The rule
Navy is agency. Teal is communion. Grey is the Self–Others axis. Nothing else is colored — not headings, not cards, not rules. Titles are type, not banners. The one inverted block is the action closer. Tufte’s short form: max[data], min[design].
What changed
| Tile | Was | Now |
|---|---|---|
| Every tile | 1.12 in navy band, teal stripe, 18-pt footer | Georgia takeaway on paper, a hairline, 12-pt grey footer |
| Framework | Four rounded tinted boxes | A coordinate cross. No boxes. Facets sit in the quadrants as type |
| Design | Four equal cards on a numbered rail | Two tracks on a 2016–2026 calendar: field study plus the 2026 model test, and the decade of nurses’ forum talk |
| Finding 1 | Thick bars, a titled dashed line | Cleveland dot + interval on 0–1. The number is a dataword. Communion’s wide interval is the loudest mark |
| Finding 2 | Colliding pass-mark labels, crop undeclared | Same slopegraph; pass marks as ticks; axis labeled “from .70” |
| Thesis / take-home | Rounded quote and card fills | A left rule; hairlines between rows |
| Digital pages (this pass) | Navy/teal/rust cards, thick HTML bars, tinted 2 × 2 boxes, teal kickers | Ink type and hairlines. Finding 1 is a Cleveland dot. The 2 × 2 is the figure, not four boxes. Pathway color only. One inverted closer |
Who the rule comes from
| Person | What the local text says | What we kept or dropped |
|---|---|---|
| Edward Tufte | max[data], min[design]; graphics as words, not boxes across the room; no data frames; smallest effective difference; Powsner and Tufte’s one-page clinical matrix. | Killed the banners and the rounded cards. Color is spent once. Did not merge Framework and Finding 1 onto one tile — the 4 × 4 hang still needs both. |
| Stephen Few | Length and 2-D position encode quantity. Bars need a zero baseline; dots do not. Tables for lookup, graphs for shape. No decoration that is not data. | Finding 1 is a Cleveland dot, not a bar. Finding 2 already was. |
| Cole Nussbaumer Knaflic | Takeaway titles, not “Findings.” One thing on stage. Audience first. | Titles stay as sentences. The filled banner went; the sentence did not. The action tile is still the one loud exception. |
| Alberto Cairo | Tufte’s rules are guidance within reason. Charts lie by cropping. Keep the interval you would rather hide. | Communion’s CI stays wide. Finding 2’s crop is declared on the axis. |
| Amanda Cox | Put an annotation layer on the graphic. Prefer “Huh? Wow!” over “Wow! Huh?” News graphics that feel like homework fail. | The .11 gap from .798 to the .910 agency bar is labeled on the mark. The timeline uses years, not step numbers. |
| Giorgia Lupi | Numbers stand in for people. Show that the data are imperfect. | Finding 3 still draws one mark per matched respondent (counts only). Untouched. |
| David McCandless | Interesting, easy, beautiful, and true. Missing the fourth tends to fail. | Two pathway colors a non-statistician can hold. Color never doubles as a verdict. |
| Mike Bostock | “Don’t get it original, get it right.” The document is the scenegraph. | Still HTML + CSS on the site, matplotlib + type on the tiles. No D3 showpiece. |
Manuscript figures
Manuscript figures stay on the APA track (no in-image titles; notes carry the legend). Finding 4 and 5 poster tiles use ink and grey, not pathway navy/teal.
Sources
- Tufte, E. R. (1992). Citation Classic commentary on The visual display of quantitative information. Notes: 01_tufte_full_readings.html.
- Powsner, S. M., & Tufte, E. R. (1997). Summarizing clinical psychiatric data. Psychiatric Services, 48(11), 1458–1461. Notes: 01_tufte_full_readings.html.
- Bostock, M., Ogievetsky, V., & Heer, J. (2011). D³: Data-Driven Documents. IEEE Transactions on Visualization and Computer Graphics, 17(12), 2301–2309. https://doi.org/10.1109/TVCG.2011.185. Notes: 02_bostock_d3_full_readings.html.
- Lupi, G. (2017, January 30). Data humanism, the revolution will be visualized. Print. Notes: 03_lupi_full_readings.html.
- Cox, A. (2012, June 1). Interview with Amanda Cox. Simply Statistics. Notes: 05_cox_full_readings.html.
- Knaflic, C. N. (2017, August 9). My guiding principles. storytellingwithdata.com. Notes: 06_knaflic_full_readings.html.
- Cairo, A. (2019–2022). Selected posts at thefunctionalart.blogspot.com. Notes: 04_cairo_full_readings.html.
- McCandless, D. (2010, July). The beauty of data visualization [TED talk]. Notes: 07_mccandless_full_readings.html.
- Few, S. Perceptual Edge white papers. Notes: 08_few_full_readings.html.