Although the link below points to a WSJ site, the column occupied Alan Abelson's front page editorial slot in this week's Barron's (Up & Down Wall Street), though it's written by Bill Alpert...Abelson must be on vacation. The relevant part starts about 6 paragraphs down regarding the threat posed by Linux.
A good complement to the remarks of Big Blue's Irving Wladawsky-Berger, who really seems like he needs to spend more time with his family.
If you can't access the link let me know and I'll post the text...I don't know which of this stuff requires a subscription and which doesn't on the WSJ site.
--QS
Edit: Screw it, here it is;
Up and Down Wall Street
Roaring Mouse
By BILL ALPERT
The good news, says stock strategist Gail Dudack, is that 2001 will be disappointing. The market won't be as bad as last year, even if it doesn't recover as quickly as people hope. The averages will finish this year more or less flat.
"Most of the pain is behind us," says Dudack, who was chief investment strategist at UBS Warburg until November, when UBS bought PaineWebber. A year ago Dudack was one of the few strategists to warn investors away from the technology stock bubble. In early March, moreover, she correctly anticipated the Nasdaq's top.
Has excess now been washed out of the market? Some, says Dudack, but not all. Speculative debt remains in the form of brokerage firms' margin loans to investors, which still exceed 1% of the stock market's value. Money continued to flow into equity mutual funds in December. The measure of bullish sentiment taken by the American Association of Individual Investors jumped to a recent 49% from 36% in early January.
"The investor on the street is still addicted to technology stocks," she notes with concern. "Those stocks are down and investors want to get back in."
The latest Federal Reserve rate cut will help, says Dudack. But it usually takes three Fed cuts to raise the market's prospects, she adds, and last week's was only the second. With economic growth hard to forecast, Dudack favors sectors with predictable internal growth: energy, health care and good-quality electric utilities.
Sometime before midyear, Dudack expects this interim rally to flag, with the market revisiting its lows. After that, she hazards, buying opportunities will appear.
Visiting clients last month, technology strategist Steve Milunovich of Merrill Lynch preached the end of the world: the world dominated by Microsoft and Sun Microsystems, that is. The pale rider that will slay those sinners, Steve revealed, is Linux.
Linux, you'll recall, is the free software built by a collective of volunteers determined to exorcise Microsoft from their PCs. About a year ago, this antinomian cult even possessed Wall Street to bid up shares of Red Hat Software to $151 and VA Linux to $320, for a combined stock market value above $27 billion.
What were slayed, however, were those stocks and those of other Linux players like Caldera Systems and Corel Software -- but not before investors had sacrificed $25 billion in stock-market value on the Linux altar. By yearend, punk revenue growth had shrunk the Linux collective's collective market cap to $2 billion. Planned IPOs of other Linux firms were spiked.
So when we saw Merrill's January 22, 2001, report on "Linux Plans for World Domination," we double-checked the date to make sure it wasn't a year old. Maybe we had fallen asleep while watching "Pinky and The Brain," that cartoon of a mouse's feckless schemes for world conquest.
With barely a nod to last year's bloodbath, the Merrill report repeats the doctrine that Linux changes everything. Proprietary approaches like Microsoft's cannot win against an "open source" phenomenon like Linux, Merrill attests. Open-source software is freely available for downloading and its source-code instructions are legible to anyone who wants to help improve it. According to theory, the volunteer army of Linux hackers will always outgun the mercenary programmers of a single corporation. The corporate software with the most to lose, says Merrill, is Microsoft's Windows 2000 (formerly Windows NT) and Sun Microsystems's version of Unix known as Solaris.
A Linux industry consultant tells Milunovich that most software will ultimately go open source. Another true believer he cites is venture investor Bill Gurley, whose Benchmark Capital has plenty of skin in the Linux game and presumably some abrasions. Among the open-source companies in Benchmark's portfolio is Red Hat, the Linux bellwether whose shares have fallen to a recent $9.
