|
I thought I would start with the question that draws out two very different ways of thinking about play. And the question is this:
Would you prefer your child learn to play the stereo or to play the piano? They both use the same word 'play' but I think these are two very different senses of playing.
You might think that playing the stereo seems like such an obvious choice. You know, it only takes about 30 seconds to learn how to do it. You hear so much these days about the importance of ease of use, that it seems like a hands down winner. Once you learn how to play the stereo you could play any music that's ever been written. What could be better than that? But in fact if you really want to get a deep understanding and a deep appreciation of music, learning to play the piano is a much better path. And I think that's because playing the piano is a much richer sense of play. And it's the way of play that we want to think about as we think about the lives of children.
When people learn to play the piano, they learn to express themselves. They learn to explore the space of possibilities of music. They learn to test the boundaries. And those are the qualities of play that we want to bring into the lives of children. As we bring new digital technologies into the lives of children, we want to think about play in that way. But unfortunately that's not the way people usually think about play when they bring technologies into the lives of children. Usually in schools there's a very different image.
I thought about this when about a year ago Newsweek magazine ran a very special issue about the power of invention. And I thought that I would really resonate with this special issue, because I am a strong believer in the idea of learning through designing and learning through invention. So this seemed like a perfect special issue to talk about; exactly the issues that I care most about. But in fact when I opened up this special issue and started paging through, well... here's the first picture of children that was in this special issue.
It's a picture of children with headphones on. Staring at the computer screens that are sort of downloading some information to them. This is really very much in the sense of playing a stereo.
It's a computer as stereo. You turn it on, it does something to you and you're also cut off from the rest of the world. What we really want to be doing is to move away from this image to much more the computational equivalent of learning to play the piano.
And to do that we really have to ask the question: who is doing the inventing? Strange, this was meant to be a special issue about invention. Somebody was doing inventing, there was somebody sitting at Silicon Valley or some other high-tech company, who'd invented the computer, who'd invented the software. But the kids themselves weren't doing any of the inventing. The important issue is: who is doing the inventing? We have a strong belief that whoever is doing the inventing is doing most of the learning and also having most of the fun.
So where you're going to have the richest learning experiences but also the richest play experiences is when you're the one who's in charge of doing the inventing and the designing.
In order to make that happen we really have to bring very different approaches to thinking about computation, and thinking about learning. And as I've looked around to try and come up with these new approaches one source of inspiration for me, has been kindergarten. In the American educational system, the kindergarten is one of the true success-stories. You hear people complain about the educational system all the time but rarely you hear people complain about kindergarten. So I think it's worth asking the question: why is it that kindergarten works so well, and can we learn lessons from that?
I think that we personally learned those lessons, 'cause at MIT we tried to run it somewhat like a big kindergarten. So right now kindergarten works well and graduate school works pretty well, we have to fix everything in between. And we want to make it a little bit more like kindergarten. What does that mean? I think there's a kind of kindergarten style of learning. The point of it is the idea of learning to make things. With my image of a classic kindergarten, I picture a bunch of kids sitting around the table, making things. Whether they're making a tower out of Lego bricks or something else. It was all about interpreting the rules but in an interesting kind of social context.
There was also kind of an element of imagination in these games. We could kind of imagine ourselves down on the battlefields but for the most part there was an abstraction. Making a painting with finger-paint. They're always making things. They're making things in a playful way and they're doing it with others. Again, this image of a bunch of kids sitting around the table working together on a sculpture, a finger-painting, that's sort of the image that we have of kindergarten and it seems to work pretty well.
This image actually dates back a long time. When Floywell started the first kindergarten in Germany about a 150 years ago, he had the idea of a set of what he called 'gifts'. There were 20 Floywell gifts, physical objects for kids to play with. And the idea was that kids would sort of learn important ideas, just from the process of playing with these specially designed physical objects. And a lot of that carries through to good kindergartens today, where they might not be using the exact Floywell gifts, but there's always a lot of manipulable materials in kindergartens. Things like Lego bricks and pattern blocks. These are all things that kids play around with. These actually are carefully designed artifacts so that as kids play around, they learn some basic ideas. They learn ideas about numbers and shapes and size and scale and colour. These kindergarten-level concepts are very important and the kids learn just through actively playing around with these specially designed concepts.
