Compiled from: Alan Kay's "The Real Computer Revolution Hasn’t Happened Yet"
Thirty-two years ago, in 1975, I was one of the lucky Americans invited to Pisa to celebrate the 20th anniversary of computer science in Italy. I submitted a paper presenting our preliminary results on the invention of personal computing at Xerox PARC. Over the years, I somehow lost that paper, but the more organized Professor Atardi found a copy of it, which has been republished as part of our ceremony today. In this speech, I would like to reflect on that paper and see how past work has influenced the present. However, I prefer to talk about future possibilities, so I wrote some historical notes to provide context for the 1975 paper and now attempt to discuss some more significant and also more subtle gifts that globally networked personal computers can bring to humanity.
One connection to the past is that the researchers who invented fundamental technologies today, such as personal computers, bit-mapped screens, overlapping windows, icons, and pointing interfaces, object-oriented programming, laser printing, Ethernet, and the Internet, were driven by the transformative achievements of the printing press. In short, the publishers of the 15th century were initially seen as a low-cost automation of handwritten documents, but by the 17th century, several of its special characteristics had gradually changed the way people thought about important ideas, to the extent that most of the significant ideas and the way people thought about them did not exist at the time of the invention of the printing press. The two most important ideas are the invention of science and new methods of social organization and politics (which in several important examples are extensions of scientific viewpoints themselves).
These changes in thought also altered the meaning of "literacy," as literacy is not just the ability to read and write, but the ability to fluently handle ideas that are important enough to be written down and discussed. One special aspect of the publisher is its ability to fully replicate an author's article, which creates a very different form of argumentation. One way to view the true printing revolution of the 17th and 18th centuries is to argue about what was being co-evolved and how the arguments were conducted. Increasingly, it was about how the real world was constructed, both physiologically and psychologically, and this argument increasingly used and promoted mathematics, attempting to shape natural language into forms that were more tightly connected logically and less like stories.
One of the realizations about computers in the 1960s was that they provided new, more powerful forms of argumentation for many important problems through dynamic simulations. That is, rather than expressing tedious points with prose and mathematical equations, computers could calculate the implications of these claims to better understand whether they constituted a valuable model of reality. Moreover, if future common cognition could include writing these new claims rather than just consuming (reading) them, we would have inventions similar to the next 500 years after the advent of printing, which could very well improve human thought.
These are indeed very ambitious aspirations! From the invention of the printing press to the great transformations of the 17th century, about 150 years passed, which means that a larger social revolution had occurred as children grew up with a different view of their ability to think, debate, learn, and communicate in a more coherent form with clear writing. Our thoughts on this come from a visit in 1968 with mathematician Seymour Papert, who invented the LOGO programming language for children and began to demonstrate certain forms of higher mathematics that, when presented in dynamic computer form, matched perfectly with the way children think.
As McLuhan pointed out in the 1950s, when a new medium appears, it is first rejected on the grounds of being "too strange and different," but if it can accept familiar old content, it is usually gradually accepted. Years (or even centuries) later, if the hidden properties of this medium lead to changes in the way people think and appear in the guise of a wolf in sheep's clothing, it will be a huge surprise. These changes are sometimes beneficial (I believe printing was beneficial, although the Catholic Church might disagree) and sometimes not (I think television is a disaster, although most marketers would disagree).
Thus, we have a sense that the ability of personal computers to mimic other media (according to Moore's Law, at a low cost) will help establish their position in society, which also makes it difficult for most people to understand what they really are. Our thought is: if we can get children to learn real things, it also makes it difficult for most people to understand what they really are. Thirty-two years later, the technologies invented by our research community have been widely used by over a billion people, and we have gradually learned how to teach children real things. But it seems that the time required for a true revolution is longer than we optimistically estimated, largely because the commercial and educational interests in old media and modes of thinking have frozen the development of personal computers at the level of "mimicking paper, recordings, movies, and television."
Meanwhile, what computers can truly do—whether simulating or arguing—has been accepted by the disciplines of science, mathematics, engineering, and design. And those interested in Papert's vision of improving the nature of children's thinking have made considerable progress over the past 30 years by actually constructing "powerful ideas" to help them learn. There is now much talk, demonstration, and education about "what children can do."
The shame of computer manufacturers (including hardware and software vendors) is that they have not created any commercial intellectual amplifiers for children. All machines and software tools are primarily aimed at businesses, and to some extent, at families.
