Public space

Public space is where smart-city technology stops being optional. Nobody opts into a bollard, a counter or a camera; you walk past it. This guide covers what cities install in the street and the square, and the measurement layer around it.

Twenty-eight verified entries in 19 cities and 15 countries carry this tag, 2009 to 2023 — and six are recorded as failed or scaled back, the highest share of any topic in this atlas. Bicycle infrastructure belongs to the smart cycling guide, policing technology to the safety guide.

What is actually being installed in the street?

Three things wear this tag; only the first is what people picture.

1. Space taken back for people. Barcelona's Poblenou piloted the modern superblock in 2016 with cheap tactical urbanism before pouring concrete; Sant Antoni took it into a market district in 2018, Vienna's Supergrätzl and Tokyo's Marunouchi trials elsewhere. Paris depaves schoolyards and opens them to residents in heatwaves; New York's bins count too, since containerising a century of pavement rubbish cost West Harlem about 500 parking spaces.
2. Counting the people who use it. Melbourne has run infrared pedestrian counters since 2009, now at 99 locations; Chicago's Array of Things did air, noise and foot traffic from 100-plus pole nodes until 2021; Zurich put 16 sensors in chairs to see how two squares were used.
3. Watching them. São Paulo's Smart Sampa matches faces across a network the city reported passing 50,000 cameras in 2026; in Rio the biometrics belong to the state police, and city hall's cameras abstain.

The halves share a pole. Shanghai removed about 40,000 poles in 2018–2020, replacing them with 16,000 masts carrying the lamp, the signals, environmental monitoring and the public-security cameras. One mast, both arguments.

Five interventions worth comparing

What does the evidence actually support?

The best-evidenced entry is also the dullest: counting people. Melbourne's counters have fed peer-reviewed work on pedestrian safety and on nowcasting retail sales — and they undercut their own headline: a 2026 analysis found 2025 foot traffic up 7.7% raw but only 1.2% comparing sensors present in both years.

Barcelona alone has independent measurement: the city's public health agency put the NO₂ fall at Sant Antoni at 25%. Poblenou's numbers are the council's own. Vienna is the discipline case: opened May 2026 having unsealed 7,000 m², its evaluation runs to September 2026, nothing is published yet, and so nothing is claimed here.

And sometimes the reading never arrives. New York's Empire Bins carry no sensors, so the only evidence is a 20.6% fall in 311 rat-sighting reports — and the city's own health department says complaint data cannot measure neighbourhood rat activity (who complains). Los Angeles grades every block for cleanliness using crews from the department that cleans them, unaudited; while CleanStat reported improvement, the City Controller found illegal-dumping reports up 300% between 2017 and 2020.

Why does so much of this get cut back?

None of the six reversals was technical. Barcelona's Green Axes built four Eixample streets by 2023 against a plan to convert one street in three; the Collboni administration ruled out expansion on cost, and a court ordered part of the Consell de Cent axis reverted, now under appeal. Buenos Aires is the only outright failure: statute confined the system to a fugitives list of 35,000–40,000 people, but the court file showed 9,392,372 biometric queries covering some 7.5 million — and the legislature's oversight commission had never been seated. Rio's Porto Maravilha sold all 6,436,722 development-rights certificates to one public fund, fewer than 10% were resold, and the concession maintaining its streets stopped work over non-payment in 2017. Cairo's bike share collapsed twice in its first year, with about two of fifteen planned kilometres of protected lane built; Bogotá's trips fell 23.2% year-on-year and 110 of roughly 300 stations were vandalised. Budgets, contracts and judges end these schemes; sensors only record it.

What should a city ask before buying any of this?

• Who is the measurement for? Counting footfall and identifying faces hang from the same poles. Decide which you are buying, in writing.
• Count or complaint? New York and Los Angeles judge public space through channels their own agencies say cannot bear the weight.
• Who enforces the redesign? Toronto's gains eroded when ticketing lapsed — under 0.3% of violations were caught.
• Is the oversight body seated? Buenos Aires had a statutory limit and nobody to enforce it.
• Can you pilot it in paint? Poblenou trialled tactically, then consolidated (why schemes fail).

The practical takeaway: the instrument here is visible to people who never asked for it — which is what makes it fragile. Buy the boring thing first — a counter you can defend, a baseline you can check — and pilot the rest in paint. A reading still needs somebody empowered to act on it (the ladder of consequence); a bench and a camera on one mast are two decisions.

Frequently asked