With full-text search, does Uncle Sam still need to classify web pages?

Added By Tony Byrne at 14-Oct-2005 | Twitter: @TonyByrne |

He's asking in an RFI called "Efficient and Effective Information Retrieval and Sharing". Having just acquired better search and fancy new clustering technology for the FirstGov search engine, could federal web content managers be spared the messy business of tagging content as mandated in the E-Gov Act of 2002? It's an important question to ask, but posing it on the "Federal Business Opportunities" website will yield some predictable responses. Since there are few classification vendors but many search and autoclassification vendors, I suspect the software industry will answer with a resounding "No." If you ask your favorite information architects, systems integrators, and true enterprise search experts, they will almost all answer, "Yes, of course you still need to tag!" As Steve Arnold recently opined in these pages: "Humans and human-like processes are needed to supplement or do certain types of taxonomy development, indexing, classification and analysis. We don't live in a "Star Trek" universe; a machine can only perform certain limited functions." What do you think? You have until 21 October to respond to the RFI.

Categories: Tony Byrne, Search and Information Access, Industry Standards