I think there’s a good case to be made that most think tanks approach taxonomies from the wrong perspective. That starts with the term taxonomy itself. If we want to be really technical, a taxonomy is a particular type of vocabulary—one that is expressed as a hierarchy, with parent and child terms.
A hierarchical taxonomy is a kind of controlled vocabulary in which each term is connected to a designated broader term (unless it is the top-level term) and one or more narrower terms (unless it is the bottom-level term), and all the terms are organized into a single larger hierarchical structure.
By contrast, an ontology documents relationships between terms. Ontologies are intended to describe an entire domain of knowledge. Here's Hedden again:
The relationships between terms within an ontology are not limited to broader/narrower and related. Rather, there can be any number of domain-specific types of relationship pairs, such as owns/belongs to, produces/is produced by, and has members/is a member of.
For the purposes of this project, I will follow Hedden in using "taxonomy" to describe any sort of knowledge management system and will use "hierarchical taxonomy" to call out things like the Linnaean system.
Deane Barker has a knack for getting at the heart of interesting problems. Here, in the space of a single tweet, he gets at the fundamental nature of content and taxonomy.
Much of information science has devoted itself to the attempt to create hierarchies of knowledge. And yet, most of those efforts have been doomed. Indeed, even the granddaddy of all taxonomies—The Linnaean classification—is the subject of fierce debate between systematists who wish to retain hierarchies and those who argue that a system without ranks better reflects the inherent messiness of biology.