As for real-world evidence, Merrill notes that Microsoft CEO Steve Ballmer recently listed Linux at the top of his competitive challenges. Unix also-rans like IBM and Hewlett-Packard have embraced Linux as a way to inch themselves up Ballmer's list, if not the competitors' list of Scott McNealy at Sun. Market research for 1999, says Merrill, showed Linux with a 24% market share and a projected five-year growth rate of 28% a year. Year 2000 data from IDC, however, weren't available.
Daunting evidence of Linux's commercial impact, however, is scarce. Merrill's own surveys of large enterprises show that computer bosses don't yet take Linux seriously. Among public Linux firms, only Red Hat narrowed its losses in its last reported quarter. Corel bailed out of the market. Since July, when Merrill started coverage of Red Hat with an "accumulate" rating at $27, shares of the Linux pure-play have fallen by two-thirds.
For what it's worth, the evidence from analogous markets hasn't shown you can get rich by giving stuff away -- or even do sustainable harm to corporate incumbents. Free financial data on the Internet doesn't seem to have killed the fancier and pricier offerings of Reuters and Bloomberg. Nor have free financial-news sites-stocked mainly with the contributions of p.r. agents-produced valuable products or sustainable business models. At least not in our unbiased view.
While Microsoft is alert to the Linux challenge, Microsoft executives have old-fashioned doubts about its sustainability. "Over the long term, I think economics matter," says Bill Veghte, Microsoft's vice president for embedded systems. "People go where the money is. They go where the R&D and innovation are. They go where they can protect their intellectual property."
The Microsoft veep wonders if the Linux community will produce innovative software, instead of just following in the taillights of Microsoft and Sun. At last week's LinuxWorld trade show in New York, conferees celebrated a new Linux release that accommodates an eight-processor computer.
Sun's Solaris software runs smoothly across dozens of processors-a capability that took five years to develop, says Solaris manager Andy Ingram. "I can't imagine someone in their right mind spending five years doing what we did, for Linux," says Ingram. "Today, Linux looks a lot like Sun's system seven years ago."
Ingram's quite an enthusiast for Linux, in fact. That's because Linux runs on Intel-based computers, where the alternative is Microsoft Windows 2000. Anything that hurts Windows is good for Sun.
Hankering for real evidence that Linux was on track for world domination, we sought out Milunovich as he took the story on the road to Merrill's customers. We chatted by cell phone. Linux is having an impact at the low end of the server computer market, said Milunovich, where it seems to have slowed sales of Microsoft's Windows NT. While Linux still can't match the number of applications available for Windows and Unix, Milunovich noted that Oracle Software and IBM had rolled out Linux versions of their databases. Still and all, cautioned the analyst, "it's going to take three to five years for Linux to really start showing up in enterprise computing."
Is there evidence that could convince him he's wrong? we asked Milunovich. That's when he handed the cell phone over to colleague John M.A. Roy. Computer operations managers, Roy reported, say they don't want another software system to learn. Comfortable with Sun's Solaris, they see no reason for switching. "They tend to be conservative," said Roy, "and they get bonuses for keeping everything up and running." Real-world data on software like Linux is not real solid, Roy explained. The Website of market researcher Orbiten, for instance, discusses the challenge of measuring volunteered R&D and free software distribution. Economic activity is easier to track when paid for by check.
Roy advises the empirically inclined to watch research from firms like IDC, which tries to count the number of servers loaded with Linux, Windows and Unix varieties like Sun Solaris. As it happens, IDC has finished that server count for the year just ended. IDC researcher Al Gillen is reluctant to release numbers to the press, he says, until he has fully briefed his subscribers. He will say this: While upgrades to existing Linux servers showed strong shipments, the growth in new Linux licenses slowed from the 1999 pace.
End-of-the-world cults are hard to dissuade, even days after their deadlines. Perhaps the year 2001 numbers will prove persuasive. This is, after all, the real millennium. |
