Alan Kay Talk at MobilFest 2007

From Viewpoints Intelligent Archive
Jump to: navigation, search
Thanks very much for inviting me to MobilFest.
Today I am going to try and say a few words about some of the mobility ideas originally came from, how they affect children and maybe a little bit about how the future is going to turn out.
What I would like to do is first to look at perhaps the most important mobile technology to ever been invented.
And we can think of this technology as wanting to encompass the entire organization of human knowledge.
We want it to be solid state and mobile of course. We want extremely hi-res and high contrast, ambient lighting display to be used anywhere in the world.
Really easy to use basic interface from getting from one place to another in it. It's got to have a wide range of the kinds of knowledge it can hold.
It needs have unlimited battery life. And we want it to be bio-degradable so it won't pollute the planet.
And of course what I'm talking about here is basic organization of knowledge which is called "The Book".
So this technology has already been invented. It already has changed basic human civilization in many ways.
It is very difficult to do better than the book using computer technology today. We can do some of the things the book can do better.
We can do some of the things a book can do, more cheaply. But, it is hard to do everything that a book can do.
So it's a good thing to think about as we compare what are new technologies are going to do.
When the book was invented people thought about the future of printing, and the best they can think of, in the 19th century when the Industrial Revolution came in, that they will be able to go from a hand wielded printing press to something run by steam. and then by electricity to really make inexpensive books and lots of different kinds of books
But in fact, what has actually happened to the surprise of most publishers and most makers of printing presses is electronic technologies came along and completely changed the dynamics.
So we didn't have something that is a slightly cheaper version of a big electrically or steam-driven web printing.
But, what we have is something that people can carry around with them and do printing on their desktops. The future here was quite unexpected.
In many ways, it still is. Many publishers are still grappling with something that they were told was going to happen 30 years ago but they didn't believe it.
Much of the same as happened with content. The Catholic Church does not try to stamp out the printing press in the 15th century because it was being used to print Bibles and that seemed to be okay.
But, within 50 years printers started making much smaller books that people can carry around; much less expensive books.
These books we're not about religious subjects but about ideas of all kinds. Many from the Greeks and the Romans.
And about a hundred years later, the huge change in thought from the Bible happened with Galileo and then with Newton.
And a 100 years after than we had huge changes in the way governments and social organization was thought of.
Whatever people thought the printing press was in the mid-fifteenth century, it turned out to be something completely different.
And most of us today think this was a good thing. I think it's a good thing. But, it certainly changed almost everything about the way the 15th century thought and did things.
McLuhan pointed out something really important, which most of us did not pay attention to and don't think about. He said, "We shape our tools and then they reshape us."
And, Thoreau said a kind of a more pessimistic version of this much earlier than McLuhan. He says, "We become the tools of our tools."
So once we make technologies we windup starting to serve them in various ways.
I can think of McLuhan's way of looking at it as optimistic and Thoreau's way of looking at it as as pessimistic.
Thoreau had an interesting comment about networks also.
When the first Atlantic telegraph cable went in, about 1865 or so, they asked him what he thought about it. He said he thought he would be afraid that he might find out that some European princess had just gotten a new hat.
So, he correctly anticipated the inability of people to use technologies seriously.
In fact, they would use it for all sorts of general human concerns and there would be a real tendency towards triviality. And that certainly happened today.
For me, I started off in math and science. I was a math major and molecular biology major in the mid-60s.
Around 1966 I saw the first real computer graphics program Ivan Sutherland had done a few years before; it is called Sketchpad.
It was a completely different use of computers that I was used to.
I was used to programming big mainframes, and here was a system that was done on a big mainframe but basically what was interesting was a 10 inch by 10 inch display on which you can draw things.
You could give them behaviours, sometimes mathematical behaviours.
This system would stimulate them dynamically. This is a completely different way of looking at computing.
My reaction to this was to think of this as if you extended it you could do all computing that way. So I came up with this idea of dynamic objects.
Next thing I saw was Engelbart's vision of personal computing, which was not unlike what we have today.
This picture was taken in 1966. There he is with the mouse he invented.
He's on a black and white display. He's dealing with hyperlinked documents.
He's collaborating with people up in Oregon from California.
He's doing something that is like the web, maybe a little bit superior to it, about 40 years ago.
