tufte-viz

name	tufte-viz
description	Use when designing or critiquing charts, dashboards, reports, or other data visualizations; choosing between visualization approaches; reducing chartjunk; improving data-ink ratio; or planning small multiples and high-density displays.
metadata	{"short-description":"Tufte-style viz design and critique with personal house style"}

Tufte Visualization Ideation

Apply Edward Tufte's principles to design clear, honest, high-density data visualizations, layered with the user's house style.

Load Preferences First

ALWAYS read preferences.md in this skill directory before producing or critiquing a chart. The user's preferences override generic Tufte defaults wherever they conflict, except on graphical integrity (lie factor, honest scales, etc.), which is non-negotiable.

Workflow

For new visualizations

1. Clarify the data story

   • What comparisons matter?
   • What is the key insight to communicate?
   • Who is the audience?

2. Select an approach using Tufte principles plus the user's preferences:

   • High comparison need -> small multiples
   • Dense data -> data tables, sparklines
   • Time series -> line charts with minimal grid
   • Part-to-whole -> avoid pie charts; prefer bar or table

3. Design with data-ink in mind

   • Start minimal, add only what is necessary
   • Every element must earn its ink
   • Default to grayscale; use color purposefully (subject to preferences.md)

4. Apply the Tufte test (see references/tufte-principles.md)

For critiquing visualizations

1. Check graphical integrity

   • Calculate lie factor if proportions seem off
   • Verify baselines and scales
   • Look for 3D distortion

2. Identify chartjunk

   • Decorative elements
   • Heavy grids
   • Unnecessary 3D effects
   • Moire patterns

3. Evaluate data-ink ratio

   • What can be erased?
   • What is redundant?

4. Suggest improvements with specific before/after recommendations, aligned with preferences.md.

References

• preferences.md -- the user's house style. Load first; it overrides generic guidance where it conflicts (except integrity).
• references/tufte-principles.md -- core principles from The Visual Display of Quantitative Information: lie factor, data-ink, chartjunk, small multiples, integrity.
• references/analytical-design.md -- extensions from Envisioning Information, Visual Explanations, and Beautiful Evidence: the six principles of analytical design, sparklines, layering and separation, micro/macro, range-frames, causality, confections. Load when designing dashboards, dense displays, sparklines, or explanatory graphics.

Quick checklist

• Lie Factor approx 1.0 (no visual distortion)
• Maximum data-ink ratio
• Zero chartjunk
• Clear labeling
• Answers "compared to what?"
• Shows causality or mechanism where relevant
• Multivariate (not over-reduced)
• Words, numbers, images integrated, not segregated
• Reveals multiple levels of detail (micro + macro)
• Layering: primary data dominates, secondary recedes
• Appropriate data density
• Consistent with preferences.md