All machines and software tools are primarily aimed at commerce, and to some extent, at families. This completely ignores the most disastrous needs in the world. What children around the world need are new ideas and ways of thinking—children's computers are essential because they are the best way to help children learn these new ideas now—and they are also much cheaper than paper for books and other traditional media.
Two years ago, several members of the research group that invented personal computing in the 1960s decided to create an extremely inexpensive personal laptop for all the children in the world—Dynabook. This initiative called “One Laptop Per Child” was initiated by Nicholas Negroponte, involving both new and old researchers, including Seymour Papert, our research institute, and many other interested designers, all aimed at bypassing the huge gap created by commercial interests.
This community has always been willing to design and build whatever is needed, regardless of whether suppliers have the tools and materials. The Alto at Xerox PARC is a great example. All the hardware and software were completed at PARC, and the small assembly line that was eventually established produced the first 2,000 of the modern personal computers. Today, most laptops are manufactured in Taiwan or mainland China, and different brands (like HP, Dell, Sony, and Macintosh) may all be produced by the same factory. So, if you want to manufacture your own laptop, just put your design in, request an order of about a million, and then fly to Taiwan! OLPC aims to create a fully functional low-cost machine, so it is interesting to see where the money you pay for a laptop is allocated.
For example, 50% of the price of a standard laptop comes from sales, marketing, distribution, and profit. OLPC is a non-profit organization that sells products directly to countries. Another 25% of the price comes from commercial software, most of which comes from Microsoft. But there is a worldwide open-source and free software community that is equivalent in many ways, especially in network and educational environments. The remaining spending items include disk drives and displays. However, the flash memory used in cameras and memory sticks may be cheaper than the cheapest disk drives (and more robust because it is solid-state). Displays are a special issue because cost is not the only concern. Displays in third-world countries need to consume less power and must be viewable in direct sunlight without backlighting. OLPC researcher Mary Lou Jepson excellently solved the display issue by inventing a new type of flat display that has a higher resolution than ordinary displays (200 pixels/inch), consumes 1/7 the power of ordinary displays, and costs 1/3 of ordinary displays.
As a result, a computer currently sells for $170, can hold hundreds of books (many of which are dynamic), with each book costing about 20 cents, and maintains an automatic mesh interconnection with other laptops. This computer has been scoffed at by hardware and software vendors, but they are now starting to produce similar low-cost products themselves (for example, Intel now has a "400-dollar laptop," and Microsoft recently announced they will sell their software for a few dollars to the third world). It is good to say they are now seeing the light, but more likely, they are just feeling threatened and responding.
One of the benefits of doing this for a non-profit organization is that as material costs decrease, manufacturing costs decrease, and all the savings will simply be passed on to the children. Meanwhile, the first phase of this project has been completed in just over two years, so many customizable materials are prepared for the next phase. If all available technologies and manufacturing techniques are put to work, it is entirely possible to manufacture a laptop for $50 or even less.
Of course, the hardware part of this project is just a small part, although creating a fully functional "100-dollar laptop" is quite challenging. There is also system software, end-user creation environments, educational content, various packaging and documentation, and most importantly, mentors are needed to help children learn powerful ideas.
I will return to the key part of the educational ecosystem. Now, let us note that for mathematics and science in first and second world countries, the proportion of elementary teachers and parents who truly understand mathematics and science is too small, and most children are left unguided through this threshold. In third-world countries, the proportion of knowledgeable mentors is negligible.
This leads to a frustrating deadlock. As I will demonstrate in a minute, it is now known how to help 10- and 11-year-old children master powerful calculus and other advanced mathematical thinking. But no child has ever invented calculus! The wonder of modern knowledge is that, with the aid of writing and teaching, many ideas that would require genius to invent (in the case of calculus, two geniuses are needed) can be learned by a broader, not particularly talented crowd. However, even for geniuses, inventing in a vacuum is very difficult. (Imagine being born in 10,000 B.C. with an IQ of 500. Not much would happen! Even Leonardo da Vinci could not invent an engine for any of his vehicles. He was smart, but he lived in the wrong era and thus did not know enough.)
