# A pattern two grids made and neither one owns

> A display shot with a camera instead of captured, its whole surface turned into interference bands that belong to neither grid that crossed to make them.

August 21, 2026 · https://justog.club/stories/the-screen-that-was-photographed

A square block of interference bands, curving and dense, fills the back of the shirt top to
bottom. Near the top of it, two pale rectangles sit inside the pattern — a window, reflected flat
on glass that the print has no glass to reflect off.

Whatever was actually on the screen is gone underneath it.

Nobody drew this pattern. It is what happens when a camera's own grid crosses a screen's grid at a
few degrees off true.

## Why does a photograph of a screen turn into this?

Because a camera sensor is a grid too, and pointing one regular grid at another regular grid,
slightly out of alignment, makes a third pattern that belongs to neither of them.

A screen is built from a fixed spacing of pixels or a fixed spacing of scan lines; a sensor is
built from its own fixed spacing of photosites. Line the two up dead straight and they mostly
cancel out. Tilt the camera even a little and the two spacings interfere, and the interference
repeats at a completely different, much larger scale than either grid on its own — bands, not
pixels. The same thing happens whenever [a screen gets shot instead of
captured](/stories/the-second-generation-of-a-screenshot); this is what it looks like
when nobody corrects for it.

## What is the pale shape near the top?

A window, caught in the glass of the screen rather than on it.

A screen's surface reflects like any other pane of glass would, and a phone held up to a monitor
photographs the room in front of it along with whatever the monitor is showing. A real screen
capture has no glass, so it can never pick up a reflection — there is nothing between the pixels
and the file. Finding one here is the second, independent proof that a camera stood in the room
and took this, rather than the screen handing over its own data directly.

## Why photograph a screen instead of just saving the file?

Because saving the file usually means finding the right menu, the right cable, or access nobody
granted, and a camera is already in every pocket.

Somebody wanted proof that a number, a message, a screen was showing something at a given moment,
and the fastest tool for that is whatever device is already in their hand. It is quicker than
asking for the export, and it works on a screen that isn't even theirs to begin with — a kiosk, a
shared terminal, somebody else's monitor held up over a shoulder.

## Does it matter that nothing on the screen can actually be read?

No — reading it back isn't the point of a photograph taken this way.

The bands don't obscure a caption anybody needed later; they obscure whatever happened to be
open at the time, and the person taking the photo already knew what that was. The picture exists
to record that the moment happened, not to preserve the content of it. Unreadable is the expected
outcome, not a failure somebody would have tried to avoid.

## Who owns a pattern made by two grids crossing?

Nobody — the bands are the mathematics of two spacings interfering, not a drawing.

Nothing in this design is lifted from a real screen, a real interface or a real photograph.
The window reflected in the glass is invented, with nothing behind it identifying a place or a
maker. A design built out of an optical effect belongs to whoever set the effect down, the same
way a fault built out of a machine's own physics does anywhere else on this site. Style has no
owner. A work does.

## Isn't this just a badly taken photo?

It looks like one, and that's exactly why it's worth keeping rather than retaking.

A badly taken photo is an accident somebody would fix given the chance — refocus, hold it
straighter, try again. This pattern isn't an accident in that sense. Any two regular grids at a
slight angle produce the same family of bands every time, which makes it a repeatable fact about
optics rather than a mistake anyone made once and could have avoided by being more careful.

## Does it print well?

Yes, with one adjustment: the bands need to stay bands, not thin down into hairlines.

Flat colour with a hard edge is the safe case, and the block itself is one — a full hard-edged
square, which here is the point rather than a flaw, since the square is standing in for the
screen's own bezel. The curves inside it are reproduced a shade coarser and a shade wider than the
source photograph's, so the weave holds them as visible bands rather than smearing them to grey.
Two inks: a blue-grey for the pattern and a pale ground for the field and the reflected window.
Charcoal, ink navy and dark stone grounds all carry it; a pale garment loses the contrast the bands
need to read at a distance.

## How do I get one made?

Describe the screen at [JustOG](/) — what's showing behind the bands, and how much of a
reflection to leave in the glass.
Pick a direction, drag the crop frame, see it composited on the real garment, and it's made to
order and shipped.

Designs other people have published are in the [shop](/shop).

Two rectangles of window, flat on glass, over a pattern that came from two grids crossing, not
from the screen.