This memoir could be inspirational for someone who is enthusiastic about programming and retro-computing. I loved Gary Kildall's humorous, conversational writing style and his detailed descriptions of what it was like to use technology before it became retro-technology. My only regret is that it was not completed.
I read that his family wanted to omit parts because they felt that Gary Kildall's alcoholism, in the writing, did not reflect his authentic self. I read people debating this on Slashdot and to me they came across as callous. I tend to agree with his family. My reasoning goes like this:
People are prescribed psych meds because "who they are" without meds is not acceptable. On the meds, their behavior and personality are slightly altered, and more acceptable.
People who are addicted to alcohol are self-medicating. Due to trauma, the absence of positive outlets, or whatever the reasons, they cannot deal with sobriety and they drink.
So in my estimation, the person who is altered by a substance is a different person than the one who is sober. It is valid to have opinions about which is the more authentic. The opinions of family count more than the opinions of strangers, and the opinions of the person in question count the most of all.
So only Gary Kildall could answer who he most authentically was, and his family certainly knows better than i do.
The memoir is short, so i will try to keep my quotes short. I quote two veins of thought:
In retro-technology, an example was the hours spent waiting for operators to process punch cards. In modern technology, the costs are represented in other ways, always externalized, normalized, and kept as "invisible" as possible while in plain sight.
To navigate you take "sights." A sight is a reference point to a celestial body, such as a star. You look through a sextant, find the proper angle, and then look at your chronometer. Next, look in the Nautical Almanac for that exact time and date. Use standard, but tedious, calculation forms to plot the result. That tells you where you are on the face of the earth.
[In the early 1960's, Gary's father] Joe often described a "machine" that he wanted to build. "You punch-in the navigation data and turn the crank, and out comes the position of your ship."
The Kildall Nautical School taught me processes that high school hadn't. Such as the ability to do mathematics of a sort and, most important, the mental tools to dissect and solve complicated problems, and to work from the beginning to the end in an organized fashion.
In 1963, at the "U-Dub," as it was called, I found myself in the evolutionary transition between mechanical calculation and digital computers. My first class in Numerical Analysis in 1964 was from a professor in his eighties. This nice old gentleman was retiring, and not too soon in my opinion, as he seemed quite feeble and at somewhat of a loss of memory. He taught the trade of computing with Marchant calculators. A Marchant calculator [1] is a big, black machine, about three times the size of a typewriter. It has a giant carriage on top with about ten holes with little mechanical numbers that pop up in the slots. There are buttons all over the front with digits, decimal points, and keyed symbols that multiply and divide to control this mechanical nightmare.
They used these Marchant calculators for Numerical Analysis, starting in about the 1940's. A business would hire rows and rows of Marchant calculator operators to compute so-called "Finite Differences."
You sat there and pushed those keys while the mechnicals made a host of noises by cranking and grinding who-knows-what, until the rattling stopped. You got a number. You put the number on a form, and eventually, with enough numbers and forms, you got a result. Sometimes the resulting number was correct. These operators also made tables to figure where the Sun, Moon, stars and planets would be at a particular time, much like the tables that solved the navigation problems that we talked about earlier. In a sense, dad's navigation "crank" was somewhat inconveniently embodied in the Marchant with a "form" alongside.
[Describing using FORTRAN in 1965:]
You walk into the keypunch room with your two-foot long cardboard box of FORTRAN STATEMENT cards. Then wait in line to get to a keypunch machine. Finally, someone tires of typing, and you get a seat.
Each card in that stupid cardboard box is like a single line on your computer screen today. I don't even want to think about it, because you young [folks] have it all too easy. ... All us oldtimers remember that the keypunch machine is like a typewriter, but it shuffles cards through its mechanism and stamps rectangular holes in a column below each character. You leaf through your box, alter each necessary card, and then relinquish your keypunch to the next anxious student. Then, take your precious box of cards to the shelf outside the computer room and wait and wait and wait.
The computer room hides the IBM 7094 with SAC-like security doors. Computer operators, trained in "Grumbly 101," sit behind that door, attending to their work and throwing down coffee without even their own notice.
I know this because I spent about 1,312,467 hours looking through that little peep-hole window to see if the operator would arise from his or her throne to go to the restroom and possibly retrieve my box of punched cards on the way back into SAC.
If you stayed around very late at night, you might get a "turnaround" of an hour or so to get the printout. The program usually didn't work, so it was back to the key-punch.
I also reviewed the Computer Chronicles Unix episode [2] hosted by Gary Kildall.
author: Kildall, Gary, 1942-1994
detail: gopher://gopherpedia.com/0/Gary_Kildall
title: Computer Connections