Instant histories of the browser wars

How can the full tale of Microsoft vs. Netscape be told while the story is still unfolding? Reviews of 'Barbarians Led by Bill Gates' and 'Speeding the Net.'

Published June 5, 1998 12:48AM (EDT)

If journalism is, as the saying goes, the first draft of history, then books are more like its final draft, or its proofs. The reporter works from the eye of the storm to meet some mad deadline; the author, typically, has the advantage of time to reflect and perfect -- and of knowing how the story ends.

But not in today's overheated technology-business book market. Eager to cash in on public fascination with the personalities and strategies of the computer industry, and desperate to grab public mind-share before the competition, book publishers in this field are beginning to rush out half-baked volumes that shout "Edit me!" from every page.

The high-profile, high-stakes battles between Microsoft, Netscape and the U.S. Department of Justice have fanned this trend to a new frenzy. In the war for the Web-browser market, the software industry kicked itself into a furious fast-forward mode of rapid development cycles and buggy releases. The publishing industry is now following suit.

The perils of this emulation are on full display in two new books that cover the Microsoft-Netscape story from different angles, "Barbarians Led by Bill Gates" and "Speeding the Net." By far the better of the two, "Speeding the Net" -- a history of Netscape by Josh Quittner and Michelle Slatalla -- is hobbled chiefly by sheer prematurity: It must end before its tale does. "Barbarians Led by Bill Gates" -- an "insider's" portrait of Microsoft by Jennifer Edstrom and Marlin Eller -- provides a more vivid tableau of the havoc that Web speed can play on the publishing process. The book is so unfinished, even a software company might think twice about releasing it.

Eller is a veteran Microsoft developer who spent 13 years coding in the trenches in Redmond, Wash. Edstrom is a writer whose surname is familiar to all journalists who've ever written about Microsoft -- she is the daughter ("estranged," according to the book's press release) of Pamela Edstrom, the company's longtime lead publicist. Together, the two promise to present, as their subtitle puts it, "Microsoft From the Inside: How the World's Richest Corporation Wields Its Power." Given the negative slant from that subtitle onward, we can safely assume that the project did not have the blessing of Edstrom mere.

And how does Microsoft wield its power? According to Eller and Edstrom, clumsily. Stupidly. Or, in the ultimate Microsoft put-down, randomly. The chief message of "Barbarians Led by Bill Gates" is: Forget the image of Microsoft as a Machiavellian colossus that has built a gleaming empire based on technical brilliance and marketing genius. Microsoft is an "essentially chaotic" enterprise that got where it is today by pure luck -- by "covering the table" so that at least some of its bets would pay off.

Eller joined Microsoft in the early 1980s and contributed some key components to the code for Windows 1.0, released in 1985 to near-universal derision. "Barbarians" is most interesting as it traces the strange saga of the rise of Windows from an "essentially useless" patch on Microsoft's dominant DOS operating system to the computing monopoly it is today.

Windows, according to Eller, was always a stepchild in Microsoft's development hierarchy, buffeted by the faddish whims of fickle execs. In its earliest days, Gates kept demanding, "Why isn't this like the Mac? Be more like the Mac!" Then, as Microsoft navigated a shifting alliance with IBM to produce the next-generation OS/2 operating system, Windows was kept on life support chiefly to provide a platform for Microsoft's Excel spreadsheet. When IBM and Microsoft parted ways, Windows 3.0 started selling millions of copies in the early '90s -- because it worked OK and was compatible with mountains of existing DOS software. Even then, the company continued to view Windows as a sideshow to the cooler operating system of the future, Microsoft's own NT. Windows 3.1 and Windows 95 triumphed by accident, Eller maintains, not by design of Microsoft's master strategists -- who were always ordering a forced march in some other direction.

