Settle — late revision
Variant: Late revision
Every metrics pipeline lies about yesterday: late events have not landed, so the tail of the line dips and someone files an incident. Here the last three points are dashed, hollow and sitting in a shaded zone whose left edge is the settling front. Two late batches arrive; the points rise to their real values on a 432ms ease-in-out — they are already on screen, so they morph rather than enter — and only after the geometry stops does the oldest one harden, dash to solid, hollow to filled. Today never hardens: one open point is the steady state. Each segment is a horizontal line placed with translate() and tilted with skewY(), so both ends move on transforms alone and it plays in Safari, which will not animate a path. Reduced motion keeps the revision, slowed, as a deliberate exception: a chart that never corrects its own dip makes the opposite claim. Use it wherever the newest data is the least complete.
The source
A complete HTML document. Paste it into an empty .html file and it plays — no stylesheet, script, font, or image to fetch.
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<title>Settle — late revision</title>
<style>
:root {
/* Neutrals — fixed, never remixed. */
--ivory: #faf9f5;
--paper: #ffffff;
--oat: #e3dacc;
--slate: #141413;
/* Knobs — the remix API. A knob exists iff declared here and used below. */
--hue: 38.8;
--lift: 0;
--chroma: 1;
--speed: 1;
/* Accents — derived from the knobs with the family's offsets baked in,
so one hue roll moves them together. Defaults reproduce the original hex. */
--clay: oklch(calc(0.6724 + var(--lift)) calc(0.1308 * var(--chroma)) var(--hue));
--clay-d: oklch(calc(0.5797 + var(--lift)) calc(0.127 * var(--chroma)) var(--hue));
--olive: oklch(
calc(0.6118 + var(--lift)) calc(0.0713 * var(--chroma)) calc(var(--hue) + 88.3)
);
--rust: oklch(
calc(0.5408 + var(--lift)) calc(0.1357 * var(--chroma)) calc(var(--hue) - 10.3)
);
--_muted: color-mix(in srgb, var(--slate) 52%, transparent);
--_grid: color-mix(in srgb, var(--oat) 70%, transparent);
--_zone: color-mix(in srgb, var(--oat) 34%, transparent);
/* Clock — 7.2s: 1% is 72ms. Late batches land at 26% and 54%; each
point moves first and hardens 144ms after it arrives. */
--cycle: 7.2s;
--t: calc(var(--cycle) / var(--speed));
--ease-out: cubic-bezier(0.23, 1, 0.32, 1);
--ease-in-out: cubic-bezier(0.77, 0, 0.175, 1);
}
* {
box-sizing: border-box;
margin: 0;
padding: 0;
}
body {
min-height: 100vh;
display: grid;
place-items: center;
background: var(--ivory);
font-family:
system-ui,
-apple-system,
'Segoe UI',
Roboto,
sans-serif;
color: var(--slate);
}
.chart {
width: min(300px, calc(100vw - 40px));
display: grid;
gap: 8px;
animation: frame var(--t) linear infinite;
}
.label {
font-size: 11px;
font-weight: 600;
letter-spacing: 0.14em;
text-transform: uppercase;
color: var(--_muted);
}
svg {
display: block;
width: 100%;
height: auto;
overflow: hidden;
}
.grid line {
stroke: var(--_grid);
stroke-width: 1;
}
.base {
stroke: var(--oat);
}
/* The unsettled region, not a highlight: its left edge is the settling
front, and it gives up one day per batch. */
.zone {
fill: var(--_zone);
animation: zone var(--t) linear infinite;
}
.settled {
fill: none;
stroke: var(--clay);
stroke-width: 2;
stroke-linejoin: round;
stroke-linecap: round;
}
.dot {
fill: var(--clay);
}
/* Each provisional segment is a 36-unit horizontal line placed with
translate() and tilted with skewY(): skew keeps x and slides y in
proportion, so the far end lands exactly where the next point sits.
