"Roman Candle" turns 20: Secrets of Elliott Smith's accidental masterpiece (slideshow)
Elliott and the friends with whom he recorded in middle school in Texas (photo courtesy of Dan Pickering)
Like many other technology professionals, I’ve followed the Microsoft antitrust saga the way some people follow soaps. So when the Supreme Court refused to hear the case, I confess I was secretly a little disappointed. The Court left us with an appellate court ruling that, well, just lacked drama. The ruling split the difference between Microsoft’s allegations of innocence and Judge Jackson’s sweeping, “Hang the swine from the yardarms!” district court decision.
Microsoft, the appellate court declared, is a monopoly engaging in anti-competitive practices, but that doesn’t make its every strategic move illegal; nor did the court call for the corporate equivalent of capital punishment. With this, the court handed the case back to the lower court with some no-nonsense instructions to find a less radical solution.
Show’s over; Microsoft will live. Everybody go back to arguing about sports.
Legal observers, however, were not fooled: the antitrust case might be over, but the door was wide open for civil suits. Sure enough, Netscape soon stepped through with a civil filing claiming treble damages — and I happily anticipated at least a few more years of wrangling over technical and economic arcana.
Alas, the complaint offers little in the way of argument. Instead, Netscape tries to play on our natural sympathy for a scrappy little company that’s been bullied to death. The lawyers can’t let more than a few paragraphs pass without reminding us that in 1995, Netscape had 70 percent of the market for Web browsers. This fact is reiterated in an indignant tone, as if that share is no more than Netscape’s right.
In other words, Netscape isn’t mad that there’s a monopoly dominating the market for Web browsers; the company’s mad because that monopoly isn’t Netscape.
Of course, you’d be mad, too. Battling Microsoft in the software market is like trying to beat the house at blackjack: The game’s rigged against you from the start. The tactics used to keep Netscape off the desktop would have made protection racketeers blush. If Microsoft couldn’t intimidate customers into signing exclusive agreements with its control of the vital Windows software, it made them sweetheart offers they couldn’t refuse. The complaint leaves no doubt that Microsoft acted in many ways, most of them illegal, to prevent third-party vendors from installing Netscape.
But there it rests. Netscape has proved that Microsoft engaged in anti-competitive behavior; it’s proven that during this time its own market share dropped faster than Milli Vanilli’s. Presumably the lawyers are hoping we won’t notice that they haven’t proved the one thing they need to: that the dirty tricks caused the collapse.
It’s tempting to go along; after all, the tricks were dreadfully dirty. But correlation is not causation, as any first-year statistics student will tell you. To believe that those shenanigans are the source of Netscape’s troubles, you need to make two giant assumptions: First, that the mere placement of an icon on a desktop — or its absence — can win or lose a market. And second, that Internet Explorer was inherently inferior to Netscape. Neither of these assumptions holds water.
The first argument requires us to accept, without proof, that the majority of computer users are either awfully lazy or awfully dim. So lazy or dim, in fact, that they will use whatever software happens to be installed on their computer, regardless. Yet if this is true, why has Microsoft lost so much money on MSN and Microsoft Money, both of which occupied prime desktop real estate for years?
Because AOL and Quicken were better, that’s why. Even though they had to make a special effort to do so, users sought out the non-Microsoft products because they preferred them. And though you may not remember this, the same thing happened to Internet Explorer in its first few years of life. Microsoft didn’t make any headway in the browser market until it came out with a decent product: Internet Explorer 4.0.
There is an argument to be made that Explorer is different because other products aren’t integrated with the operating system in the same way. But I can attest, from years as a computer consultant, that from the point of view of the consumer there’s little difference between the browser and MSN — it’s damn near impossible to get either of them off your machine once they’ve been installed. Yet users keep right on switching to AOL, because it’s better.
Which brings us to the second assumption that Netscape needs us to make: that Microsoft’s technology was inferior to Netscape’s. After all, if Internet Explorer had been better than Netscape, Microsoft would have won anyway. And in that case, there are no damages for Netscape to collect on.