Eller divides Microsoft's employees into two classes: the "really 'smart' people, generally Ph.D. physicist types who were masters at designing new architectures," and the "master hackers" who did the grunt work of actually creating new products. The ethereal "smart" types rarely sat down at keyboards to write actual code, but periodically issued edicts to the hacker developers to throw out everything they'd done so far and start anew. (Microsoft technology guru Nathan Myhrvold is portrayed as the epitome of this species -- an ineffectual "big picture" guy who kept pointing Microsoft in the direction of buzzword-of-the-moment technologies like RISC chips and set-top boxes.) Meanwhile, the master hackers would be at work, often unauthorized, pursuing genuinely useful projects -- like Windows 3.0's protected memory scheme -- that would come to the company's rescue when all the strategic maneuvers of the "smart people" went awry.

At this point, conspiracy theorists may wonder whether "Barbarians Led by Bill Gates" might not be a spin effort by the Microsoft publicity machine, after all. At this exact moment in the company's history -- with antitrust lawyers bearing down on every move it makes, looking for evidence of illegal tactics in Microsoft's effort to crush Netscape -- what could be more useful to Gates than the suggestion that he is hardly omniscient, that he can barely figure out where the computer industry is going next month, let alone plot a devious four-year campaign to eliminate a competitor?

But "Barbarians" is too raw and "essentially chaotic" itself to be a calculated PR move. The authors don't know how to shape a chapter or develop a theme; the best they can sustain is amusing, dishy anecdotes (like a funny foot-in-mouth bit in which Eller points at a buggy routine and says to Gates, "Who was the jerk who wrote this brain-dead piece of shit?" -- only to learn that the answer was Gates himself). Because the tale follows Eller's own patchwork career, it sometimes lurches in odd directions: when we most want to hear more about the evolution of Windows, for instance, we're getting a primer in the dead-end technology of pen computing.

The book's listlessness is a shame, because a developer's-eye view of Microsoft is certainly a useful corrective to the company's own propaganda. "Barbarians Led by Bill Gates" contains some illuminating quotations from Microsoft developers interviewed by the authors to cover the period after Eller left the company in 1995 (at least I think that's the date -- the book is sketchy on such details); these might make useful source material for some future author's effort to write the full, complex saga of Microsoft's history. Otherwise, "Barbarians" is an act of sheer opportunism, rushed to market full of minor errors. (As Netscape conquered the Web, we learn, "Bill Gates and his teeming hoards were scurrying to play catch-up." Hoards the man has, that's for sure; but only a horde can scurry.)

Even the book's title is a mess: Gates' minions are portrayed sometimes as conniving, or egotistical, or small-minded, but hardly "barbaric." Perhaps realizing that the title demanded some justification beyond its feeble punning homage to a business-history bestseller, the authors fill their final page with some convoluted musings on history:

Who knows what thoughts passed through the head of [Netscape's] Jim Barksdale or Jim Clark as they watched their hard-won empire crumble. One thinks of Roman plutocrats in the last days, watching nervously the council fires of Vandals or Visigoths, those hungry barbarians encircling their villas or camped outside their city gates, waiting with sharpened sticks to pillage, and plunder, and rape.

Then Edstrom and Eller think twice about this inept comparison: After all, Netscape, unlike Rome, was built in a day and hardly constitutes a classical culture overrun by unlettered mobs. So they stagger off into a different analogy between Netscape's "encircled men" and the besieged fortress of Troy, before finally collapsing on the book's finish line with a feeble quip about bewaring "geeks bearing gifts." At such moments, the only barbarism in sight is the authors'.

With "Speeding the Net," the language -- and the reader -- is in far safer hands. Quittner covers the computer industry and the Net for Time magazine, Slatalla is a columnist for the New York Times' Circuits section, the husband-and-wife team has written both nonfiction and fiction together and their writing has an easy confidence and playfulness that's priceless in covering technology. Because, let's face it, the subject isn't nearly as sexy as its current -- and likely fleeting -- popularity implies: Full-length company histories too easily devolve into annotated lists of deals, executive boasts and boardroom shake-ups. The challenge is to find the story in the company.

