# Four tries at the same photograph

> A two by two grid of four near-identical photographs of the same two children, one figure changing posture across them until the last panel settles.

August 21, 2026 · https://justog.club/stories/the-photo-where-one-of-them-is-crying

Four photographs, same white border, same two seated figures, arranged in a strict two by two
grid. Look for the difference and it takes a moment to find.

Across the four attempts, one figure shifts. By the fourth panel it has settled.

## Why print four attempts at the same photograph instead of the one that came out best?

Because the four together are the record of how long it actually took, and the fourth panel alone
would just look like luck.

Anybody who has tried to get two children to sit still and face a camera at the same instant knows
roughly what these four cost — not in any exact count, but in the general shape of it: try, wait,
try again. Printing only the keeper erases that. Printing all four keeps the price attached to the
result.

## What's actually different between the four panels?

One of the two figures changes posture between them, in small enough steps that the four read as
close cousins of each other rather than four distinct scenes.

The differences sit at a scale that rewards a second look rather than announcing themselves. That
closeness is deliberate — a photograph taken four times in quick succession doesn't produce four
wildly different pictures, it produces four almost-identical ones, and the print matches that
rather than exaggerating the gap between panels for clarity.

## Why does the fourth panel get to be the one everyone keeps?

Because it's the one where the shifting figure has settled, which is the only thing the four
panels were actually waiting for.

None of the first three is labelled a failure anywhere in the design — all four carry the same
white border, the same size, the same weight. The fourth isn't rewarded for being correct; it's
simply where the sequence stopped, because a fifth attempt stopped being worth asking for, in the
same way a joint celebration stops once [a cake iced by one hand, minutes apart](/stories/the-cake-with-two-names-on-it)
is finished rather than redone until it comes out even.

## Is the joke that the photos look almost the same?

Yes — the four panels are close enough to each other that spotting the actual difference is the
entire point, in the same way [the difference in a photo where one child faces the camera and the
other doesn't](/stories/only-one-is-looking-at-the-camera) rewards a second look rather
than a first one.

A grid of four dramatically different photographs would tell a bigger, blunter story and a less
familiar one. Four photographs this close together is what an actual run of attempts looks like,
and the near-sameness is doing more work than any obvious difference could.

## Does this need to be four panels, or would two work?

Two shows a change; four shows a process. The extra panels are what turn "before and after" into
something that reads as effort rather than a simple contrast.

A two-panel version — first attempt, last attempt — would still communicate that something
improved between them, but it loses the sense of a run that had to be sat through. Four is close
to the minimum that still reads as a grid rather than a pair, without asking a reader to count
past what a shirt can hold — [a twelve-panel version of the same idea, built for a whole
group instead of two](/stories/the-one-good-one-of-everybody) marks its single usable
frame rather than its last one, because with more people in the shot there usually isn't a clean
progression to show.

## Do I need permission to print a sequence of photographs like this?

If the photographs are your own, no. If a real sequence belongs to somebody outside your
household, that permission is theirs to give, the same as any other family photograph.

This design is built from an invented pair of silhouettes rather than a specific family's actual
photographs, which is most of what lets it work for any two children rather than one particular
pair. A studio or a designer working from somebody else's real photo strip is licensing a real
sequence, not inventing one, and that's a different job with a different owner.

## Does it print well?

Yes, and the grid does most of the work. Four equal white-bordered panels in two rows read as a
contact sheet immediately, before a viewer even looks at what's inside them.

Keep each panel's border weight identical — a thicker or thinner border on any one panel reads as
an error rather than a deliberate difference. The figures inside stay flat and hard-edged, with
only the smallest shift in posture carried between panels; anything larger stops reading as the
same photograph taken four times and starts reading as four different photographs.

## How do I get one made?

Describe the sequence at [JustOG](/) — how many attempts it actually took, and what finally
settled.
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).

The other three are still on the camera roll, further down.
