The dead tube on the noodle stall's menu board

Six rows. Small photographs, faded to a colour that isn't quite the food's any more. Names and prices on clip-in numerals. SOLD OUT slapped diagonally across the braised tofu. One tube behind row three has gone dim, and nobody's swapped it yet.

This is the part of the street nobody prints.

Why is a menu board the interesting object here, not the skyline behind it?

Because the menu board is what customers actually look at, and the skyline is what customers walk past without noticing.

Everybody who's stood at a noodle stall has read a menu exactly like this one — scanned the faded photos, checked which numeral got taped over, decided based on the one dish that still looks appealing under bad light. Nobody at that stall is looking up at the buildings. The stall is the most-used location in the genre, and almost none of what's printed from it is the object people are actually reading while they stand there deciding what to order.

Why does a faded photograph read stronger than a rendered one?

Because fading is evidence of time passing, and a rendered image can claim it without proving anything.

The mackerel in row one hasn't been that colour since the panel was installed — years of a fluorescent tube behind acetate has pushed every photograph toward the same washed amber-and-cyan cast, uniform across all six, which is exactly what happens to backlit signage nobody's replaced in a long time. A design can render that colour shift deliberately, but the honesty of it depends on consistency: every photo shifted the same way, at the same rate, because they've all sat behind the same ageing light for the same stretch of time. That consistency is what a viewer reads as real rather than styled.

What the clip-in prices add

Numerals that slot into a track rather than being printed fixed are their own piece of evidence — a board built to be corrected constantly, by hand, by whoever's behind the counter that night. A price that's been bumped, tilted slightly off the row, or replaced with a numeral one shade off from the rest, says the stall has been open long enough to need adjusting more than once.

What does the dead tube do that the working rows don't?

It's proof the stall keeps operating imperfectly rather than stopping to fix itself.

Row three, dimmer than the other five, isn't a design flourish — it's the kind of thing a real proprietor notices, means to deal with, and doesn't get to for another few weeks because the stall's still open every night regardless. The SOLD OUT tab does the same job from a different angle: a board that's still being used, tonight, by someone actually running the counter, rather than a polished sign standing in for the idea of one.

Is it fine to invent a stall like this, or does it need to be a real one?

Invented is not just fine here, it's the stronger version — a real restaurant's menu, name and photography belong to that restaurant, and copying an actual stall's signage is copying somebody's livelihood rather than borrowing an aesthetic.

None of the six dishes, the prices or the stall's identity need to reference anything real to land. A menu this specific — six dishes, one sold out, one dim tube — reads as found rather than staged because of the specificity, not because it's actually somebody's real board. Invent the dishes, invent the prices, and the format still does everything it needs to do.

Does the format work outside a noodle stall specifically?

Yes — any real place with a lit menu and a counter runs on the same object, which is the same argument this genre's convenience-store signage makes in a different country.

A fry stall, a late bar's drinks board, a market stand with prices written on masking tape over the last set of prices: the mechanics repeat. A board meant to be corrected, ageing under artificial light, run by somebody who's open later than they'd probably like to be.

Doesn't a menu board feel too mundane for the genre?

That's the point, not a weakness in it — the genre's actual argument is that ordinary commerce is where the interesting texture lives, not in the architecture above it.

A skyline is scenery. A menu somebody's actually reading while deciding what to eat for nine dollars is a document about a real transaction, priced, corrected by hand, currently missing one dish and running one dim tube behind the glass. That's a smaller claim than a skyline makes, and it's a truer one, which is the same reasoning behind letting a chest print carry one specific object rather than a whole scene.

Does it print well?

Yes, with the ageing colour cast doing most of the work already, and the clip-in numerals worth keeping crisp against it.

A short, warm-shifted palette — amber, faded cyan, a worn cream ground — holds together as a limited set of flat tones rather than six independently rendered photographs, which is the same case for a short palette printing more reliably than a literal one. The clip-in numerals and the SOLD OUT tab should stay hard-edged and legible against the softer, faded photography around them, since that contrast between crisp type and worn image is what sells the board as a real object rather than a photo of one.

Amber, warm cream or rust suit the base closely, echoing the panel's own backlit cast. Three or four inks cover it: the faded photo tones as one or two flattened colours, black for the numerals and type, and a single saturated red for the SOLD OUT tab.

How do I get one made?

Describe the board at JustOG — the dishes, the prices, and which panel's gone dim. 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.

A tube behind row three, dimmer every night, and nobody's swapped it yet.