Earth to Larry: too late.

Four years ago, this newbie with the all-star cast of investors might have been just another big idea with dubious prospects in a sea of bloated expectations. But these days, Cometa's plan to blanket the nation with 20,000 high-speed wi-fi networks, or "hot spots," and wholesale the bandwidth to broadband retailers has worked the high-tech world into a lather. And as such, the company and the wi-fi market it hopes to exploit are starting to look like a hype bubble--part of the same unfortunate fugue state that landed us in this rotten economic crisis in the first place.

The biggest question facing Cometa and other wi-fi hopefuls is, of course, how to make money selling this stuff. Wi-fi is yet another example of a grassroots technology for which early adopters have grown accustomed to paying little or nothing. As with ISPs, wi-fi service providers like Boingo Wireless and T-Mobile will struggle to charge enough for their service to pay for the considerable costs of building a network.

It becomes a risky numbers game, a classic landgrab. "For us, the key for profits is having enough subscribers," says David Hagan, president of Boingo, one of the pioneer wi-fi service providers, with 800 hot spots in hotels and airports around the United States, and a staunch believer in the build-it-and-they-will-come philosophy of customer acquisition. "It is all driven by our footprint."

Boingo's expensive philosophy is precisely what Cometa intends to avoid. (Boingo raised $20 million in venture capital a year ago and says it has two years' worth of cash left.) By fashioning itself as a wi-fi service wholesaler, Cometa cleverly plans to sidestep the considerable customer-acquisition costs by letting broadband carriers--AT&T, Verizon Communications, and AOL Time Warner spring to mind--bear the lion's share of the load. Before anyone had ever heard of Cometa, the brains behind the operation were running numbers on various business models, fully aware of the danger of getting caught in the hype.

"We looked at all those failed models," says Mark Christensen, a vice president and director at Intel Capital and one of the key players behind the formation of Cometa. "We don't plan to own the customer. If you have to own everything, the model is difficult to make pencil out."

Cometa is relying on the widespread corporate adoption of wi-fi, a trend that is still nascent. The company's executives call their target end-user the "windshield warrior," a term they use relentlessly. Picture a modern Willie Loman stopping at a burger joint to download his email, submit sales orders, and catch the score of the Knicks game. Cometa's stated goal in creating its 20,000 hot spots throughout the United States is to provide wi-fi access within a five-minute walk of any urban point or a five-minute drive of any suburban site. But experts say it will take far more than 20,000 sites to accomplish that ambitious goal, especially since it requires several hundred hot spots just to cover a decent-size college campus.

Given the players, it seems appropriate that Cometa is making a bet on the enterprise market. After all, home networking enthusiasts have been known to be stingy. But analysts predict that growth in the small office/home wi-fi market will far outstrip the enterprise wi-fi market. According to the Synergy Research Group, a market research firm, the home market already accounts for 58 percent of the $1.8 billion wi-fi equipment market and is growing faster than the enterprise market (see "Market Monitor"). But Cometa's management summarily dismisses the question of whether the company is backing the wrong horse.

"Enterprises are wrestling with wireless for employees that work out of their car," explains Dean Douglas, vice president of telecommunications at IBM Global Services and another of the high-tech triumvirate that conceived and created Cometa. "These people have a lot of dead time during the day, and they don't have an efficient way to pass along data. Giving them somewhere to go and have high-speed access is something that these organizations are trying to do."

The main stumbling block for corporate wi-fi deployment has been very legitimate concerns about security. Hacking into a wi-fi network is much like hacking into an ordinary local area network, with one big difference: you don't have to physically plug into the network. With IBM (site installation), Intel (wi-fi chip manufacturing), and AT&T (broadband access) running the show, however, corporations are likely to feel more comfortable exchanging sensitive data wirelessly than through, say, companies called HereUAre or Surf and Sip. And Cometa executives are confident they can drive standards through the industry that will allay any security fears that corporations may have.

But despite its impressive pedigree, Cometa is just another startup, feeling its way in the dark of a promising yet untested market. Rose Klimovich, the Cometa representative at AT&T, says it best when she admits: "We have to figure out the right way to make money. We will see over the next year or two whether Cometa has figured that out. It hasn't been tested yet."

Indeed it hasn't. So while there may not be much else to get excited about these days in the world of tech startups, observers would do well to view Cometa with the newly gained skepticism the wireless market demands. Even Mr. Brilliant, Cometa's own CEO, cringes at the amount of attention his young company has received. "I don't like seeing all the hype," he says. We don't either.

Dan Briody is a freelance writer in New York.

Through late July, energy and technology services stocks were down an average of 56 percent in 2002, almost double the Nasdaq's 29 percent decline. In fact, the performance has been so dismal that nine of them appeared to be trading below their cash on hand in midsummer. Beacon Power (Nasdaq: BCON), for example, a maker of flywheel energy storage systems, had $0.67 a share in cash as of its last quarterly report, yet was trading for less than a third of that--$0.21 a share--in mid-July. The list also includes former market darlings like Active Power and Capstone Turbine (Nasdaq: ACPW, CPST). Still--and this is also seemingly contrary to logic--acquisitions don't seem to be looming on the horizon as a savior for the industry.

"Cash is no longer a floor or even a trigger for a rebound, because valuation is not driving these stocks," explains Christine Farkas, energy technology analyst at Merrill Lynch. "Many of these companies are just concepts--and the market thinks they're going to burn through their cash with little to show for it." Apparently, investors have decided there's a better way to create energy than burning cash. Go figure.

Many startups complain of MEN's unassailable market position, control over the pace of change in wireless, and potentially anticompetitive practices. Take Don Listwin. The CEO of Openwave Systems, a wireless software firm, complains that the threesome kept his company from taking part in the development of a wireless instant-messaging standard last year. "MEN shut us out," he says of MEN's endeavor, Wireless Village. "They don't care if the pie is smaller, as long as it's still their pie."

Naturally, Motorola has a different take on the creation of Wireless Village. "The motive behind it was to get together and quickly drive some standards," says Craig Peddie, general manager at Motorola. "I guess Openwave did ask if they could join, and we politely told them they could wait a bit. I guess that didn't sit well with them." It wasn't until after MEN had developed and adopted their instant-messaging standard that they opened the standards body to other companies, including Openwave.

