Machine-readable page content
Canonical: https://semiotic.nteract.io/examples/aesthetic-policy-studio
Aesthetic Policy Studio
01 · OUT OF THE BOXA capable baseline, not borrowed identity.
Semiotic’s balanced profile makes legibility and hierarchy count most. It allows the familiar default palette, but records that no organization-specific palette decision has been made. That is a useful default—and an intentionally incomplete design system.
Benefits resolution leads at 91%. Permit turnaround is lowest at 72%. All five services use the same Q2 snapshot and percentage scale.
Mark hierarchy100Primary data marks lead while reference scaffolding remains visible and subordinate.
Palette voice20The chart inherits a ubiquitous categorical default without an organizational palette decision.
Color economy1001 color role(s) are used against a policy ceiling of 7.
Type hierarchy100Title-to-supporting-type scale is 1.33×.
Theme coherence1000 of 0 explicit chart color(s) fall outside the theme vocabulary.
Editorial emphasis35No machine-visible editorial emphasis rule is declared.
THE NON-NEGOTIABLE FLOOR
Aesthetic disagreement begins after comprehension.
01Marks lead; grids recedeBoth outputs keep reference lines visible but subordinate and maintain at least twice as much contrast in the weakest data mark.
02Encoding remains honestBars retain a zero baseline. The tighter dot-plot domain is explicitly named as a comparison window rather than presented as magnitude from zero.
03Meaning survives styleThe data, title, description, summary, labels, and accessible table are identical or semantically equivalent in every output.
The receding grid is the clearest example. Stone and Bartram’s guidance supports a boundary—reference structure should remain visible without competing with the data—not a single fashionable grid color. Semiotic measures that relationship. Both organizations may choose radically different voices, but neither gets to make the apparatus louder than the evidence.
02 · TWO LEGITIMATE INSTITUTIONSOpposing weights. Different outputs. Same facts.
Each profile scores the same two candidate treatments. The higher-scoring treatment is emitted. This is not a model guessing which chart looks expensive; it is a reconstructable weighted decision over named, machine-visible evidence.
What leadership chose to rewardMark hierarchy5.0×
Palette voice0.5×
Color economy4.0×
Type hierarchy2.0×
Theme coherence4.0×
Editorial emphasis0.0×
SELECTED TREATMENTContinuity bar
Policy fit100/ 100 · floor 78
Benefits resolution leads at 91%. Permit turnaround is lowest at 72%. All five services use the same Q2 snapshot and percentage scale.
A zero-baseline bar chart and one durable color maximize routine comparability across a reporting estate.
The other treatment under this policy74 / 100The evidence is still valid; it simply fits this organization’s declared visual priorities less well.
What leadership chose to rewardMark hierarchy3.0×
Palette voice5.0×
Color economy0.5×
Type hierarchy4.0×
Theme coherence2.0×
Editorial emphasis5.0×
SELECTED TREATMENTEditorial dot plot
Policy fit100/ 100 · floor 78
Benefits resolution leads at 91%. Permit turnaround is lowest at 72%. All five services use the same Q2 snapshot and percentage scale.
A labeled comparison window, stronger headline, and one scarce accent create an editorial reading without altering the evidence.
The other treatment under this policy76 / 100The evidence is still valid; it simply fits this organization’s declared visual priorities less well.
THE DECISION RECORDEvery selection remains auditable.
Northstar rewards one-color economy and estate-wide coherence; it assigns no score to editorial emphasis. Fieldnote heavily rewards authored palette, typographic hierarchy, and a selective focal accent. Neither policy alters the underlying feature measurements.
Candidate treatmentNorthstarFieldnote
Continuity bar100 selected76
Editorial dot plot74100 selected