Sunday, January 24, 1999

	Search online   merchants   for this book.

Not so long ago, in the not-so-distant past, a "computer" referred to a human being at a desk, armed with a mechanical calculator or a slide rule. So Paul E. Ceruzzi reminds us in the introduction to this ambitious work before going on to show, in impressive if sometimes incomplete detail, how we got here from there.

Ceruzzi, curator of the National Air and Space Museum's Department of Space History, aims to show how political, economic and social factors directed technological developments. The book's best segments succeed brilliantly at that, skillfully knitting together different threads of history to demonstrate how accidental progress can be. Consider, for instance, his prehistory of the personal computer, which shows how such different factors as U.S. Air Force ICBM development, collegiate pranks on time-sharing computers and the first wave of programmable calculators produced a combination of tested, cheap technology and willing, technically oriented users that allowed crude, kit-built personal computers like the Altair 8800 to find success.

Ceruzzi also shows a knack for illuminating analogies and examples. At one point he estimates, for instance, that the IRS's file-transfer system of the 1960s – a courier toting reels of computer tape on a cross-country flight – would have been conveying an average of 30,000 bits per second, about what a typical home computer can send over phone lines today.

But while Ceruzzi does an excellent job of explaining why categories of products soared or tanked in the market, he leaves bigger questions unanswered. The book's treatment of IBM raises the most such questions. Ceruzzi sensibly devotes a large portion of the work to IBM's role in both developing and marketing technology but doesn't expend many pages digging into the social context of Big Blue's success – for instance, the "nobody ever got fired for buying IBM" mentality which, with IBM's place taken by Microsoft, still holds sway. Ceruzzi also seems to avoid taking a stand on the question of whether the company's market position bogged down technological advances, although his comments on the minicomputer market pioneered by Digital Equipment Corporation suggest that he may have an opinion to share:

"DEC did not dominate in minicomputers in the same way IBM dominated mainframes . . . . DEC's competitors did not feel they had to answer every product announcement, or offer software-compatible products. Technological innovation, at low cost and in a compact package, mattered more." Now that we've bought enough Microsoft products to put Microsoft in a position of even greater market dominance than IBM, an examination of this historical lesson seems warranted.

The other key absence in this work is some sense of the people who made all this happen, given Ceruzzi's repeated emphasis on the importance of individual creativity and entrepreneurship. As he states unequivocally in the first chapter, "The 'computer age'. . . was not just invented; it was willed into existence by people who wanted it to happen." He does not, however, go on to explain what made a Ken Olsen push Digital into leading the minicomputer market.

The book's thematic organization, in which Ceruzzi picks his way through one broad topic at a time (software, minicomputers, integrated circuits, networking and so on), works only for the first few decades of the story. Later on, a detailed account of the Unix operating system doesn't come until the last 30 pages or so, several chapters after we've been introduced to the integrated circuits that arrived at about the same time.

And some topics, such as the versions of the C programming language that dominate software development today, seem to have gotten lost in the shuffle. Ditto for the open-source software that, in the form of the Apache and Sendmail Web and e-mail utilities, runs much of the Internet. That last omission is puzzling: A major theme of the book is the success of open, easily extended hardware and software over closed, proprietary systems.

Of course, this linear narrative's problems could have been avoided entirely if it were published as a hypertext document on the Web. But Ceruzzi himself may not be ready for that, judging by a footnote in which he declines to cite online postings by Netscape founder Marc Andreesen on the grounds that "there is no way of knowing how long this material will be preserved."

You could say the exact same thing about the books on Ceruzzi's desk – yet because those postings live only on a computer disk, they somehow don't really exist in the way that paper documents do. But that attitude is another story, and book, entirely.

Back to the top