It moves both ends with transforms alone — no path morph, which
Safari will not animate. The angle tweens in degrees while the dot
tweens in y, so mid-move a join can drift by a third of a unit —
inside the dot's radius, which covers it. */
.seg line {
stroke-width: 2;
stroke-linecap: round;
stroke: var(--clay);
}
.seg .dash {
stroke-dasharray: 3 4;
}
.seg .firm {
opacity: 0;
}
.s56 {
transform: translate(196px, 56px) skewY(21.25deg);
animation: s56 var(--t) linear infinite;
}
.s67 {
transform: translate(232px, 70px) skewY(26.57deg);
animation: s67 var(--t) linear infinite;
}
.s78 {
transform: translate(268px, 88px) skewY(18.43deg);
animation: s78 var(--t) linear infinite;
}
.s56 .dash {
animation: soft1 var(--t) linear infinite;
}
.s56 .firm {
animation: firm1 var(--t) linear infinite;
}
.s67 .dash {
animation: soft2 var(--t) linear infinite;
}
.s67 .firm {
animation: firm2 var(--t) linear infinite;
}
.pt circle {
stroke: var(--clay);
stroke-width: 1.5;
fill: var(--paper);
}
.p6 {
transform: translate(232px, 70px);
animation: p6 var(--t) linear infinite;
}
.p7 {
transform: translate(268px, 88px);
animation: p7 var(--t) linear infinite;
}
.p8 {
transform: translate(304px, 100px);
animation: p8 var(--t) linear infinite;
}
/* Hollow means "still counting"; the fill is the point being vouched for. */
.p6 circle {
animation: fill1 var(--t) linear infinite;
}
.p7 circle {
animation: fill2 var(--t) linear infinite;
}
.key {
display: flex;
flex-wrap: wrap;
gap: 4px 14px;
font-size: 11px;
color: var(--_muted);
}
.key span {
display: inline-flex;
align-items: center;
gap: 6px;
}
.key i {
width: 16px;
border-top: 2px solid var(--clay);
}
.key .k-dash {
border-top-style: dashed;
}
/* Everything here is already on screen and changing value, so it morphs
on ease-in-out. The geometry settles first; the ink hardens after. */
@keyframes p6 {
0%,
26% {
transform: translate(232px, 70px);
animation-timing-function: var(--ease-in-out);
}
32%,
100% {
transform: translate(232px, 52px);
}
}
@keyframes p7 {
0%,
26% {
transform: translate(268px, 88px);
animation-timing-function: var(--ease-in-out);
}
32%,
54% {
transform: translate(268px, 66px);
animation-timing-function: var(--ease-in-out);
}
60%,
100% {
transform: translate(268px, 58px);
}
}
/* Today never settles. It rises as the batches land and stays hollow —
one open point is the steady state, not a glitch. */
@keyframes p8 {
0%,
26% {
transform: translate(304px, 100px);
animation-timing-function: var(--ease-in-out);
}
32%,
54% {
transform: translate(304px, 80px);
animation-timing-function: var(--ease-in-out);
}
60%,
100% {
transform: translate(304px, 62px);
}
}
@keyframes s56 {
0%,
26% {
transform: translate(196px, 56px) skewY(21.25deg);
animation-timing-function: var(--ease-in-out);
}
32%,
100% {
transform: translate(196px, 56px) skewY(-6.34deg);
}
}
@keyframes s67 {
0%,
26% {
transform: translate(232px, 70px) skewY(26.57deg);
animation-timing-function: var(--ease-in-out);
}
32%,
54% {
transform: translate(232px, 52px) skewY(21.25deg);
animation-timing-function: var(--ease-in-out);
}
60%,
100% {
transform: translate(232px, 52px) skewY(9.46deg);
}
}
@keyframes s78 {
0%,
26% {
transform: translate(268px, 88px) skewY(18.43deg);
animation-timing-function: var(--ease-in-out);
}
32%,
54% {
transform: translate(268px, 66px) skewY(21.25deg);
animation-timing-function: var(--ease-in-out);
}
60%,
100% {
