The Dynabook That Never Was (and Always Is)

In 1972, Alan Kay wrote a short paper with a long title: A Personal Computer for Children of All Ages. In it he described a device he called the Dynabook — about the size and weight of a notebook, with a crisp screen, a keyboard, a battery, the ability to connect to other machines, and, crucially, a programming environment dynamic enough that an ordinary person could create with it.

If you've ever held a laptop or a tablet, the description will sound eerily familiar. Kay sketched the hardware of personal computing a decade before it existed. And yet Kay, who is still alive and still talking, will tell you bluntly that the Dynabook has not been built — that the devices we carry are, at best, its body without its soul.

To understand that claim is to understand one of the most important and least absorbed ideas in computing.

A medium, not a gadget#

The Dynabook was never fundamentally about portability or even about computers in the ordinary sense. It was about the computer as a medium — and Kay meant that word in the full, McLuhan-soaked sense it carries in the medium is the message.

Kay's claim, which he has made for fifty years, is that the computer is not a tool like a hammer or even a printing press. It is a metamedium: a medium that can dynamically simulate the details of any other medium, including media that have never existed because no static material could host them. Paper can hold text and pictures. The computer can hold text, pictures, sound, simulations you can run, models you can interrogate, and notations that respond to you. It is, potentially, the richest medium for thought ever invented — on the order of writing and mathematics, not on the order of television.

And here is the move that makes Kay radical: a medium is only realized when ordinary people can author in it, not merely consume it. Print mattered because people learned to write, not just to read. So the Dynabook's defining feature was never the screen. It was that a child could make things in it — simulations, tools, art, ideas — fluently, the way a literate person writes a letter without thinking of it as "programming."

Smalltalk and PARC#

This is not idle philosophy; Kay tried to build it. At Xerox PARC in the 1970s, the pursuit of the Dynabook drove an astonishing burst of invention. To give the imagined child a medium to author in, Kay's group — Dan Ingalls, Adele Goldberg, and others — created Smalltalk, a language and environment built around the idea that everything is an object communicating by message, in a world that is alive and editable while it runs.

To make that environment usable, they built the things we now take for granted: the bitmapped display, overlapping windows, pop-up menus, the modern graphical user interface. The famous story of Steve Jobs visiting PARC and carrying the GUI off to the Macintosh is true, but it captures only the surface. Jobs took the interface. The deeper cargo — the computer as an authorable medium for thought, the system you sculpt while it lives — mostly stayed behind.

Why Kay says it never shipped#

So why isn't your tablet a Dynabook? Kay's answer is uncomfortable and consistent: because we kept the consumption and dropped the authoring.

The device a child holds today is spectacular at delivering media to them and nearly mute as a medium for them. The default relationship is reading, watching, scrolling — not making. Programming, the act of authoring in the medium, has been walled off into a profession, exactly the opposite of the "children of all ages" who could all write in it. We took a metamedium and used it mostly as a better television and a better mail-order catalog.

Kay puts it in his characteristically bracing way: the computer revolution hasn't happened yet. The pop-culture version — devices, apps, connectivity — happened enormously. The revolution he cared about — a civilization-scale jump in how ordinary people think, comparable to the spread of literacy — did not.

Always is#

It would be easy to file all this under "disappointed visionary," but that misreads both Kay and the situation. The Dynabook is not a product that failed to ship. It's a standard — a way of asking, of any piece of technology, whether it makes its user more capable of thought or merely more occupied.

By that standard the Dynabook is being partially built all the time, in pieces, by people who took the vision seriously: in Scratch, the heir to Papert's constructionism; in the live, malleable environments that descend from Smalltalk; in Bret Victor's experiments in making thought immediate and tangible; in every spreadsheet that turned an ordinary person into a modeler. None of these is the Dynabook. All of them are arguments that it remains possible.

Kay likes to quote himself: the best way to predict the future is to invent it. The Dynabook is his standing reminder that the future of personal computing was specified, in remarkable detail, half a century ago — and that most of it is still unclaimed, waiting for people willing to build the soul and not just the body.