divine SinglePoint Search is based on the SinglePoint product that was first launched in December 2000 by Northern Light (NL), a company that divine purchased in January (see the NewsBreak at http://newsbreaks.infotoday.com/nbreader.asp?ArticleID=17271). The Northern Light information portal product was an outsourced solution that was run on dedicated hardware at NL. The newly branded divine SinglePoint Search, a browser-based application, is offered for internal enterprise implementation (for Sun/Solaris now and for AIX and Windows XP in the future), as a hosted application, or in a hybrid implementation. The new product supplies a number of enhancements and delivers access to the divine (Northern Light) taxonomy, which provides more than 17,000 pre-existing subject concepts and requires no customization.

According to Hank Barnes, senior vice president of marketing and strategy for divine's Software Services group, the integrated search function is a key feature of the enterprise solution. Users can search all available internal data sources and the Internet. They can then choose which sources they would like to search and receive a single results list with the relevance of all resources ranked to a uniform standard. The product can integrate with divine's content management systems (Content Server and Participant Server), other content or document management systems, and internal file systems. It can also crawl existing Web sites. This functionality is included in the standard licensing fee for the product. For additional fees, enterprises can license the rights to content offered by divine, such as sources in the Special Collection (which had previously been provided by NL on a pay-per-view basis) and from providers like Investext.

Barnes feels that taxonomy and auto-categorization should be an integral part of an enterprise search solution, not an option that's tacked on. However, users can opt for management and regular updates to the divine taxonomy for an annual subscription fee. Professional searchers who have used Northern Light are most likely familiar with the reputation of the taxonomy, which was built by librarians over the course of several years. It was assembled by using a number of trusted sources, including the Library of Congress subject headings index, the SIC/NAICS system, the Dewey Decimal System, and an additional 1,000 industry-specific taxonomies. The team has optimized the taxonomy for auto-classification and continues to update and train it on a regular basis.

Barnes stressed the savings for an enterprise that's able to implement an out-of-the-box solution, rather than create taxonomies from scratch. He said that divine would work with customers that might want to merge a custom taxonomy into the divine SinglePoint taxonomy.

"Where divine's taxonomy differs from other solutions available today is in its automatic classifications that result in immediate effectiveness and in its consistent and ongoing support from the team of divine librarians," said Barnes. "Using a tools-based approach can easily result in organizations spending significant time developing the taxonomy and auto-classification capabilities using trial and error. By implementing divine's subscription-based service, enterprises ensure that their taxonomy remains up-to-date and effective at a fraction of the cost of doing it themselves."

In a report entitled "Taxonomy, Thesaurus, Tagging: Balancing Automation and Editorial Review," Laura Ramos, a director at analyst research firm Giga Information Group, said: "As the deployments of portals, enterprise content management (ECM), and second-generation search solutions proliferate and the amount of documents and Web pages managed by these systems increases, the need to deploy second-generation search tools that encompass content organization and classification technology becomes more important. Without ways to organize content, categorize it automatically, and provide relevant navigation, applications that focus on collaboration and communities can fail to provide effective user experiences or to deliver returns to the business."

The divine SinglePoint Search engine indexes and searches any file type and offers support for Microsoft Office, HTML, Adobe PDF, and XML. The search functionality includes Boolean syntax, fielded searching, +/- syntax, and natural language queries. The search engine does relevance ranking based on 26 different factors, with weighting adjusted according to the query. Results are presented to users in custom search folders.

divine hopes its new product provides better search results for enterprise customers without the time-consuming investment in developing an information taxonomy. divine focuses on Global 5000 and high-growth middle market firms, government agencies, and educational institutions. It currently serves over 20,000 customers.