DaimlerChrysler, Toyota, and Ford are betting the farm on the death of the combustion engine. Their money is riding on the fuel cell – in particular, the proton exchange membrane (PEM). But it’s not the only game is town. Not by a long shot.

Six years ago, the U.S. Department of Energy challenged the best and brightest scientists of the nation's government labs to develop the solid oxide fuel cell (SOFC). Thanks to government backing and lab breakthroughs, SOFCs are gaining steam.

SOFCs produce energy electrochemically, and are cleaner and more efficient than current combustion technologies. Compared to PEMs that run on pure hydrogen, SOFCs run on natural gas, diesel, and biomass (plant material) fuels that are easier to obtain than hydrogen. Those that push hydrogen as the cleanest alternative to fossil fuel often fail to mention that it does not occur naturally – it has to be cracked out of existing compounds like water. The extraction takes energy, often using dirty sources such as coal or gas burning power plants.

A leading SOFC developer is the Lawrence Berkeley National Labs (LBL), operated by the University of California for the Department of Energy. With a $500 million annual budget and a solid global reputation in battery technology, electric vehicles, and new energy solutions, LBL is making strides to solve SOFC's main sticking points, and are wise to the pitfalls.

The global fuel cell industry is projected to be worth more than $18.6 billion in 2013, according to Allied Business Intelligence (ABI), a technology think tank. It says that the three main market applications will be automotive, stationary, and portable generators.

LBL officials believe that the portable generator market provides the best near-term market opportunities. The portable generator market is worth $45 billion today, with 44 percent coming from the U.S., 30 percent coming from Europe, and 20 percent from the rest of the world, according to ABI. "Many expected the market sector to take off after the California power crisis,” says Atakan Ozbek, an ABI research analyst. “This didn’t happen but there is mounting pressure to find alternative energy sources.”

Indeed, recent wide-scale power grid meltdowns in New York City and parts of California sent the world a wakeup call.

In Spain, for example, power companies Endesa and Iberdrola limit their domestic customers to 10 kilowatts per household. This is their way of ensuring that they do not suffer from the type of outages experienced by the rest of the world. Effective in preventing outages, the measure means sacrifice for customers, who are unable to run several electrical appliances at the same time, and are often forced to buy generators.

Breaking the price barrierSo what’s stopping SOFCs from owning the portable generator market? “The adoption of new technology such as SOFCs is a gradual process,” says Steve Visco, an LBL principal investigator. “You have to start with relatively high-cost applications until you can mass produce the technology.

“Green is good for the marketing department," Mr. Visco adds, "but unless you can price fuel cells competitively with today’s technology, nobody is going to buy them.”

Price has been holding SOFCs back. Building costs currently hover around $4,000 per kilowatt (the energy pros' yardstick) according to the Solid Oxide Fuel Cell Alliance, compared to just $400 per kilowatt for gas turbine generators. LBL is hoping to break below that magic barrier, by rethinking – and re-pricing – all aspects of SFOC production.

Materials offer one cost-cutting opportunity. SOFCs consist of three components: a cathode, an anode, and an electrolyte, or catalyst, sandwiched between the two. The electrolyte is made up of a thick plate of yittria-zirconiam, a ceramic material backed with porous nickel. The raw materials cost upwards of $60 per kilowatt, not including the other components and manufacturing costs. The nickel is expensive and the ceramic is difficult to manufacture and manipulate. With material scientists Craig Jacobson and Lutgard De Jongh, Mr. Visco reduced the amount of yittria-zirconia from 500 microns to just 15 microns (25 microns is the width of a human hair) and they swapped nickel for stainless steel.

The bottom line: costs for raw materials were cut to $137 per kilowatt. Typically, the raw materials make up about one-third of the cost for the whole unit so they are on par with gas turbine technology. The big difference between the two technologies is that while gas turbine generators cost $400 per kilowatt to build, they typically come in 15 megawatt configurations – too costly for anybody except power producers.

They also reduced the thickness of the catalyst, which allowed them to lower the operating temperature by 19 percent to 650 degrees Celsius. That, in turn, increased the unit's efficiency.