There's a problem there that once you go beyond kindergarten, the approach becomes very different, becomes somewhat more like this, where you just have someone that's pouring information into someone else's head. It's this transmission model of learning.
People think that learning is about just pouring information in someone's head.
Now again, there might be some reasons why this is happening. I think part of it is that although things like Lego bricks and so on, are good for learning kindergarten level concepts like numbers, shape and size, they aren't necessarily good for learning more advanced concepts. So all you have around you are some of the traditional manipulative materials. They're good for kindergarten but in fact they don't work for learning some of the more advanced concepts you need to learn later in school. Schools have tended to abandon that approach and take the more traditional approach of filling out worksheets, listening to teachers.
But here's where we think we really can make a difference. We really wanted to take a look and say "how can we make everyone of all ages learn in this kindergarten style". And we think this is where computation can make a big difference.
We think computation used in the correct way, can really help us learn ever more advanced ideas but still using the kindergarten approach.
So we wondered if you can learn these more advanced ideas, just with modelling clay, finger-paints and Lego bricks. You can if you give the right types of physical materials. So we've been developing a new generation of manipulative that we called 'digital manipulatives'. They bring together the digital and the physical. So the idea is to have all the advantages of the physical world. Your kids grow up in the physical world. They have all types of intuition and experiences and passions with the physical world. We want to leverage all those advantages of the physical world but at the same time we wanted to use the digital world to extend the range of things that kids can design and the range of things that can learn through their interactions with physical objects.
So when we're talking about some of the variety of things we've been working on, we tend to take things that are already in the culture of childhood. Things like blocks and beads and balls and badges. These are objects that are already in the lives of children. We look at how we can augment these traditional objects to give them new capabilities. So that children still can play out their passions with these objects, but learn a rich range of new ideas in the process of doing so.
And we see that by combining the digital and the physical, it opens up a new range of ideas that kids can be learning about. The learn about ideas like 'process' and 'feedback' and 'emergence'.
These might sound like sophisticate ideas, and sure enough in the past, they've been taught at university courses, where you need advanced and sophisticated mathematics. But I think that's because we haven't had the right materials.
Our belief is that if we just have the right materials that we can put in the hands of children, they can start learning about these concepts, the same way they learn about number, shape, size and colour.
At the same time they can learn about how the design builds an event. So we probably want to sort of rethink how children learn, but also what they learn. The things that kids learn at school today were sort of decided upon as what was easy to learn with traditional materials and traditional paper and pencil. We think our new materials are able to widen the range of what it is that kids might be able to learn in this kindergarten style.
So let me get a little bit more concrete, and give some examples to get a better idea. The project that we worked at the longest period of time is what we called 'programmable bricks'. This is where there is a long collaboration with the Lego toy company, and there's a whole group at Media Lab that have worked on this. And in addition to myself, Fred Martin at the lab is very involved in the design of these programmable bricks and the idea is to put computation directly into the familiar Lego brick. And in fact the Lego toy company two months ago introduced the product based on this technology that they're now selling in stores. Kids can actually write programs to control the things they build out of Lego. They can build anything from an amusement park to an automated factory, but make it come to life by building behaviours the same way that they built structures in the past. So the same way they could build towers and mechanisms in the past, now children can use these programmable bricks to build behaviours into the machines that they build.
Let me just give a quick example. This is the next generation programmable bricks that we've worked on at the Media Lab. These are called Crickets. So here are two little creatures that have these Crickets in them. They're supposed to communicate with each other by infrared light. Right now it looks like they're not doing very much but we turn them so they see each other. They're so happy to see each other, they go into a little dance. If I block the communication, they'll stop dancing. When they see each other again, they'll start dancing again.
