Live-blogging this talk. As usual, I’ll procrastinate on cleaning up these notes until a later date …
David Mindell, the Dibner Professor at M.I.T., came to campus today to talk about the interesting challenges that occurred in coordinating humans and machines during the first six lunar landings in the NASA Apollo space program.
During the 1960s, engineers and astronauts worked together with mixed results to design and build a “man-machine system,” combining the attributes of human (reliability, judgment) and computer (reliability). NASA and a small group of engineers at MIT’s Instrumentation Lab collaborated to design and build the Apollo guidance and control system, the focus of the talk.
Astronauts were involved in the design of the system, and a question repeatedly arose: how much to automate the flight to the moon? Some engineers were convinced that computers could run the entire mission, with no input from the astronauts. NASA, of course, could not condone full automation, as the astronauts played a political role in the projects as exemplars of American prowess. The computer design and the software then emerged to reflect a philosophy of aiding the pilots in critical functions and at critical moments, while not actually replacing them. Tensions in this philosophy emerged in real-time during the final moments of each of the six lunar landings
Former faculty advisor for Eden Medina
* first six human lunar landings (there also may be more)
* listening to Buzz Aldrin as the lander touches down on the moon, as running out of fuel … “The Eagle has landed” (from 1960s, Apollo 11)
* why was there so little fuel left, why did it happen this way
* working on a book – walk us through the main points of the new book (out of previous book, “Between Human and Machine” about 1916-1948 era of cybernetics)
* modest goal of trying to rewrite the history of computing machines to include telephony, gunfire control, military machines … logical end is the moon landings
* astronaut’s tools in command module using optical telescope, aligning the stars with inertial guidance system … built by the same company who built gun-sight computer in WWII … institutional and personal continuity between systems
* why was there a digital computer in the Apollo capsule … one of the first places where human lives were dependent on technology
* Wright brothers oriented the pilot as central to the flight, not the aircraft … aviator is a creation of the 20th century
* Roscoe Turner, barnstormer … era of heroic pilots, flew with pet lion (Gilmore)
* Collins wrote that he flew with a slide rule and computer … heroism and adventure drawn out of pilot experience by mechanism and computer
* Society of Experimental Test Pilots (SETP) … asked what is their role … users talking to each other … first annual awards banquet on Oct. 4, 1957 (the same day as Sputnik) … hear the the first hero of space race is a machine
* rise of systems engineers, feedback engineers … Dean Wooldridge and ??? Ramo (“R” and “W” in TRW) ..
SETP view – “head-shrinkers” decided humans had reached their limit and were being designed out of the cockpit … “icy B.M.” (triple entendre)
* “dinosaur” is an early space prototype where humans could fly the space ships off the launch pad
* Vernor Von Braun – human beings are outrageously slow and cumbersome … vision of a moon ship absent of pilot … once in space, lots for human to do
* Mercury mission has an interesting path of rhetoric (“We Seven” is an adaptation of the articles) … language differences (“capsule” v. “spaceship”)
* Kennedy’s famous speech to joint session places astronauts in a passive role (“send a man to the moon”)
* first contract for NASA was for a computer (mars probe)
* envisioned two buttons – “take me to moon” and “take me home”
* astronauts had no interest in having a computer on board … planned for the computer to be turned off when they got into space
* after Scott Carpenter’s urine bag leaked, causing corrosion in the electronics and completely shutting it off on a Mercury mission, the decision was to make on board computer 100% reliable
* astronauts didn’t think data entry was “manly” flying … but felt good about joysticks, even if controls were entirely computer mediated
* almost missed the end-of-decade deadline due to software issues … reached peak just as Neil Armstrong’s mission to land him on the moon
* software had to be, literally, hard-wired into the hardware … called on weavers to do the wiring
* last 200 pages of the book are about the last 10 minutes of the landing … “landing was a 13 on a scale of 10”
* During the descent, vision is sacrificed for fuel conservation (pilots laid flat and can’t see land) … only during the last part does the trade-off switch (at 1 minute 40 sec, 7400 ft)
* “1202” error occurred four times during landing
* Neil Armstrong, while trying to assess whether to go on, missed a moment of recognition, forcing the ship to fly further over craters, wasting fuel
* NASA take: at critical moment, computer fails and human saves the day … system engineers were offended (“It was a software error only if you consider the crew part of the software.”)
* All six missions turned off automated landing … story was that the automated targets all had too many boulders … pilots liked challenge
* Many astronauts still bitter about the last spot in the Apollo missions taken by a geologist (Jack Schmidt?)
* pilots eventually won – space shuttle was created to land on a runway (“wheels and wings”) … land like a regular pilot, rather than be fished out of the sea like a “bag of soggy cats”
“The reason we sent people to the moon was to send people to the moon. It makes it a human story.” … not because they can do something better than a machine. … The rhetoric of space flight should be about human experience. … Frustration talking to Neil Armstrong, because he has nothing to add to the national story. That’s a loss, after spending billions of dollars to send him. … Human presence provides sense of trust, safety
What gets lost when things get digitized?