Stories about MEN's alleged abuse of their dominance abound in the wireless industry. There are even accusations of possible collusion: the claim is that MEN are purposely holding back the technological progress of wireless networks so they can cash in on their existing product lines.

One startup claims Ericsson admitted that wireless carriers would benefit from its technology, but instead of offering it to the carriers, insisted that the carriers could "muddle through the next two or three years" without it. "I couldn't believe they were admitting that to us. We were amazed," says the CEO of the company, who requested anonymity. Ericsson declined to comment on this story. Nokia says such claims are overblown and unfounded. "Of course we have the upper hand because of our customer relationships," says Jaakko Myllymдki, general manager of marketing and sales for IP mobility networks at Nokia. "But I doubt that there is anything that we can really control. We can't tell the carriers what they should and shouldn't do."

Whether this is just startups sniping at larger rivals or business as usual, MEN, along with Lucent Technologies, Nortel Networks, and Siemens (a far less catchy acronym), are effectively the gatekeepers for new technology in the wireless industry. MEN's position is strengthened by its global reach: Motorola, Ericsson, and Nokia are based in the United States, Sweden, and Finland, respectively; all three sell into worldwide markets. And at a time when so many economies are struggling around the world, the need has never been greater for wireless carriers to have access to next-generation technologies that increase efficiency and that might get businesses and consumers spending on telecommunications again.

With so much need, the dominant positions of MEN are anything but guaranteed. Hordes of startups want to unseat them and gain the chance to sell their products to wireless carriers. And now that one of the main tools MEN used to achieve inordinate power, vendor financing, has dried up, the empire is ripe for a coup.

In the late '90s, wireless operators couldn't buy network equipment fast enough. As carriers built their networks for the anticipated avalanche of wireless subscribers, the value of wireless equipment contracts more than doubled, from $22 billion in 1999 to $47 billion in 2000, according to the investment bank U.S. Bancorp Piper Jaffray. This tremendous growth was a boon for MEN. The big three combined owned nearly 55 percent of the market in 1999, and raked in $96 billion in revenue.

LOAN SHARKS

The equipment makers then used their cash stores to further lock in their customers; MEN offered vendor-financing deals, lending billions of dollars to carriers to buy more equipment. "They were like banks," says Ed Curran, executive director of network purchasing at Verizon Wireless. He adds that Verizon did not participate in the money handout.

Vendor financing became popular in Europe, as struggling wireless carriers there borrowed their way into corporate bondage. In June 2001, for instance, Nokia had more than $3.5 billion in credit commitments to carriers. In a one-week lending binge, it lent nearly $2 billion to Hutchison 3G and Orange. All told, the nine largest manufacturers of telecom equipment had more than $25 billion at risk in vendor finance by summer 2001, according to McKinsey & Company, an IT consultancy.

This quickly inverted the usual vendor-customer relationship. Many carriers found themselves shackled to their vendors and their products, unable to buy equipment from competitors. In addition to borrowing money from Nokia in summer 2001, Orange also borrowed from Ericsson, bringing its total debt to $1.3 billion. Carriers became beholden to their vendors, unable to green-light a new technology without vendor approval.

"The vendors were saying, 'Look, I'm financing your network, so why would I lend you money so that you can buy equipment from someone else?' " explains a vice president at Cisco Systems, another company eager to break the hold of the incumbent equipment makers. "It became a source of tension."

Then, as now, many startups had products that target wireless base stations, the million-dollar hardware at the bottom of every cell tower. It was the main source of income for MEN, and they had no intention of letting anyone else in. "You could not get into the base station; that's where MEN thrive," says Peter Lojko, CEO of WaterCove Networks, a Chelmsford, Massachusetts, mobile data startup. "They would not give that up."

NETWORK OUTAGE

But in summer 2001, everything changed. Just as MEN had tightened their lock on carriers, Wall Street lost patience with wireless carriers. Investors called for them to stop building networks--and start showing profits. Network equipment sales tanked as carriers canceled order after order, sending equipment providers into a tailspin--one made worse by vendor financing. Nokia lost $3 billion of its $100 billion market capitalization when Turkey's Telsim Mobile Telecommunications Services, the country's second-largest carrier, defaulted on a $719 million loan. And Lucent was stung badly when Winstar Communications, a wireless broadband provider, defaulted on a $2 billion line of credit.

MEN had little choice but to make dramatic cuts. Ericsson and Motorola laid off some 78,000 workers combined--enough people to populate a small city. Research and development also suffered. This spelled opportunity for startups.

"In almost every case, the startups continue to innovate much faster than the incumbent equipment providers," says Bruce Sachs, a partner at Charles River Ventures and an investor in WaterCove Networks and Flarion Technologies, two of the firms battling it out with MEN. "The incumbents are no longer onto the next big thing. Startups have as good a chance as ever to walk into a wireless operator and make a deal happen."

Carriers are in need of any technology that can save or earn them money, even if it is from an unproven startup. MEN are vulnerable when it comes to making networks more efficient and upgrading them with IP-based technology, the same technique of data communications that underlies the Internet and some advanced voice communications. The adoption of IP networks is widely viewed as the critical step for carriers to offer mobile data services and unlock new revenue streams. Yet MEN have resisted building new IP-based efficiencies into networks because efficient networks require less of their network equipment. They have focused instead on existing voice networks.

"There is a huge gap between the expertise of the hardware incumbents in the voice market and their level of expertise on the data side," says Ray Dolan, CEO of Flarion. By offering a mobile data product that rides on top of existing networks, essentially relegating MEN to the role of voice equipment providers, Flarion aims to be the next Qualcomm: an underestimated startup that hit the big time. "We are at a crossroads," says Mr. Dolan. "The solutions the hardware incumbents are proposing for wireless voice and data, 3G [third-generation wireless services], is just like ISDN was on the wireline side." ISDN, a colossal failure, was proposed in the mid-'90s by equipment providers to telecom firms looking to provide data-communication services. "You really think the carriers want to do that again?" asks Mr. Dolan.

