Gateway 2000

Navigation Bar Subscribe to Government Technology Conferences & Events Search Magazines and Supplements News Home Home Solution Center

The Visionaries:


and Interface

Alan Kay Alan KayGT: I understand you're interested in extending Disney into projects that use technology for good purposes. What would those be?


Kay: Disney is about fun. If you take a look at various kinds of activities, there's soft fun and hard fun. Soft fun is what happens when the environment does most of the things for you. Hard fun [is] like playing baseball, playing a musical instrument, as opposed to listening to it.

The hard fun activities tend to be a lot of hard work and the payoff is that things get better if you are trying to grow. People need both of these.

At the end of the 20th century, because of mass media [and television], soft fun is a very big industry.

My interests have always been in hard fun. So I've always been interested in designing environments on the computer that encourage children of all ages to stretch and grow. That's basically what I'm doing here.

Disney is a big enough place and is interested in all aspects of fun-producing environments, and I'm interested in the ones -- I don't want to use the word "education" because that's a little too connected with school -- that are analogous to sports and music.

GT: What's your view of how technology is currently being used to educate children.

Kay: Well, again, the technology of the baseball bat or the violin or piano are interesting examples -- where it's pretty clear that for exercise you don't need external technology to do it. Music started with people singing, so you don't need external technologies specifically to perform music.

For the essence of many things, you don't need technology to really get you thinking about interesting things unless you consider language a technology.

The thing I've always been interested in is the amplification possibilities when you take something that's already interesting and already part of deep human interest and use some technology to amplify the reach. So musical instruments become very important.

I'm more interested in an analogy to the bicycle than I am to the automobile. [With the bicycle] you can still exercise flat out, and it amplifies that, whereas the car removes the exercise. It's mainly about getting from A to B, it's not so much about the side effects.

The computer is a little more complicated. I thought it was complicated to explain to people 25 years ago when they didn't know anything about it. Now it's like what Will Rogers said: "It's not what you don't know that will hurt you, it's what you think you do know."

Things have become much more complicated because people have made decisions about what a computer is vis-a-vis children and schools. Most of those decisions I think are way off.

If they tried to use these analogies, particularly to musical instruments, it would lead them to asking questions that would more likely give them good hints about what they should be doing. But basically they aren't asking those questions.

GT: So we need to take the metaphor of how we learn music and apply it to how we should learn with computers?

Kay: One of the things people should really ask about the computer is: "What is its music? What is the thing that it is actually amplifying and extending?" A nice analogy is that the piano is an amplifier of human musical impulses, because you can play polyphony on it. It is an amplifier of thinking about how musical structure works. The concept of harmony was largely thought out on the keyboard back in the 17th century. The keyboard allows you to think about music in a different way. A lot of what harmony is, is a visual analogy.

Another interesting thing about the piano is that it can act as a prosthetic. You put a prosthetic limb on a healthy one and it starts atrophying. The piano is a prosthetic because it always plays the right pitch. It's not very expressive. You can't make a sound louder after you have played it. So, it's possible to be much more mechanical on the piano, much more than you can get away with on a violin. A great violinist with his ears completely stopped up can't play a tune. So it's not just learning muscular movements, but it's a constant dynamic feedback from what the player is actually hearing and adjusting to.

You should be doing that on the piano, but because the piano is always playing the pitch it was tuned to, you can play the piano without even listening to it. So pianists have to work harder to be as musical as violinists and flute players are.

In fact, wind instruments tend to be the easiest to be musical on because the phrasing is built in. You have to take a breath. With a violin, you have to decide to phrase, but some of that is still natural because of how the bow works. Whereas on the piano, you have to make acts of will for everything.

It's always a good question of any technology to ask: When is it an amplifier and when is it a prosthetic? That's another question people should be asking about computers.

Other important questions to ask are: What is the music of the computer? What is it giving you the most leverage in doing? Is this music just for some people? Our society has no trouble with someone who decides not to be a musician. It's thought to be for some and not for others.

After the piano, the second big analogy to put the computer against is the printing press. There, we do have a lot of trouble in our society if people don't learn how to read and write. It used to be just for the few. But after the printing press, society said it is no longer just for the few, it's for all. Schools, as we know them today, were set up to teach reading and writing to children.

The real question about the computer is: Do we think of it as something for a small number of people or do we think of it in terms of the printing press? There are some good arguments as to why we should think of it in terms of the printing press. Those arguments are very similar to the arguments that I and other people have made for why math and science should be like reading and writing.

GT: What do you mean?

Kay: Reading and writing are considered an integral part of our civilization. I believe that the next interesting things that came along after reading and writing were math and science. I think in the 20th century they are not optional. It's not to learn the facts of science that's so important, but for people to learn the epistemology of science: The stance about the world and about knowledge that science takes, about argument and how you settle argument, how you find out new things and so forth, are vitally important and directly connected to our political process.

Now obviously, the schools don't believe this as well as state legislatures and most parents. So the computer is in this interesting place where this rapid activity connected with it right now is mostly in one or two spheres, somewhat overlapping.

One of the spheres is in vocationalism. It is amazing how early in the grades you can go now, even down to second- or perhaps first-grade, and find parents who are worried whether their kids are going to get a job when they leave high school. And the computer is seen as a critical technology for getting a job. That vocationalism has been an incredibly distracting thing in education in general.

GT: Are you aware of something that would help make the computer a viable tool in schools so that it does not become a distraction?

Kay: Well, one of the things we invented 25 years ago was something called the overlapping window interface. One of the reasons the computer is where it is now is because there are interfaces on every computer. So that's an example of a media invention that helps get the computer everywhere.

GT: In other words, users can look at more than one set of information at the same time?

Kay: But it's not that. It [the interface] makes a complete difference about what you have to remember. You may never have used the interfaces that were prevalent before that, like MS-DOS. Most people simply didn't use computers when those interfaces were around. You had to remember all the different commands -- 500 or 600 of them -- literally.

You still have the same number of commands on your Macintosh or Windows 95 system, but most of them are in front of your face and fairly obvious as to what the possibilities are and what they can do. There's quite a bit of cognizant psychology in the design of that interface. And it's more than 25 years old now. It was originally done in 1971.

But I believe the biggest problem has not been getting books into schools, but in getting kids to be fluent readers and writers. The biggest problem has not been to get the words "math" and "science" into schools but to get them to actually happen in the classrooms.

One of the reasons it hasn't happened in the classroom is a vanishingly small percentage of elementary school teachers who know anything about science or math. Generally speaking, it's only the stuff that teachers feel confident about that gets discussed at all. So if they do math or science, it tends to be in a very rigid, lock-step, non-intellectual, very goal-oriented fashion, which is antithetical to the way kids learn things. Often kids will learn something useful from whatever the teachers' real interests are.

It's an interesting thing that, in 1997, we would probably ask for the dismissal of any teacher who couldn't read or write, but we are completely of a different mind, in fact we don't even want to find out, how much the teachers don't know about math or science. They are not even tested for it. That's completely wrong-headed.

GT: Can we expect something like the computer to make things better?

Kay: In the sense that the computer is like a book, you would have to say the answer is no. But the interesting thing is the sense in which a computer isn't like a book. A computer can present the same kinds of information -- can present things more dynamically. But it's still an open question if, for instance, you imagine this scenario for five or 10 years from now. A kid stumbles on something [interesting] on the Internet. As the result of just poking around in it, the system starts intuiting what their level of knowledge is. Like an adult, [the computer] starts to engage the kid. So the system is now making an effort to understand where the kid is coming from and trying to get the kid to go deep on this thing he is interested in.

The analogy to this is a really good writer who tries to do this with a general audience. The writer tries to make the front part of the writing draw the reader in and make them feel comfortable, and then he starts scaffolding concepts they may or may not know, all of which is going to lead a thousand words later to what the writer's central point is.

The art of writing is that the writer has to do that for a reader they don't know. It also requires the reader to be fluent, so it's difficult to accomplish with someone who isn't a good reader. That's why books without mediating adults don't work very well for kids.

So it's an open question whether the active nature of the computer plus a lot more work over the next few years can actually bridge the gap that adults have given up on now.

GT: Is this type of intuitive computer possible?

Kay: It's somewhat possible. But one question is whether present day adults should be allowed to get away with what they are now getting away with [regarding] their kids. There's a question of when you have kids and you can afford infinite babysitters, should you? There's moral questions here.

I think it will be possible 10 years from now for much computer media ... to be somewhat like an intelligent person you meet on a plane flight. You start conversing with that person who, if interested in engaging with you, will find the level at which they can engage with you. They will get you interested in things at that level.