The idea I got from that was, "Boy, this is hard to do on time sharing. So maybe we should do it on a desktop computer."
So this wasn't the first personal computer, but it was a pretty modern looking one.
It had a pointing device, multiple windowed screen, and so forth.
Because this Flex machine that Ed Cheadle and I did was aimed at non-computer professionals, I started visiting people who have been working with non-computer professionals.
The most interesting person I found was Seymour Papert who been working with children.
Papert was a mathematician like I was. He started doing some really profound things with kids.
Not just having them make pictures on the screen, but actually thinking about these pictures in a mathematical way, using an advanced form of mathematics called, Differential Geometry,
that was actually paradoxically much easier for children to understand.
So Seymour would take a young child and get them to close their eyes and walk in a circle with their body and ask them what they were doing. They would say, "Well, I'm going a little and turning a little, over and over."
In LOGO, going a little forward and turning a little is "TURN 5" and over and over again is "REPEAT". And so if you tell a Turtle to do this, then the Turtle would make a perfect circle.
It doesn't need Cartesian coordinates. It doesn't even need a center because the circle is that geometric figure that has constant curvature.
I thought this is the best idea that anybody ever had for computing. Which is, here is an area where it actually went beyond the book to be able to embody a special kind of powerful mathematics in a way that very young children could learn it. So I got very excited about that.
And my reaction to this was to draw a cartoon on the plane home that showed a couple of kids sitting out in the middle of a grass field having just programmed their own game of Space War and learned about "F = ma" and other parts of physics.
Their little DynaBooks here are communicating with each other with wireless communication. The research community I was embedded in at that time, the ARPA Research Community, had been working on the ARPAnet. I had also started on a wireless version.
I have this notion that finally there is a real reason for thinking ahead, that there would be machines like this DynaBook, and we can make them inexpensively enough for children. We could try a lot of Seymour Papert's ideas on them on it.
Then right after that this idea and a bunch of others from other people got concentrated at Xerox PARC.
Here's a cardboard model I made of the that cartoon idea and we decided to build a desktop computer again, but one that was kind of like what the DynaBook was going to be. We invented much of the modern technology that you used today.
We had object-oriented programming, dynamic animation, the windows user interface, desktop publishing, the ethernet, the laser printer. We invented part of The Internet and so forth. So there's a whole suite of technologies here.
Looking ahead, to just a few years ahead, that set of inventions done at Xerox PARC was done by only two dozen people.
The reason two dozen people were able to do all that is the technology had changed from discrete transistors to integrated circuits that had enough components on them so that relatively small numbers of people can build computers all by themselves.
The next revolution along these lines is happening now and is just starting to be seen in the in the marketplace.
But, by about 10 years from now, not just displays will be made out of conductive plastics, but actually the entire computer. and these conductive plastic, some of them, can be made by something like an inkjet printer sitting right on your desk.
You can imagine now a high school student, or college student, or computer scientist, or somebody anywhere in the world, who gets an idea for a computer, both hardware and software, can make the entire thing just on their desktop.
This is going to be an absolute revolution in both the cost of what it takes to make computers but also who will be able to make them.
So if we think about what is behind all this, it's as Mick Jagger would say, "It's only Math. and Science."
Essentially what's happening here is, instead of requiring 200,000 people to make an Egyptian pyramid out of old material, what we basically need are 5 to 5 t0 15 people who really know how to deal with the both the physical and informational universe.
That is sufficient to allow entirely new inventions to be made and spread around.
If we look at these three slides together and think about the Third World or the non-developed world, what this says is that the development in the non-developed world can primarily take place through education by learning new points of view and new ways of using the points of view.
That the need for huge steel mills or a billion-dollar integrated circuit factories to make computers or in the realm of biology of rather large installations to make fermentations of various kinds of bacteria and stuff.
This is going to change to something that's more like a kitchen science. The key to it is going to be the points of view and the kinds of knowledge that people can exert.
What we're talking about here is an education revolution here.
Now, most people in the regular world prize IQ. Leonardo is very justly celebrated for having a high IQ.
nBut, we have to ask what it would be like if you were born with an IQ of 500, but in 10,000 BC. How far would you get?
How far did Leonardo get with his huge IQ, born when he was born?