If children have already learned how to read, they can, to some extent, bypass adults—whether at home or at school—by going to the library and learning from reading. There are many such examples, and it is likely that a considerable part of the printing revolution occurred gradually this way. However, for a child, if there is no adult help (or at least no adult cooperation), it is very difficult for the child to learn to read, and we see again the critical importance of guidance. When Andrew Carnegie established thousands of free public libraries in the United States, each library had a special room where librarians taught reading to anyone who wanted to learn!
Of course, children can learn a lot through experimentation and sharing knowledge without special guidance. But we do not know of any examples that include humanity's great inventions, such as deductive mathematics and science based on mathematical experience. To put it metaphorically: what if we made a cheap piano and put it in every classroom? Children would certainly learn to do something with it on their own—this could be fun, possibly genuinely expressive, and it would certainly be a form of music. But this would miss out on what great musicians have invented over centuries in music. For music, this would be a shame—but for science and mathematics, it would be a disaster. The special processes and prospects of the latter (especially in science) are so critical and subtle that not teaching them as "skills which allow the art" is very harmful. As Ed Wilson pointed out, our societal interests, motivations, communication, and the genes of invention are essentially those of Pleistocene humans. Our so-called modern civilization is largely created by inventions such as agriculture, writing and reading, mathematics and science, and governance based on equal rights. These are all difficult to invent and best learned through guidance.
Therefore, we must find ways to address the tutoring issue, not just in third-world countries but also in first and second world countries. This fall, we could easily produce 5 million OLPC laptops, but no matter how much money we spend, we cannot cultivate 1,000 new teachers with the necessary knowledge and skills (partly because it takes years for humans to learn and practice the knowledge they need). This is one of the reasons education lags severely behind scientific, technological, and other intellectual advances.
Sometimes, when very young children are in a good learning environment, it can be astonishing to see what they can do. The most important principle in early childhood learning is to figure out what they can do, what kinds of ways of expressing ideas work best for them, and what kinds of social environments stimulate their inner desires to gain competence in the world they live in.
First-grade teacher Julia Nishijima, whom we met 15 years ago at a school, is somewhat unusual because she is a natural mathematician.
We believe she has never formally studied mathematics and has not taken a calculus course. But she is like a gifted jazz musician who has never taken formal lessons. She truly understands the music of mathematics.
She has an innate mathematical worldview, which is one of the most interesting projects we see in her classroom. She lets children choose a shape they like, with the idea being to create the next larger shape with the same shape using only these shapes.
Here are diamonds, squares, triangles, and trapezoids.
Julia then has her students reflect on their creations. She views mathematics as a science, allowing children to create structures with interesting mathematical properties and then analyze them. Six-year-old Lauren noticed that making the first tile required one tile, and the total number of tiles was one. Making the next shape required three more tiles, bringing the total to four.
Five more tiles were used to make the next one. Soon, she saw, "Oh yes, these are just odd numbers, each differing by two, I add two each time, and I get the next one."
The sum of these numbers is a square number, at least 6 times 6 (she is not quite sure if it is 7 times 7).
Lauren discovered two very interesting developments that mathematicians and scientists in the audience would recognize.
Then the teacher had the children bring their projects to the front of the classroom and place them on the floor so they could all look at them, and the children were very surprised because all the progress was exactly the same!
Each child filled out a form that looked like Lauren's, meaning that all these growth patterns were exactly the same, and the children discovered a universal law about growth.
Mathematicians and scientists reading this will recognize that the odds are generated as a first-order differential relationship, giving a smooth, consistent series—in the mathematics of computers we use, this is represented as increasing incrementally:
Repeat: odds increase by 2
The total number of tiles is generated by a second-order differential relationship (because it uses the results of the first-order relationship):
Repeat: total increases as odds increase
Incremental increases are easy for everyone because it simply involves putting things (that is, adding things) into a pile of things.
Mathematics is essentially "careful thinking about how representations of ideas could imply other representations of ideas," and the most important process to help anyone learn how to think mathematically is to place them in many situations where they can more carefully utilize their current way of thinking. We can see that children are able to find a very good way to think about both growth and change. "Incremental increase" is a very powerful concept because many changes in the physical world can be simulated through one or two "increments," which is a representation that children can easily understand.
Now, what is a truly good way to help children think about the Pythagorean theorem?
In the lower left is Euclid's proof of the Pythagorean theorem, suitable for high school students. It is elegant and subtle, illuminating other areas of geometry, but it is not a suitable initial proof for most young people.
