FODA as a Taxonomy Development Methodology

The Framework

Feature-Oriented Domain Analysis (FODA), originally designed for identifying commonalities across software product lines, can be repurposed as a rigorous method for developing taxonomies in domains with competing definitions.

Why This Matters

When a field has multiple overlapping definitions of a core concept (like “data quality” or “information architecture”), comparing them becomes difficult. FODA provides a structured approach: rather than arguing which definition is correct, it identifies the dimensions along which definitions differ.

The result is a feature model, a tree structure showing mandatory vs. optional features, alternatives, and dependencies. This makes explicit what was previously implicit variation.

How to Apply

1. Context analysis: Systematically collect all definitions in scope (via literature review)
2. Domain modeling: Identify basic domain knowledge that underlies all definitions
3. Feature identification: For each definition, identify which features it exhibits
4. Feature model construction: Organize features into a hierarchy showing relationships
5. Classification: Map each definition to its feature configuration

When to Use This

FODA is particularly valuable when:

• Multiple competing definitions exist with no clear consensus
• Definitions vary in scope, terminology, and focus
• You need to position a new definition relative to existing work
• A field would benefit from shared vocabulary for comparison

Limitations

• Requires substantial initial corpus of definitions
• Feature identification involves interpretation (different analysts may structure features differently
• The resulting taxonomy reflects the analyst’s judgment about what constitutes meaningful variation

Related: 04-atom—data-quality-consensus-gap