Devlog · 22 October 2026

What the sources said

Most of the prototype's numbers started as estimates from general knowledge. A research pass compared them with published sources. Some held up. A few were wrong in ways that changed the design.

Method

Four researchers, each working through one slice of the data: aircraft, airports, airlines and markets. Each had to open the page behind every figure and record the address, a short quote and any caveat. Anything they could not find was recorded as not found. The records are in the repository under docs/research and the checked figures are shown on the aircraft, airport and airline pages.

One limit matters. The pages were read through a tool that summarises them, so a figure should be rechecked against the original before any release.

Held up

Of 123 aircraft figures checked, 66 agreed with the sources and 30 were close. The coordinates of all 38 airports agreed with published positions to within about 2.5 km. Populations for 28 of 34 markets were within a plausible distance of the 1960 census once the area definition was matched.

Changed

Seats on the early jets. The game used about 104 mixed class seats on the first Boeing 707. Airline layouts in the period ran to roughly 109 to 112 on early aircraft, and 142 on a 707-120B in an international layout, according to Boeing's own airport planning document. The seats now follow the sources.
Runways. Lisbon's longest runway was about 2,000 m until 1962, when a longer one came into use. That matters in a game where a jet needs a long runway, so the 707 cannot use Lisbon before then. Prestwick, Frankfurt, Zurich, Heathrow and Oslo also gained dated runway steps.
Rome. The data closed Ciampino to airlines in 1961 when Fiumicino opened. Sources say Ciampino stayed open for charter, executive and some domestic flights. It now stays open.
Jet orders. The game's order books were too small. Pan American ordered 20 Boeing 707s and 25 DC-8s, TWA 33 707s and BOAC 15 707s and 19 Comet 4s, over several years. The start books now include more of them, scaled to what can be delivered inside the window.
Calibration. Two of the seven reference figures are now taken from published statistics: the ICAO world passenger load factor for 1958 (58.2 percent) and the combined operating margin of the US certificated carriers (about 5 percent). A third is interpolated between sourced Atlantic passenger counts for 1951 and 1961. The other four remain unsourced and are labelled that way.

Wrong

The most important finding was about the island scenario. The game had a small private operator flying between Land's End and the Isles of Scilly, named as a placeholder. The sources say otherwise for 1958 to 1963. British European Airways flew the service with de Havilland Dragon Rapides from 1 February 1947 until May 1964. At the 1957 peak that was up to 20 round trips a day, and the airport's history page reports 36,000 passengers in 1953. A small independent, Mayflower Air Services, joined from 1961 with two Rapides. In the House of Commons on 13 May 1963 the Minister of Aviation said the service cost BEA a small annual loss, and announced helicopters, which began in 1964. I found no recurring subsidy.

The scenario now uses a fictional operator and says so on the title screen. Opinion: a scenario about the real independent, which fought a state airline with two ageing aircraft and lost one in 1963, would be a better story than an invented one, and it is on the list.

Not found

Fleet counts by type for 1958 for any airline. The starting fleets are still estimates.
A North Atlantic load factor, the New York to London carrier shares, the jet share of capacity and a jet fuel price for the period.
Historic runway lengths for many airports, including Boston, Washington, Miami, Montreal, Toronto, Shannon and Dublin.
Sailing frequency, fares and crossing time for the Penzance to Scilly ferry in these years.