But there was a snag. The cathode is fused to the electrolyte at very high temperatures, above 2,000 degrees Celsius. Stainless steel, however, tends to oxidize or break down above 1,200 degrees. Also, Mr. Visco fears that the stainless steel will be less robust than nickel and will degrade more quickly than other fuel cells. The LBL group is working a way around the problem as it goes through the patent process.

There are other doubts: “We are not sure how long our fuel cells will last,” Mr. Visco says. “Everybody shoots for a 10-year life span.” LBL's cell, he adds, will go through extensive testing.

From the basement to the pavementThe lab is also hoping to create the underlying technology for the small portable power unit (about the size of a water heater) market and is working with companies such as Ikerlan, a domestic appliance manufacturer in Spain, and the Columbus, Indiana-based Cummins Power, maker of energy technology products including emission solutions and electrical power generation systems.

They are all players in an expanding market. Several companies including Siemens Westinghouse, Acumentrics, and Cummins are producing fuel cells, mostly for evaluation purposes. Kleiner Perkins Caufield & Byers, the venture capital firm behind Netscape and Google, has also invested in a SOFC company, but because the company is in "stealth mode," those details are hush-hush.

Fuel cells are also ideal auxiliary and backup power generators for office buildings, apartment blocks, and even private residences. Ikerlan is hoping to sell SOFC generators to augment household power supplies. Cummins, on the other hand, would use the technology to provide a near-silent alternative to RV power generators, banned in the evening at some campgrounds because of their racket. Canada's Ballard Power, best known for producing fuel cells for the auto industry, just released AirGen, its one-kilowatt fuel-cell generator. The big drawback is the cost: $5,995 (hydrogen cylinders not included).

If LBL irons out its wrinkles, the market could have a 10-kilowatt gas-powered fuel cell generator for less then $4,000. Quiet and efficient, the unit would be small enough to fit in a garage or basement. Such progress stands to recharge the ailing power industry and cutback on its biggest threat: the blackout. Back-up generation, however, is just the near-term use for the technology. There are all forms of other applications, from refrigeration container trucks to replacing the good old combustion engine itself.

“The auto industry is the last market to adopt a new technology – not the first – because it is the least forgiving and the most highly developed,” says Mr. Visco. “But right now it is the one industry in most need of SOFCs.” The price of its fuel cells, which cost currently cost $1 to $4 million to produce, will scare away even the most ardent environmentalists.

That's why SOFCs could pull an end-run on the fuel-cell market. If the technology behind LBL’s next-generation fuel cell performs as advertised and its costs hold up, SOFCs are poised to have a lasting impact on the portable and stationary generator sector. It doesn’t end there. By proving its value among the little fish, the technology may be able to target the biggest fish of all – and sneak under the hoods of cars.

You could call him the real Agent Mulder. Like that indefatigable character from The X-Files television series, Jonathan Trent, an astrobiologist with NASA, is driven by a curiosity to find life-forms that can exist on other planets. For the last 15 years, he has dived into oceans, climbed volcanoes, and scoured the world’s hot springs to study extremophiles, rare microorganisms that can survive in extreme heat, cold, and even acidic conditions. “These are the types of extraterrestrials that we should be looking for, not funny creatures with bumps on their heads,” says Mr. Trent. “The question I have been trying to answer is, How do these organisms survive in these conditions?”

In 1991, his search for “extreme” organisms turned up a breakthrough that may revolutionize a branch of nanotechnology known as molecular manufacturing. In this nascent field of study, scientists assemble structures and materials using atoms and basic molecular forces as primary building blocks.

If molecular manufacturing delivers on its promise, tiny microchips could be created atom by atom – preserving Moore’s law well into the next century. New medical devices, like nanobots that roam our blood stream and deliver drugs, could become a kind of backup immune system.

But nanotechnology has made far more progress in press releases and the pages of science fiction novels than it has as a viable science. Thus far, molecules have proven too small to manipulate with precision, even with atomic-force microscopes. For the most part, scientists have been able to do little more than prod, magnetize, or push them around a surface. This means that they cannot really build anything; it is the difference between stacking up cement blocks and actually building a house. If molecular manufacturing were house construction work, we would still be learning to build the wooden frame.

