The operations room is the part of Project Cybersyn that everyone has seen, because there is one colour photograph of it, taken in 1972 or 1973 and carried out of Chile after the coup by its chief designer, Gui Bonsiepe. The photograph is the designers’ copyright and is not reproduced here. What follows instead is the room in words, from two kinds of witness: Stafford Beer, who conceived it and described it in a lecture in New Delhi in December 1974, fifteen months after it ceased to exist, and the designers and historians who have described the object itself. They do not quite agree, and the differences are worth keeping.

What Beer wanted

Beer’s complaint about the rooms in which governments and boards made decisions was that they were “shrines to institutional pomp and circumstances” whose latest scientific aid was “the ball point pen.” He had wanted for years, he said, “to create an environment of decisions, a place where a creative group of people really could undertake creative thinking”. In Santiago he got the chance. The room was conceived, in his words, “as a decision environment for a creative group that would certainly include workers.” There were to be no tables and no paper. The people in the chairs were “the brainpower”, and they were to be served, as a brain is served, by a nervous system reaching into the whole of Chilean industry: the daily filtered indices from the Cyberstride program, the algedonic alarms, and the pictures.

He also said what it was not. “The room I describe was built and was working. But it was simply a prototype, and the economy of Chile was never actually commanded from there.” Smaller and plainer rooms, he said, had been set up at other levels of the system “to act as foci for information and regulatory activity.” Eden Medina’s reading of the archive is that these simpler rooms, with blackboards instead of screens, were planned for the factories but never built.

The walls, by Beer

Beer counted eight walls. “The room is octagonal,” he said: “a room, ten metres wide, inside a larger room,” the ring between the two filled with “the sixteen projectors and the digital logic racks required for Datafeed.”

On the first wall was “an illuminated, animated, screen eight feet high depicting the neurocybernetic model”, the five-level diagram of the viable system, which the room manager could set to show whichever level of the economy the meeting was about: an industry, a branch, or the whole. Into it were slotted icons showing the mean indices, so that one could see “how low the productivity, here, how high is the latency there.” The flow lines did not carry arrows; they moved, at one of three speeds. If an algedonic alarm was present it flashed on this wall.

On the second wall were two screens, one for the day’s Cyberstride output at the meeting’s level, one for the alarms coming up from the level below. “In this actual room, those signals had to be hand posted,” Beer admitted; with the right interface equipment, “perfectly available”, the computer would have driven them.

On the third wall was Datafeed. Beer did not believe in data banks from which any fact could be retrieved on demand, because retrieval, he said, was a problem that had not been solved and probably could not be. Datafeed was instead “a visual data bank”: three active screens under large index screens, each active screen fed by back projection from five carousel projectors of eighty slides apiece, so that the three screens “between them command twelve hundred different pieces of visual information.” The carousels were selected from store by the room manager for the meeting’s level of recursion. On them were the quantified flow charts of every plant and industry, photographs of the plants themselves, “and how much information the experienced brain can gain simply from looking at a photograph of a works”, and investment plans and projections in the same iconic style.

The fourth wall was the door to the machinery. The fifth was “set aside to contain an algedonic meter”, of which more below. The sixth was the entrance. On the seventh was “a huge animated model of the whole economy”, the CHECO simulator, of which Beer said: “in the two years that we had, this was never made to work. It certainly worked visually but the computer drive behind it was experimental and fragmentary.” The eighth wall was to carry CHECO’s ten-year projections of the effects of a decision tried out on the seventh. “As I said, we did not achieve this result.”

In the middle were seven swivel chairs. “The fact of seven is drawn from experimental psychology,” Beer said; “it appears that this is about the largest number for a creative group.” In the arms of the chairs were buttons. “Anyone can slam home a button, and gain control of datafeed,” and then, with the index screen and another set of buttons, show the group whatever he wanted. “The digital logic for doing this is quite complicated, but it certainly worked.”

The walls, by the designers

The room was designed and built by the Industrial Design Group of INTEC, the state technology institute, under Bonsiepe, who had taught at the Ulm school of design in Germany before coming to Chile, with Rodrigo Walker, Fernando Shultz and Pepa Foncea among others; the control electronics were ordered from Electrosonic in England and were late. Medina, who interviewed the designers and read the project’s papers, gives the room six walls, not eight. The designers had wanted a circle, could not build one, and chose a hexagon, which gave five wall spaces for displays and one for the entrance. Entering, a visitor found the first wall to the right opened into a small kitchen. The second wall held the four Datafeed screens, one large above three small, each in its own fibreglass cabinet; the large screen showed the button combinations needed to change the three below. Each chair’s armrest carried ten buttons, including a hold button that gave its user the screens until released. The chairs were white fibreglass with orange cushions, on a brown carpet, in a room panelled in wood. The number seven was taken from George A. Miller’s paper of 1956 on the span of short-term memory, “The Magical Number Seven, Plus or Minus Two”. Paper and writing were banned.

