MLC Isometric Heatmap

Chapter 6: Zero Is Not Nothing

The distinction between a measured zero and an absent measurement is the most critical principle in this library — and the one data visualizations blur most frequently.

Value Rendering
Number Column reflecting the value's height and palette color
0 Flat tile on the floor in zeroColor or the theme's base color
null Nothing — only bare floor grid lines, no 3D geometry, no tooltip

A zero means: the sensor or machine was online and recorded zero output. A null means: we do not know. Conflating the two transforms an outage into a deceptively calm night.

That a null cell receives no tooltip is deliberate: there is nothing to report, and an empty tooltip would invent information where none exists.

How the Shapes Handle Missing Values

prism: the gap is an empty slot in the grid, unmistakable.
cylinder: identical semantics with cylindrical geometry.
ribbon: the curve breaks and restarts cleanly rather than interpolating across the void.
flatribbon: the floating ribbon also pauses — drawing a continuous line would invent fictitious telemetry.

This final point is paramount: for continuous geometries it would be trivial to bridge across gaps automatically. The resulting image might look smoother, but it would be dishonest.

Holes in the Terrain

In a 3D terrain mesh, missing values create clear voids in the surface. This is the most visually striking way of stating: "data is missing here".

A terrain relief with a cut-out. The edges remain clean because no polygons are stretched across missing data points.

Licence: CC BY-NC 4.0