He designed many, many machines, but he couldn't invent a single engine for any of them that would make them work.
He was just born in the wrong time and his IQ wasn't strong enough to surmount the general knowledge of the time.
So, knowledge tends to trump IQ and always has, but basically there's so much different kinds of knowledge.
People have had knowledge of different kinds for hundreds of thousands of years. And so it's not so much the knowledge that is important, but the outlook.
I'm going to come back to this because the outlook that we have about what we think knowledge is, how we think we can get it, how we trust the knowledge, and how we distrust the knowledge, how we teach the knowledge, and so forth, is actually the key to making progress here.
In the last couple of years we've seen the start of real computers for children.
Thanks primarily to Nicholas Negroponte who was the spearhead behind the one laptop per child XO.
This kind of got some companies first to be against the idea, but then to develop versions of their own.
Like the Intel Classmate here, and there are others appearing on the scene.
Now that Nicholas has a topic of conversation and gotten other people to realize that children are a huge potential market and a huge untapped resource for making the world better.
Companies like Nokia, who make a million phones a day, more than 400 million phones a year, are another example of a another technology that is actually going to kind of converge with inexpensive communications technologies with displays with enough pixels to allow thoughts about education to seriously be thought about in the next several years.
This is something that's happening right now. The OLPC has just started their first real mass build a couple of days ago. And I'm sure that there are some talks about the OLPC at your conference.
When we look at this idea of children's computers and look at what is easy and what's hard,the easiest thing is actually to make the computer itself.
Even though it's kind of a monumental task over the last few years it was still done by a few dozen people and perhaps a hundred people at Quantum, the company that's making it.
So it's not that difficult to do technology today. The system software and so forth is also a fair amount of work, but it's definitely doable.
Where things start to get harder here is in thinking about the environments for the children, and really harder to start thinking about how we can teach children non-trivial ideas.
They got almost intractable when we start asking questions, "But, who is going to help the child learn these things? Where is the mentor? Where are the teachers? Where are the informed parents?"
And this question right now is somewhat unanswerable in the 1st and 2nd world.
In the developed world we don't have enough elementary school teachers who understand real math and real science.
We certainly don't have enough well-informed teachers in the developing world.
So, I believe this is actually the the problem that is going to have to be solved in order for all the rest of the stuff to make sense.
Of course most of the attention so far is being directed at the easy parts of the problem and the most physical parts of the problem and and so forth.
The more invisible parts of the problem are just starting to surface now that these machines are starting to show up in various countries and people are wondering, "What can we do with them?"
So we take a look again at the difference between IQ, knowledge, and the outlook of it, one of the big changes in history was kind of the outlook from thinking of most causes and effects that are important, as having a kind of a spiritual or a magical background.
Actually, our human brains are very set up by nature to assign superstitious, spiritual, and magical causes to things.
Most of humanity for most of the time we've been on the planet has had various theories about the world that are kind of propelled by angels if you will.
The Greeks started a revolution in thinking by trying to connect thoughts to get together in a kind of a way. That's kind of like articulating gears so when your turn one gear it actually causes a lot of other things to happen. In Greek mathematics it's very much like that.
But, of course there's more to a revolution in thinking then just creating gear-like causes and effects.
Gear-like causes and effects gives us engineering, but it doesn't give us science.
So we have to deal with what's wrong with our brains in order to get the science. A good example of this is to think about the assertion that the shape and sizes of these tabletops is exactly the same.
But, we can't see it. This one looks long and skinny. This one looks fat.
But, we can kind of prove it by picking up one of the table tops here and you can rotate this guy around. If I do it just carefully you can see it just fits nicely here.
In fact I originally got this table talk from over here and rotated it and put it over there.
This is an interesting problem here: That we actually are fooled in a way that we cannot see even after it's proven that we were fooled.
I have done this hundreds of times and I still see this as long and skinny and this is short and fat.
Another thing about our brain is as it's basically about dealing with now and simple changes, and now and simple extrapolations.
So almost everything we think about when there's a change, it has this simple incremental way of thinking about it.
But, usually we're actually embedded in something like this particular in the last few hundred years and particularly in the last 50 years.
And we just can't see this guy without something special. Just as an artist would have to measure those tabletops in order to draw them accurately, we have to do measurements and use mathematics in order to be able to see something that looks like this is actually one of these guys.