MEN, of course, are hoping to avoid that mistake. For instance, to gain IP-network expertise, Ericsson has partnered with the router giant Juniper Networks, and Nokia has struck a similar agreement with Cisco. Though these alliances could be viewed as an acknowledgement of weakness, they are also proof that MEN are dynamic and worthy competitors. "If I have an embedded position, I'm going to play to my strengths and hide my weaknesses," says Jeff Corley, director of strategic alliances and partnerships for Lucent, referring to wireless equipment vendors. "If my weaknesses are IP, I'm going to try to slow that up."

FUD POISONING

One stalling tactic cited by many startups is that MEN are spreading FUD--fear, uncertainty, and doubt--about the startups' organizations and technologies. The tactics vary from questioning a startup's financial solvency with carriers to spreading misinformation about a new product's capabilities. And as these seeds of discord take root, MEN are buying time to develop rival products, the startups claim. In some cases, if carriers add a new technology to their existing base stations, or "boxes," without consulting MEN, the warranties on those boxes will be nullified automatically, a paralyzing tactic. "Everyone making base stations is saying you will void the warranty if you crack open the box," says Mike Dolbec, general partner at Orange Ventures, the VC arm of the European wireless operator Orange. "It's their last beachhead. It's a way for them to protect their market position," says Mr. Dolbec.

And while some startups claim that MEN work together to stifle competition, the only real claim they have is that incumbents are opposed collectively to carriers' making the switch to IP networks, a stance that could backfire eventually. "A lot of this has to do with the pacing of the market," says the Cisco vice president. "On one level, it works for a while, but it is a very dangerous game. Sooner or later your customers realize what you're doing."

Others are more bold. "Are MEN holding back wireless? Absolutely," says WaterCove's Mr. Lojko. "The capabilities are there, and going without them is hurting the carriers' revenue." As for the charge of collusion, the Cisco vice president chooses his words carefully: "I don't think the proof is necessary, if their interests align enough." Marty Cooper, the cell phone's inventor and the outspoken CEO and cofounder of ArrayComm, a smart-antenna company, calls it "accidental collusion."

Thus far, government has held back. The European Commission is watching the situation, but it feels that competition is healthy in Europe. "If we were to find that the market is being dominated by two or three big players that have the power to slow down the rate of change, we would look close," says Michael Tscherny, a spokesperson for the European Commission. The U.S. Federal Trade Commission would not discuss whether it was monitoring the situation.

As for the wireless carriers, they are retaking control of their networks' destinies. The pressure to make their networks profitable, coming mostly from Wall Street, is at last steering them toward cost-efficient solutions, regardless of vendor. In the past, startups often were thwarted by carriers when attempting to pitch their products. In most cases, carriers redirected startups to their equipment providers for vetting. The ostensible purpose: carriers had to be sure that a new technology worked with their existing network. In practice, this process became a killing field for new products.

Now, however, some carriers are wising up and listening to startups. "If you have something that the operators need today, you get them to go beat the crap out of MEN and tell them to make it work," says Doug Smith, CEO of Cyneta Networks, a maker of mobile data-networking hardware. Carriers have also beefed up their testing labs to conduct more in-house reviews of new technologies, slowly weaning themselves off MEN.

All this suggests that MEN's dominance could soon be over. For instance, MEN eventually opened its Wireless Village standards body to include 157 companies. Ultimately, the charge that an exclusive MEN's club exists may depend on whether a company is a member. "It's all a matter of perspective," says Kevin Wagner, director of technology at Openwave, tempered by its newfound membership in Wireless Village. "Any time there's a group that you're not a part of, you view it as closed. But if you're in the inner circle, things seem just fine."

It will always be hard for startups to unseat established companies. And it should be. With carriers staking their reputations on the quality of their networks, it is easier for them to do business with long-standing suppliers. But more often than not, the best technology wins--once enough carriers demand it. Let's hope that happens soon.

Amid all the post-bubble lawsuits and finger-pointing, the current surge in Sell recommendations could be the biggest slap in the face to individual investors yet. As the chart below shows, analysts have reversed the golden rule of investing: they had nary a Sell recommendation at the peak of the market and are now piling them on during what might actually be the trough.

The portion of Sell recommendations has almost quadrupled from 0.7 percent of all stock recommendations in October 1998 to 2.7 percent today. Ratings look best suited as a lagging indicator, useful only for retrospective discussions about what we should not have done. (It's not clear to us that they were ever predictive.)

Some banks have been more adept than others with this too-little-too-late brand of advice. Morgan Stanley, for example, currently boasts a whopping 36 Sell recommendations in technology stocks, and zero Strong Buys. Even Merrill Lynch, once home to the King of the Bubble, Henry Blodget himself, proudly displays 16 Sell recommendations for tech stocks. But most of the other banks are still gun-shy, issuing one or two Sell ratings, and bunching the rest up as Holds. That's poor consolation: many analysts kept the Buy on while stocks were falling, only to downgrade to Hold once they'd hit bottom.

As part of its settlement with the State of New York, Merrill Lynch now attaches a chart to its analyst reports that indicates its ratings distribution across all stocks the firm covers, and also specifically across those companies with which the firm has investment banking relationships. As of May 31, Merrill had Sell ratings on 5.9 percent of the technology stocks it covers. But of those with investment banking relationships with the firm, only 2.1 percent were Sells. The firm has also promised to tie analyst pay directly to the accuracy of their reports--and not to investment banking assists. That's encouraging, but until investment banks truly separate their research groups from their dealmakers, analysts will always have incentive to be more bullish than otherwise.

But after Mr. Waksal's arrest for insider trading and the subsequent investigation of Ms. Stewart, shares of Martha Stewart Living Omnimedia (NYSE: MSO) slipped 45 percent, from $19.01 on June 6, when the story broke, to $10.40 on June 26. That's a paper loss of $264 million for Ms. Stewart, who owns 30.7 million shares. Mr. Waksal, known for his rampant carousing, has turned out to be anything but a cheap date.

