From Chapter 2:

To compete effectively in tomorrow's global economy in spite of economic turmoil, the enterprise must integrate many data sources - databases, email, marketing and policy documents in a variety of formats, and Internet, Intranet, and Extranet Web sites - into a single coherent fund of knowledge, accessible from anywhere, at any time. The world has moved beyond data processing and e-commerce into the uncharted waters of the Semantic Web. A trivial example of the challenge we face can be seen by entering a natural-language query into any search engine on the World-Wide Web, then noting the degree of relevance - or not - in the results obtained. The volume of responses often runs into the hundreds of thousands for a query of any complexity, and we can only suspect that the best possible answers may have eluded us. Truth, trust, and relevance are not necessarily the first attributes that come to mind when describing the Web experience.

A more insidious situation pertains within organizations. Of those organizations who have invested in making intellectual assets available through information technology, only the most far-sighted and fortunate have been effective in doing so. Consider the difficulty of searching for an item of information that may be in an email, a document in the file system, or in any one of the corporate databases, and you will see the pervasive nature of this problem, which has been dubbed federated search.