In the lower right is a very different proof: perhaps the original proof of Pythagoras. We have seen many elementary school children prove this by playing with triangles and squares. Demonstrating arrangements, using 3 triangles to enclose square C into a larger square, copying the larger square, removing square C, rotating the two triangles, and noticing there is space for squares A and B, moving them. Success! This proof has a heartfelt method and feeling, a powerful simplicity, perfect for beginners, and provides a solid foundation for later viewing this idea more abstractly and subtly.
The rule of thumb here is to find ideas and expressions that allow "beginners to act as intermediates," that is, to let learners immediately start engaging in real activities in some authentic form.
The way calculus views many ideas is so powerful and important that we want children to start thinking along these lines at an earlier age. Thus, we created a real form of calculus that can be thought about by young minds, and computers bring it to life in many delightful ways.
One project that 9, 10, and 11-year-olds around the world love is designing and creating a car they want to learn to drive. They first draw their car (and often equip it with large off-road tires like this).
So far, this is just a drawing. But then they can "dive into" their drawing to see its properties (such as the car's position and direction) and behaviors (the ability to move in its direction or change its direction by turning). These behaviors can be extracted and removed from the "world," generating a script—no input needed—that can be set to "tick" by clicking a clock. The car begins to move according to the script.
If we put the pen of the car on the "world," it will leave a trail (in this case, a circle), and we see this is Papert disguised as the logo turtle—a turtle dressed in "clothing," observing, writing, and controlling it in a simple way.
To drive, children find that changing the car number after turning will change direction.
Then, they draw a steering wheel (something like the car, just looking different) and see if they can immediately turn the steering wheel after the car turns, which might allow the steering wheel to influence the car.
They can obtain the forward direction they are heading and put it into the script. Now they can drive the car with the steering wheel!
Children have just learned what variables are and how they work. Our experience shows that they learn a lot from this one example.
They soon find it difficult to control the car. They need to introduce a "gear" in the connection between the wheels and the car. They can get the needed advice from teachers, parents, friends, or children thousands of miles away through an online tutoring interface. They open the expression in the script and divide the number output from the steering wheel by 3. This scaling makes the steering wheel's rotation have less impact. They have just learned the real use of division (and multiplication).
Quite a bit of what is happening is "just doing," so reflecting on what just happened is also a good idea. One way to do this is to let objects leave traces, showing what they are doing over time.
If the speed is constant, then the trajectory of the points is evenly spaced, indicating that the same distance is covered in each small segment of time.
If we increase each tick of the clock, we will get a pattern similar to the second image. This is a visual pattern of uniform acceleration.
If we make the speed random each time (in this case between 0 and 40), we will get an irregular distance pattern for each tick.
Using randomness in two-dimensional space is quite interesting. Here, we can randomly move a car forward and then randomly turn. If we put down the pen, we will get a series of "Drunkard’s Walks."
From one perspective, over enough time, a little randomness will lead to a large area being accessed, which will greatly change the probability of collisions (this probability is still low, but now "more likely").
Thus, we might guess that random traversal in scripts using feedback principles can accomplish surprising things.
So far, our examples have been fundamentally mathematical because they deal with the relationships between ideas expressed by computers. These ideas may resemble the real world (models of speed and acceleration) or may not (the car has no support in the examples of speed and acceleration but does not fall because there is no "gravity" in the computer world unless we model it). Sometimes we can fabricate a story that resembles the real world, and we can even have a reasonable guess. But for most of human history, guesses about the physical world have been far from reality.
Natural science truly began when people started to carefully observe and measure the physical world, first precisely mapping navigation, exploration, and trade, and then observing more phenomena more carefully with better instruments and techniques.
Another good example of "high attention low cost" is measuring the circumference of bicycle tires for fifth-grade students. Many philosophical golds of science can be found in this observational activity. Students used different materials and got different answers, but they were very sure there was an exact answer, measured in centimeters (partly because school education requires them to get accurate answers, not real answers).
One of the teachers thought so too, because on one side of the tire, the tire diameter was 20 inches. The teacher "knew" that the circumference was π*diameter, "π is 3.14," and "inches multiplied by 2.54 converts to 'centimeters,'" etc., and multiplied to get the tire's "exact circumference" = 159.512 cm. We suggested they measure the diameter, and they found the diameter was actually more like 19¾ inches (without inflation)! This was shocking because they were all set to believe almost everything recorded, and they did not think to independently test what was recorded.
