Metadata - Words or relationships

After a fun filled flight that consisted of a number of delays I arrived in San Francisco to hear the last presentation of the day by Erik Hartman. He was speaking on classifying web content. Below are some points that I took home:

1. Metadata can be grouped as optional, required or automated. The optional files are the challenging ones to get filled out.

2. He described the KM challenge as: “You cannot manage knowledge, you can manage people that have knowledge.” He also presented an equation for KM that basically is:

Knowledge = Information x (characteristics, skills, and attitudes of people)

3. He spent a lot of time talking about the differences between taxonomy and ontology. What I heard as the difference is that taxonomy relies more on hierarchies and ontologies focus on relationships. He made comments like (I hope I captured these correctly):
- Relationships still need a language
- People want the relationships, more than the hierarchies.
- “People don’t want a drill, they want a hole in the wall.”
- We’re not after the semantic web, we want the pragmatic web.
- Ask two experts and you get a different taxonomy. So how do we expect any taxonomy to fit all users?
In pondering these last comments I started thinking about our current path. We’re focusing on the words now, but in the future the relationships between the words is where we will find the value