We find it difficult to disagree, at least on the timing. Dell could take advantage of the likely disarray as HP and Compaq Computer attempt to integrate staff, strategy, and product road maps. Initially, Dell is more interested in capturing the recurring revenue from ink cartridges than in cranking out its own printers. The company's current strategy of merely reselling HP and Epson printers--with no ink to follow--doesn't thrill them too much. "Increasing demand for someone else's ink cartridges isn't particularly attractive," explains T.R. Reid, a Dell spokesperson.

Mr. Neff doubts that Dell will make its own printers, suggesting instead that it will resell a leading vendor's product under the Dell name. In any case, when the Dell machine gets humming, look out, HP. Still, not all in the industry think it's a strategy worth pursuing. Most printers are sold at a loss, and it takes a long time to penetrate an installed base as large as HP's. "This is not the gold mine that people think it is," argues Ashok Kumar, senior research analyst at investment bank U.S. Bancorp Piper Jaffray. "It only looks good on paper." Get it?

Each time the Red Herring 100 rolls around, it falls upon the Investor section of this magazine to distinguish a good company from a good stock. "The simplest way to put it is that you've got market perception vs. fundamental reality," says Tobias Levkovich, senior U.S. institutional equity strategist at Salomon Smith Barney. "Good investors exploit the gap."

The gap, mind you, is a moving target. And getting a bead on it isn't easy. "We're obsessed with change," says Pip Coburn, a Red Herring columnist and global technology strategist at UBS Warburg. "If we can't find positive change, we wouldn't buy a stock." We agree. What follows are five key elements of change for determining the right time to load up.

Preparing for change: building a healthy balance sheet. As the current economic climate will attest, the advantages of a strong balance sheet are difficult to overstate. Strength comes in many forms--cash, low debt, or even richly valued stock. Look for companies with a pristine ledger; such financial flexibility can allow them to persevere in lean times and to capitalize on opportunity when it presents itself. The world watched America Online use overvalued stock to transform itself into a media powerhouse by purchasing Time Warner two years ago. But with $22.7 billion in debt, $719 million in cash, and a stock price languishing around $19 a share, AOL Time Warner's buying days are over (NYSE: AOL). On the other hand, Nokia (page 46) sits atop its market with $1.3 billion in cash and only $154 million in debt, ready to handle a scrappy upstart or a market-altering technology should the need arise (NYSE: NOK).

Creating change: strategies for reinvestment. What a company chooses to spend its cash on is a good an indication of its future. Technology companies should spend as much as 15 percent of revenue on research and development. But R&D isn't always at the heart of a company's success. "A lot of investors are focused on whether a certain technology will be the next big thing," says Tom Telford, portfolio manager of the American Century Technology Fund. "But what matters is whether people will pay for it." In its last four quarters, Sun Microsystems spent $1.9 billion--13.7 percent of its revenue--on R&D. But Dell Computer spent only $458 million, a meager 1.4 percent of revenue. Both companies are successful, and both know their strengths (Nasdaq: SUNW, DELL). Sun makes money from innovative technology, Dell makes it from innovative marketing and manufacturing.

Sustaining change: mapping macro trends. Be impressed by growth, yes, but try to determine whether it's sustainable or just the result of a short-term trend. When the electronics business started to head south two years ago and device makers moved to outsource manufacturing, many investors, including Red Herring, saw electronic manufacturing service providers, like Flextronics, as obvious winners. But the initial rush to outsource, which prompted Flextronics to expand rapidly, has been thwarted by withering end-market demand. As a result, Flextronics's low margins have caught up with it (Nasdaq: FLEX). "These guys are like a supermarket," says Chris Wolfe, U.S. equity strategist at J.P. Morgan Chase. "I'd rather own Safeway." Ballard Power Systems, on the other hand, stands to gain from a very long-term trend: the move to sustainable energy. Ballard makes fuel cells for automobiles, a market with at least a decade of growth ahead of it (Nasdaq: BLDP).

Missing the change: the valuation barometer. Perhaps the most important question when considering a stock is, did I miss the boat? With technology's fervent following, the answer is often yes. Consider Synaptics, the touch pad maker and star IPO of the first quarter, which rose 39 percent from its January debut of $13 to $18 in late April. Its 2003 price/earnings ratio of 42 means this ship has already set sail (Nasdaq: SYNA).

But investors have moods, and bad moods can create opportunity. "There used to be lots of strong companies that were not good investments," says Paul Wick, managing director at the investment management firm J. & W. Seligman & Co. "But given the carnage, there are very few companies where valuation is an impediment." In some cases, a stock will offer investors a second chance. After rising 78 percent from October to January, the graphics chip maker Nvidia has been dragged down by a U.S. Securities & Exchange Commission investigation, giving it an extremely compelling 2003 P/E of just 15 (Nasdaq: NVDA). Most analysts say the storm will pass, leaving Nvidia unharmed.

Preventing change: monopolizing a market. There's one way to keep change working in your favor: by controlling it. Though rare, monopoly control of a market can ensure that pesky startups don't spoil the fun. Steven Milunovich, technology strategist and enterprise hardware analyst at Merrill Lynch (and also a Red Herring columnist), says that the only way to make money from a new technology over an extended period of time is through monopoly protection, an opinion shared by many equity strategists. "Someone who claims they have a long-term outlook in tech investing is full of it," says Mr. Wolfe. "Because in ten years, there will be a whole new Cisco, a whole new Dell, and so on." Besides Microsoft, only a few companies enjoy monopoly protection in the Red Herring 100. The telecom companies SBC Communications and Singapore Telecommunications are graced with a particularly favorable brand of monopoly: a government-sanctioned one.

One thing that will never change about tech investing is that there is no shortage of opinions. Mr. Wolfe, for example, doesn't agree with Mr. Wick that technology valuations are attractive. "I don't doubt that there are good tech companies out there, just not many good tech stocks," complains Mr. Wolfe. And the debate goes on.

