for tomorrow's information providers as maps for the traveller. And once topic maps have become ubiquitous, they will indeed constitute the GPS of the information universe.

2. Knowledge structures and information management

Before