And that is where Mr. Trent and his team at NASA’s Ames Research Center in Mountain View, California, come in. The scientists have discovered a substance that can act as that much-needed frame to hold atoms in place and organize them into different structures. The discovery will, they say, make the manufacture of nano-scale devices like microprocessors, memory chips, or drug-delivery modules as easy as mixing a solution in a test tube. Best of all, this substance could enable such items to be produced in large quantities economically.

Mr. Trent’s discovery has not gone unnoticed. The research center’s Nanotechnology Lab, one of five regional nanotech research centers in the United States, is the largest recipient of the government’s $700 million nanotech investment fund. Moreover, NASA is already in licensing discussions with at least one major biotech firm to use it.

The miracle substance is the byproduct of a microbe that has survived the last 3.4 billion years in some of the world’s harshest environments. It has survived the ice ages, the extinction of the dinosaurs, and now quite happily thrives in sulfuric hot springs, where temperatures reach 185 degrees Fahrenheit. This little microbe’s resilience has helped scientists begin to overcome the obstacles to manufacturing nano-scale devices.

Until now, molecular manufacturing has been constrained by what scientists like to call the “fat fingers” problem. As anybody who has chased a grain of rice around with a pair of chopsticks can testify, it can be very difficult to pick up or place really tiny objects with relatively big tools.

There are, roughly speaking, three basic approaches to molecular manufacturing: binding molecules by mixing them with sticky molecules like fats (chemistry), binding them through genetic engineering (biology), or binding them using physical methods like bombarding them with electrons (physics). But because the scale involved in doing these manipulations is so small, many scientists have concluded that it would be best if nanotech devices organized themselves like chemicals, replicated themselves like living organisms, and functioned as predictably as machines. Overcoming these hurdles surely would require Mr. Trent’s express focus on these problems, along with the integration of several different branches of science and a healthy dose of creativity.

Then, as often happens in scientific discovery, serendipity took over. Mr. Trent and his team unearthed a protein called Heat Shock Protein 60 (HSP 60) while studying Sulfolobus shibatae, an extremophile microbe that lives in the sulphuric hot springs in Yellowstone National Park. This protein assembles itself into a ring structure called a chaperonin, which fortifies the cell membrane so the microbe can live in extreme conditions.

Closer inspection revealed other fascinating features. First, a chaperonin molecule has an empty core, like a stack of Lifesavers candy, which makes it an ideal manufacturing frame. But the diameter of the core is too narrow to hold enough nanosize particles for practical manufacturing purposes. Chad Paavola, a molecular biologist at NASA, genetically tweaked the protein to widen the “hole” and remove unwanted strands of protein. He then engineered the protein so that it would grow amino acids on the inside of the opening so that nanoparticles would stick there. Otherwise, nanoparticles could bounce out of the chaperonin opening.

Once they were able to increase the diameter of the hole, the molecule could be used to trap and contain nano-size substances like gold, magnetized particles (for storage devices), or toxins that target cancer cells. And chaperonins use little hooks around their outside edges to attach to one another and organize into rows and columns. This ability to self-organize is key: it solves the “fat fingers” problem. Furthermore, because Sulfolobus shibatae lives in warm environments, the chaperonin could be used in manufacturing processes that reach temperatures as high as the boiling point of water.

After getting acquainted with these microbes in the lab, NASA biologists employed a common biotech trick: they inserted foreign genetic code into the rapidly replicating Escherichia coli bacteria. In this case, they fooled the single-celled organism into producing industrial quantities of chaperonin. “Normally you have to separate out the bacteria and the protein in a costly process,” says Mr. Trent. “But HSP 60 is heat resistant, so you can warm up the solution until you burn off the E. coli.” And all that remains is very useful nano-scale scaffolding.

But the work did not end there. NASA scientist Andrew McMillan discovered that the chaperonin could be morphed from its tubular shape into various other shapes, like a honeycomb, which could provide the lattice structure required to assemble a nano-size memory chip.

According to nanotechnology expert David Goldhaber-Gordon, assistant professor of physics at Stanford University, each individual piece of Mr. Trent’s method has been done before. Panasonic and Sony, for example, are working with a protein that binds to magnetized iron particles, which they hope will provide nano-scale storage mediums.

“However, what is unique is combining work such as protein design with self-assembly techniques to make a hybrid organic/inorganic substance,” says Mr. Goldhaber-Gordon. “And it is the versatility of the chaperonin that is new here. Once you have the scaffolding, you can attach pretty much anything you want.”