UWB is hardly a new technology. Used for decades by the military to communicate wirelessly without being detected by opposing forces, the technology is just now getting noticed for its wider commercial potential. Proponents consider it a low-cost, low-power, high-speed alternative to a host of current wireless technologies. For example, in a wireless local area network environment, UWB could transmit data at speeds of 500 Mbps at a fraction of the cost of a network based on 802.11b, a popular wireless specification that typically sends data at 11 Mbps.

Because it uses a wide, flat signal that pulses rapidly, UWB is different from traditional radio communications systems. "You can eliminate a lot of the traditional radio functions, like an oscillator, filter, and mixer, and that cuts down on cost," says Chris Fisher, vice president of marketing at XtremeSpectrum, based in Vienna, Virginia, and one of several semiconductor companies hoping to cash in on the FCC's recent decision. In addition to UWB chip startups, other well-known names in wireless are licking their chops at the potential market. "We look at this as an excellent opportunity," gushes Jeffrey Belk, senior vice president of corporate marketing at Qualcomm. Other possible benefactors are equipment makers that incorporate UWB chips into their new products, like cell phones.

But the conceivable threat of harmful interference from UWB has rocked the wireless community, which fought bitterly against UWB in one of the ugliest clashes the FCC has ever witnessed. More than 900 companies filed concerns with the government agency; some went so far as to claim that airplanes would fall out of the sky because of interference with FAA guidance systems, and that the blood would be on the hands of the UWB startups. "Saying that something will interfere with GPS is the modern-day equivalent of shouting 'Fire!' in a crowded theater," says Ralph Petroff, president and CEO of Time Domain, a UWB chip-set company based in Huntsville, Alabama, that is backed by Sony, Siemens, WorldCom, and Marconi. In a letter to Donald Evans, the U.S. secretary of commerce, and Michael Powell, chairman of the FCC, representatives of AT&T Wireless, Cingular Wireless, Qualcomm, Sprint PCS, and Verizon Wireless implored the administration to restrict the use of UWB further, claiming that "people in an office building trying to use their cell phones to report a fire or other emergency could well have their calls blocked" by UWB.

But after extensive testing, the Bush administration and the FCC determined that it was all a lot of hot air, and the technology was safe after all, provided it was operated between 3.1 Ghz and 10.6 GHz, and under a certain power level. "Any precedent-setting, disruptive technology faces resistance, but you have to see through the fury of the moment and filter for the truth," says Michael Gallagher, deputy assistant secretary at the National Telecommunications and Information Administration, the Washington, D.C., organization that advises the president on all telecom matters and which oversaw the testing of UWB.

All of which makes one wonder: why are the wireless carriers so afraid of UWB, especially as the FAA and DOD have been using it without incident for 40 years? Perhaps because they know how effective UWB is. "This is about competitive concerns," says Maura Colleton, managing director of Qorvis Communications, the Washington, D.C., public affairs firm that lobbied Capitol Hill on behalf of XtremeSpectrum. "These companies are protecting their existing markets and their ability to go into future ones."

Indeed, UWB mightily threatens a number of existing markets. The 802.11 community will be in immediate danger, and Bluetooth, the emerging wireless standard for device-to-device communications, may be rendered obsolete before anyone actually gets to use it. But far more significant is the effect that UWB could have on the next-generation networks that mobile carriers have spent so lavishly on to develop over the past three years. "I believe the wireless carriers' objections really stemmed from a financial and a political perspective, more than from spectrum interference," says Martin Rofheart, the cofounder and CEO of XtremeSpectrum. At the current power restrictions delineated by the FCC, UWB is not yet approved for commercial use beyond networks of a couple hundred feet at the most.

The real reason that carriers recoil in horror at the mention of UWB is that if the power restrictions were ever loosened by the FCC, which is scheduled to revisit the restrictions by next April, UWB would become a viable competitor to existing cellular networks. In regions like South America or China, where wireless networks are less established than in the United States and Japan, UWB could be a key part of the wireless infrastructure.

The incumbents aren't giving up their fight, however. Following the FCC's ruling, many carriers issued press releases critical of the decision, demanding that the agency revisit the issue and tighten the restrictions. But buoyed by their improbable victory in this fight, the UWB startups are confident of their future. "There is almost this Pavlovian response toward new technologies by incumbents--the fear factor," says Mr. Petroff. "But this was a big win for the little guys, because nothing is so powerful as an idea whose time has come, and this . . . this is a good idea." Good enough to scare the pants off the entire wireless industry.

New technologies will begin to address these limitations. Companies like ArrayComm, which uses smart antennae to focus radio waves more efficiently, and Conductus, which uses superconductors to reduce the cost of networks, are producing incremental advances in wireless efficiency. Other companies, like Schema, are helping carriers to use their networks more cost effectively by employing evolutionary algorithms--highly complex spectrum analysis--to optimize spectrum usage. For their part, Seven Networks and Vaultus are both working to make wireless technology more useful in the workplace, something that could bring significant revenue to wireless carriers.

But few technologies have emerged that can provide a clearly perceptible jump forward. Wi-fi, based on the 802.11b standard, is still just for local area networks. And ultra wideband, also known as UWB or digital pulse wireless, is bogged down in a fierce regulatory battle at the U.S. Federal Communications Commission. On the bright side, however, Flarion Technologies' orthogonal frequency division multiplexing, an alternative data transmission technology for existing network standards, has the potential to become the new transmission standard for wireless data. If it works, the woes of wireless may soon end.

But here at Red Herring, hope springs eternal, and nobody is looking forward to the big thaw more than we are. We take very little solace in the fact that the Nasdaq, which lost 10 percent, had a rough month too, considering that the Dow managed to eke out a gain of 2 percent. We take solace only in the fact that spring is upon us--and, of course, that things really can't get much worse. So for the time being, frozen by the fear of change, we are sticking with our picks, which in January nearly wiped us out (last month we swapped in five new stocks, which then lost an average of 39 percent).

In our February review, we have but one stock to crow about: Applied Materials. Although it gained a meager 1 percent, it is our pride and joy. It's sad when your only winner saw its sales decline for the fifth straight quarter, but the semiconductor equipment maker did manage to post a profit, excluding charges, that surpassed analyst expectations by 2 cents a share. All of which serves as yet more proof that the semiconductor business has finally turned a corner.

