Research & data

01 · FRAMEWORK

The question

Ray Oldenburg's The Great Good Place (1989) describes the third place: the café, the library, the bar, the bench that is neither home nor work, where you meet people you did not arrive to meet. Later research — Jeffres, Bracken, Jian & Casey (2009); Mehta & Bosson (2010); Adelfio and colleagues (2020) — measures what those places do for a neighbourhood's social life.

PLUB turns that into a map. It finds the places, scores them with a stated method, and shows the evidence behind every number. What it will not do is quietly fill the gaps. That constraint is the product.

02 · SCORING METHOD

How a place is scored

The Great Good Place index is the average of three ranked signals — access (↦, distance to a rail point), scene (⁘, how many third places are nearby) and character (◇, independence from chains) — each ranked against every other place in the catalogue, plus the third-place traits from the research. Two more signals, work (≋) and notable (✦), are shown but held out of the index: too few places carry them yet to rank fairly.

The work readiness score is a different animal, and it is not a grade. It is the weighted mean of the seven things that would tell you whether you could work somewhere — Wi-Fi, seating, power, noise, long-stay policy, price, step-free access — over the ones that are actually documented. It is published only when at least half the weight is documented, and it always names the dimensions it could not score. A dimension nobody wrote down stays unknown; it is never given a zero.

Each fact carries the kind of evidence behind it: on the map, the venue says, visitors report, or one named person's dated on-site report. The Bridge 1958 is a worked example of the last one.

03 · REFUSALS

What PLUB refuses

• No scraping. No Google, Facebook, Instagram, X or private-group data. Commercial platform fields are published as absent, with the reason.
• No invented geometry. No guessed building heights, no inferred centroids, no street connectivity that the open map does not carry. Unknown buildings stay flat.
• No invented facts. A missing opening hour is missing. A venue is not scored for a Wi-Fi network nobody wrote down.
• No paid places. Nobody has paid to appear here, and nothing in the catalogue is sponsored.
• No private notebook. Your visits, notes and photos stay on your device until you choose to back them up or share them.

04 · PROVENANCE & LICENCES

Sources & licences

Places: OpenStreetMap (ODbL-1.0) and Overture Maps (CDLA-P-2.0). Neighbourhood atmospheres use the Sanzo Wada colour collection, credited digital conversions. Research: Oldenburg (1989); Jeffres, Bracken, Jian & Casey, Applied Research in Quality of Life 4 (2009); Mehta & Bosson, Environment and Behavior 42:6 (2010); Adelfio et al., Applied Spatial Analysis and Policy (2020).