ON THE MARK

What entropy actually looks like

A letterform can stand for a name. It can't stand for an idea. Before we could design a mark for Entropy, we had to be precise about what entropy actually is, physically and informationally, and only then ask what it looks like when you draw it.

THE IDEA

In thermodynamics, entropy is a measure of disorder: the number of ways a system's parts can be arranged and still look the same from the outside. Left alone, every system drifts toward the arrangement with the most disorder, because there are vastly more disordered states than ordered ones. That drift is irreversible. It's the reason heat spreads instead of concentrating, the reason a dropped glass shatters but never reassembles, the reason time has a direction at all.

Claude Shannon borrowed the same word and the same mathematics for information theory: entropy as a measure of uncertainty, or noise, in a signal. A clean signal carries structure you can predict. A degraded one carries less and less of it, until what's left is indistinguishable from static.

Both readings point at the same thing from different angles: order is a temporary, effortful state. Disorder is where everything goes if you stop doing work to prevent it. For an engineering studio, that's not a bleak name, it's the whole job description. Production systems don't stay correct on their own. Someone has to keep pushing back against the drift.

THREE WAYS TO DRAW IT

Three physical metaphors, one idea

Particle dispersion

A tight, regular grid, a lattice at rest, breaks rank. Reading outward from it, the squares become circles, drift, shrink, and fade into scatter. The most literal illustration of the thermodynamic reading: order as a held breath, disorder as what happens when you let go.

Heat death

A bright point sends out rings that thin, break, and fade with distance, dissolving into an even scatter of dust. Maximum entropy isn't stillness, it's the loss of any structure left to read. The most precise illustration of the physics, at the cost of a weaker read at very small sizes.

Signal degradation

A clean waveform enters, jitters, fragments, and trails into isolated noise. The informational reading, and the one that speaks most directly to signal-to-noise in trading systems and consensus protocols, not just physics.

THE DECISION

We chose particle dispersion

Heat death is the truest physics, but a dashed ring that thins to nothing is the first thing to disappear at 16 pixels, and a mark that only works at full size isn't a usable mark. Signal degradation has the most direct read for what this studio actually builds, but its payoff is in the fragments on the right, and fragments are exactly what a favicon-sized render loses first.

Particle dispersion survived the stress test. The grid gives the eye a fixed point to anchor on even when the mark is shrunk down to a browser tab, and the scatter around it still reads as scatter, not as noise, at every size in between. It's also the most direct visual match for the word we started with: dispersal.

It's the mark everywhere Entropy appears: the navigation, the footer, the brand section, and the favicon in your browser tab.

The diagrams above are the original sketches: flat colour on a dark tile, built to compare three ideas quickly. The mark actually in use went a step further, rendered by hand, in red, on white, with real dimension. That's a separate decision with its own reasoning: see the logo design →

THE ORIGINAL SKETCH
96px
48px
32px
16px