Even a technology that was merely equal to Netscape’s is not sufficient to make its case, because the software market is so dominated by network effects. For those who aren’t familiar with the term, “network effects” refers to the fact that many products become more valuable to their users with each additional user who adopts them. It’s no good having a fax machine unless someone else you know does too. These effects often allow a single product, like Windows, to dominate its market, because users want their software to be compatible with everyone else’s. In the case of the market for browser software, the network effects stem not from the users but rather from the Web designers, who prefer to write code for a single platform. This is why most people think that over time one browser would have predominated even without the market mischief.
This is key. If the products were equal, then it was 50/50 which firm would dominate — which means that the “lost” revenues for which Netscape wants damages weren’t probable, or even particularly likely. That’s why Netscape needs us to believe that Explorer was so bad that no one would have used it unless they had been, well, tricked. Many of those pushing this idea cite wide support in the technical community, a large chunk of which hates both Microsoft’s practices and its products.
Back in my days as a consultant, I often had to fight for projects with the client’s Unix administrators. I spent long hours passionately defending my platform against opponents whose arguments ran along the lines of “Unix is just better.” Eventually, inevitably, each opponent opened his mouth with a self-satisfied look that told me it was coming: The Question. At every client, every Unix guru asked the same damn thing when he wanted to forestall argument: “Would you rather run your heart/lung machine on my platform or yours?”
Then one day, my back to the wall, I had an epiphany. “Yours,” I replied. “But I’m not building myself a heart/lung machine. I’m building an application server for Marketing. And your design costs five times as much.”
He closed his mouth.
I tell you this not to illustrate my rapier wit, but to demonstrate that you can’t talk about a technology being “better” without discussing what it is you want that technology to do. The Ferrari Testarossa may embody the finest in modern automotive engineering, but it’s not much good for hauling the kids to soccer practice.
Technology people like certain things in a system. Above all they prize reliability, followed by powerful, elegant code. Because Microsoft doesn’t do well under this set of standards, they leap to the conclusion, as Netscape wants us to, that Microsoft’s unsavory practices explain its market power. But that is begging the question. If Microsoft’s software is so bad, how did it get that power in the first place?
Unsurprisingly, ordinary consumers don’t value the same things tech people do, any more than you buy the same groceries as Julia Child. For ordinary users reliability takes a back seat to two features that neither Apple nor Sun can match Windows on: the availability of software and, most especially, the price. You just can’t buy a system from either company for $1,000 — but you can buy a really decent PC.
This price disparity has held for the entire history of the PC industry. While low price can’t overcome an awful product — no matter how little I charge for it, people aren’t going to listen to any hip-hop album cut by me — it is decisive if the products offer otherwise acceptable tradeoffs. That’s how Microsoft made such inroads into the browser market, though you wouldn’t know it from the complaint, which somehow omits one of the most powerful features Microsoft put into Internet Explorer.
I mentioned before that, as with my imaginary hip-hop album, Microsoft literally couldn’t give away the first three versions of Internet Explorer. It was received wisdom that Microsoft would never catch up, right up until Microsoft released IE 4.0. Oops. Most users couldn’t see much difference between it and Netscape — and while Netscape cost $40, IE 4.0 was on the house.
Netscape couldn’t compete. Oh, eventually Netscape stopped charging for its browser, which slowed the decline in market share. But it’s hard to see how that helped, really. It’s the old economist’s joke: “We’re losing money on every unit, but we’ll make it up in volume!” Netscape’s core product cost an enormous amount to develop and generated no direct revenue — and with Microsoft breathing down its neck, Netscape didn’t dare stop building the next generation of product.
Nor were losses limited to browser revenue. When Microsoft became a viable competitor, the other revenue streams on which Netscape had counted — such as selling server software that was as tightly integrated with its browser as IE is with Windows, and leasing prime space on its product to portals like Yahoo — became more competitive and therefore much less valuable. With a good competing free browser on the market, Netscape simply couldn’t make enough money to survive.