NASA’s approach is promising because any particle can be mixed in with the chaperonins, which will provide the framing upon which other nano-scale devices can be built.

Whether it is the scary killer bugs in Michael Crichton’s techno-thriller Prey or the tiny medical devices that roam our bodies eating cancers in K. Eric Drexler’s visionary book Engines of Creation, nanotechnology is billed as both the miracle and the nightmare science of our age. But so far it has made little progress as either. In fact, it will be a few years before we have even a nano-scale computer chip. What we are seeing now is the creation of the materials or building blocks that will perhaps make these things possible. Society will have to decide whether that is good or bad.

Like any living, breathing American secretary, this digital creation has to understand English, answer the phone, schedule meetings, and reply to email. (No running out for Krispy Kremes just yet.) The assistant, much like a school child, is expected to learn over time and pass exams each year.

What has any of this got to do with the military? The Defense Advanced Research Projects Agency (DARPA) wants SRI to build a digital secretary so it can apply that knowledge to create applications suited for the battlefield. DARPA envisions soldiers who can simply ask their computers for information rather than spend time operating clunky keyboards and push-button screens.

The software secretary is an extension of an idea that has been floating around research institutions for some time under a variety of different guises. The late Michael Dertouzos, former director of Massachusetts Institute of Technology’s (MIT) Laboratory for Computer Science, believed that computing should be human-centric. He was fond of saying that people were slaves to their computers – forced to spend their lives feeding them data (see Lab Rat, February 22, 2001). To steer away from that predicament, the lab wants to make computers “ubiquitous,” wherein the machines are so simple they fade into the background of our lives. An early step is creating intelligent assistants and natural language processing applications.

Likewise, the Xerox Palo Alto Research Center (PARC) has been working on the concept of ubiquitous computing since the 1980s. Last year, IBM had a pervasive computing program code-named Planet Blue (see Lab Rat, October 19, 2000) that was designed to find new ways for today’s emerging technologies. Other DARPA-funded research institutions are running similar programs: Carnegie Mellon University's (CMU) Project Aura, the University of Washington's Portolano project, and the University of California at Berkeley's Endeavor Expedition are each developing human-centric computing architectures.

“We are going to have to develop systems that understand spoken language because the basic form factors – the laptop, the phone, and the PDA – are not going to change,” says Norman Winarsky, vice president of strategic development at SRI.

What is different this time, according to Adam Cheyer, chief architect of the CALO project, is the approach. SRI will pull together brains and experience from more than 22 research establishments including Yale University, MIT and companies such as Boeing. Together, they will develop a project called Cognitive Assistant that Learns (CALO), which is derived from the Latin calonis, meaning “soldier’s assistant.”

“It’s the largest artificial intelligence project every funded by DARPA,” says Mr. Cheyer. “We are going to bring together about 150 of the world’s top artificial intelligence experts here to create a software system that will make fewer errors, be more flexible, and have the ability to learn,” he says. “It’s a really different way of working.” SRI hopes to integrate a number of different AI programs including an email assistant, a scheduling assistant, a Web master assistant, and a space-planning assistant, into a sort of Microsoft Office-like suite of intelligent applications.

Specifically, the system will be made up of multiple artificial intelligence algorithms and systems covering such areas as reasoning and action, physical awareness, cyber awareness and multimodal dialogue or language. Agent-based software will come from CMU, while adaptive learning and natural language (understanding) algorithms are the responsibility of the SRI team.

Language is one of the toughest pieces of the puzzle. “One reason language is difficult is enormous the amount of common sense reasoning behind a sentence,” says Mr. Cheyer, “as well as social cues such as intonation. There is simply too much information.”

The new software systems will be developed and tested on a distributed network used by the 22 research establishments involved in the project. All participants will have the ability to access the software from a personal computer, PDA, or mobile phone.

Such software could, for example, examine an incoming telephone number and compare it to its user’s contact management software. It then might decide to reroute a call from the boss or family member to a cellular phone. It might even interrupt a call in progress, say, if the boss is on another line.

A software-based scheduling assistant might trawl the Internet to find the cheapest flight, book it and automatically cancel any meetings that occur on site during a business trip. If the tasks it carries out meet with the user’s approval, the software should evolve to carry out these tasks on its own.

It is a tall order, but DARPA hopes that research projects such as these will give rise to a next generation of military technology. And that usually leads to new consumer devices and products. Global positioning system (GPS) technology, the laser, and the Internet were all developed with agency funding. “Today’s software is brittle and stupid,” says Mr. Cheyer. “We want to change that.” But if CALO fails its yearly test, DARPA will not allow it to move up to the next level. If only that were the case for all software.

In 1991, his search for "extreme" organisms turned up a breakthrough that may revolutionize a branch of nanotechnology known as molecular manufacturing. In this nascent field of study, scientists assemble structures and materials using atoms and basic molecular forces as primary building blocks.

If molecular manufacturing delivers on its promise, tiny microchips could be created atom by atom--preserving Moore's law well into the next century. New medical devices, like nanobots that roam our blood stream and deliver drugs, could become a kind of backup immune system.

Little Progress

But nanotechnology has made far more progress in press releases and the pages of science fiction novels than it has as a viable science. Thus far, molecules have proven too small to manipulate with precision, even with atomic-force microscopes. For the most part, scientists have been able to do little more than prod, magnetize, or push them around a surface. This means that they can't really build anything; it's the difference between stacking up cement blocks and actually building a house. If molecular manufacturing were house construction work, we would still be learning to build the wooden frame.

And that is where Mr. Trent and his team at NASA's Ames Research Center in Mountain View, California, come in. The scientists have discovered a substance that can act as that much-needed frame to hold atoms in place and organize them into different structures. The discovery will, they say, make the manufacture of nano-scale devices like microprocessors, memory chips, or drug-delivery modules as easy as mixing a solution in a test tube. Best of all, this substance could enable such items to be produced in large quantities economically.

Mr. Trent's discovery hasn't gone unnoticed. The research center's Nanotechnology Lab, one of five regional nanotech research centers in the United States, is the largest recipient of the U.S. government's $700 million nanotech investment fund. Moreover, NASA is already in licensing discussions with at least one major biotech firm to use it.

The miracle substance is the by-product of a microbe that's survived the last 3.4 billion years in some of the world's harshest environments. It's survived the ice ages, the extinction of the dinosaurs, and now quite happily thrives in sulfuric hot springs, where temperatures reach 185 degrees Fahrenheit. This little microbe's resilience has helped scientists begin to overcome the obstacles to manufacturing nano-scale devices.

Fat Fingers

Until now, molecular manufacturing has been constrained by what scientists like to call the "fat fingers" problem. As anybody who has chased a grain of rice around with a pair of chopsticks can testify, it can be very difficult to pick up or place really tiny objects with relatively big tools.

There are, roughly speaking, three basic approaches to molecular manufacturing: binding molecules by mixing them with sticky molecules like fats (chemistry), binding them through genetic engineering (biology), or binding them using physical methods like bombarding them with electrons (physics). But because the scale involved in doing these manipulations is so small, many scientists have concluded that it would be best if nanotech devices organized themselves like chemicals, replicated themselves like living organisms, and functioned as predictably as machines. Overcoming these hurdles surely would require Mr. Trent's express focus on these problems, along with the integration of several different branches of science and a healthy dose of creativity.

Then, as often happens in scientific discovery, serendipity took over. Without even considering the problem, Mr. Trent and his team unearthed a protein called Heat Shock Protein 60 (HSP 60) while studying Sulfolobus shibatae, an extremophile microbe that lives in the sulphuric hot springs in Yellowstone National Park. This protein assembles itself into a ring structure called a chaperonin, which fortifies the cell membrane so the microbe can live in extreme conditions.

Closer inspection revealed other fascinating features. First, a chaperonin molecule has an empty core, like a stack of Lifesavers, which makes it an ideal manufacturing frame. But the diameter of the core is too narrow to hold enough nano-size particles for practical manufacturing purposes. Chad Paavola, a molecular biologist at NASA, genetically tweaked the protein to widen the "hole" and remove unwanted strands of protein. He then engineered the protein so that it would grow amino acids on the inside of the opening so that nanoparticles would stick there. Otherwise, nanoparticles could bounce out of the chaperonin opening.

