Taxonomies and change: the nature of the beast

An interesting problem was posed to a mailing list I am a part of…

Imagine that you have been using a single hierarchy to structure and organize your information for years, and it has been very successful up until now…

But now it is time to move to a different content management system, and not only that – business has changed (of course), and not every way of organizing and understanding the information could possibly have been anticipated. (Or perhaps you did anticipate some, but for practical matters limited the amount of metadata you might apply to content.) So you have new ways that users want to search and navigate, but never considered these at the start. What do you do?

Well, there are (at least) three issues behind this problem:

  1. First of all, can your infrastructure expose faceted navigation? (If not, you can consider bolting on a search interface that leverages entity extraction or metadata)
  2. How stable are top level terms and how flexible are the ‘core’ organizing
  3. Can content be retrospectively indexed with metadata?

Before a facet was conceived, there was nothing captured that could represent that organizing metaphor. We now need to map new terms to content or update terms that were already applied to content.
In a meeting today to review a faceted taxonomy for an insurance company, we
were asked about the implications of change.
The answer is that there will always be new and evolving terms, but the high
level characterization of those terms (the top term, facet or meta-data
field) should stay somewhat stable. What do sales people, customer service
reps and claims processors need to understand? Well, certainly Product for
one. Coverage issues for another, etc. There are certain organizing
principles that naturally characterize information and that should stay
somewhat stable.
If there are completely new processes or business characteristics that
emerge over time, can those be described in the current framework (adding a
new hierarchy to an existing facet and selectively exposing that to the
user), or do we need to create new metadata facets and populate those?
The challenge is always about exposing the metadata to the UI both for
tagging and navigation and of course going back and adding the metadata that
was not captured before the requirement was identified.

There are strategies for dealing with this situation. We need to consider whether metadata can be derived from content or whether it needs to be applied to the content. In either case, search and navigation tools can then expose that information to help users find what they need to find to accomplish their task.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: