1999 Interview with Alan Kay

In 1999, I wrote a new afterword to the MIT Press edition of Tools for Thought, originally published by Simon & Schuster in 1985. I re-interviewed Doug Engelbart, Alan Kay, Bob Taylor. Here is the unedited text of my interview with Kay.

Alan Kay still believes in powerful ideas, still directs his energies at creating a new medium for stretching young minds, and does it with a small crew of remarkable minds under the aegis of Disney, just as he and a small group of remarkable minds in the employ of Xerox invented the networked personal computer thirty years ago. Kay was running Atari Research Lab at the time I first met him, when I wrote Tools for Thought in 1983. He was Atari’s vice president of research and development: “Atari Labs was my only shot at running a large organization,” Kay recalls: “I decided I didn't want to do that. At Apple and Disney, the group has been roughly size and scope of the  group at PARC in the 1970s. I’m doing the same thing today I started to do in the 1960s -- trying to discover ways to help children grow up to think better than most adults do today.”  The Dynabook was  the vision that Kay brought to PARC in the early 1970s, a vision of a handheld device with high resolution screen, network capabilities, and sophisticated graphic software, which drove the development of the first  point-and-click interface -- the personal computer as it is known today. Even after Xerox “fumbled the future” by failing to market PARC’s inventions, it took decades for the global computer and communications industries to approach the devices Kay and colleagues envisioned a decade before the first Macintosh.  “Our specs for the Dynabook were  a little less ambitious today’s technologies, in some respects, and  more ambitious in terms of others, especially in terms of the software. The original Dynabook idea was 95% software. 5% was the lightweight package for delivering the software. The state of the art of software for children of all ages that embodies truly powerful ideas is still not there yet --not where we envisioned it would have to be. The closest thing I know of today  to what we saw in 1970s is what we are doing at Disney.”  Kay tends to talk about what “we” thought about personal computers when “we” first started thinking about them, referring to the ARPA projects in the late 1960s (especially the creation of ARPAnet)  and the Alto, Smalltalk, and Ethernet projects at PARC in the 1970s, when the world’s entire braintrust in networked personal computing worked together on outlandishly futuristic inventions that ended up changing civilization.  “Commercialization didn't follow many of the things we thought were important: one of the things that happened in a half hearted way over the last thirty years was the effort to make authoring and consumption symmetrical. Everybody who consumes the content of a medium can create and distribute content for others. To a limited extent, if you have a PC you can author a few things. For most people it's text and maybe something simple with graphics or spreadsheet calculation. We were interested in what there is about computers that no other medium can do. The computer’s powers of representation were the same as a book, and more. The computer medium subsumes the printing medium, plus it has a dynamic aspect and an ability to read its own writing. What capabilities of this new medium would enable people to think in an altogether different way, not possible with previous media?   “We thought that just as the press enabled people to argue about things not within the scope of traditional oral discourse,  a dynamic book would allow people to argue about science and politics and other things in ways the book didn't enable. Nobody had been able to argue for democracy in terms of stained glass windows, but you can argue for theology and monarchy very well with pre-printing media. The printing press and authors such as Tom Paine made it possible to make something important politically in regard to democracy. It was no accident that the  press and science preceded our modern form of democracy. The particular way of thinking necessary for public discourse and self was apparent to scientists -- the freedom and ability to argue. A literate public is not an oral culture plus awriting system. The epistemological stance that goes with literacy is different from the way of thinking about knowledge that goes with orality. The act of learning to read, to deal with abstract symbols, leads to a different way of thinking about things -- a way of thinking that is necessary for science and democracy.”  “This idea of a new kind of literacy and way of thinking that could be enabled by a technology that didn’t yet exist was highly important when I realized it. I realized a lot of it from seeing Seymour Papert's stuff in 68. The goal of dynabook was to guess what this new literacy would be like and make a form of it that children could learn, and put it into a children’s computer. It would have to be portable because kids are mobile. The individual units would have to be in communication. Wireless networking was built into specs. Kids want to design and create, so we wanted bitmapped screens and ways to create graphics. Because we didn't want to throw out the book, we wanted high resolution screen to do readable fonts, a capability that would become possible when PC displays put a million pixels on a screen. The typical large screen on a portabl etoday is around 800,000. Within the next 5 years, pixels per inch will go up and displays will hold still well enough for PCs to be as effective visually as print on paper. In fact, an economical high resolution screen is being developed at Kent State, using Xerox technology first developed 30 years ago.”  “In the late 1960s Nicholas Negroponte and I were in a very small minority who thought the media aspects of the computer were really important. Again looking at the history of the Gutenberg press -- the ability to print things was not important as the rhetoric invented to take advantage of what press could do -- the argumentive essay and scientific essay, 150 years after printing technology wmered. Would we have to wait for a century and a half for people to bumble into the power of computers as media, or could we take a lesson from history and invent the media of the future. The overlapping window interface was a media invention in response to the problem of how to deal with abstract ideas on a computer in a way kids could understand.”  Kay pointed out that when argumentive essays were invented, they were invented in imitation of Euclid's elements. Descartes and Bacon and others were starting to think that argument from authority wasn't sufficient -- a method of reasonin as precise as geometry was needed. The greatest product of the post-printing way of thinking thing accomplished in the 17th century was Newton's Principia. The Federalist papers in the 18th century were another world-changing application of the new rhetoric that widespread literacy made possible. A powerful set of mental tools was developed when people applied formal sytems that had to do with numbers and space to written argument. Science and democracy were the most important things that could only be built with these thinking tools.  “Along with understanding the computer as a medium is an understanding that the content of computer is language. Bit patterns are only numbers because they are made to act like numbers, but they are bascially symbols. When distinguishable symbols that represent things can be according to a consisten language, the computer acts as language machine. It had occurred to Von Neumman in 40s that computers would be capable of simulating the physical universe in ways that not even mathematics could do. Back then, they were simulating artillery, but he could see astronomy calculations turning into a langgage machine of a mathematical form based on differential equations. Other complex non-linear phenomena, not mathematically tractable, could be represented by computers in ways that could lead to new understanding”  “Papert realized that there are a couple of important things about way computers do dynamic math that take things like calculus that seem difficult to college students and make them look like toys that much younger children can play with and learn from. The turtle is a way of having vector calculus in an inertial coordinate system. The kinds of playing you could do with Papert’s inventions enabled elementary schoolkids to work with the same cognitive and epistemological tools that science uses. I went batshit about this in 1968.”  Kay acknowledges that it has been a difficult problem, translating Papert’s insights about “powerful ideas” and Kay’s understanding of computers language machines into something that could affect the minds of millions of children: “ Once you have this insight, the hard work is finding out what the form of it is, how does it match up with interests and motivations of kids, teachers, society.” He added that revolutionizing schooling is not the only means affecting millions of minds. From the beginning of the dynabook quest, Kay has thought in terms of “subversive media” that bypass established institutions.  “The printing press was subversive because people discovered quickly that you could print many things, not just bibles. Over the years, there have been a lot of different forays into simulation media. The Vivarium project at the open school in mid 1980s lasted seven or eight years. Papert has continued to experiment, and so have others at the Media Lab. In each experiment, some things are learned, some things don't work, some things are learned, and other things are not understood. The personal computer and early Internet industries were selling to customers who didn’t fully understand what they were buying, which made them something like a teenager just beginning to get interested in music -- unsophisticated taste, combined with real buying poer.The result was the entrenchment of poor defacto standards. MS DOS wasn’t the right way to present computation power to new users, and it became a standard.”  “The Web is the MS DOS of the Internet. Engelbart had thought out what an open hyperdocument system should work like. The creators of the Web did not understand what Engelbart had done, and lacked intellectual curiousity to find out. Saving grace of systems is that none of them prevent them from doing their own thing. You can invoke a program from MS DOS or Web that has nothing to do with standard. sS people  have remarked about television, it doesn't tell people how to think, but because it is so total, it makes it difficult to think about things other than what it is presenting. People who are not strong thinkers find it hard to think betwond MS DOS or the WEB. Most people don't know that Net and Web are two completely different things at two different levels.”  Computers in classrooms couldn't do great graaphics, but an Apple II plus a teacher who knew something and Logo was all you needed. Elementary shcool teachers who really know math are 1 in a thousand. Never got beyond getting kids to make pictdures using the turtle. Today it's even worse because computers are used as a symbol that a school is being progressive. If you look closely, it's bascially air guitar.  Reforms through the school system requires many different things to be done. ANd it is possible that bureaucracy and politics and teacher education all make it too difficult. After sputnik, money pured into teaching math and science, and it wasn't successful. Don't think prospects are rosy within that environe t. There will be plenty of empty gestures.  Other way of doing it is to get kids interested. Frank Oppenheirmer's Exploratorium. Playpen instead of museum. Frank said you don't understand, in order to have contact with science, you need to get into yourself the notion that the world is not as it seems. Our senses and learned perceptions are not waht the world is on close examination. Science was invented only a few years ago. I made exhibtits to show taht the world is not what it seems. Brownian notion of thousands of kids in 500 exhibits might bring a kid to the revelation of that idea. One of the best ideas I had ever heard. Clearly applicable to the Internet: a thousand different things that could be interesting to particular kids -- each one is a string that if they pull on it, it brings them into stuff that is meaningful to them and good for them. Literature as opposed to formal schooling. Some book gets you thinking in a different way, you are a different person. In fact, you can do both ways. Over the years, Seymour and I and other people have concentrated at times on schools and times when we concentrated on subversive media. This is a good time for subversive media. Why I am at a media company.  PCs and the Internet in the pure form were done pretty well. On top of them are various kinds of generally bad defacto standards done by people who were more interested in acting than thinking. Dual things going along. You see that in publishing as well. We knew it was going to happen. Thing that people who want to take high road is to take the several thousand examples of interesting high content out on the net and start understanding what to do to get kids who are not intrinsically interested in ideas to be interested in it. Not original thought: all scientists, engineers, physicians are less than 1 percent of population. Carriers of the mechanics of our civilization are a small percentage. Industrial reveolution enabled them to have influence on everybody. Not particularly good for politics. Don't have to have majority of population understanding ideas of the civilization, but you do need that to have a decent political system.  Personal Computer For Children of All Ages -- 1972. Also that year: A Dynamic Medium for Creative Thought. Both Xerox publications.  From my standpoint, things are ten years behind where they could have been.  The change is more of a fad phenomenon. A critical mass of people getting excited -- it spreads like a disease. Herd stuff. Doesn't mean people understand it. It would have been hard for us idealists in the 1970s to think people would go for something as bad as MSDOS. We should have known better from VHS killing beta at end of the 1970s.   If you look back on history, we shouldn't be surprised, It';s the way lots of things have happened. There has never been a time in history when the middle of the bell curve has so much disposable income. That's important in trying to guess how things will go. There are always pop cultures, but generally not as powerful as now. Two parts of interest in Disney. I always loved what imagineering has tried to do: exquisite sense of place. We use Disneyland as metaphor for UI in PARC in seventies -- fun being there even when you didn't have a goal. Other UIs in 1960s were only wothwhile for goals, but not fun, not explorable. The other thing is that Disney is a trusted brand name. In a pop culture, brand is often a substitute for people thinking about the content they are buying. Parents will buy Disney stuff for their kids. One of my main questions: could we do high content stuff? You bet! We will see results next year.