Saturday, April 05, 2008

building complexity

Introduction to LogoWorks by Marvin Minsky (1994)

I had a few LOLs and AHA moments whilst reading this article.The scratch program in conjunction with Barry Newell's Turtle Confusion booklet does provide us with the opportunity to teach the concept of state and how to build complex structures from simpler structures.

State is an important fundamental concept that has come into being as a "fundamental" following the invention of the computer, which Alan Turing ran in his head before it came into being as a physical thing. Which goes to show that the fundamentals change.

One possible reason why kids building things declined in popularity:
The golden age of construction-sets came to its end in the 1960's. Most newer sets have changed to using gross, shabby, plastic parts, too bulky to make fine machinery. Meccano went out of business. That made me very sad. You can still buy Erector, but insist on the metal versions. Today the most popular construction set seems to be LEGO -- a set of little plastic bricks that snap together. ... It is probably easier for children, at first, but it spans a less interesting universe, and doesn't quite give that sense of being able to build "anything." Another new construction toy is FischerTechnik, which has good strong parts and fasteners. It is so well made that engineers can use it. But because it has so many different kinds of parts, it doesn't quite give you that LOGO-like sense of being able to build your own imaginary world.

About the time that building-toys went out of style, so did many other things that clever kids could do. Cars got too hard to take apart -- and radios, impossible. No one learned to build much any more, except to snap-together useless plastic toys. And no one seemed to notice this, since sports and drugs and television-crime came just in time. Perhaps computers can help bring us back.
How not to explain to a Martian how things work:
A Martian szneech once mindlinked me; it wanted to know what literature was. I told it how we make sentences by putting words together, and words by putting letters together, and how we put bigger spaces between words so that you can tell where they start and stop. "Aha," it said, "but what about the letters?" I explained that all you need are little dots since, if you have enough of them, you can make anything.

The next time, it called to ask what tigers were. I explained that tigers were mostly composed of hydrogen and oxygen. "Aha," it said, "I wondered why they burned so bright." The last time it called, it had to know about computers. I told it all about bits and binary decisions. "Aha," it said, "I understand."
The importance of understanding state:
When Turing was quite young, he realized that what a computer does only depends on the States of its parts -- and on the laws that change their states. Except for that, it doesn't matter how the parts are made. Then Turing asked what programs are -- and realized that you could think of programs as just sets of states -- or rather, ways to pre-arrange how a computer will, later, change its States
Computer programs are societies and algorithmic processes work independently of what materials they are composed of:
This must be the secret of those magical experiences I had, first with those construction sets and, later, with languages like LOGO. There's something "universal" about the ways that big things don't depend so much on what's inside their little parts. What matters is more how the parts affect each other – and less about what they are, themselves. That's why it doesn't matter much if money's made of paper or of gold, or houses out of boards or bricks. Similarly, it probably won't matter much if aliens from outer space had golden bones instead of ones of stones, like ours. People are missing something important, who don't appreciate how simple things can grow into entire worlds. They find it hard to understand Science, because they find it hard to see how all the different things we know could be made of just a few kinds of atoms. They find it hard to understand Evolution because they find it hard to see how different things like birds and bees and bears could come from boring, lifeless chemicals -- by testing trillions of procedures. The trick, of course, is doing it by many steps, each using procedures which have been debugged already, in the same way, but on smaller scales

1 comment:

Mark Miller said...

I got a Meccano set when I was around 8-10, but I don't think I ever used it past the first time I tried it. It was intimidating, because it had a lot of small pieces. Its manual only came with complex designs. The most frustrating part was putting the model together, because if you didn't tighten the nuts and bolts enough, parts would start falling off the more you worked with it. They'd work themselves loose.

I got a metal Erector set around this age, and used that. The model designs were just a little above my level. The pieces were larger, and there were fewer kinds of parts. Even so I often needed help from my mom to make anything. Again, using nuts and bolts to put stuff together was difficult.

Tinker Toys were the easiest to work with at that age. There were just a few kinds of parts, and they fit together well. It was simple enough that I could build things myself.

I remember FischerTecnik as a child, so it's not new. My mom got me a 2nd hand set once, but I don't remember using it much.

As a teen I built models from kits. I had to cut pieces (on the wood models), glue them myself, and such. There were a lot of glue-together plastic model kits when I was young. They seemed simple enough at first, but it took skill and patience to put them together. Most of that was knowing how to apply just the right amount of glue. Too little and the model would fall apart. Too much and you'd create a messy, sticky model with plastic melted at the seams.

A lot of the Lego stuff now is essentially snap together kits. A lot of the pieces are specialized towards creating one thing.

Interesting perspective in the article though. It sounds kind of like a conversation you and I may have had with Alan Kay once. I remember him talking about "Take something that looks huge, find some simple structures that will simplify it, and build everything else out of that."

I have noticed that there are fewer opportunities to experiment with basic elements. My mom got me a couple of 35-in-1 and 50-in-1 electronics kits from Radio Shack when I was young. I don't think they sell those anymore, or just one model. They were pre-packaged layouts of electronic components like transistors, capacitors, resistors, diodes, etc. They were all laid out on a board with spring connectors. They came with flexible, insulated wires of varying lengths that I could use to wire them up. They taught me a little about electronics, but not much. The manuals that came with them contained instructions for building some basic electronic circuits (though I had no idea what they did--things like a "Weatstone Bridge", or something), but mostly fun or interesting things like building a transistor radio that really worked, or a police siren that sounded real. All it took was putting in a couple batteries, and time and patience to follow directions. At least when I got to wiring up breadboards in one of my computer science courses I was used to the idea. :)