Many observers call that predatory pricing, a highly illegal tactic whereby a company sells its product below cost in order to drive its competitors out of business so it can then jack up prices even higher. There are extensive arguments and counterarguments as to whether Microsoft’s behavior fits the description, but ultimately they’re irrelevant, because Netscape doesn’t mention pricing in the complaint.
Predatory pricing is the first of three main arguments advanced to explain how Microsoft unfairly competed against Netscape. The second argument is that Microsoft abused its control over the OS code, pre-empting a long standards battle by incorporating key code into Windows. This argument is compelling, which is why, as readers may recall, the court ordered Microsoft to separate the browser from the OS.
Many critics go further, alleging that Windows programmers deliberately inserted code to sabotage Netscape’s performance, but most such rumors seem to peter out into the friend-of-a-friend-of-a-friend of urban legend. And ultimately Microsoft’s technical escapades don’t matter either; the complaint makes no mention of them.
The third argument, of course, is the leverage that Microsoft exerted on its distribution network. As we’ve seen, while this argument is true, it is not sufficient to explain Netscape’s decline. Yet the complaint presents only this, the weakest claim. It’s hard not to ask yourself: What the heck were the lawyers thinking?
The answer is that they didn’t have much choice. The appellate ruling sharply limited Netscape’s case. While the appeals court did establish the grounds for the suit by ruling that Microsoft had behaved illegally, it sharply restricted the grounds on which Netscape could sue. In particular, the court specifically excluded predatory pricing as a cause of action, which threw out Netscape’s most powerful argument — price — and severely undermined its claims about the browser/OS integration, which, after all, is just another way of giving away the product for free.
Moreover, the civil trial, unlike the antitrust action, will probably be a jury trial. Long technical arguments, which didn’t fare well even in front of the highly educated district court judge, risk alienating the jury. And, as the O.J. trial showed, jurors often ignore evidence they don’t understand. Not to mention that arguing predatory pricing means telling jurors you deserved to win so that … they could pay $40 for something they currently get for free. Not compelling. On the other hand, not only are predatory tactics easy to explain, but they also play into our twin national dislikes: big corporations and sharp dealing.
Stuck between a rock and a hard place, Netscape repeats its sad tale over and over, in the hopes that we won’t ask how, exactly, an icon on a desktop could have been responsible for so much destruction. But we have to ask, because the law demands that lawyers not only tell you how a competitor caused their client harm, but also prove it. Which they can’t, at least not with this argument.
It isn’t fair. Netscape gave us a revolutionary product that has touched the lives of everyone in America, and in exchange we took away its market and made the company a minor subsidiary of AOL Time Warner. But we can’t fix things by exchanging the rule of law for a popularity contest. I know that if it were me I’d be mad as hell — but nevertheless this is one fight that Netscape deserves to lose.
- – - – - – - – - – - -
First of a two-part series on the legal battle between Netscape and Microsoft. Read Part 2.
Megan McArdle is a former technology consultant who recently graduated from the University of Chicago's Graduate School of Business.More Megan McArdle.
Elliott and the friends with whom he recorded in middle school in Texas (photo courtesy of Dan Pickering)
Heatmiser publicity shot (L-R: Tony Lash, Brandt Peterson, Neil Gust, Elliott Smith) (photo courtesy of JJ Gonson photography)
Elliott and JJ Gonson (photo courtesy of JJ Gonson photography)
"Stray" 7-inch, Cavity Search Records (photo courtesy of JJ Gonson photography)
Elliott's Hampshire College ID photo, 1987
Elliott with "Le Domino," the guitar he used on "Roman Candle" (courtesy of JJ Gonson photography)
Full "Roman Candle" record cover (courtesy of JJ Gonson photography)
Elliott goofing off in Portland (courtesy of JJ Gonson photography)
Heatmiser (L-R: Elliott Smith, Neil Gust, Tony Lash, Brandt Peterson)(courtesy of JJ Gonson photography)
The Greenhouse Sleeve -- Cassette sleeve from Murder of Crows release, 1988, with first appearance of Condor Avenue (photo courtesy of Glynnis Fawkes)