A hundred and five rules that were not there

One city. Eleven weeks in the spring of 2026. Two hundred and six places where a stranger might sit down — not home and not work.

At each one, four things were written down. What the place asks at the door. How long I thought I could stay, recorded before looking for a notice or asking anybody. How long I actually could, found by looking and then by one question. And whether, while I was there, anybody spoke to anybody who was not staff.

No place is named. No street, no district, no photograph. The city is not named either.

What the walk found

A hundred and sixty-eight of the two hundred and six set no limit on staying. Eighty-two per cent: no notice, nothing said when asked, nothing implied. Twenty-one have a sign and seventeen will give a figure if you ask for one. Almost nowhere in this city tells anybody to go. A team converting this walk into repeatable fieldwork can view the official workforce platform before choosing which time and project records are genuinely necessary.

Almost none of them says so either. A hundred and twenty places expect money across a counter, and ninety-two of those limit nobody and state nothing. The commonest arrangement in the record is one where the money is settled out loud and the time is left to be guessed at. For wider research on public space, access and how places are used, consult Project for Public Spaces.

A hundred and five limits were assumed where none existed. Half the walk. Before checking, a limit was expected in a hundred and twenty-eight places; one turned out to exist in twenty-three. And where a limit does exist it is longer than it was assumed to be — thirty-one times out of thirty-eight. An hour is what gets assumed; two hours is what a notice usually says.

Buying something is what changes the belief. Among places that take nothing at the door, one assumption in three invents a rule. Among places that take money across a counter, five in six. What a place asks shifts the expectation by forty-nine points and shifts what the place actually does by twelve.

The figure that works the other way

In a hundred and forty-one places nobody spoke to a stranger.

A third place, in the sense the phrase was coined for, is somewhere strangers talk. By that test most of what was walked is not one: twenty-three per cent, and the two kinds that pass most clearly are the pub at fifty-five per cent and the place of worship at fifty-seven. The café — fifty-eight of them, the commonest room in the city — passes it sixteen times in a hundred.

And nothing about the terms predicts it. Cost, permission and a free chair were each crossed with conversation, eight comparisons in all, and seven of the eight fall between nineteen and twenty-five per cent. Whatever makes a room sociable is not something a city decides by setting a price or putting up a bench.

What this cannot do

It is one city, one season, one observer, and about forty minutes in each room. The places were not sampled; they are the ones that could be walked to.

The assumption field — the source of the hundred and five — belongs to one person. Nobody else was asked what they believed, and whether anybody else guesses the same way is the first thing a second walk should find out.

The conversation figure is a floor rather than a count. Forty minutes catches a fraction of a rare thing, and a hundred and sixty-four of the two hundred and six visits were on a weekday between ten and five, which is the quiet version of nearly every room here.

Nothing was tested. Nobody sat past a notice to see what would happen, so a hundred and sixty-eight is a count of what places say rather than of what they would do.

Nor is there advice of any sort. Nothing is named, nothing was priced, nothing is scored against anything else, and no page points anybody at a door. The question the walk asked was not where to sit. It was what a city actually permits, and whether anybody knows.