Stocks that didn't post gains, but also didn't completely embarrass us, include Electronic Data Systems, Comcast, and Nokia. EDS posted a 20 percent increase in fourth-quarter earnings over the same quarter last year. But investors were disappointed that sales of $5.9 billion were a bit shy of analyst estimates of $6.1 billion.

Comcast continued to ride the roller-coaster of news coming out of its proposed $72 billion acquisition of AT&T's cable assets. The U.S. Department of Justice is looking over the deal, but most experts expect the transaction to go through. Comcast slid 2.8 percent.

Nokia lost 11.9 percent as it got dragged down by the same wireless hysteria that torched Nextel Communications for 36.4 percent. At least Nokia has its 40 percent market share in mobile phone sales to bolster its stock price in the face of a mad sell-off. Nextel has $15 billion in debt and a delayed announcement of fourth-quarter results. The wireless boutique is expected to restructure its debt and take a charge, which would make it more attractive as an acquisition target, as will its diminutive stock price of $5 a share.

The rest is ugly. Cisco Systems blew away earnings expectations of 5 cents a share by reporting 9 cents a share on sales of $4.8 billion in its second quarter (ended January). But CEO John Chambers offered a dour outlook, sparking a sell-off. The stock shed 25.7 percent. At these levels, analysts are recommending that investors take out a second mortgage and put it into Cisco. We'll settle for holding our current stake.

Riverstone Networks, our clear loss leader after shedding three quarters of its value during the month, led the telecommunications meltdown. After fending off allegations of an investigation by the U.S. Securities and Exchange Commission, the company warned that it would miss fourth-quarter estimates substantially. Broadcom and the Marvell Technology Group also suffered from the same fears plaguing the telecommunications sector, as the two lost 27.4 and 23.7 percent, respectively. It has been a long, harsh winter for the Red Herring portfolio, but we think we see the sun emerging from behind the clouds. It's about time.

Write to portfolio@redherring.com.

It's hard to blame companies for fantasizing about their own allegorical attics--especially big companies, where it's not inconceivable that the value of some intellectual property has been obscured over time. While experts warn that usually only 1 percent of a company's patents hold 99 percent of the value of an entire portfolio, the exercise can still be well worth the effort. When Lou Gerstner joined IBM as CEO in 1993, one of the first things he did was direct the research labs to start commercializing the latent intellectual property sitting in IBM's massive patent portfolio. Today the company owns more than 20,000 active patents, from which it generates $1.5 billion in annual revenue.

With companies clamping down on expenses, capitalizing on these assets is a relatively low-overhead operation that can increase revenue without coincident expenses. Texas Instruments, the company that patented the integrated circuit, found that licensing intellectual property was borne out of necessity. "We made a conscious decision to start realizing more of a return on our R&D dollars when things were bad in the mid-'80s," explains Fred Telecky, senior vice president and general patent counsel at TI. "It worked." The company generated $340 million in semiconductor licensing fees last year.

Intellectual-property licensing is a $150 billion market, but discovering the value in a portfolio and transforming it into revenue is harder than it sounds. So companies like IpValue Management, a recently launched venture that specializes in managing the intellectual property of companies, are being formed. "Some organizations are sitting on gold," says David Pecaut, president of the iFormation Group, the parent of IpValue. "And people are starting to get a lot more comfortable with the idea of IP licensing as a legitimate source of revenue." British Telecommunications' advanced research and technology division, BTexact, recently signed with IpValue, hoping to extract more revenue from its portfolio of 14,000 patents.

But the industry that has emerged around intellectual-property management has yet to find sure footing. Theoretically, a market downturn should be a boom time for firms engaged in this kind of management, as companies look for inexpensive sources of revenue. Tell that to Aurigin Systems. The intellectual-property management company that Mr. Rivette founded has gone into financial disarray after the entire senior management team and board of directors jumped ship in January, leaving Aurigin's future uncertain. And IpValue, the newest entrant to the business, still has but one lone customer.

The biggest issue is the intangible and volatile nature of intellectual property. The values change with people's moods--just like the stock market--meaning that both value and volatility must be considered when purchasing a patent. Graphics processors, for example, are among the most volatile forms of intellectual property, and their value has declined 37 percent in the last six months. The technology patents with the most value now are in storage area network firms.

Of course, when it comes to turning intellectual property into revenue, you have to own something of value. "The problem I have with all this talk about 'Rembrandts in the attic' is that if they're in the attic, you're already in trouble," says Jerry Rosenthal, vice president of intellectual property and licensing at IBM. Mr. Rosenthal may be preaching with the righteousness of the recently converted, given IBM's own turnabout, but his point is valid nonetheless. "Odds are if we didn't find a patent valuable at the time we secured it, no one else will either. You only make money on things that people want." That, of course, is the hardest part.

Nextel is an undeniably attractive target. Its stock is about as cheap as it's ever going to get, clocking in at just $3.51 on February 19, down from a 52-week high of $34 (Nasdaq: NXTL). And the Virginia-based wireless operator has two of the things that wireless carriers covet most: spectrum and business customers.

With the spectrum shortage in the United States getting desperate, carriers now are willing to merge or acquire just to gain spectrum to improve service quality and coverage. And in December, the Federal Communications Commission paved the way for wireless consolidation by voting to remove the spectrum cap, which had limited wireless carriers from owning more than 45 MHz of spectrum in any given geographical area. By the time the cap is lifted, most of the expected mergers should already be under way. Nextel has 10 MHz of geographically contiguous spectrum in the United States, a wide swath of bandwidth that covers the country's major urban centers.

Add to that the $70 average monthly revenue per user (ARPU) that Nextel derives from its business customers, who are addicted to the Direct Connect, or "push-to-talk," feature, and the company is looking mighty fine. ARPU figures for more consumer-focused carriers hover around $50, and none of the other major U.S. carriers have figured out how to court the more profitable business market. Nextel's lifetime ARPU is a cool $3,430, compared with $1,768 for its competitors. Nextel's customers could be the most desirable in the industry, according to Sean Butson, vice president of wireless services and towers research at Legg Mason, an investment bank.

