In most organizations, information is located in a variety of different data stores, from file servers, groupware systems, relational databases and legacy systems to external sources such as the Internet. The ability to index the information in those corporate data repositories and to allow users to search against it is among the most critical functional components of any knowledge management (KM) system. But even if the foundation of an organization's KM strategy starts with text indexing and retrieval, it soon becomes clear that it is only a starting point.
With that in mind, vendors of search-oriented software are enhancing their systems to provide more value for KM. Case in point: Excalibur Technologies (www.excalib.com). With the release of RetrievalWare 6.7, Excalibur has extended the product's capabilities beyond search and retrieval to include features that are essential to KM, such as user contribution capabilities and expert identification. While RetrievalWare may not yet offer soup-to-nuts technology for far-reaching KM initiatives, we feel it provides an excellent foundation for those applications that require strong information retrieval capabilities as well as support for user contribution and basic collaboration.
Starting with search
That does not diminish Excalibur's traditional strengths in indexing, searching and retrieval. RetrievalWare certainly provides an effective means for users to find information relevant to their needs, regardless of document type or format.
The product offers a single interface for searching for information across multiple data sources, seamlessly presenting returned results within a single result list. Right from a Web browser, users can search and retrieve information--including text, word processing documents, PDF files, news feeds, images, messages and data from e-mail and groupware systems, and information from relational databases--without knowing or caring where the information is located. Information can be presented to users in a familiar folder hierarchy, making the software easy for users to learn.
RetrievalWare's robust information retrieval engine relies on two core technologies, adaptive pattern recognition processing (APRP) and the semantic network. Those technologies provide highly accurate search results, allowing users to locate information based on meaningful content. As a result, RetrievalWare is more error-tolerant, both with respect to input data and query terms. It does not require the use of specialized search languages--an important consideration for organizations as they seek user acceptance of a new technology.
Broadening the appeal
Realizing that there are many other aspects to KM besides information retrieval, Excalibur has equipped RetrievalWare to provide functionality for user contribution and knowledge sharing. The product now includes Knowledge Contributor, a desktop client application that allows users to add documents (or directories or entire directory trees) to the system via simple drag-and-drop operations. It also presents a list of the documents that have been added to the system. That raises the value of the RetrievalWare knowledge store, because users have access to the expertise that other users have chosen to contribute to the system, as opposed to only the information that an automated engine can locate.
In addition, the product offers categorization capabilities. Users can create categories for documents on a particular subject--categories that other members of the organization can navigate and/or search on, thereby enabling users to share knowledge. It provides an option to maintain categories over time, updating them as new information is added and removing documents as they become outdated.
Another key enhancement is the Experts Directory. That utility connects people within an organization to specific subject categories, allowing users to search for individuals that have expertise in particular subject areas. Thus, people can track down key individuals within the organization to leverage their expertise--a difficult proposition in extremely large organizations where people aren't necessarily acquainted with employees in other departments or locations.
The FileRoom allows organizations to incorporate indexes for paper-based information into the knowledge store--a feature we haven't seen in many other KM products. Organizations can integrate their paper-based knowledge assets into the same system that employees use to track down all other knowledge assets and external information. That is particularly valuable to organizations that still maintain large volumes of paper records or that want to integrate paper-based records management systems into their KM strategies.
RetrievalWare's search and KM capabilities are built around a solid architecture designed to handle the rigors of enterprise deployments in large distributed organizations. For example, Ford Motor Company is using RetrievalWare to make marketing research information accessible to key decision makers throughout the organization, and United Airlines is using RetrievalWare to provide up-to-date maintenance records and instruction manuals to its engineers worldwide.
The central component of RetrievalWare is the scalable Search Server, a proprietary search engine that handles processing for indexing and searching, as well as system administration. The server is architected to make efficient use of computing resources--even when searching large databases. The 32-bit server is available for Windows NT and for various Unix platforms. The NT-based server takes advantage of NT's administration utilities, user and group definitions, and security. The server can be configured to honor security established in Unix environments or in Microsoft Exchange and Lotus Notes groupware.
The RetrievalWare server provides "synchronizers" that allow access to different data repository types, including file servers (Windows NT, Novell and Unix), Lotus Notes, Microsoft Exchange, relational databases (Oracle, Sybase, Informix and Microsoft SQL Server), the Internet, and Usenet Newsgroups. Synchronizers for FileNet and Documentum repositories are under development. Document formats supported include Adobe PDF, HTML, SGML and XML, as well as all popular word processors, spreadsheets and publishing systems.
The server also includes the Excalibur Internet Spider, a multimedia Web crawler that efficiently explores intranets and the Internet, based on interest profiles that can be specified at the user, departmental or organizational level. The Internet Spider can monitor unlimited numbers of internal and external Web pages, actively gathering specific documents from them.
Overall, Excalibur RetrievalWare is a solid information retrieval solution, and we are happy to see the company's commitment to enhancing the product to provide higher value for KM strategies. But additional enhancements are warranted in a number of areas.
For example, as good as RetrievalWare is at harnessing information from multiple sources, the product still neglects the vast volumes of data stored within mainframe systems. For organizations that are still using their legacy systems as primary production applications, such as financial institutions and insurance companies, RetrievalWare does not yet offer the ability to integrate that information into user searches.
In addition, although Excalibur recognizes the importance of user contribution and collaboration to KM, RetrievalWare still has room to grow in that area. Comment capability, for example, would be a useful addition, as would e-mail integration to allow users to contribute documents by simply mailing them. The latter feature would also o