This led to issues of expansion under different pressures, etc. But most people still believed there was an exact circumference. Then, one of us contacted the tire manufacturer (who happened to be Korean), and we had many interesting email exchanges until an engineer replied, saying, "We actually do not know the circumference or diameter of the tire." We squeeze them and cut them to a length of 159.6 cm ±1 mm tolerance!"
This really shocked and impressed the children—tire manufacturers do not even know its diameter or circumference! This gave them a more powerful idea. Perhaps you cannot measure things accurately. Aren't there "atoms" below? Don't they shake? Aren't atoms made of shaking matter? And so on. A similar question is, "How long is the coastline?" The answer is partly due to the scale and tolerance of measurement. As Mandelbrot and others interested in fractals have shown, the length of a mathematical coastline can be infinite, while physics tells us that physical measurements can be "almost" as long (very long).
There are many ways to leverage the powerful idea of "tolerance." For example, when children do gravity projects and come up with models of how gravity affects objects near the Earth's surface (see the next project), it is very important for them to realize that they can only measure within a pixel on the computer screen, and they can also make small slides. Measuring completely literally will lead them to overlook what uniform acceleration is. So they need to tolerate very small errors. On the other hand, they need to remain highly vigilant about differences beyond typical measurement errors. Historically, Galileo could not accurately measure how a ball rolled down an incline, and Newton could not know the actual situation of Mercury's orbit when observed closely.
Children discovering, measuring Galileo's gravity, and mathematically modeling it
A good example of "real science" for 11-year-olds is studying what happens when we throw objects of different weights.
Children think heavier weights fall faster. They think a stopwatch will tell them what happens.
But it is difficult to judge when the weight is released and when it reaches.
In every class, you usually find a "Galileo child." In this class, a little girl realizes: actually, you do not need a stopwatch; you just need to drop both the heavy and the light and listen to see if they hit at the same time. This is the same insight Galileo had 400 years ago, and it is clear that in the previous 80,000 years, no adult (including very smart Greeks) had thought of it.
To get a more detailed understanding of gravity near the Earth's surface, we can use a camera to capture the dynamics of gravity falling.
We can see the position of the ball frame by frame, spaced 1/30 of a second apart. To make it easier to see, we can pull out one frame every five frames and place them side by side:
Another good way is to draw each frame, draw the unimportant parts, and then stack them. When children do this, most of them will immediately say "acceleration," as they recognize that the vertical spacing pattern is the same as the horizontal spacing pattern they played with cars months ago.
But what kind of acceleration? We need to measure.
Some children will measure directly on the expanded frames, while others prefer to measure on the stacked frames.
These semi-transparent rectangles help observe because the bottom of the ball can be seen more accurately. The height of the rectangles represents the speed of the ball at that moment (speed is the distance moved in unit time, in this case, about 1/5 of a second).
When we stack these rectangles together, we can see that the differences in speed are represented by the small bars exposed, and the heights of these bars appear to be the same!
These measurements show that acceleration appears quite constant, and children wrote scripts like this for their cars months ago. The quickest realization is that because the ball is vertical, they must write this script so that the vertical speed increases, and the vertical position y changes. They drew a small circle to simulate the ball and wrote the script.
Now, how to prove this is a good model of what they observed? 11-year-old Tyrone decided to treat it like he did with his car months ago: leave a dot copy to show that the path of his simulated ball matches the position of the real ball in the video perfectly.
Here is what he had to say as he explained what he did and how he did it:
To make sure I was doing it right, I looked for a magnifying glass that could help me judge whether I had—if the size was correct.
When I finished, I clicked the basic category button, and a small menu popped up, one of the categories was geometry, and I clicked it.
Here it has many things related to the size and shape of rectangles. This way I could know how high it was. I kept along this process until I got them all aligned with their heights.
I subtracted the height of the smaller dot from the height of the larger dot to see if there was a pattern that could help me. My best guess was correct: to prove it was right, I decided to make a dot copy (so the ball could move at the correct speed and acceleration).
An 11-year-old child's investigative work!
In the United States, about 70% of college students do not understand gravity when studying it near the Earth's surface. This is not because college students are smarter than fifth graders, but because most college students learn these ideas in contexts and mathematical methods that do not suit their way of thinking. We find that over 90% of fifth graders can understand changes through this better context and expression.