For would-be acquirers, there are two obstacles. The first--and ugliest--of Nextel's warts is its long-term debt of $16.5 billion. It would take a healthy balance sheet to absorb a blow like that. The second is that Nextel also uses a proprietary network technology, called iDEN (integrated digital enhanced network), which isn't compatible with CDMA or GSM, the two network protocols used by all other U.S. carriers. While the company has indicated that it may move to CDMA in the near future, it has yet to commit. As it stands now, an acquirer would need either to run two parallel but incompatible networks or to migrate customers from iDEN to CDMA or GSM, a move that would be both disruptive and technically challenging.

Neither obstacle is insurmountable. AT&T Wireless, fresh from its spin-off from parent AT&T, has strong financials. And spectrum-starved Verizon, which uses CDMA, will likely be interested if Nextel continues to cozy up to that protocol.

Even using conservative estimates, an acquirer could conceivably pay $10 a share for Nextel, says Frank Marsala, an analyst at the investment bank Robertson Stephens. That's almost three times the current stock price. We find that math hard to argue with, despite Nextel's imperfections.

Today, Mr. Cooper finds himself on the other end of the cell phone business: he's pitching smart-antenna technology (which reduces signal interference and the number of dropped calls) to his old colleagues at Motorola. He's frustrated simply because his pitches are being rebuffed at every turn. Mr. Cooper says, "3G cellular just won't work without smart antennas, so this game is not over by any means. We're hoping the dam will break soon."

ArrayComm isn't the only company pushing technology to help eliminate wireless inefficiencies. Other startups are finding that, although their technology is desperately needed, it is not welcomed by established wireless hardware vendors, like Ericsson, Lucent Technologies, Motorola, Nokia, and Siemens.

At a time when dropped calls and spotty coverage are the norm, one would think the wireless carriers couldn't get enough of new technologies that speed data transmission, focus radio waves for spectral efficiency (rather than broadcasting signals in a 360- degree radius), and map network optimization plans (plotting out prime locations for cell towers and other hardware). What's even more surprising is that some of these technologies, a number of which were spun out of Lucent and Motorola themselves, have been around since 1987.

The resistance is mostly because more efficient technology threatens these same companies' more lucrative base-station market (worth a whopping $138 billion in 2001). With powerful hardware vendors standing in the way, many of these promising technologies are going nowhere.

"The hardware vendors have a disincentive to sell these products, either because they will sell fewer base stations, or they are working on something themselves," explains Charles Shalvoy, CEO of Conductus, a superconductor company that says its products can expand a transmission or cell tower's capacity by more than 80 percent. Mr. Shalvoy's argument is this: the more efficient the network, the less hardware is needed to run it and the fewer base stations get sold. And that could cost the hardware vendors a bundle in new revenue.

Vendor BenderThe effort to keep the new efficiency technology down has taken more than a few ugly turns. Most of the efficiency startups are too small to go straight to the wireless carriers, so they generally need to go through hardware vendors to get a contract. Carriers prefer to let their hardware partners test and qualify new technologies for them. And, not surprisingly, going through the vendors is often a dead end.

Among the other startups clamoring to be heard are Schema, a software company that focuses on optimizing network traffic; ScoreBoard, which promises network optimization through analysis and planning; and Superconductor Technologies, a company that makes materials that provide better circuit performance for electronic components. But the news isn't all bad. Metawave Communications, which makes smart-antenna systems, has been able to persuade Lucent to use its smart antennae in a limited deployment, and is even working with the company to develop a next-generation antenna. Conductus has had some success with smaller carriers, like Dobson Communications, and has even gotten third-parties like Nortel Networks and Motorola to concur that its technology is compatible with their networks. But the process is slow and painful.

However, even when efficiency companies are able to grab the attention of a major carrier, the hardware players get fiercely territorial. One startup, which requested anonymity, had successfully convinced a carrier to retrofit its base stations with the startup's new technology. But on hearing of the changes, Ericsson threatened to void the warranty on the existing base stations. The carrier had no choice but to back out.

Stories like this abound in the David- and-Goliath startup culture of the telecommunications industry; in the wireless arena, it gets particularly cutthroat. Take, for example, ArrayComm's experience: it presented its smart-antenna technology to a major hardware vendor, only to have the vendor turn around and tell its carrier customers that the technology didn't work. The hardware vendor then recommended its own solution. Mr. Cooper says the vendor's claims were completely bogus. But carriers typically listen to their vendors. The vendors also have financial reasons for dragging their feet: many of them are financing one or more of the carrier customers through the tough economic times.

"At Motorola, we were experts at this stuff," says Mr. Cooper. "If you weren't leading in a technological field, you explained how the other guys' stuff didn't work. It was a way of stalling." One startup even accused Ericsson of not promoting smart-antenna technology it had already developed in-house for fear of undermining Ericsson's own base-station sales.

Carriers maintain they have every intention of bringing the best technology to market. Despite these broad claims of commitment, companies like Sprint PCS are still only in the early stages of evaluating smart antennae from ArrayComm and Metawave. In contrast, Lucent, one of Sprint's top suppliers, has a smart-antenna product of its own, due out by summer 2003. "We could have put smart antennas into our technology sooner," says Paul Mankiewich, chief technology officer of the wireless network group at Lucent. "But the demand wasn't there yet. It's a fundamental build-or-buy decision."

For their part, hardware vendors say the complaints of startups are exaggerated at best. Hakan Eriksson, vice president of research at Ericsson, says it wouldn't be in Ericsson's best interest to keep new technology down, even if it were possible for the company to do so. "If we did block the efficiency technology, maybe we could sell more base stations for a short time, but in the long run, we would be risking our entire business," Mr. Eriksson explains, noting that the competition among hardware vendors is very tight. If one were to drag its feet in adopting new technology, a competitor would surely seize the opportunity in the open market.

The large number of layoffs have also made it difficult for hardware vendors and carriers. In-house research departments don't have the resources to innovate as they did in the past. This leaves big players all the more dependent on technologies developed by nimble startups.

