As the Internet has grown, so has the amount of informal information available on it, and latterly access to formally published content has emerged. The potential of the Internet as an information resource is clearly enormous. Realising this potential, however, can be difficult: building software tools that can help a user to do this is one of the most important challenges facing computer scientists in the late 1990s. As users, we need rich, personalised, added- value services that help us find, retrieve, manage, understand, interpret and use the wealth of valuable information that is so obviously present both on and via the Internet.