Walker called the ability to choose one’s own route through the slides “like a hypertext”, twenty years before the web. Medina’s judgement is that the comparison does not quite hold, and that the room’s modern look concealed a clunky technology: slide projectors, carousels and relays, with “a lot of human labor” behind the wall to keep the illusion up.

Which count is right? The hexagon is the shape in the photograph and in the reconstruction built in Santiago in 2023 with the original designers’ help, so six it was. Beer’s eight may count the hexagon’s six walls together with two he had planned for: the CHECO projection and the algedonic meter. His lecture was given from memory, in India, a year after the room was gone.

The empty wall

The fifth wall, in Beer’s count, was for “an algedonic meter”, and the idea behind it was the least built and most far-reaching part of the project. Beer called it Project Cyberfolk, or the People Project. He had noticed, he said, that people who could not argue a policy could still say whether they were happy or unhappy with the state of things, and he borrowed Aristotle’s word for that kind of social happiness: eudemony. A meter to register it had to be “simple and cheap”. The device he took to Chile was “a low-voltage analogue device” of two intertwined discs, one orange for happy and one grey, which a group, “perhaps a whole village”, would be entitled to set; the settings could be read electrically and summed for a district or a nation. “The equipment that I took had ten stations, and one summation machine.” It was to be tried first in a single factory, with a meter in each section and the sum shown at the gate and in the managing director’s office. The idea for the wall was that the President’s economic committee might explain its policies on television from the operations room “while the nation registered its reactions on algedonic meters summated on the fifth wall — which the audience could see.” Beer saw the danger himself: the villages gathered to reply “would doubtless be suborned by political parties. Such questions need much thought.”

The journalist John Bartlett, writing in 2019, says the device was built by Beer’s son Simon from old radio parts, was tried in two cities, and was never deployed; Espejo told him the team used handsets of the kind in their own meetings, with the aggregate shown on a screen in front of the speaker. In the version of the project that the Chilean government published, Cyberfolk did not appear.

30 December 1972

The President saw the room once. On Saturday 30 December 1972 Fernando Flores brought Allende and General Carlos Prats, the minister of the interior, to see it; Allende sat in a chair and pressed a button. It was midsummer, the electronics behind the panels had overheated, and the projectors misbehaved. Roberto Cañete telexed Beer on 3 January 1973 that the room’s electronics had “behaved awfully” during the visit. Walker remembered that “the projectors went crazy; you would push for one thing and something else would appear.” Allende, by Flores’s account to the Santiago paper The Clinic in 2003, said it would be good to have it working, and to keep at it; Prats was interested in what such a room might do for a military command. Beer had drafted a speech for the President to inaugurate the room. It was never given. On 8 September 1973 Allende asked for the room to be moved into La Moneda; the plans, Beer said, were nearly ready on the 11th.

After

After the coup the military questioned the team and other Chilean computer people about the room. One soldier took a knife to the slides and put it through every one. The room was abandoned or taken apart; Medina’s two accounts say both. In 2023 a full-scale reconstruction, seventy-two square metres of hexagon with seven working chairs, was built at the Centro Cultural La Moneda for the exhibition How to Design a Revolution, with Bonsiepe, Shultz, Walker and Foncea advising and students from MIT building the logic for the buttons.

Sources

• Stafford Beer, Cybernetics of National Development, Evolved from Work in Chile, the Zaheer Science Foundation lecture, New Delhi, 5 December 1974, section “The Operations Room” and “The People Project”. Every quotation attributed to Beer above. In copyright; quoted in short phrases. archive.org
• Eden Medina, Cybernetic Revolutionaries: Technology and Politics in Allende’s Chile, MIT Press, 2011, prologue and chapter 4, for the hexagon, the kitchen, the Datafeed cabinets, the ten buttons, Miller’s paper, the ban on paper, Walker’s “hypertext”, the visit of 30 December 1972 and Cañete’s telex of 3 January 1973, and the knife; and “Designing Freedom: The Cybersyn Operations Room”, Cabinet 46, 2012, the published excerpt. cabinetmagazine.org
• “Project Cybersyn”, 99% Invisible, episode 230, 2016, transcript, for Medina on the labour behind the illusion and Bonsiepe’s photograph. 99percentinvisible.org
• John Bartlett, “Allende’s Strange Plan to Connect Chile, Long Before the Internet”, Americas Quarterly, 11 July 2019, for Simon Beer’s device and Espejo’s handsets. americasquarterly.org
• Michael Brindley, “Designing a revolution”, MIT News, 16 October 2023, for the reconstruction. news.mit.edu
• Metaphorum, “Cybersyn”, for the eudemony meter as the Beer society describes it. metaphorum.org