Now this might seem like a cute little trick but it's actually much more than that. If this is all that you could do with it, I don't think it will be a very interesting toy. Unfortunately, a lot of electronic toys in the stores, that's sort of what they do. They're one trick toys. This is one thing it does but the idea is that children could keep giving new behaviours to this. And children have to design the behaviours themselves. So in fact the kids that did this, they had to think about when the red creature wants to tell the white creature to dance, it sends a signal. But at first if it just says dance. does the white creature even know what dance means? So kids have to start thinking about the nature of communication, that if you want to communicate you have to have a model of what's in the mind of the listener. So one robot has to know what the other robot already knows. And this is a lesson not just for programming robots, but our own communication. When I talk to you I'm constantly thinking about 'what do they already know?' I'd have to develop a model of what you already know. When they build up those communicating robots they start to get a better idea, and start to think in deeper ways about the nature of communication in general.
So we've had kids over to work on all sorts of these artificial creatures where they explore things about sensing, control, behaviour, feedback. When we got started on this, we started actually with older kids, with MIT students. So the first video I want to show is an example of the robot design competition, that was run at MIT. Started by Fred Martin and some other people and you'll see this is now an annual event. Every January at MIT, students use this technology to have a robot competition. What you will see now is an example where the robot was supposed to push ping pong balls from one side of the table to the other. These are autonomous robots. These are not controlled by a joystick or by a remote control, they are programmed ahead of time. One team here has made a robot that is gonna throw the ball to the other side of the table. The MIT students have much more complex mechanisms. I'll show in a minute some of the things that younger kids have done. But this just gives the spirit of some of the types of things they can do. It sort of mixes together lots of different disciplines: they learn about mechanical design, electrical design. So you see one threw the ping pong balls to the other side. The other one has been patiently gathering the balls and brings it back. This is a major spectator sport at MIT. This is the equivalent of the big football weekend at MIT. This is not speeded up - it's real live action here.
You can see from the tape the passions that come out.
The next video shows how we start to take these ideas and shift them to younger children. Of course, they're not going to work at things of the same complexity but the same ideas come across, where children start learning about sensing a behaviour of how you can build things that can interact with the world and then behave in the world. One thing that I think a lot of children find when they start to work on these robots is that just a couple of simple rules already lead to surprising behaviours. And that alone is a very important lesson to be learned.
The complex behaviours that we see in the world must have a few simple rules underlying it. So if we just put a few simple rules there, all of a sudden you might get some seemingly sophisticated behaviour.
The next videotape shows a variety of different creatures that we've done, with some of the labels that the people have used to describe the creatures. This one just has a couple of simple touch sensors. Just a simple program that says: if your touch sensor is hit, reverse direction. The next one has a light sensor with a simple rule: on in the light, off in the dark. But the next one is a slight change that says: forward in the light, backwards in the dark. And it gets trapped on edges. So already from two simple rules something a little surprising happens. The next one is forward in the light, turn in the dark.
One thing that we learn is although in some ways this might seem like an engineering task. You know, building robots. But kids as you can see from the simple little creatures, it really cuts across and they start asking questions about how's it that real creatures behave, and in what ways are the sensors that are on these robots similar and different to the sensors that we have? When they see an animal in the world - their pet dog, their pet cat - they start asking the question about how is it behaving. How are its behaviours similar or different to the behaviours that I can put in my robot? These aren't easy questions to answer. They are very, very difficult questions to answer, questions that people at universities continue to argue about. But I think it's very good for kids to start asking these questions, and we see that happening on the different sites where we use these technologies.
The kids just start asking questions that just never were in the discourse of children before. It's a way that they just can start thinking about some of the most fundamental philosophical questions of the time about: 'how it is that minds work, how it is that behaviours come about?'
