really important.
• o •
Okay, so I'd built that tic-tac-toe system basically by putting together electronic gates. The idea was to put the gates together into a system of transistor circuits that would never let you beat it. And as I said, I came up with the rules by playing all possible games.
But in the eighth grade I did something altogether different. I came up with a machine I called the Adder/Subtractor. This would be the closest thing to an actual computer I'd ever designed. I can say this because I designed it so it would do something—you could add or subtract numbers, and the result would show up on an electric display—but also because it wasn't made up of just a set of logic gates like the tic-tac-toe machine. Addition and subtraction are logic, just like tic-tac-toe; based on inputting Is and Os, you can calculate what Is and Os come out.
The Adder/Subtractor wasn't more complicated in terms of size or construction time than the tic-tac-toe machine, but this project actually had a goal that was closer to real computing. A more important purpose than tic-tac-toe. We're taught to add
and subtract in school, but nobody teaches you tic-tac-toe. It's not that important. Adding numbers could put a man on the moon; tic-tac-toe couldn't.
My project had a function, a real function that was useful. You could input numbers, add or subtract one, and see your answer.
This Adder/Subtractor was about a foot square. I had a plastic board filled with holes and store-bought connectors I could plug down into the holes to form connection points. I plugged the connectors in where needed and soldered transistors and other parts to them.
I had ten little switches to represent Os and Is, and another set of switches to represent more than 0s and Is. So if you wanted to add 3 plus 2, on one row you would have to toggle the right-most two switches (which is equivalent to 0000000011, the binary number representing 3) on the top row. Then, to represent 2,1 had to toggle the next to last switch to the right on the bottom row. In binary, that is 0000000010. The answer would show up in lights, the lights I had attached. In this example, two lights would be on—representing 0000000101, which represents 5. This would all be assuming that you had the Adder/Subtractor-. in "add" mode instead of "subtract" mode. What was impressive about this was that I knew so many levels of electronics, logic, binary number theory, soldering, and all the experiences of my life so far just added up. I could explain to judges how binary numbers worked, how you added and subtracted them, and then I could explain how gates were made of diodes and transistors. I would then show the right combination of gates that made a one-bit adder (something that could only add 0 and 1). I could show them a simple modification I did that could do subtraction as well. I also told the judges how I'd solved a nonworking problem in the electronics of a gate, switching from resistors to diodes. That's real electronics know-how.
On the one board were ten Adder/Subtractor circuits side by
side handling carries and borrows (remember arithmetic) so you could add or subtract larger numbers—any number up to 1,023.
But here's the thing. I took it down to the Bay Area Science Fair one night, to set it up before the day of judging. Some people showed me where to put it and asked me if I'd like to tell them about it. I told them no, figuring that I'd just tell them the story on judging day. By then Id gotten kind of shy. Looking back, I think I may have turned down the judges without knowing it.
When I showed up on judging day, all the projects already had their awards. The judging had already happened somehow! I had an honorable mention, and there were three exhibits that had higher awards than mine. I saw them and remember thinking they were trivial compared to mine, so what happened? I then looked in the fair brochure and those three projects were all from the school district that was