Users are LosersWhile the wireless struggle rages on, consumers continue to suffer with poor-quality wireless service, especially in the United States. And analysts fear it will only get worse. Data is expected to swamp the airwaves in coming years, and the interference and coverage problems that consumers now face will increase unabated, says Christine Loredo, an analyst at the Strategis Group, a telecommunications research firm. Ms. Loredo says that merely adding base stations, which improve coverage and capacity, will not be the answer.

According to Strategis, in 2001, U.S. wireless carriers increased the number of cell towers 19 percent, to just more than 1 million. But wireless-tower makers have begun scaling back, and the annual growth figures will fall to 8 percent by 2003. Last year, the number of subscribers grew 22 percent, and total wireless minutes of use ballooned 75 percent. With radio spectrum already stretched to the limit, too, networks will have to get more efficient one way or another.

Startups are trying to change fundamentally the relationship between vendors and carriers, requiring hardware sales to be based on quality of service, rather than just quantity of hardware. With their leverage, carriers could demand minimum performance levels for their networks within a given geographical area, shifting the onus for network efficiency to the hardware vendors. According to Conductus's Mr. Shalvoy, this could light the fire under the hardware vendors, forcing them to adopt new technologies more rapidly. Some of these new technologies may soon find their way to market. Meanwhile, network-efficiency providers are still waiting for the revenue to flow.

In the past six months, tech companies have lopped off a total of $1.7 billion in R&D spending (see "Red Alert," below). Of course, many customers aren't buying right now, so some budget trimming is in order. But when things turn around, those with new products coming out of the R&D pipeline are going to be better positioned to cash in than those that scrimped during the downturn.

It would be nice to know which companies will be on the first roster. But that's more difficult than it sounds. Ask a hundred Wall Street analysts how to factor R&D into stock selection, and you'll get a hundred different answers. Try to get them to quantify it, and they immediately start to hedge. "This stuff gets very subjective," laments Alvin Kressler, a semiconductor analyst at the investment bank Friedman, Billings, Ramsey. "I try to look at it from a qualitative standpoint, rather than a quantitative one."

Consider the following: if you'd noticed that Apple Computer maintained its R&D spending through its unprofitable years of 1996 and 1997, and you bought the stock as a result, you'd have been on board for the release of the iMac in 1998--and been amply rewarded. On the other hand, if you thought Lucent Technology's spending of $4 billion a year in R&D since 1998 was a portent of future success, you'd have been dead wrong. You also would have lost 90 percent of your investment.

So how can you factor R&D into an investment decision? Michael Murphy, founder and editor of the California Technology Stock Letter, uses a "price/growth flow" model. Instead of merely looking at a company's trailing 12-month earnings per share--the denominator in a trailing price/earnings ratio--Mr. Murphy calculates a company's R&D per share and adds that to earnings. It's a shorthand way of tracking commitment to R&D. It's also the best we've got. Unfortunately, like most short cuts, price/growth flow has shortfalls. Primarily, it provides no indication of whether it's tracking effective R&D spending, like Apple's, or futile R&D, э la Lucent.

Measuring such success is a challenge. Companies often cite the number of patents filed as an indication of the quality of their R&D spending, or even as a predictor of success. But using a patents-per-dollar of R&D ratio "tends to be more about the efficacy of your lawyers, rather than your R&D department," deadpans Vadim Zlotnikov, an analyst at the research outfit Sanford C. Bernstein.

Turning patents into marketable products is a problem that has plagued countless tech companies. Xerox is famous for it--the company's Palo Alto Research Center (PARC) developed one disruptive technology after another yet failed to capitalize on them. In fact, many of its inventions, like the laser printer and the mouse, are now the domain of competitors.

"You have to be able to measure the yield of your R&D dollars to avoid becoming another Xerox PARC," says Robert Morris, director of IBM's Almaden Research Center, the envy of the industry. "But it's not easy. Getting market feedback from R&D is like a drunk man driving a car--by the time you see your mistake, it's too late to correct it."

One of the best ways to measure R&D efficiency is to look at the percentage of sales that comes from products introduced in the last three years. That provides a general indication that the firm is releasing new products and not coasting on existing ones. "A telltale sign that a company is milking rather than growing is when you see them growing services and cutting back R&D," says Mr. Zlotnikov.

Another measure to keep an eye on is a company's R&D as a percent of sales, which varies from industry to industry. Of course, most R&D spending has come down over the past six months, a natural consequence of the revenues that are now circling the drain. But few companies have been foolish enough to dramatically cut R&D as a percentage of sales. In fact, as revenues have tanked, the industry average has increased from 11 percent to 17 percent, as firms have been loath to cut R&D.

Intel, for example, pared its R&D spending from $998 million in fourth quarter 2000 to $919 million two quarters later, but R&D as a percentage of sales ballooned from 11.5 percent to 14.5 percent. It even managed to introduce silicon-on-insulator technology, a breakthrough for next-generation processors. "If you're in a rotten, stinking environment, you have to make every department pay a little," says Sam May, an analyst at U.S. Bancorp Piper Jaffray. "But I'd make the R&D people pay a little less than others."

On Wall Street, the fear is very real that reduced R&D will gut the future of technology companies. Over the last two years, 3Com has averaged $160 million a quarter on R&D. In its latest quarter, the company spent only $86 million, a precipitous drop that has analysts on alert. "They're cutting costs where they can, but there's a risk of missing product cycles and losing share," frets Seth Weber, a Merrill Lynch analyst who has a Neutral rating on the stock (Nasdaq: COMS).

Of course, not all technology companies are the same. Dell Computer has made a living by not spending on R&D--the company outlays about 1.5 percent of sales on R&D--yet has produced phenomenal returns for investors (Nasdaq: DELL). Others, like Cisco Systems, have healthy R&D programs but also use acquisitions as an adjunct to in-house research, making any attempt to gauge the effectiveness of their R&D "spending" that much more difficult.

Ultimately, if you want to open the can of worms that is R&D, be prepared to do some heavy lifting. It's a world fraught with anomalies, misleading statistics, and unqualified results. Nobody said this was going to be easy.