How it is that machines are like animals? How're they different from animals? Important questions (like these arise) just from the very early stages with the simplest of programs and the simplest of designs. Kids start asking what we see as some of the most important questions, questions that would have never been addressed by young kids before. So it's a way to start to turn upside down some of the traditional curriculum in schools, to really have these new ideas that just weren't successful before but suddenly become accessible to young kids once they have these new materials.
I'll jump ahead to another project that we're working on and come back to the videos later. We started out with these type of creatures: you have a single computer controlling the machine, then there's another creature and some other machine. What we've really focused on the last couple of years is not just what you can do with a single computer, but imagining you had a dozen or a hundred of these little computers. So as you can make these computers ever smaller and lighter and cheaper and give them communication capabilities, kids can literally have 100 little computers on the desktop. They can have their own personal little Internet on the desktop. What might you do with that? Now that's some of the questions we're asking. Or thinking what new things you might start exploring when they literally can have a whole network of little computers that all interact with each other on their desktop.
Let me give one example of the type of thing we've been working on with this technology. What we have here is the metaphor of beads on a necklace. Making necklaces is one of the most popular craft activities, especially among young girls. There are whole bead stores in the United States, I assume here as well, where you can get all sorts of beads and make different necklaces. What we have here is something not quite as small and as pretty as beads right now, but they're somewhat like beads. Let me explain: each of these beads has a little computer inside and each bead can communicate with it's neighbouring beads. So light gets passed back and forth along the necklace. If I move the beads apart you can see that they only pass the light from one neighbour to the other. So the light just stays from one bead to the next bead. If I'm quick, I can grab the light and move it across. So I can grab the light, move it here. You can see the light just goes to the neighbouring bead. Now each of these beads has the exact same program in it. Just says: pass the light to my neighbouring bead. So it just makes the light go back and forward along here. But of course, we could have beads with different programs. The whole idea here is to let kids decide the programs. So here I have another bead that I call the eater bead. This one eats the light. So I bring this one down. You see when the light comes, it gets eaten and the light is gone. Luckily I have another bead here that's the feeder bead. This one can feed the light in. So now we have a new pattern. I this one doesn't see the light for a while it generates a new light, and as it goes down it gets eaten at the other side. If I get rid of the eater bead, we have the initial pattern again. But since we have the feeder bead if I break the line here, this will generate another one and we have two patterns of light going on back and forth.
This might just seem like a game but in fact this is the physical instantiation of a type of game that engineers and hackers have played for the last 20 years called cellular automata on computer screens, where they give simple rules to lots of little elements on the screen and see the patterns that arise from those types of simple interactions. And this type of work has led to some of the deepest insights of how complex patterns arise from simple interactions. The way it's traditionally been done hasn't been really accessible to kids. But the idea is that if we can embed it in physical objects that are familiar to kids, like beads, they can start to get a deeper understanding of how simple rules and simple interactions can lead to a complex phenomenon and complex patterns. It's the same way that children always put beads into necklaces to get different static patterns that put red-blue-red-blue and start learning about patterns and symmetry. Now they can start learning about dynamic patterns. And that's where the computer plays a big role - how the dynamic world works hasn't been very accessible to kids, but maybe these ideas can enter their lives through these new types of toys.
The next video is a short TV clip that came about from a workshop we ran last summer with a group of girls. A group of girl scouts did a two week workshop on these, building their own scientific instruments to do their own scientific investigations. You see it starts off with a hamster house they built to keep track of how the hamster moves. We started this at one of the after school learning centres called computer club houses. But even more important is using a new approach to educational learning that we're developing. So it's aimed at kids who generally don't have either access to technology or access to the people who can help them think about how they can use the technology. So we've had now ten of these clubhouses in inner-city areas. Having started in Boston, we now have them in some other US cities, one in Germany and we just opened one in Colombia. So we're trying to see this as a way to connect the kids throughout their whole lives, not just in school but outside of school as well, for them to share these experiences and make this a day-long experience, being able to do these types of investigations.