Far too many politicians, both past, present, and I'm afraid in the future, are going to be fooled by this.
So again our common sense is to aggregate things. So when we think about what can we make from bricks, the simple way of aggregating bricks is in a pile or in a kind of a stack.
If we make big aggregates that way, we get Egyptian pyramids or big walls.
But what we don't get, and what some of the smarter civilizations in history like the Greeks didn't get, is non-obvious combinations of these bricks.
Partly because the powerful idea that there could be non-obvious things hadn't really surfaced yet.
Here's a non-obvious way of dealing with bricks, and you can sort of see why it's difficult to invent the notion of an arch.
BeCause in order to make an arch you actually have to make a lot more than is here when you get done.
There's a whole scaffolding that goes in. The bricks are laid on top of it. You drop the keystone in. You put some more bricks around this wall and then you subtract out the scaffolding and everything stays up.
And it's both light and powerful. So you can make very big structures, structures the size of the Pyramids from about a millionth the amount of material by having one of these architectural ideas.
Much of modern thinking is actually a shift from simple aggregations of things to new kinds of architectures.
Anthropologists have studied human beings for maybe a 150 years now.
One of the things that they found are that all human groups, maybe they've looked at three or four thousand human groups, have a set of traits that is the same from group to group.
Every group has gotten good at coping. That's why they still around.
They cope partly through cultural knowledge.
They're social. They have a language a culture.
They have fantasies. They tell stories. They have tools, art, technologies, and so forth.
A whole bunch of things over here that are common.
(Every culture has a somewhat different.) Actual cultures have different, somewhat different actual religions. They have different actual stories.
But, the general traits here are the same from human being to human being.
Once you lay out these human universals it's interesting to look into various cultures around the world for things that aren't universal.
And you find things like this. For instance, the notion of progress is modern.
It's not a not an old idea at all because the old days you were just trying to survive.
You weren't trying to progress at all and almost all of the strategies in traditional cultures are for for coping rather than progressing.
Most people have lived and died on the earth without ever learning to read and write.
Without having a deductive, abstract mathematics, model-based science, and so forth.
Equal rights is an invention. There's no traditional culture that anthropologists have ever found that has this idea, particularly for women.
So, you get these rare ideas over here that are almost certainly human inventions.
if you think about it much of civilization depends on learning these things rather than these. And, much of modern education is about trying to teach these things.
Because they're generally more difficult to learn and they're more difficult to teach because they're less natural than these human universals.
So a lot of what we should be thinking about when we're trying to deal with education and make positive changes in education is, "How can we do this stuff better?"
"How can we find more of these guys better?"
Another McLuhan idea here is what I really like. He says, "Don't worry about whether it's right or wrong, just try to find out what is going on."
What he might mean here is that right or wrong we think of as absolutes but they're always relative to our belief structures.
If there's one thing that we've learned over the last four or five hundred years is that things we believe a few hundred years ago have turned out not to be so.
Many of the most disastrous things we've done over the last few hundred to a few thousand years have been because of incorrect beliefs.
So science is a way of trying to get around our tendency to believe in thing and trying to find out what's going on.
So this is great general advice from Marshall McLuhan.
In the wonderful OLPC XO, we have a real chance to do things that are like LOGO, like Mitchel Resnick Star Logo and, like the Smalltalk stuff that we did at Xerox PARC long time ago.
So we can actually experience some of these ideas by using our XO.
I'm using the E-toys software that runs on the XO. Here to have this little car here that is going to be driven by this little script here.
So the speed is always going to be 30. I'm going to increase the car's position by the car's speed each time.
If I let it run here I get a trail of dots because the speed is constant and the dots are 30 apart.
If I change the speed each time... for instance, if I say let's increase the speed by 30 each time and let the car run. What does it look like?
The car just takes off and I get this pattern that shows that in this unit time we went this far, in this unit time we went that far plus 30, and this time we went that far plus 30, and so forth.
So I get a kind of a visual pattern of what the accelerated speed looks like.
If I do that with some children a couple of months before a physical experiment we can get some interesting results.
So here's a a bunch of fifth graders.
The objects that you think will fall to the earth at the same time.
Okay. Not pay any attention to what anybody else dropping the same speed.