"Speeding the Net's" subtitle is "The Inside Story of Netscape and How It Challenged Microsoft," but Microsoft is barely present for two-thirds of the book -- it just looms in the background as "a constant ache" and spur to the young Netscape rebels' long hours. Quittner and Slatalla seem to understand that, as a story, the "browser war" is both too arcane and too unfinished to serve as a book's spine. So they do their best to orient the bulk of their book as a chronicle of a revolutionary moment in Internet history: the Web's jump-start ignition by the 1993 invention of Mosaic -- the first browser people could use on PCs and Macs to view both pictures and text on the Web.

"Speeding the Net" lingers in its opening chapters with Mosaic co-creator Marc Andreessen and his cohorts at the University of Illinois' National Center for Supercomputing Applications (NCSA), painting a series of vivid pictures from lives of young hackers: splicing a cable line to deliver TV to the desktop of Andreessen's Silicon Graphics workstation; discovering the nascent World Wide Web; hacking together the first version of Mosaic on a diet of Mountain Dew and Pepperidge Farm cookies; then watching in awe as the world pounds on the NCSA's servers to download the cool new browser.

This is a great story, and though it's been told before, Quittner and Slatalla give it a fresh-minted glow: The student programmers' creative impatience charges each page. The immediacy holds as Andreessen and his cohorts -- estranged from NCSA after the center tried to cut them off from their vital e-mail lifeline -- move west to join Jim Clark in Silicon Valley in founding Mosaic Communications (as Netscape was originally named). "Speeding the Net" gives us a cubicle-level view of the atmosphere at the newborn start-up company, where the developers punctuated their punishingly long hours with Ping-Pong-ball gun firefights and radio-controlled toy car races.

Netscape had on its side both youthful energy and upstart pluck -- but it was as competitive, in its own way, as Microsoft. Its first goal in the market was to build a better browser and crush Mosaic, still owned by NCSA. "Speeding the Net" reminds us that Netscape's browser supremacy wasn't inevitable -- it came about because the original Netscape products were fast and usable and relatively reliable. Yet at the same time they were building the browser that would sweep the Net, Netscape's developers thought of themselves as a legion of "The Doomed": doomed because they could never hope to meet their impossible deadlines; then, even if they did, doomed once more by the inevitable intervention of Microsoft.

The story reaches its giddy apogee with Netscape's legendarily insane initial public offering in August 1995, when the market bid this small stock up to the stratosphere without anyone's being entirely clear on how the company expected to make a profit. As if following the arc of the Netscape saga itself, "Speeding the Net" grows more confused from this moment on: As it concentrates less on crazy kids and more on corporate strategy, the pictures grow fuzzier, and the authors provide less perspective. And so we learn about CEO Jim Barksdale's plan to sell collaboration tools for use on corporate intranets -- but not about how integrating those tools into Netscape's browser bloated later releases, thus helping open the door to the Microsoft competition.

As "Speeding the Net" approaches the present showdown between Microsoft, Netscape and the government, its narrative grows less colorful, and the lively developers introduced in its first half drift into the background. Racing to a cliffhanger as the government takes on Microsoft and Netscape plunges into an uncertain future of giveaway browsers and open distribution of its source code, "Speeding the Net" has no choice: It must end in media res -- precisely where, according to the annals of epic poetry, the story should begin.

Certainly, Quittner and Slatalla hang from their cliff more gracefully than Edstrom and Eller do. But "Speeding the Net" still leaves you with a feeling of incompleteness nearly as disheartening as the void at the end of "Barbarians Led by Bill Gates."

Both books -- and their readers -- are victims of a publishing industry that apparently can no longer wait for a story to arrive at its natural end before cranking up the presses. The publishers, of course, just want to cash in on a news story while it's still hot, and who can blame them for that? Still, having hustled these volumes out prematurely, the publishing houses are now a lot less likely to back new takes on the subject once the time is riper.

So we're left with technology-business books that are almost as sketchy and context-free as the technology news we get from newspapers and magazines and Web sites. And that's a scandal. Books should aspire to tell a definitive story -- to provide at least a good, thorough edit of history. We're unlikely to get it anywhere else.

By Scott Rosenberg

Salon co-founder Scott Rosenberg is director of He is the author of "Say Everything" and Dreaming in Code and blogs at

MORE FROM Scott Rosenberg

Related Topics ------------------------------------------

Books Business