I'm going to show one last video that goes into a little greater depth of one of the types of investigations. A girl at one of these after school learning centres. This is Cathy who was really interested in birds. We always tell kids to start with something they're really interested about, passionate about. Cathy cared about birds but she said there was a problem that she had a bird feeder in her backyard. And she would put different bird food in, she'd go away but then when she came back the bird food would sometimes be gone and she didn't get to see the birds. So she thought maybe she could use this new technology in order to keep track of what birds came to her bird feeder.
You got some sense of first of all the passion that Cathy brought to her project and the comparison she makes with other types of projects she has done in the past, which she talks about dismissively while mixing some powder with some liquid: 'We're here to talk about my bird feeder that I made.' This type of design and invention project, where kids are doing their own invention, creates a sense of ownership which goes beyond mere motivation. This sense of ownership is what led her to do the type of investigations that she did. To figure out what birds were coming, what the focal length of the camera was, to set it up. That type of personal ownership I think is critical for kids doing investigations. And what we find is that this new technology provides an opportunity for kids to follow up on more of their own fantasies and their own dreams and their own ideas. And 10,000 years ago, kids could wonder what birds eat and maybe try to follow them and see. But they had no way of doing it. And it's only today that we're finally putting the technology in the hands of kids, so that if they want to do these types of investigation, they can. It just wasn't possible before. And we see opening up a new range of possibilities for kids, expanding the space of what it is that kids could design and invent and then in the process expanding the set of things that kids can learn through those design and invention activities.
Let me just wrap up by saying that these are some ongoing efforts that are part of the beginnings of a new initiative at the Media Lab. Just last week, we announced a big, new initiative that's called the Centre for Future Children. It's a new building built next to the current Media Lab that will roughly double the size of the current Lab. In the Centre for Future Children we're looking at some of these questions about how technology can enter meaningfully into the lives of children around the world - there's a real effort to look at kids in all sorts of different cultural settings, from one-room school houses in the developing world to inner city areas in the United States, where some of these videos were taken. The new building will be open around four years from now, so we just starting to ramp up as part of this. We think that we need to rethink all sorts of things in the world of children, learning and technology.
To start with, we have to rethink technology. We think that too many of the technologies that are out there today weren't really developed with kids in mind.
And we try to rethink what are the meaningful ways to bring to bring technology to the lives of children. We come up with things like these programmable bricks that aren't out there. We have to rethink what it is that's going to be most meaningful for kids to use. In some ways, I think a lot of the computers that are out there today really were developed for the TV-generation. They even look like televisions. And most people using computers don't really take full advantage of the possibilities of the computer, to let them invent, explore, design things. We think the next generation of children will be able to take advantage of those capabilities of the computer.
So one thing I want to rethink in the technology is to design a new generation of technology that's worthy of a new generation of children.
- 'Cause we think a new generation of children will need more and demand more than what we're providing today. In addition to rethinking technology, we need to rethink the process of learning. We need to move away from that caricature that I showed before, pouring information into someone's head. To take this idea of what it's like to engage children in design projects, enabling them to learn through a process of creating things in the world. In addition to rethinking technology and rethinking learning, perhaps the most ambitiously of all, we need to rethink childhood itself. The role children play in the world today is not engraved in stone. It's a construct of our culture.
And we really think that as society changes, as new tools and materials enter the lives of children, children could start playing a new role in the world around them.
Just last week at MIT, we held an activity called the Junior Summit, where a hundred kids from 54 different countries came to the Media Lab for a week after several months online talking with each other. And what we can start to see is that given these new technologies, these kids started to come with ideas and carrying through projects and started to play roles in their communities that just weren't accessible before. So we see this as a major challenge: to rethink technology, to rethink learning, to rethink childhood. We think without doing all of the above, we really won't do a good job at any of them. We hope to be working with many of you in the years to come to try to make those dreams a reality.
|