transform: translate(268px, 58px) skewY(6.34deg);
}
}
@keyframes zone {
0%,
32% {
transform: translateX(0);
animation-timing-function: var(--ease-in-out);
}
37%,
60% {
transform: translateX(36px);
animation-timing-function: var(--ease-in-out);
}
65%,
97% {
transform: translateX(72px);
}
98%,
100% {
transform: translateX(0);
}
}
@keyframes soft1 {
0%,
34% {
opacity: 1;
animation-timing-function: var(--ease-out);
}
37%,
97% {
opacity: 0;
}
98%,
100% {
opacity: 1;
}
}
@keyframes firm1 {
0%,
34% {
opacity: 0;
animation-timing-function: var(--ease-out);
}
37%,
97% {
opacity: 1;
}
98%,
100% {
opacity: 0;
}
}
@keyframes soft2 {
0%,
62% {
opacity: 1;
animation-timing-function: var(--ease-out);
}
65%,
97% {
opacity: 0;
}
98%,
100% {
opacity: 1;
}
}
@keyframes firm2 {
0%,
62% {
opacity: 0;
animation-timing-function: var(--ease-out);
}
65%,
97% {
opacity: 1;
}
98%,
100% {
opacity: 0;
}
}
@keyframes fill1 {
0%,
34% {
fill: var(--paper);
animation-timing-function: var(--ease-out);
}
37%,
97% {
fill: var(--clay);
}
98%,
100% {
fill: var(--paper);
}
}
@keyframes fill2 {
0%,
62% {
fill: var(--paper);
animation-timing-function: var(--ease-out);
}
65%,
97% {
fill: var(--clay);
}
98%,
100% {
fill: var(--paper);
}
}
/* The geometry snaps back at 97–98%, once the frame is fully dark. */
@keyframes frame {
0% {
opacity: 0;
animation-timing-function: var(--ease-out);
}
3%,
92% {
opacity: 1;
animation-timing-function: var(--ease-out);
}
97%,
100% {
opacity: 0;
}
}
@media (prefers-reduced-motion: reduce) {
/* The revision is the reading — a chart that never corrects its own
dip makes the opposite claim — so the points still move, on a 1.6x
clock every part shares so segments and dots stay joined. */
.chart,
.chart * {
animation-duration: calc(var(--t) * 1.6);
}
}
</style>
</head>
<body>
<div
class="chart"
role="img"
aria-label="A line chart of orders per day whose last three points are dashed and hollow and dip sharply; two batches of late data arrive, the points rise to their real values, and the older two turn solid while today stays open"
>
<span class="label">Orders / day</span>
<svg viewBox="0 0 320 120" aria-hidden="true">
<g class="grid">
<line x1="0" x2="320" y1="30" y2="30" />
<line x1="0" x2="320" y1="60" y2="60" />
<line x1="0" x2="320" y1="90" y2="90" />
<line class="base" x1="0" x2="320" y1="119.5" y2="119.5" />
</g>
<rect class="zone" x="214" y="0" width="120" height="119" />
<polyline class="settled" points="16,78 52,70 88,74 124,62 160,66 196,56" />
<g class="seg s56">
<line class="dash" x2="36" />
<line class="firm" x2="36" />
</g>
<g class="seg s67">
<line class="dash" x2="36" />
<line class="firm" x2="36" />
</g>
<g class="seg s78"><line class="dash" x2="36" /></g>
<circle class="dot" cx="16" cy="78" r="2.6" />
<circle class="dot" cx="52" cy="70" r="2.6" />
<circle class="dot" cx="88" cy="74" r="2.6" />
<circle class="dot" cx="124" cy="62" r="2.6" />
<circle class="dot" cx="160" cy="66" r="2.6" />
<circle class="dot" cx="196" cy="56" r="2.6" />
<g class="pt p6"><circle cx="0" cy="0" r="3.2" /></g>
<g class="pt p7"><circle cx="0" cy="0" r="3.2" /></g>
<g class="pt p8"><circle cx="0" cy="0" r="3.2" /></g>
</svg>
<div class="key">
<span><i></i>Settled</span>
<span><i class="k-dash"></i>Late data still landing</span>
</div>
</div>
</body>
</html>
The variant family
Same idea, one axis moved. Compare them side by side before you commit.