Bedrock Computing Written June 2017 My former high school classmate, Terry Rockhold, wrote me a long letter in October 1967.  We were now college juniors.  He was at Case Western Reserve University in Cleveland, and I was at Oberlin College 34 miles to the southwest. Terry reported that his father was still on strike at a factory outside Green Camp, Ohio.  “I have a hunch they are going to get some new union leaders in elections this month!” And he was following a municipal election in which Carl Stokes was attempting to become Cleveland's first black mayor.  According to Terry, in the Democratic primary Stokes was expected to get 97% of the black vote (“that's what I call unity!”), but not much of the white vote.  The previous time he ran, he received only 3% support from white voters.  His primary opponent this time was incumbent mayor Ralph Locher. Racial tensions had been rising in Cleveland.  “Stokes signs (orange with black letters) are all over the place,” Terry reported.  “When you see a bunch of Locher signs (red and blue), you know you're passing through a nationality area (Italian, Slavic, etc.)  Council candidates who support Stokes have orange signs with black letters; those supporting Locher, red and blue.  Right now I'm picking Stokes (perhaps hoping is a better word), but it's going to be awfully close.” Stokes did win the primary.  And in the 1967 general election the next month against Seth Taft, Stokes was able to add 20% of the white vote to his black base, achieving a narrow 50.5% majority. Every college had at least one computer by 1967, but it was still the Stone Age.  Oberlin's number cruncher was a big IBM mainframe that lived in a climate-controlled cave in the basement of the physics building.  I think it served the college administration's needs, but we physics students could use it too. However, we couldn't simply log on from a laptop.  We had to write out our program using the Fortran language and take the coding sheet to another room in the basement. That room housed keypunch machines like this one.  We would laboriously type in each line of our program, and the machine would noisily punch holes in IBM cards, one card for each line. Then we'd carry the stack of cards to the door of the computer's lair and hand it to one of the acolytes who dwelt therein. (Many of those acolytes had been hijacked from the college radio station.  Once upon a time, Oberlin College's best and brightest nerds had hung out at WOBC, building and maintaining the electronic equipment.  But then the computer age arrived.  To tend to the IBM's needs, the college stole away WOBC's nerds and actually paid them.) The attendants would feed our stack of cards to the computer, which would read each card by sensing the locations of the holes and then attempt to run the encoded program.  After maybe an hour, we'd get our cards back along with a printout of our results.  Usually we'd made some mistake, so we'd have to do some more keypunching and resubmit the stack. Over at Case, Terry Rockhold was beginning to work toward an MBA degree.  In his October letter, he enclosed a spare IBM card.  “As you might guess from the card,” he wrote, “I spent yesterday playing with, uh, using the computer.  It was my first time, and I really got a kick out of it. “It was a program to read the hours worked, hourly rate, number of exemptions, and voluntary deductions of a worker and to output a wage statement.  The first time I ran it through the computer, the output was exactly like I wanted it, but I got six diagnostic messages in the program.  I found that I had put a PRINT message in the wrong columns and that I had forgotten to put a ) at the end of one of my FORMAT statements. “I retyped the two cards, ran them through the computer, and got one diagnostic message.  After talking to one of the other guys, I found out what was wrong and was ready to run it again, when the computer broke down.  That put both printers out of action for about twenty minutes. “Finally, I got the thing run through perfectly and left.  It was then that it dawned on me that I had been so wrapped up doing the blasted thing that I had forgotten to go to one of my classes!” Another friend, Jack Heller, had recently transferred from Oberlin to Ohio State.  “I'll major in electrical engineering and advanced typing,” he wrote me that month.  “So now, besides math I take chemistry, engineering graphics (glorified drafting), and engineering mechanics (computer programming). “At OSU, Fortran is processed using PUFFT (Purdue University Fast Fortran Translator) and other languages such as FAP, OSUY, and AGWYG (AlGo Where You Go, yuk, yuk). “They have a 7904 here which is voluntarily shut down every Friday for maintenance. “It takes five hours for your program to be run, and you only get one chance a day.”