The Popularity Contest: Taxonomy Development in the Petabyte Era

Forget taxonomy, ontology, and psychology. Who knows why people do what they do? The point is they do it, and we can track and measure it with unprecedented fidelity. With enough data, the numbers speak for themselves.” (1)

Recently Chris Anderson wrote an article for Wired magazine called the The End of Theory. The thesis of the article in a nutshell is that the impending petabyte era of data storage signals the end of the traditional scientific method of discovery. No longer are we bound to the outdated model of observation, hypothesis and measurement. Computers (developed by Google & IBM) “can throw the numbers into the biggest computing clusters the world has ever seen and let statistical algorithms find patterns where science cannot.”

Subject Matter Experts and Taxonomy Development

I was recently in a meeting where it was said that a lack of subject matter expertise is a disadvantage in taxonomy development. This is understandable; it makes sense to assume that the more domain expertise a taxonomist has, the better the final information product will be. However, my experience has shown that this is not always the case. So exactly how does the role of subject matter expert (SME) fit into taxonomy development?

Cleaning up the Other Bucket

The Other category: also known as General, Miscellaneous, My Stuff, or, too often, the shared drive.

The Other category is the junk drawer for all those taxonomy terms that just don’t seem to fit anywhere else. Reaching into it is taking a chance as you never know what you may find—the yo-yo you can’t seem to throw away, a pair of rusty scissors threatening to impale you when looking for loose change, batteries, baggy ties, the old cell phone with numbers in it of people you don’t even remember. When it’s time for cleaning up, you can either throw it all away or sort it out and put everything in a more appropriate place.

Tips for Conducting Keyword Research - Part 1

Last week Stephanie wrote about (post) the importance of considering specific facets of search engine optimization in helping taxonomists guide clients in choosing the right keywords. To further that discussion, I thought I’d put together a series of posts to speak in more detail about using keyword research as a tool for determining (or at least being consciously aware of) the language being used by those searching for your content, products and/or services online.

Preparation - Creating Your Master List
The first step in the process is the groundwork. I always allocate a certain amount of time up front to plan and prepare the list of initial keywords to be used as a basis for conducting keyword research. You need to have an inventory of words or phrases to get started, so why not put some thought and effort into generating a solid list to work from. From my perspective, the better the plan, the better the results. So let’s get to it.

DAM and Controlled Vocabulary

I worked on a project recently highlighting findability issues with unstructured content and the need for appropriate tagging using values from a controlled vocabulary.

At the heart of this project was Digital Asset Management (DAM), a rapidly growing area as more multimedia content is being distributed online, particularly for marketing purposes. The inherent problem with digital assets is the potentially large amount of information about what a piece of content is but the lack of information describing what that content is about. Unlike other content, which may contain text or be located with surrounding textual context, digital assets do not typically contain text, especially any which is structured for discovery by search engines. Any textual and searchable elements must be associated to digital assets through the use of metadata. Metadata describing what the content is, including attributes like video length, number of pixels, and file size, can be associated to the content and is often automatically attributed through business rules.

What the asset is about, however, is not inherent. It must be associated to the content either manually or automatically by loading the content once business rules have been thought out and established.