On Types of Typologies

A defense of the Myers-Briggs Type Indicator — not as science, but as a useful descriptive tool. The argument: MBTI needn't give you new information or carve personality at its true joints; it just condenses what you put in into discrete, human-processable bins, and even badly-drawn bins beat a 'correct' continuous model for that purpose. The European-countries analogy carries it: national borders are unscientific and error-ridden (Belgium 'making NO SENSE'), but 'Andreas is Greek, Franz is German' conveys far more usable information than a precise two-factor (latitude/longitude) coordinate. Five-Factor is built for research; MBTI for human-processable typing — and if you wanted Five-Factor to compete there, you'd have to bin it into Harry-Potter-meme 'OCIDN' types too.

Quality 67 · Strong. 67 — high-Strong. A clever, memorable reframe with a genuinely useful distinction (discrete human-processable bins vs continuous scientific models) and the illuminating European-countries analogy. A sharp insight rather than a major contribution, which places it high-Strong.

Claude’s paradigm shift 50 · Moderate. 50 — Moderate. The 'unscientific typologies are still useful as human-processable bins' framing is a fresh, memorable reframe of the MBTI-vs-Five-Factor debate.

Real-world impact 2 · Minor. 2 — a within-blog insight; mildly memetic, no material reach.