Now that the children have "captured gravity," they can use it to explore other physical conditions and create games. If we redraw this ball as a spaceship and create one that can land on the moon, then for a 12-year-old (and most 11-year-olds), making the classic game "Lunar Lander" becomes quite easy. The gravity script is standard, accelerating the spaceship downwards towards the moon. Children can add a rocket engine script controlled by a joystick to accelerate the spaceship upwards. Note that in each case, ySpeed increases in one direction or another.
Children made some nice decorations, such as when the spaceship crashes with too high a downward speed, and when the rocket engine of the spaceship starts, flames are displayed.
Many physical phenomena can be simulated with "increments," including inertia, orbits, springs, etc. But now let us look at another powerful idea: a method that allows progress to be made without sufficient information to formulate a complete plan.
Sometimes we have enough information to formulate a foolproof plan. But most of the time, things do not go as smoothly as expected (even if we have a "foolproof" plan), and we end up having to seek new information, make new corrections, and sometimes new plans.
The ability of all animals and other mechanisms to gather information is quite limited, and their ability to infer the future is also quite limited. For example, the simplest bacteria may be harmed or killed by excessive (or insufficient) acidity. They have evolved molecular machines that help them detect when some dangerous substances begin to affect them, and those swimming animals exhibit a rolling behavior that fundamentally (and randomly) changes their swimming direction. If things are "good," they do not fall; if "bad," they fall again.
This general strategy of perceiving "good" and "bad," and doing something that might make things better, is very common in biology and is now present in many human-made machines and motors.
An interesting activity is to have children blindfolded and find their way outside the school building just by touch. The simplest strategy is to walk along the wall, always going in the same direction when lost.
We can make our car do this, drawing a different color as a "touch sensor," and writing a script that looks like:
Then we ask the children to make a car and a road, letting the car go to the center of the road instead of outside. There are many solutions. Here are two 11-year-old girls who did well together.
They found that if they made the edges of the road with two different colors, then there were only three situations: when the sensor was in the middle, or when the sensor was on one of the two sides. Their car, road, and script looked like:
We can see that this is a better script than the one we showed them. They decided their robot car would only move forward when it was in the middle. This means it can safely navigate any turn (the first example cannot always do this because the turning radius is 5).
Now let us simulate typical animal behavior used to track chemical signals in the environment, capable of perceiving the relative concentration of chemicals and being able to remember past smells and concentrations well, deciding whether to continue forward or try a different path.
We will choose a salmon swimming upstream to spawn at the place of its birth.
In our model, we will avoid the dramatic leaps backward off waterfalls and focus on how the salmon guides itself by sniffing out a special chemical from the spawning ground and remembering the concentration it last sniffed.
To simulate the chemicals in the water, we will use a color gradient, with darker colors indicating a higher concentration of the chemical. Etoys not only allows us to perceive the color beneath objects but also the brightness of objects. Thus, in this simulation, lower brightness means "closer."
Below we see the salmon successfully finding the darkest corner, and on the right, we see the path it has taken. The script under the "river" has a turtle that follows the salmon's position, drawing a trail on different playgrounds. The supports and scripts under the "river" make the salmon's body undulate like a fish.
This simple "try, test, and if possible, continue doing it; if not, do it randomly" pattern can be found in most organisms, starting with bacteria, and one abstraction of it is the working principle of evolution.
Although every phenomenon in the world has its uniqueness, the best way to understand most things is to view them as members of species with common characteristics and behaviors. Since computers are very good at quickly and cheaply replicating things, we can use them to turn an individual model into a model with many participants. For example, we can introduce any number of salmon into the model. This means we can model population ecology and individual ecology.
Ants are a great example that children can learn from and imitate. They use their ability to perceive and follow gradients to mark paths by setting scent trails, helping other ants find "interesting things" (usually food), thus communicating with each other. Ants are "social animals," and their behavior often resembles that of a larger organism, with its "cells" capable of independently sensing, thinking, and acting.
