Knowledge organization - Classification and User‑Centered Strategies

Facet Analytic and Analytico-Synthetic Approaches What Is Facet Analysis? Facet-analytic classification represents a major shift in how we organize information. Rather than forcing every subject into a single linear hierarchical structure, facet analysis breaks complex subjects down into their component parts—called facets—and allows these parts to be combined flexibly. S. R. Ranganathan introduced this revolutionary approach in 1933 through his Colon Classification system. His work fundamentally challenged the traditional view that all subjects could be arranged in a single hierarchical tree. The British Classification Research Group later refined and developed these methods, helping establish facet analysis as a standard approach in library science and information organization. Understanding Analytico-Synthetic Methodology The core of facet analysis is the analytico-synthetic process. Here's how it works: Analysis is the first step: you take a complex subject and break it down into its basic, irreducible concepts. Think of it like deconstructing a compound object into its ingredients. Synthesis is the second step: you recombine these basic concepts to describe the actual information package you're organizing. This reconstruction allows you to represent the same subject in different ways depending on which concepts are most important. The key insight is that by identifying the fundamental building blocks of subjects, you can construct descriptive systems that are far more flexible than traditional rigid classifications. For example, consider the subject "steel manufacturing in Japan during the 1970s." Through analysis, you'd identify the basic concepts: the material (steel), the process (manufacturing), the location (Japan), and the time period (1970s). Through synthesis, you could combine these concepts in different orders depending on what your users most want to search by—some might search by location first, others by time period. The PMEST Formula Ranganathan identified five fundamental facets that appear repeatedly across different subjects. This framework is called the PMEST formula, and understanding it is central to facet-based classification: Personality is the distinguishing characteristic that makes the subject what it is. This is the most specific defining feature. For example, in "steel manufacturing," steel is the personality—it's what makes this manufacturing process different from other types of manufacturing. Matter is the physical material that the subject is composed of or acts upon. In some subjects this overlaps with Personality, but it can be distinct. For instance, if you're classifying books about furniture, the matter might be wood, metal, or plastic—the materials used. Energy describes any action, process, or operation that occurs with respect to the subject. This is often a verb or process. In "steel manufacturing," the manufacturing process itself is the energy—the action being performed. Space represents the geographic or spatial component of the subject. Where does this subject occur or where is it located? In our steel manufacturing example, Japan is the Space facet. Time is the period or era associated with the subject. When is this happening? The 1970s would be the Time facet. These five categories appear in many different contexts because they represent fundamental ways that humans think about and distinguish subjects. When you classify something using PMEST, you're asking: What is it? What is it made of? What action occurs? Where? When? The order of facets matters too. Ranganathan established a citation order—the sequence in which facets should typically be combined—that helps ensure consistency. Generally, Personality tends to be most specific and comes first, followed by the other facets in a logical sequence. <extrainfo> The PMEST formula isn't the only possible framework for facets, and different classification systems may emphasize different facets depending on their domain. However, PMEST remains the most influential and widely recognized framework in information science. </extrainfo> Modern Approaches: User-Oriented Views and Folksonomies From Expert Classification to User-Generated Tagging Early research into how people actually search for and organize information revealed something surprising: users often prefer simpler, more intuitive methods over formal classification notations. This finding led to the development of user-oriented classification systems, which are designed based on empirical studies of how real people behave and what they prefer. One major challenge with traditional classification systems—even well-designed faceted ones—is that they require users to learn and navigate complex notation systems. The notation may be precise and logically structured, but users can find it intimidating or cumbersome. Folksonomies: A Collaborative Alternative The rise of the internet introduced a radically different approach to organizing information: folksonomies. Rather than having librarians or subject matter experts assign classification terms, folksonomies allow users themselves to create and assign indexing tags to information resources. When you tag a photo on social media, bookmark a website with custom tags, or label a post on a collaborative platform, you're participating in a folksonomy. The term combines "folk" (regular users) and "taxonomy" (classification system). Folksonomies have several important characteristics: Bottom-up rather than top-down: Users determine what terms matter, rather than librarians imposing a pre-designed structure Flexible and evolving: Tags can be created on the fly and change as language and needs evolve Potentially less precise: Without controlled vocabulary, different users might use different terms for the same concept, making searching less predictable Highly democratic: They reflect what many users actually think, rather than expert consensus The advantage of folksonomies is that they're intuitive and require no training. The potential disadvantage is inconsistency and lack of synonym control—when one user tags something "automobile" and another tags it "car," those tags don't connect automatically. Cognitive Considerations in Classification Design When designing classification systems, we must consider how humans actually use them. Cognitive considerations refer to understanding how users perceive, process, and search for information. A key insight is that even well-designed, logically-consistent classification notations can fail if users don't understand them or don't think to use them. For example, if your classification system uses precise notation like "621.3 → 72.4 → 19," but users expect to simply type "electric motors," the notation is useless no matter how logically perfect it might be. Modern systems increasingly balance formal organization (which ensures consistency and prevents information loss) with user-centered design (which ensures people can actually find what they need). Folksonomies succeed because they match user mental models, even if they sacrifice some organizational precision. Universal Organizational Principles: The LATCH Framework Five Universal Ways to Organize Information Richard Saul Wurman developed the LATCH framework to identify the fundamental ways that humans organize information. According to Wurman, virtually all information organization falls into one of five categories: Location organizes information by where something is or occurs. Examples include maps, geographic listings, or arranging books by library floor and aisle. Alphabet uses alphabetical ordering. This is perhaps the most familiar organizational principle—dictionaries, phone books, and alphabetical name lists all use this approach. It's intuitive to most people because we learn alphabetical order early, but it tells us nothing about the meaning or relationships of items. Time sequences information chronologically. Timelines, historical narratives, and chronological event lists all use Time organization. This is natural for narratives and historical information but less useful for static subjects. Category groups information by type or classification. This is what traditional classification systems do—putting all books about biology together, separating history from fiction. Category is powerful for finding similar items but requires a coherent organizational scheme. Hierarchy orders information from most important/general to least important/specific (or vice versa). Outlines, org charts, and the traditional Dewey Decimal system all use hierarchy. Hierarchy is effective for some subjects but can be limiting when subjects don't naturally fit a ranked structure. The LATCH framework is useful because it shows that most organizational challenges can be addressed through one (or a combination) of these five approaches. When designing an information system, asking "Which of these five principles will best serve my users?" is often a productive starting point. Notice that PMEST (from facet analysis) and LATCH represent different perspectives. PMEST identifies what content goes into a classification system, while LATCH identifies how to arrange any kind of information once you've decided what it contains. <extrainfo> Different organizational principles work better for different types of information and different user goals. A library might use Category (by subject) as its primary principle but also provide Alphabetical backup searching. A museum might use both Location (where artifacts are displayed) and Time (chronological development). Understanding this flexibility helps you design more effective information systems. </extrainfo>