That might bridge the gap for the responsibilities that parents and teachers are shirking right now. Morally, it's an interesting situation. Any town in which the citizens don't pick up stray trash probably can't hire enough workers to pick it up.

Part of the problem is the attitude you get into when you turn everything over to service organizations, when you outsource everything. Parents who outsource parenting probably should be censured, whether or not technology can do something about it.

I don't think computers are going to bridge the social breakdown that occurs when kids are brought up in a vacuum, where adult humans are absent. I think television is already that thing.

For those of us who like hard fun, one of the things is trying to understand what the great inventions of the last several thousand years really mean in the world of a child. There have been some very smart people who have thought about that over the past several years. There's not a complete set of answers, but there are some very good answers for a number of different areas.

The other part is whether by embedding some of these ideas into computer media, you can bridge some gaps.

GT: You were involved with the development of the PC back in the late 60's at Xerox's PARC research center. In November 1997, "Computerworld" magazine said the PC is suffering a midlife crisis in corporate America. That the industry has "lapsed into listlessness" and is no longer innovative. Is that a fair assessment? What's wrong with today's PC industry?

Kay: When was it innovative? I think the stuff we did at Xerox PARC was innovative, but that was years before there was a PC. When the IBM PC and the Apple II came out, it was years after we had done the stuff that's like the machines today. So MS-DOS was a huge retrogression from what we had at Xerox PARC.

I would say that it would be very hard to point at innovations in the computer industry.

GT: Why isn't the industry innovative?

Kay: What business do you know of that wants to buy something that's innovative? The place where some innovations happened in the last 15-20 years is in the workstation area. That's because engineers tend to buy workstations based on value rather than on price, because they can actually assess what features they actually want.

I think the aiming of the PC into business created a huge industry and made a few people very rich, but there's nothing creative about it. I'm willing to be proved wrong, but I can't think of anything. It has actually made things very boring and difficult.

GT: State and local governments are still in the process of giving many workers their first PC. For these workers, the transition from paper to Windows and a mouse is not easy, despite the user-friendly GUI environment. Having worked on the original user-friendly interface, how do you rate today's GUI, and what changes can we expect that might make the computer intuitive to use?

Kay: I don't think the GUIs today are user-friendly. The stuff we did at Xerox PARC was reasonably friendly, and the Mac was reasonably friendly, but I think it would be very hard to sustain arguments that the current-day Macs or PCs are user-friendly. That's partly because the companies that developed these ideas really didn't understand the cognitive psychology basis of them. So they tended to just add features in order to make next year's marketing deadline.

Just adding features into something willy-nilly means you are just going to end up with a hodgepodge again. I think that's what we have.

GT: Can we expect the computer industry to make GUIs more intuitive?

Kay: If a good idea came around, what is the probability that it would get incorporated into any of the current systems? Remember, it was government that decided that MS-DOS computers were going to be the standard. So government was one of the biggest malefactors in anything having to do with user-friendliness.

They had a perfect chance to buy Macs when Macs were a huge difference in productivity, but they didn't do it. The reason is that government's various rules and predilections to technology said, "buy price." They weren't willing to pay anything for value, and now they are complaining. So I don't have much sympathy for that. I remember talking to an admiral who was complaining to me about computing, and he was the very person who had led the drive to buy low price and ended up with bad systems. Then he started complaining about them.

GT: Have you any thoughts on how we can exploit technology to improve governance in the next millennium?

Kay: If you think of technology as an amplifier, you have to remember the saying: "Garbage in, garbage out." If you don't put very interesting stuff into a microphone and throw it through a 300-watt amplifier, you are just going to get a bigger version of the crud that went in.

The problem with the technology is that people blame it when things go wrong, and they want it to save them as well. In fact, it's like blaming a car when somebody drives it off the road. The real problem is that most people in schools are not learning to think very well.

It's generally true from person to person and from organization to organization. You take people who are not very fluent at thinking and throw a lot of technology on to them that amplifies this sort of inept behavior, you're going to get a mess. It's like children playing with matches.

The dilemma of the car is that suddenly human beings are forced to decide when they are going to exercise, whereas before the car, nature forced them to exercise. So every time technology comes in, we are forced to decide when we're going to use it and when we're going to exercise the thing that the technology is nullifying.

