Terry Tales

Calculated Progress

Approximately a lifetime ago, my uncle Willie gave me a simple four-function pocket calculator with an LED display. I was one of the first kids in high school to use one, and it was a hinge point of modern history. Between my leaving high school and entering college, only three months later, the slide rule became obsolete. At our engineering department orientation, the dean advised us to throw out our prized ‘slipsticks’ because we wouldn’t be able to keep up without an electronic calculator.

That was just the beginning of many computing revelations. Intel had released the first 8-bit microprocessor a couple of years earlier and the Apple II personal computer would go on sale a couple of years later. I began writing programs as a college sophomore, first with Basic and then doing more complex assignments in Fortran. I still needed my Rockwell pocket calculator for math and physics homework, but the real computing power clearly resided in the university mainframe.

I won't even try to describe the stone-age programs that we wrote for that Xerox Sigma 7, but I have to say a few words about the ASR-33 teletype machine that lowly sophomores used to communicate with it. The teletype is a hybrid of the mechanical typewriter and the Morse code telegraph system. Pushing a clunky mechanical key at one teletype will cause the printed letter to appear on the paper in another teletype, or on multiple teletype machines located in newsrooms across the world. That was the original newsfeed, and the chattering sound of the teletype machine accompanied TV news programs for many decades.

Imagine that chattering multiplied by forty. That was the environment in the computing lab. There were teletype machines lined up against all four walls and a double row lined up back-to-back in the center of the room. Forty students pounding away at crude keyboards to input data into the mainframe and then reading the output on a continuous roll of paper as it printed out of the teletype machine. The incessant din wasn't really great for concentration and the amount of wasted paper that merely showed error messages was almost scandalous.

Believe it or not, there was an even lower-tech way to input data to the mainframe and receive its output. Before the teletype became common, people would use a keypunch machine to put code onto paper punch cards and Siggy-7 used a card reader to scan them. The 3” x 7” cards typically had twelve rows of 80 columns and thus offered 960 bits of data per card. As I remember, one program line in Basic or Fortran would fit on each card. A 500-line program would require a three-inch thick deck. There were many people still using card decks in 1976 and some decks were more than a foot thick.

The user would write a program on paper, then sit at the keypunch and transcribe it onto cards. But the user wasn’t allowed to interact directly with the mainframe. Instead, the cards were bound with rubber bands and dropped into the incoming bin. A computer tech would load the deck into the card scanner and run the program. The output was printed on fan-fold paper, wrapped around the card deck, and put into the outgoing bin. It was never quick and results often came back the next day. It was common to hear someone cursing with frustration after waiting hours just to get an error message. And Siggy-7 always stopped reading at the first error. Multiple errors meant multiple runs. Debugging a really complex program could take weeks. And woe to the hapless fool who neglected to number the cards and then dropped the deck! My first computer science class included one programming assignment on punch cards so I could properly appreciate the advanced convenience of the teletype. And it took only once to learn that lesson, never again!

But the lowest tech of all was the telephone, invented almost 100 years earlier. Each teletype machine was paired with a rotary-dial telephone that connected the user to the mainframe. The phone was programmed by dialing one digit at a time to send electrical pulses down a copper wire. It provided our link to the space-age technology of Siggy-7 just like it provided the teletype link between newsrooms. Each computer programming session was initiated by dialing into the mainframe and then placing the telephone handset into the rubber cups of an acoustic modem. Remember the cute little beeps and boops that R2D2 used to communicate with C3PO? Well, the teletype machine quite literally talked to the mainframe over a phone line in a similar fashion. Except that the modem tones weren't cute at all. The lunatic squealing and chirping of those modems still haunts my dreams.

The ASR-33 teletype, limited to 300 baud, was soon replaced by green-screen terminals, but the modem didn’t disappear. It simply migrated into the computer. My first IBM personal computer could send data at an astonishing 19.2 kilobaud on its dial-up modem. And I was still using a phone line for my internet connection thirty years after first meeting Siggy-7. Of course, that DSL modem could run at more than 3 megabaud and even supported video streaming.

Modern apps are orders of magnitude more sophisticated than our ancient Basic programs, and today's ultra hi-res screens can show the most trivial images in beautiful detail. There are thousands of solitaire games and millions of clickbait posts to fill our leisure time without resorting to real-life social interaction. And endless newsfeed items allow for obsessive-compulsive doomscrolling. But don't forget, even though today's smartphone is primarily used for updating social media, it can also make phone calls to real people. Just like the ASR-33 talked to Siggy-7, you can literally talk with another human being. Or is it an AI chatbot?