What to do about Google...

February, 2005

The inspiration for this article was a question submitted by one of our members, the project manager of a "Taxonomy Center of Excellence." She asked:

"How do we position ourselves against Google, which is in essence,
anti-classification/taxonomy? (OR do we embrace it... and if so, how?)"

Google is something of a red herring. Search engines are here to stay, and no one — not even the most ardent indexer — is ready to throw them out. The real issue is the cost-effectiveness of two different publishing models: "dump-search-sift" versus "select-package-promote. "

In "dump-search-sift," everyone dumps content of questionable quality into publicly accessible repositories. Lacking any other alternative, people use a search engine to find the needles in this enormous haystack and then spend hours sifting through the results. In "select-package-promote," writers compose their works according to quality standards, editors select and package documents to meet the needs of a specific audience, and publishers promote (or announce) the publication or service to intended users.

As we show in this article, there are times when Google works well. But because it's ubiquitous and deceptively easy to use, Google creates unrealistic expectations for many users. It also fosters an over-simplified model of information work. Everyone uses Google, but everyone uses it in a different way depending on their job role, personality, and other factors. Furthermore, most professional employees play more than one role, often within the same work day.

Competing with Google isn't feasible. Embracing it — i.e. recognizing when it works well and when to use other tools instead — is more productive. As an illustration, we look at six different job roles in our own company — Manager, Author, Editor, Instructor, Taxonomist, and User (a catchall category) — to see how they use five different kinds of tools:

  • Google (public search engine)
  • Ultraseek (site-specific search engine)
  • internal knowledge base (data-oriented search engine)
  • A - Z index (topic browse)
  • Google desktop (local search engine)

More ... (members only) How to become a member

Created on March 4, 2005