The circle is the oldest shape for cyclical data. Long before bar charts and line graphs, humans tracked seasons, tides, and planetary motion on circular instruments. The sundial. The astrolabe. The zodiacal wheel. The compass rose. The circle is not a stylistic choice. It is the correct answer to a particular kind of question: what happens, and when does it happen again?
A year of code is exactly that kind of question. The sprint cycles come back. Hacktoberfest lights up October; the December freeze dims it again. January brings its burst of resolution; February settles into the grind. You have lived this sequence before. The question is whether you can see it clearly enough to learn from it.
THE INSTRUMENT BEHIND THE LAYOUT
In the 2nd century CE, the Alexandrian astronomer Claudius Ptolemy described stereographic projection: a method for mapping the curved surface of the celestial sphere onto a flat disc. Every star, constellation, and coordinate line could be transferred from sphere to plane while preserving the angles between them. You could hold the sky in your hands.
The physical instrument that followed was the astrolabe. Medieval Islamic astronomers - al-Farghani, al-Battani, al-Zarqali - refined it into the most sophisticated scientific instrument of the pre-modern world: a brass disc engraved with a precise map of the visible stars. Rotate the overlay to match the current hour and latitude, and the position of every star in the sky was immediately legible. A thousand years of scholarship, encoded in a circle.
The European planisphere followed. In the 17th century, the Dutch cartographer Frederik de Wit produced hand-painted celestial planispheres mapping the northern and southern hemispheres of the sky - scientific instruments and objects of beauty simultaneously. In the 1680s, Jean-Dominique Cassini took the concept to its most literal extreme: a planisphere built into the floor of the Paris Observatory, twelve metres across, that visitors could walk across.
Today the planisphere is still manufactured as a rotating star-finding tool. Set the date and hour, and the window shows exactly which stars are above your horizon. The principle is unchanged from Ptolemy's original description. The circle persists because the problem persists: cyclical data belongs on a cyclical surface.
TWELVE MONTHS, TWELVE POSITIONS
A calendar year has twelve months. A clock face has twelve hours. The coincidence is convenient, but the deeper logic is geometric. Both divide a cycle into equal arcs. Both use position around a circumference to encode the passage of time.
The Planisphere layout places January at the top and proceeds clockwise, so that months orbit the circumference the way hours orbit a dial. Each month occupies thirty degrees of arc - a clean division of the full circle, a slice of the year made visible as a slice of the disc. Contributions within each month radiate inward toward the centre, with density encoded as stellar brightness: a heavy day sits near the circumference like a bright star near the horizon, a quiet day recedes toward the centre like a faint star near the zenith.
The result is that the year becomes a single legible object. You do not scan left to right, hunting for the month you want. You read the whole thing at once, the way you read a clock face. Twelve and six, three and nine: the eye finds its bearings immediately.
This geometry does something a sequential layout cannot: it places opposite months in direct visual relation. March sits across from September. June faces December. A developer who wonders whether their summer slowdown is real can compare it to the top arc at a glance - no scrolling, no memory required. The question becomes spatial rather than analytical, and spatial questions are the ones human vision answers effortlessly.
THE RHYTHM THE GRID CANNOT SHOW
The standard GitHub contribution grid is a rectangle: 52 columns of weeks, 7 rows of days, colour-coded from pale green to saturated green. It is honest and it is useful. It answers the question "did I commit today?" It does not answer the question "what is the shape of my year?"
That question matters more than it first appears. Most developers do not work at a constant rate. They work in rhythms: the intensive sprint before a release, the quiet aftermath when the adrenaline clears, the slow weeks in August when half the team is on holiday, the sudden cluster in October that Hacktoberfest always produces. These rhythms are the actual texture of a working life. They are more interesting and more revealing than any single day's count.
The grid represents those rhythms as a sequence, which means you have to hold them in working memory to perceive them as a pattern. You scan from January to December, you notice the dark patch in August, you notice the cluster of activity in March and again in October, and then you have to synthesise those observations into a shape. The shape exists in your mind, not in the image.
The Planisphere externalises the shape. The year is laid out in space, not in sequence. The March cluster and the October cluster are visible simultaneously, positioned symmetrically on either side of the disc. The dark patch in August sits at the bottom of the circle, a void in the lower arc. The eye does not need to synthesise. It sees.
BRIGHTNESS AND DARKNESS
Astronomers have always talked about the sky in terms of brightness and darkness. Hipparchus devised the magnitude scale in the 2nd century BCE: first-magnitude stars are the brightest, sixth-magnitude stars are the faintest visible to the naked eye. The language of stellar brightness became the standard vocabulary for describing the distribution of light across the sky.
The Planisphere layout borrows this vocabulary. Dense months glow. Sparse months dim. The visual effect is of a star field with concentrations and voids - a natural distribution, uneven in the way that real skies are uneven, rather than the uniform grid of a spreadsheet. Where the grid assigns colours on a fixed scale, the atlas assigns brightness on a relative one. The brightest point is always your most productive period; the rest is scaled accordingly.
This makes the layout particularly good at one task: revealing which months are consistently bright and which are consistently dark. When multiple years are overlaid, the structural patterns of a working life become unmistakable. If October is always dense and February always sparse, the circle makes that rhythm visible as a spatial fact. You stop asking whether the pattern is real; you can see that it is. The more interesting questions begin from there: why does Q1 always start strong? Is the post-launch quiet in May a recovery or a drift? Does the conference season correlate with a burst of new thinking, or does it drain the weeks around it?
Codex Stellarum renders those questions as light and shadow on a disc you can put on a wall. The numbers are there, encoded in the brightness of each point. But what you read is not numbers. You read the shape of how you work.
THOSE WHO CHOOSE THE CIRCLE
Developers who select the Planisphere layout tend to share a preoccupation. They are not primarily interested in their peak day or their total commit count. They are interested in how they work over time - in whether the rhythms they sense in their own practice are real, and whether the year, viewed whole, has a shape.
There is something clarifying about discovering that your working patterns are seasonal: that you always slow down in summer, always resurface in September, always burn bright in the weeks before end of quarter. These are not failures. They are the rhythm of a practice that has found its shape over years of iteration. Seeing them in space, not in sequence, is the first step toward understanding them.
The circular format rewards that kind of attention. A rectangle presents data as a list. A circle presents data as a pattern. The Planisphere does not invite you to count your contributions. It invites you to recognise them.
Ptolemy's stereographic projection was admired for two thousand years because it solved a genuine problem: the celestial sphere is real, but it cannot be examined directly. The projection made it examinable without making it false. The Planisphere layout does the same for a year of work. Once a year has passed, it becomes opaque: twelve months of decisions and effort compressed into vague impressions. The layout returns it to you as a legible object - a disc you can hold, a pattern you can read.
The circle persists because the problem persists.
Find the shape of your working year. Enter your GitHub username at codexstellarum.com and preview your Planisphere atlas - free to try, prints from £25.