I think we should do is to drop the shot put ball and the sponge ball, because they are two totally different weights. If you drop them at the same time, they need to drop at the same speed.
In every classroom, there is at least one Galileo kid. Here. It was a little girl, who jumped to the same conclusion that Galileo did 400 years ago which is well...
What we should do is actually just try dropping a heavy one in the light one and we can hear whether they hit the ground at the same time or not.
So obviously Aristotle did not ask a child and Saint Thomas Aquinas did not ask a child either because he believed what Aristotle said.
So perhaps Galileo was the first adult in history to be able to think like children can think.
Kind of an interesting common theory. Then we can look at this closer by taking a movie.
Here's a movie of the ball dropping and we can see when it's dropping continuously it's hard to see what's going on.
Even when we single step it, it's hard to see what's going on.
But if we look at every fifth frame here we can see a little bit more what's going on.
If we stack up those fifth every fifth frames we can see a pattern here that when the children see that they say, "Oh! Acceleration."
So the question is, "How to measure the acceleration?"
What they do here is by measuring from the bottom of one ball on this frame to the bottom of the ball here are measuring from the bottom here to here.
So they get something like this.
And if they stack these these up carefully, this gets harder and they get older, they stack these up like this.
Then, they say, "Oh, the excess from here to here is this strip and the excess from the yellow on to the next one is this and the excess from next one to this one, and that's constant."
So this pattern is a pattern of constant acceleration.
Let's listen to how Tyrone actually did this.
And to make sure that I was doing it just right, I got a magnifier, which would help me figure out if the size was just right. After I'd done that, I would go and click on the little basic category button.
And then, a little menu would pop up. And, one of the categories would be geometry. so I click on that.
And, here it has many things that have to do with the size and shape of the rectangle. So, I would see what the height is. And I kept going along the process until I had them all lined up with their height.
I subtracted the smaller one's height from the big one to see if there was a kind of pattern anywhere that could help me...
In order to show that it was working I decided to leave a dot copy, so it would show if the ball was going the exact right speed and acceleration.
This is pretty cool. He decides to drop a dot from his simulated ball and shows that it matches up with the stacked frames here.
Another way to to do this is to run a movie and the animation at the same time.
By the way, 70% percent of college students fail to understand Galileo and gravity.
But, most fifth graders that we've worked with using this way of doing it, this kind of mathematics and this kind of approach, and doing their own simulations on the OLPC machine, are able to understand it much much better than most college students.
So, a payoff here is this little gravity script that they come up with can be used to make things move the way gravity makes things move on things more interesting than balls, like a spaceship.
They start that script running on the spaceship, it will be drawn down to the surface of the moon, and will crash.
But, we can add a motor here. so we can think of gravity is a velocity eater and the motor is a velocity producer.
You have a little script here to show a flame from the rocket ship when the motor is on.
This little script decides whether there's a crash or not if the ship is going too fast when it hits.
If I start the game and grab on to the joystick here, you see I can oppose gravity here and land it carefully and not get a crash this time.
This is a game people used to pay money with.
Another way of thinking about things is that simple cause-and-effect relationships, like gears, don't scale very well.
So you can build a clock, perhaps out of a thousand gears. But then it gets really complicated to build more complicated things out of gears.
And simple connections of wires between things starts getting really complicated.
This is an actual room, I won't say where, and there's a sign here that says, "Do not touch any of these wires."
The research group I was part of, the ARPA/PARC research group, realized that you can replace any mess like this with a simple message sending network like an Ethernet or an Internet.
And all things large and small that have to be connected together can be connected in a very simple way.
This simple way is actually a looser coupling way than tightly tying wires together.
This is another outlook that we have to think about.
So here's a really simple example of using the gravity thing we just discovered.
So the gravity is working on that green line, you can think of it as being a lid on top of these atoms. The atoms are being moved around because they're hot.
Temperature is motion. And, they're exchanging momentum with the green lid.
And so they're keeping that lid up.
We have a thousand of these going so this is a using a version of Mitchel Resnick's Star Logo that we made in Squeak.
If I turn that turtle count into 500 here, the weight of this lid is being pushed upwards by half the number molecules.
So what happens is we have a pressure decrease that lowers this thing about halfway down.