A striking example of the "new powerful argument" is that self-written computer simulations can help clarify a very complex and difficult-to-understand threat—such as a slow but deadly epidemic like AIDS—in ways that go beyond simple claims and assertions. One issue in understanding AIDS is that the virus has an incubation period of up to five years or longer. According to normal human common sense, when the most proactive actions can be taken to stop the infection, "nothing happens." In many traditional societies around the world, common sense has been valued, and not taking any action has resulted in devastating consequences years later and continues to do so. One of the simplest investigations that a creative system capable of displaying thousands of elements can accomplish is to create a simulated community, trying different scenarios of infection, slow-moving incubation, and hopeless situations. In the slow incubation period, it can initially appear that "nothing seems to happen," but ultimately everyone dies. Because learners have built the simulation themselves, they have created dynamic mathematics and modeling that they care about. They can choose initial conditions, and the emotional impact of the same disastrous outcome will profoundly affect their views on the epidemic.
Now there are thousands of Etoys projects completed by children in many countries around the world in their native languages, including: the United States, Canada, Mexico, Argentina, Brazil, France, Germany, Spain, Japan, Korea, China, Nepal, and more. The low-cost OLPC laptops and other products it inspires will soon reach millions of children. Therefore, just as the advent of books in the 15th century transformed learning potential in much of the world—as McLuhan pointed out, the potential to transform itself—has now arrived.
About 40 years ago, we began our research aimed at helping children—and thus helping humanity—learn and absorb "science in the large." We believe science is all processes that help "make the invisible visible." The so-called invisible refers to things that we humans cannot see for various reasons, not only including scientific objects that are usually of interest to us because they are too small or too far away, or emit wavelengths we cannot perceive, but also those ideas and objects we cannot see because our mental organs cannot think them or reject them (because they "cannot possibly be true"), and so on.
I include all "serious arts," whose purpose is to awaken us and make us aware that what our consciousness presents to us is not reality but a story that may be far from reality and sometimes even dangerous. What science does, rather than changing or fixing our noisy mental organs, is to add additional processes in our minds (outside the society of scientists) to discover many of our errors and attempt to reduce their scale and variety.
As Thomas Jefferson said: "The moment a person forms a theory, his imagination sees, in every object, only the traits that favor that theory." The larger scientific society plays a role of a "super scientist"—far beyond the scope of what any individual "knows." In this super-organism, the debugger of ideas is better and more skeptical than most people in their own minds. The super-organism has more views on how the universe works than any individual, and these views are very useful (even if some motivations may not be scientific). Therefore, without needing to anthropomorphize science, we have ample reason to say that "science" is smarter, more knowledgeable, and more visionary than any individual, and is a "better scientist."
A larger society can also behave smarter than most people and is less likely to make catastrophic decisions and unnecessary aggressive actions. In democratic societies, especially in democratic republics, the purpose of education is to engage all citizens in the most powerful thinking processes, dialogues, and debates that have already been invented, and representatives of democratic republics must be elected by the entire society. Jefferson again:
I know of no safe repository for the ultimate powers of society but the people themselves; and if we think them not enlightened enough to exercise their control with a wholesome discretion, the remedy is not to take it from them, but to increase their discretion by education.
H.G. Wells said, "Civilization is in a race between education and catastrophe." Perhaps the word "education" is too vague here. I want to replace it with "a race between education of outlook and catastrophe," because it is not knowledge itself that makes the greatest difference, but the outlook or point that provides a context in which rational thinking actually matches the real world and serves humanity. For example, in the 20th century, an environment that some people considered a non-human pest was allowed to establish because we exterminated the pests, so logically deciding to exterminate humanity in this dreadful environment. This is not uncommon in human history, having occurred more than once in the 20th century, and is happening now. Slavery is another product of a dreadful environment and expediency, which still exists in various forms around us today.
The first step of science is the astonishing realization that "the world is not as it seems," many adults never take this step, instead treating the world and their inner stories as reality, often leading to catastrophic consequences. The first step is an important step, best taken by children (most people realize this when they are very young). From then on, taking another big step to include ourselves (that is, all of humanity) as suitable subjects of study: trying to transcend our own stories, better understand "what we are?" and ask "how can we mitigate our shortcomings?"
While today's world is far from peaceful, there are now larger groups of people living in peace and prosperity than at any time in history. The enlightenment of some has brought about a community of worldviews, knowledge, wealth, commerce, and energy that helps those who are less enlightened perform better. The first part of this true social revolution was driven by the printing press, and this is no coincidence. The next revolution of thought—such as thinking and planning of the entire system leading to new significant changes in viewpoints—will be driven by the true computer revolution—it may come just in time to overcome catastrophe.