| Knowledge discovery, as the term alludes to, is based on searching for information. For a good implementation of SharePoint Portal Server, the following major design goals must be kept in mind: -
Timely Unless users get their search results relatively quickly, they will not use the tool. You likely know from your own experience how annoying it is to wait for the results of a query. -
Accurate Search results should contain the specific information that has been searched for. That is, if you wonder why a specific document was included in the search results, you likely will not trust any of the results. Another example of inaccurate results would be references to documents that cannot be read by the user due to security constraints. Revealing the existence of information, even though the user is not allowed to see it, can have some serious business implications and must be avoided. -
Relevant Users will likely look at the top results first. Therefore, it is important that the most likely expected results ”or what one may call the best matches ”are returned first. Knowing what is relevant and what is not in the end will be determined by the user who is searching for documents. But as you will see, SharePoint Portal Server uses some advanced heuristics that generally work very well. -
Comprehensive If a document does match a specific search request, it must be shown as a search result. Imagine that an author is searching for his document on imaging, for example, and his document does not show up in the results! This will certainly affect his confidence in any search results returned by SharePoint Portal Server. To these four "common sense" design goals, another three have been added that make SharePoint Portal Server a very rich, flexible, and easy to use tool within an organization: -
Extensible Many of the documents one would like to search for might be dispersed across the enterprise. They might be stored for example on a Web site, a file share, an Exchange Public Folder, or in a Lotus Notes Document Library. And while the place where information resides may vary, the format in which the information is provided may also be different. Providing flexible and extensible architecture therefore is essential. -
Self-serviceable While a typical systems administrator has excellent skills to maintain a system, this person likely will not know which information is relevant for a specific department. That is much more expected from the end-user community, the folks who understand the business. Such business information, for example, includes knowledge of the competition, or the end-user community's most frequently visited sites. It is much more efficient, and thus more cost effective, if the end-user community can directly define the information that should get indexed. -
Scalable Searching for information is done at all levels of an organization: on an individual desktop, within the department, or throughout the enterprise. Providing a scalable architecture that allows the reuse of the same technology for department and enterprise is a clear benefit.  |
