Alan Turing is often considered the father of modern computer science. He designed a machine that could decode the encryption of the fantastically complex German Enigma machines during World War II, enabling the Allies to read the messages.
I saw my first computer program in the late fifties. My brother Stephen, who was a math major at the University of Michigan, was taking a FORTRAN class, and I helped him desk-check a program. FORTRAN is a scientific programming language, and at that time most computers were used for scientific purposes.
I went to work for IBM in October 1961 when they were beginning to install 1400 computer systems. The 1401 was the first practical business computer, meaning that you didn’t have to be a mathematical genius to learn how to program it.
In order for a customer to install a 1401 system, a room had to be dedicated to the computer components, with a raised floor and air conditioning. The components included a Central Processing Unit (CPU), a card reader/punch combination (IBM punched cards held 80 characters of information), a printer (which spewed out paper at 600 lines per minute), and possibly tape drives. Soon it was also possible to install a physically large disk drive that could store a few million characters. Each component was connected to the CPU via cables, hence the need for a raised floor with the cables running underneath.
A programmer had to write instructions telling the computer everything it was to do. For example, if the program called for reading a card, the programmer added a READ instruction. Other instructions moved data around within the computer’s memory, compared contents of data fields (equal to, greater than, less than), printed reports, stored information on punched cards, tape or disk drives, and did calculations, including add, subtract, multiply and divide.
On early 1400 models, the multiply/divide instruction was optional, thus saving the customer some money. For one of our customers, I had to write a program that multiplied (by successive addition) and divided (by successive subtraction). The first time I tested dividing, the computer punched a card for each result—clunk, clunk, clunk—about one a second. Then it occurred to me that I could reduce the number of subtraction instructions by shifting the subtraction from the units position to the tens position, etc., as the subtraction at each position was completed, which greatly reduced the number of instructions the computer had to execute. The clunk, clunk, clunk became much faster. Computers were slow in those days.
Those of you who are of a certain age will remember the panic about the year 2000 date problem. It was feared that on January 1, 2000, computer records containing the year, used by millions of computer programs, would have the year changed from 1999 to 1900 instead of 2000. The reason for this was that punched cards contained only 80 characters and early computers had very limited central storage capacity, and so programmers designed files with two-digit years (such as 99) instead of four-digit years (1999).
It was true that by the year 2000 computers had become much faster and had much more storage capacity than those in the 1960s. Punched cards became obsolete. However, many old programs were still being run and had never been updated. My wife, Bonny, who was an executive at Xerox, had to make sure that all the programs and files used by Xerox all over the world were checked, and if necessary, updated.
I was told at one point that computers would continue to become faster, smaller, with increasing capacity, at a phenomenal rate, more or less forever, and that prediction has come true so far. Computers don’t need to be located in their own room anymore, and now my audiologist talks about updating the firmware in my hearing aids. Although I’ve also been told that nobody can predict the future, this seems to be the exception.
Now it is possible to store all the videos and movies ever created, easily, and retrieve them at a moment’s notice, and YouTube is certainly attempting to do that. The internet, a relative newcomer, has made it possible to transmit this information to practically everyone in the world at the same time, as Netflix does.
Like all innovation, however, computer design had a slow start and gradually became faster. We need to thank Alan Turing and all the computer designers and programmers since then, because today with a few clicks on our iPhone we can retrieve information on all the baseball players who have batted .300, or stream the latest episode of our favorite sitcom.
I was fortunate in 1961 to attend a technically oriented college where we learned the basics of computing from the very beginning. In printing class, for example, I had to set a few lines of text from a California Job Case. With all the letters and symbols to make it print professionally, used for hundreds of years. But very tedious, taking me hours to set all the little pieces in place for a few lines of story. I also carved a block image, that could have been a letter in the alphabet, for block printing like Gutenberg. Both of these were lessons on how far we have come.
They had several linotype machines that were used, perhaps, for over 50 years to create the text for daily newspapers. AD own, Jerry Bolton, was a linotype operator for many years. When I read what he wrote, using Microsoft Word, I can see the skill he had from having to type newspaper articles without error because the fonts were actually cast in lead and any mistakes had to be melted down and re-typed, a wasteful and slow process that would cause you to get fired for not being accurate or too slow the first time first time.
We were fortunate to have, I believe, a 1100 IBM computer at school. The first program we had to write was in binary language, basically, like Morse code, ones and zeros. It was very difficult to do, but that slow computer could handle binary very fast. Much faster than the computer our high school class visited from a major corporation in 1959. As we entered the building, it was very hot. Filled with thousands of vacuum tubes that were all lit up to do calculations for the company. There were people that did nothing except replace those vacuum tubes that were constantly burning out. A simple calculator that came shortly after could do the same task held in your hand or at your school desk.
From binary language, we had to write assembly language which was the next step up. And then, lo and behold, FORTRAN! Opening the powers of that little computer to doing mathematical, practical work. For those of you know and use Excel, Excel is based entirely on FORTRAN.
When I was at Stanford studying for my master's degree, we had several computers, some of quite different architecture, to write programs for. They were developing the IBM 360 time shared computer. It basically could do multiple programs at the same time by taking little bites from each program and then, quickly move on to the next. The next stage was connecting 360s between universities… The start of the Internet.
They were also developing something with power for working with words. The first word processor that was sent off to Northern Illinois University as WYLBUR. I used it for 100 pages of my doctoral dissertation in 1973. Wonderful to use and powerful even at that time.
So, we all should be grateful for the ability to compute on so many levels as we work on our creative endeavors using the power of computing. We are standing on the shoulders of giants. People of vision who have taken us where we never thought we would go. It is mind-boggling where we will go from here.
Ron