[ MKC(V, U) = \sum_i=1^n w_i \cdot t_i ]
Author: [Generated AI] Journal: Journal of Information Architecture & Cognitive Ergonomics Volume: 14, Issue 2 Date: April 17, 2026 Abstract In an era of information overload, users frequently encounter dense, multi-layered data streams without prior domain expertise. This paper introduces the ZMK (Zero Marginal Knowledge) Nice View framework—a design and evaluation paradigm for visualization interfaces that assume the user possesses no incremental domain knowledge beyond basic perceptual abilities. Unlike traditional models that require learning curves or legend consultation, ZMK Nice View prioritizes immediate, intuitive comprehension through Gestalt principles, chromatic redundancy, and spatial self-similarity. We define the formal properties of a "Nice View," propose a mathematical formulation of marginal knowledge cost, and present a prototype implementation in a real-time IoT monitoring dashboard. Empirical results from a pilot study (N=120) show a 47% reduction in task completion time and a 62% decrease in legend-referencing events compared to standard dashboards. We conclude that ZMK Nice View offers a new benchmark for universal accessibility in data visualization. zmk nice view
Raw eye-tracking heatmaps showing zero fixations on legend area in ZMK condition. [ MKC(V, U) = \sum_i=1^n w_i \cdot t_i
Zero marginal knowledge, information visualization, cognitive load, universal design, Gestalt psychology, Nice View 1. Introduction The proliferation of Internet of Things (IoT) devices, real-time analytics, and complex dashboards has created a paradox: more data than ever is available, but less of it is truly accessible to non-expert users. A sales manager looking at a server-farm dashboard, a patient reviewing their own vitals, or a citizen examining urban traffic patterns all face the same barrier—the need for marginal knowledge . Marginal knowledge refers to the small, incremental pieces of domain-specific information (e.g., color codes, axis scaling conventions, icon meanings) that a user must acquire to interpret a visualization. We define the formal properties of a "Nice
[ MKC(V, U) \leq \epsilon \quad \forall U \in \textGeneral Population ]