You can see this. If I put in say, 2000 here, I get an explosion and some nice shockwaves.
But, eventually it settles into an equilibrium that's about twice what it was with a thousand.
You can see some simulations here of Boyle's law done with this very nice simple computer simulation that children can easily make up to understand how aggregates of lots of things actually work together.
This extends our notion of outlook.
So the outlook of Newtonian physics, and so much of the 18th century, and some of 19th century physics, and a lot of the ways we teach mathematics and science today, is kind of this very tightly bound cause-and-effect relationship.
But, the latter part of the 20th century in many, many ways has gotten much more biological.
So we can think of this as a transition from gears to biology and this transition is not just in the world of molecular biology, but also the Internet is made this way.
And, object-oriented programming is made this way. It's kind of a no-centers, distributed, loose-coupling way of modeling things.
This way of thinking about things is the probably the most comprehensive, most powerful way we have of thinking about complexity today.
So it's natural to try and think as hard as possible about how we might help children learn this.
One of the ways of doing this is by having them make ever more complex models on the computer and to see the unintended emerging properties that come from dealing with many objects loosely coupling, working together.
We've all heard about the blind men and the elephant trying to figure out what it is. If there's an elephant in modern education today, that element is Mathematics.
The last couple of years I've been traveling around talking to teachers at various levels in the United States.
And in the United States, at least, if there's a common denominator to why things aren't happening in both math and science, it's basically, at the root is that the adults who teach children generally lacked fluency in mathematical thinking.
And I don't mean knowledge of mathematics. Some adults have some knowledge of mathematics.
But, the ability to think heuristically, to be able to represent things that haven't been represented before, to think about relationships between them.
And, all of the things that are kind of the background knowledge and skills of mathematics and of modern science are pretty much lacking for most of the K-12 teachers that I've talked to.
Mathematical, heuristic thinking, or problem-solving heuristic thinking, or looking at things from multiple points of view heuristic thinking, is one of the biggest problems that we have to face in in the future, at least in the United States.
Another problem, again in the United States, but I think affecting everywhere, is this idea of "amusing ourselves to death".
My friend, Neil Postman, wrote a great book about it.
If he were to write that book today, I think he would re-title it: "Distracting Ourselves to Death."
Because one of the ways we're using electronic media is to make an enormous amount of different kinds of experiences that attract our cave-person brains, in a way.
Many of these experiences are keyed into these human universals of things that we want to do. We want to communicate with other people, we want to go to the theater, we want to tell stories and so forth.
And now we've been able to multiply that with the power of the Industrial Revolution and it's produced an incredible set of distractors.
For instance, there is now a television show in the United States that is a contest for people playing air guitar with the game, "Guitar Hero".
So, this has gone from being a fantasy game to something that is now an actual pursuit by people who are very, very far from learning anything about actually playing a playing guitar.
This is a real disaster. A nice quote here is, "The bull wears itself out on the cape and fails to see the sword."
If we're going to make big changes in education, we have to somehow get aware of the cape, and concentrate more on what's important with our limited amount of time and our limited ability to use our brains.
And, finally, maybe one of the biggest trends happening today, again because of electronic media, and because of everybody becoming aware of everybody else, is this thing that goes to the root of what it means to be a human being, and especially a human being in a traditional society.
As it's for most people they're trying to live their life in the present, and the meaning of life is too abstract for them.
You could think about a lot of what our impulses towards civilization have been, since the Greeks and particularly in the last few hundred years, as trying to go from just experiencing being alive, which everybody has for the last 200 years.
They're trying to get more of a handle on what the meaning of life could be.
I think we know enough today to realize that the meaning of life is something that we can make up.
It's part-and-parcel of what it means, for example, to say that people should have equal rights and to try and teach children to believe in equal rights and to try to act as though equal rights are something that is naturally in the world rather than something made up.
So I think this change from naturalistic living to a more artificial, but perhaps richer, perhaps nicer, perhaps more productive, perhaps more progressive form of living is the kind of thing I, at least, would like to strive for.
And I hope everybody else who is interested in helping children learn powerful ideas will get interested in also.
So thank you very much again. I'm sorry not to be there but I was glad to make this video for you.
I hope it helps you think about what it means to have a personal computer for every child and on the planet. Thank you.