Suppose things are more enlightened. Then you could certainly imagine computers helping people deal with complexities that are beyond our unaided brain, precisely as writing helps us deal with complexities that are beyond an oral society.

It's been noted by many people, especially anthropologists, that a literate society is not the same as an oral society with writing. Literate societies think about the world in a completely different way. Scientific society thinks about things in a completely different way than a religious society.

If you want to see what computers could be doing for people, one of the things to look at is the change the computer has made in various branches of science. There are whole branches of science that are being looked at now because they were completely intractable before the computer came about. If more people were like scientists, they could be using the computers more wisely than they do now.

For instance, the World Health Organization says 30 million people have HIV now. A couple of years ago, their prediction for the year 2010-2015 was somewhere in the order of 40 to 50 million people will have died worldwide from AIDS. This is in spite of the fact that by understanding how epidemics work, and particularly AIDS, you can cut that down tremendously.

The truth is that most people don't understand exponential progressions, which have this horrible innocence about them during the only period when you can control them. That's an example that can be used both in schools and in state legislatures, let alone Congress.

I'm really amused that Congress has decided not to allow personal computers in their chambers. First, they were only thought of as note-taking devices, which is ludicrous. The second thing is the possibility of them understanding something more complex than their unaided brains is moot now.

What you should be doing is dealing with complexity with the aid of tools, just as we deal with transportation through the use of tools. The same with agriculture. And these people who are our elected representatives don't understand any of this. They don't want to learn how to use the tool.

To me, the bottom line is that basically most of the stuff I've read in the last decade or so about computers and why they should be here and why they shouldn't be there is just completely off the mark. So it's not a question of saying, 'no, it should be this way.' It's more a question of getting back to these fundamental things I mentioned at the beginning. They have to understand what the thing is an amplifier for. Why is it worth spending as much time to learn how to read and write to become fluent in it?

The user-friendly computer is a red herring. The user-friendliness of a book just makes it easier to turn pages. There's nothing user-friendly about learning to read. The reason is, we want people to change as a result of it. We don't want them to be oral people.

User-friendly means the absence of gratuitous stupidity. You don't want to make it hard to turn pages of a book. Make it as easy as possible. But in fact, you want to write as well as you can. You don't want kids to read dumb, poorly constructed writing. But there's an element of this stuff that is difficult, and it's difficult for a reason, because it's carrying big ideas. Those are the things that we want society to have the will to inculcate into kids -- the will to spend many years necessary to change themselves in order to be able to deal with the stuff fluently. Short of that, what technology is going to amplify is just pap. Because that's all that's going to be going on in people's brains.

Government for a
World Long Gone

To Put the
Vision in Place

Excellence: Do it
Right the First Time

Capital City
of the Future

When the Buck Stops,
Spend Ideas

and Interface

Return to
the Beginning.
"It was government that decided that MS-DOS computers were going to be the standard. So government was one of the biggest malefactors in anything having to do with user-friendliness."
"It is amazing how early in the grades you can go now and find parents who are worried whether their kids are going to get a job when they leave high school. And the computer is seen as a critical technology for getting a job."

"I think the aiming of the PC into business created a huge industry and made a few people very rich, but there's nothing creative about it. It has actually made things very boring and difficult."

"I don't think computers are going to bridge the social breakdown that occurs when kids are brought up in a vacuum, where adult humans are absent. I think television is already that thing."

"If you think of technology as an amplifier, you have to remember the saying: Garbage in, garbage out. You are just going to get a bigger version of the crud that went in."

Alan Kay


Alan Kay is perhaps best known for making computers easy to use. The graphical user interface, object-oriented programming, Smalltalk, Ethernet, laser printing and client/ server computing evolved from work carried out in the early 1970s by Kay and his associates at Xerox Corp.'s Palo Alto Research Center (PARC).

Kay's interest in children and education led him from Xerox to Apple to Walt Disney Imagineering, where he is now a Disney Fellow. A former professional jazz guitarist and composer, he is now an amateur classical pipe organist.

Navigation Bar


Home | News | Publications | Conferences | GovCenter | Media | Search | Subscribe
©Copyright e.Republic, Inc. All Rights Reserved. Privacy Statement Questions or Comments?

Subscribe to Government Technology Conferences & Events Search Magazines and Supplements News Home Home Solution Center

Click Here!