Douglas Engelbart is not among the famous pioneers of modern computing, but he has not been totally forgotten. Many know of him as the inventor of the mouse, which is true, although hardly indicative of the broad range of work that he has done. A few know of him as having developed systems of linked electronic files that allowed workers to collaborate, using the computer as a facilitating medium. This early use of what Ted Nelson later called “hypertext” has become common over the World Wide Web, although in a form that Engelbart did not espouse and, indeed, is still ambivalent about.

This is the first comprehensive account of Engelbart’s work. The author places the developments mentioned above in the context of a life dedicated, not to developing artificially intelligent computers as companions to humans, but to co-evolving computers and humans to work in symbiosis with one another. This view has been at odds with the computing mainstream. It is explicitly opposed to the goals of the artificial intelligence (AI) community, which enjoyed generous funding from government agencies and universities, even as Engelbart received modest support from the same sources. It was at odds with the drive to make computers “user friendly,” specifically by using a desktop metaphor, with overlapping windows, icons, and mouse control. Engelbart preferred training the user at the same time as developing a better computer interface. His vision remains at odds with the current culture of the World Wide Web, with its emphasis on flashy graphics, and its flat file structure that imposes no order on the way one moves between its nodes. Engelbart preferred structuring information in a controlled, hierarchical way, much like the now-dormant Gopher navigation system.

All three facets of modern computing have come under heavy criticism recently. AI has failed to deliver on many of its promises, and the Defense Advanced Research Projects Agency (DARPA) withdrew much of its financial support for AI after 1986. The baroque and cumbersome Windows interface has come under fire from critics such as Donald Norman and Thomas Landauer, who have an easy time showing that “user-friendly” is a meaningless phrase when applied to the modern personal computer. The World Wide Web, though phenomenally successful, cries out for help for the individual seeking to navigate it, even as portals like Yahoo and search engines like Google are making a valiant effort to help.

In that context, one might expect Engelbart to have finally gotten the credit he deserves, and, in part, he has. He is the recipient of the 1997 Lemelson-MIT Prize, the National Medal of Technology from President Clinton in 2000, and the George Stibitz Computer Pioneer Award from the American Computer Museum, to mention only a few awards. Still, he and his work are not well understood. Bardini quotes another computer pioneer who says that Engelbart is “bitter” over the way things have turned out. I do not think that is accurate. It is more that he is frustrated and exasperated at how the world failed, and continues to fail, to see the correctness of his ways. I once heard him speak, and I recall the self-deprecating way he described how frustrated corporate executives and funding officers were when dealing with him. His stories were amusing and had no trace of bitterness, but it was clear that he would have preferred a more favorable reception. Some of the most poignant parts of this book deal with the painful demise of Engelbart’s Augmentation Research Center (ARC) as it was bypassed by flashier developments at Xerox PARC, at Apple, and on the Internet.

The real strength of Bardini’s book is the way he excavates the philosophical, social, and technical foundations of Engelbart’s work. Thus, we learn not only of the research that led to the mouse’s invention, but of debates going back to the 19th century on telegraph keying systems, typewriter keyboards, and other methods of sending and receiving information electronically. The author brings these ancient stories back to life and shows how they are relevant to current debates. He does the same with the history of hypertext, and with early attempts to make it easier for humans to interact with computers. The book is also strong in retelling a familiar tale, of how Engelbart’s work migrated from his lab to Xerox PARC, then to Apple, and finally to the masses through Microsoft Windows. The net effect was to negate much of what was appealing about his ideas in the first place, but if one asked a representative of Microsoft or AOL, one would be told that the “marketplace” had spoken.

My one quibble with the book is that the author does not give adequate credit to the real origins of the personal computer. He devotes a lot of space to Xerox PARC, its Alto computer, and the influence it had. But it was the announcement of the Altair computer, from a tiny Albuquerque hobby shop, MITS, in 1975, that sparked the PC phenomenon. It was the Altair’s creators who chose an Intel microprocessor as the brain of their system. It was to Albuquerque that Bill Gates and Paul Allen moved (from Cambridge, Massachusetts) to be closer to the future of computing. And it was the Altair, not the Xerox Alto, that paved the way for the IBM PC, the machine that really put computers on the desktop, and that owed nothing to Xerox PARC research. Engelbart, like Ted Nelson and probably many others at the time, looked on the spread of the IBM PC with horror--it was a standalone machine that was fairly hostile to networking, and to them networking was the most important thing about computing. Perhaps the need to reconcile these two incommensurate models of personal computing--the standalone model of the Altair, IBM PC, and Microsoft, and the networked model of the ARC, Xerox PARC, Arpanet, and the Web--is the cause for the frustration we experience with computing today.