Unknown Pleasures pulsar ridgeline plot - Quarto/R/ggridges reproduction

_brand.yml

color:
  palette:
    black: "#000000"
    white: "#FFFFFF"
    dark-gray: "#222222"
    gray: "#333333"

  foreground: white
  background: dark-gray
  primary: white
  secondary: gray

unknown-pleasures.qmd

---
title: "Unknown Pleasures: The Pulsar Plot"
author: "Reproduced in R"
respect-user-color-scheme: true
papersize: us-letter
format:
  typst:
    keep-typ: true
---

## The Original Visualization

The iconic cover art of Joy Division's 1979 debut album *Unknown Pleasures* is one of the most recognizable images in music history. But before it was album art, it was science.

### Harold D. Craft Jr. and CP 1919

The visualization was created by **Harold D. Craft Jr.**, a radio astronomer and doctoral student at Cornell University.
[Craft's thesis: *"Radio Observations of the Pulse Profiles and Dispersion Measures of Twelve Pulsars"* (September 1970)]{.aside}

The image depicts **80 successive radio pulses from CP 1919**---the first pulsar ever discovered.
[CP 1919: "Cambridge Pulsar at 19 hours 19 minutes right ascension"---the naming convention used before pulsars were well understood.]{.aside}
Each horizontal line represents one pulse period of approximately 1.337 seconds, stacked vertically to reveal patterns in the signal.

The data was collected at the **Arecibo Radio Observatory** in Puerto Rico.
[Craft worked under Frank Drake, famous for the Drake Equation estimating extraterrestrial civilizations.]{.aside}

### What is a Pulsar?
[A teaspoon of neutron star material would weigh about 6 billion tons on Earth.]{.aside}

Pulsars are rapidly rotating neutron stars---incredibly dense remnants of massive stars that have exploded as supernovae. They contain roughly 10^57 protons and neutrons compressed into a sphere about 10 kilometers in radius. As they rotate, they emit beams of radio waves like a lighthouse, creating the regular pulses detected by radio telescopes.
[CP 1919 rotates once every 1.337 seconds---relatively slow for a pulsar. The fastest known pulsar spins 716 times per second.]{.aside}

```{r}
#| label: setup
#| echo: false
#| message: false

library(ggplot2)
library(ggridges)
library(brand.yml)

# Load brand colors
brand <- read_brand_yml("_brand.yml")
```

```{r}
#| label: generate-data
#| echo: false

set.seed(1919)  # A nod to CP 1919

n_lines <- 80
n_points <- 300

# Generate data frame for ridgeline plot
pulsar_data <- data.frame(
 x = numeric(),
 y = numeric(),
 height = numeric()
)

for (i in 1:n_lines) {
 x <- seq(0, 10, length.out = n_points)

 # Create a base signal with some gaussian pulses
 # Main pulse in center with some variation
 center <- 5 + rnorm(1, 0, 0.3)
 width <- runif(1, 0.3, 0.6)
 amplitude <- runif(1, 0.5, 1.5)

 # Main pulse
 signal <- amplitude * exp(-((x - center)^2) / (2 * width^2))

 # Add secondary features (interpulse, noise)
 signal <- signal + runif(1, 0, 0.3) * exp(-((x - (center - 2))^2) / (2 * 0.4^2))
 signal <- signal + runif(1, 0, 0.2) * exp(-((x - (center + 1.5))^2) / (2 * 0.3^2))

 # Add noise
 signal <- signal + abs(rnorm(n_points, 0, 0.05))

 pulsar_data <- rbind(pulsar_data, data.frame(
   x = x,
   y = factor(i, levels = 1:n_lines),
   height = signal
 ))
}
```

```{r}
#| label: unknown-pleasures-plot
#| echo: false
#| fig-width: 8
#| fig-height: 10
#| fig-dpi: 150
#| column: page
#| out-width: 100%

# Filter to lines 2-80 to hide bottommost line (avoids artifacts)
plot_data <- pulsar_data[pulsar_data$y %in% 2:n_lines, ]

ggplot(plot_data, aes(x = x, y = y, height = height, group = y)) +
 geom_ridgeline(
   fill = brand$color$background,
   color = brand$color$foreground,
   scale = 2,
   linewidth = 0.5,
   min_height = 0  # Don't draw below baseline
 ) +
 scale_y_discrete(expand = expansion(mult = c(0, 0.05)), drop = FALSE) +  # 5% padding at top
 scale_x_continuous(expand = c(0, 0)) +
 coord_cartesian(clip = "on") +
 theme_void() +
 theme(
   plot.background = element_rect(fill = brand$color$background, color = brand$color$background),
   panel.background = element_rect(fill = brand$color$background, color = brand$color$background),
   plot.margin = margin(30, 30, 30, 30)
 )
```

### Path to Pop Culture
[Peter Saville removed the original caption and axis labels, reversed the colors (white lines on black), and rotated the image 90°.]{.aside}

1. **1970**: Craft's PhD thesis at Cornell
2. **1971**: Published in *Scientific American* (January issue)
3. **1974**: Appeared in *Graphis Diagrams*
4. **1977**: Included in *The Cambridge Encyclopaedia of Astronomy*
5. **1979**: Designer **Peter Saville** adapted it for Joy Division's debut album
[Joy Division took their name from the WWII term for groups of Jewish women forced into prostitution by the Nazis---a deliberately provocative choice that courted controversy.]{.aside}

When asked about finding his scientific work on an album cover, Craft said: *"I said no, I had no clue. So I went to the record store and, son of a gun, there it was."*

## Technical Notes

This reproduction uses:

- **R package**: `ggridges` version `r packageVersion("ggridges")` by Claus O. Wilke [The ggridges package was inspired by the "joy plots" trend that emerged after data scientists rediscovered this visualization style.]{.aside}
- **Visualization approach**: `geom_ridgeline()` for stacked line plots with filled areas
- **Data**: Simulated Gaussian pulses to mimic the original pulsar signal characteristics
[The hidden line algorithm was common in early computer graphics---each line is drawn front-to-back, with filled areas masking lines behind.]{.aside}

The original visualization by Harold Craft used a "hidden line" technique where peaks in the foreground obscure the lines behind them---exactly what `geom_ridgeline()` accomplishes with its fill parameter.

## References

- Craft, H.D. Jr. (1970). *Radio Observations of the Pulse Profiles and Dispersion Measures of Twelve Pulsars*. PhD thesis, Cornell University.
- Scientific American (2015). "Pop Culture Pulsar: Origin Story of Joy Division's Unknown Pleasures Album Cover"
- Wilke, C.O. ggridges: Ridgeline Plots in ggplot2. https://wilkelab.org/ggridges/