The press says Apple's dead? Two can play that game. Why Windows Is Doomed...

If they weren't so unfair, the exaggerated headlines tolling Apple's troubles would be downright hilarious. Computer stores have a slow Christmas? "Apple Battered; Fans Lose Faith." Some VP departs over an ego tussle? "Apple Brain Drain Escalates." Apple creates an absolutely amazing laptop/ desktop/ palmtop computer? "New Apple Machine Is Too Little, Too Late."

Meanwhile, Windows 95 is a technological mess. Plug-and-play doesn't work right, multimedia requires hours of fiddling, maintenance is a horror. And no wonder--no single operating system could possibly work flawlessly on machines from dozens of unsupervised clone makers. "Windows must work with a plethora of hardware, each device made by a company fighting to differentiate itself," says Fortune columnist Stewart Alsop. "[That's] a less than artful way to make an operating system that works smoothly and intuitively ."

Yet where are the Windows doomsayers?

Right here. In this purely intellectual exercise, I'll use exactly the same reporting standards we've seen applied to Apple. The statistics you're about to read are real. Only the conclusions are twisted.

Or are they?

---Windows Market Splintered; Trouble Seen Ahead

Market share is everything. Big market share attracts software companies, and the platform prospers. The Software Publishers Association (SPA) puts Windows' share at 67 percent.

Ah, but "Windows" is several different products running different software on different computers. Windows 3.1, Windows 95, Windows NT, Windows CE--individually, each holds a market share not nearly as dazzling as the total.

According to research firm International Data Corporation (IDC), the fastest-growing operating system of all is the corporate-oriented, headline-grabbing success story known as Windows NT. This superstar's 1996 market share? Try 3 percent. "NT's shiny image is beginning to tarnish," says Information Week, "through problems with underlying technology, licensing squabbles, and mixed messages from Microsoft on which operating system to implement where."

Windows 3.1, on the other hand, has an enormous market share. But according to PC Week, Microsoft plans to discontinue Win 3.1 this summer. Which leaves Windows 95--whose installed market share, according to the SPA, is a mere 18 percent.

Suddenly Apple's 16 percent installed base doesn't look so bad.

---Windows Upgrade Strategy Fails

Microsoft's business plan is simple: sell Windows installed on each PC, then sell an upgrade every other year.

Microsoft has sold only 46 million copies of Windows 95, a stunning 27 percent fewer than forecast. Meanwhile, according to IDC, 13 million people--despite $200 million worth of ads-- ignored Windows 95 last year, deliberately buying Windows 3.1 instead.

"Microsoft is always saying, 'You've got to upgrade,' " says IDC's OS analyst Dan Kusnetzky. "But today's businesses want a platform that runs reliably and isn't going to change." Microsoft's upgrades-forever strategy, in other words, "won't fly for a long period of time."

"Microsoft keeps making new operating systems to fix its old mistakes," agrees Alsop. "Few of the programs designed to work on previous versions of the operating system really work on the current version, forcing [upgrading] customers to replace their old software"--and hardware.

If Windows users don't buy into the upgrade-treadmill concept, Windows can't keep flying high. And with each new version of Windows, Microsoft splinters the PC market further. Suppose that, five years from now, there are six incompatible versions of Windows, each with a fraction of the market; software companies' decisions about which platform to write for won't be quite so automatic.

---PC Boom Over; Windows Crashes

Microsoft got rich by selling Windows on each new PC. This strategy worked great during the PC boom of the early nineties.

But the ride is over. "The industry as a whole is headed for a second straight year of declining growth," says IDC; PC sales growth in 1996 was down a frightening 35 percent from 1995. PC manufacturers, the primary source of Windows sales, are dropping like flies: Acer, Blackship, Canon, DEC, Packard Bell, Tandy, Zenith, and others have either closed their PC divisions or are hemorrhaging money in the millions. And it's only going to get worse. IDC predicts that "the 1997 growth rate will be less than half that of both 1994 and 1995," and BusinessWeek says sales growth could stall completely as early as 1998. Fewer people are buying Windows software, too; the SPA's 1996 data shows an 18 percent drop in sales of Windows programs.

Without the steadily increasing torrent of cash, there can be no steadily increasing torrent of Windows.

---Internet to Quash Windows Sales

What really keeps Bill Gates up nights, though, is the Internet. The bigger it gets, the more irrelevant Windows becomes.

NCs (network computers) are cheap machines whose software and data sit in a central location (the Internet or a com-pany server), eliminating the hassles of upgrades, conflicts, and backups. IDC expects NCs to multiply at a rate of 86 percent per year through 2001.

And not one NC will ever need a copy of Windows.

The real ugliness will begin, though, when Intel itself turns its back on Microsoft. Because PC sales are dropping off, Intel will eventually be forced to support NC computing. "If Intel does not aggressively participate in the Internet appliance business," says IDC, "it will miss the greatest growth opportunities in the decade ahead. The leading PC suppliers will not wait for Intel for very long."

Trust me on this: you don't want to be in the same room with Bill Gates when Intel stops doing Windows.

---Windows Too Expensive, Say Users

IDC predicts that in the coming years, "users will increase their focus on total cost of ownership [TCO]." In English: Money counts. Unfortunately, when you include the costs of training, upgrading, and troubleshooting, Windows 95 PCs are among the most expensive on earth. As Intel's own Web site puts it, "Gartner Group has estimated an annual TCO of $13,187 per year [per Windows PC]."

As this horrific truth spreads, more companies will question why they're throwing away such outrageous sums. Especially when they're buying less and less productivity: tests by

Byte Byte, February 1997 and InfoWorld show that the aging Pentium chip's speed and capabilities are falling further behind those of rival chips. It's only a matter of time before PC buyers replace their Wintel boxes with machines the Gartner Group says have far lower TCOs, such as NCs--or Macs.

---Can Microsoft Be Saved?

The cold, hard numbers tell the story. Plummeting sales, flawed and expensive technology, a splintered and resentful customer base--the Windows ship is steaming toward an iceberg. Without Windows, Microsoft loses the inside advantage it has writing Windows applications. And when that happens, Microsoft's hyperinflated stock price will crash; the Wall Street Journal will begin running "Customers Dump Windows" articles; and the great tide of public opinion will turn to any company that's got superior technology to fill the void.

Aren't statistics great?