Once they were able to increase the diameter of the hole, the molecule could be used to trap and contain nano-size substances like gold, magnetized particles (for storage devices), or toxins that target cancer cells. And chaperonins use little hooks around their outside edges to attach to one another and organize into rows and columns. This ability to self-organize is key: it solves the "fat fingers" problem. Furthermore, because Sulfolobus shibatae lives in warm environments, the chaperonin could be used in manufacturing processes that reach temperatures as high as the boiling point of water.

After getting acquainted with these microbes in the lab, NASA biologists employed a common biotech trick: they inserted foreign genetic code into the rapidly replicating Escherichia coli bacteria. In this case, they fooled the single-celled organism into producing industrial quantities of chaperonin. "Normally you have to separate out the bacteria and the protein in a costly process," says Mr. Trent. "But HSP 60 is heat resistant, so you can warm up the solution until you burn off the E. coli." And all that remains is very useful nano-scale scaffolding.

But the work didn't end there. NASA scientist Andrew McMillan discovered that the chaperonin could be morphed from its tubular shape into various other shapes, like a honeycomb, which could provide the lattice structure required to assemble a nano-size memory chip.

According to nanotechnology expert David Goldhaber-Gordon, assistant professor of physics at Stanford University, each individual piece of Mr. Trent's method has been done before. Panasonic and Sony, for example, have been working with a protein that binds to magnetized iron particles, which they hope will provide nano-scale storage mediums.

"However, what is unique is combining work such as protein design with self-assembly techniques to make a hybrid organic/inorganic substance," says Mr. Goldhaber-Gordon. "And it's the versatility of the chaperonin that's new here. Once you have the scaffolding, you can attach pretty much anything you want."

NASA's approach is promising because any particle can be mixed in with the chaperonins, which will provide the framing upon which other nano-scale devices can be built.

Whether it's the scary killer bugs in Michael Crichton's new techno-thriller Prey or the tiny medical devices that roam our bodies eating cancers in K. Eric Drexler's visionary book Engines of Creation, nanotechnology is billed as both the miracle and the nightmare science of our age. But so far it has made little progress as either. In fact, it will be a few years before we have even a nano-scale computer chip. What we are seeing now is the creation of the materials or building blocks that will perhaps make these things possible. Society will have to decide whether that's good or bad.

Niall McKay is a contributing editor to Red Herring. Write to letters@ redherring.com.

In the mid-'90s, Mr. O'Connor lobbied the government to build a wind farm in the shallow sand banks of Dublin bay, near the capital. But his superiors opposed the idea. So he quit his government job in 1997 and founded Airtricity.

The wind might be changing for Mr. O'Connor. The Irish government needs wind: the European Union has made it clear that Ireland needs to reduce its high levels of greenhouse gases, so the government is no longer in a position to give ideas like his the cold shoulder.

Early next year, Airtricity will begin construction on the Arklow Banks project: the same 200-wind turbine plant in the Irish Sea that Mr. O'Connor proposed earlier. It will be the largest offshore wind farm in the world, generating 520 megawatts--enough power for more than half a million homes. Its location, off the coast of Arklow in County Wicklow, is ideal because it's shallow, only 4 miles from the shore, and just some 20 miles from Dublin.

Airtricity has already built two wind farms and has planning permits for six others, each capable of generating 400 MW of power. These wind farms dwarf those of other wind-energy providers in the region. The company has already won 14,000 business customers and has sales in excess of $35.9 million per year. "This year the company will be profitable," says Mr. O'Connor.

But building Ireland's first private--and only--green utility company hasn't been easy: he needed to convince Dublin's business community that wind energy is not only good for the environment, but is a good moneymaker as well.

The economics of wind farms is straightforward. It costs approximately $1 million to install 1 MW of capacity. Electric power fetches about $0.035 per kilowatt, so a 1-MW tower can earn around $35 per hour, or $314,000 per year. So theoretically, such a tower can pay for its own construction within four years. But the wind doesn't blow all the time, so the yield rates in each region vary. Areas with more wind yield more power than their less-blustery counterparts do.

"On mainland Europe, the typical yield would be in the region of 25 percent," says Sheila Layden, managing director for Gaoithe Saor Teoranta, an engineering company that designs, builds, and manages wind farms in Ireland. So a 1-MW wind turbine actually produces only about 250 kW of power. "But here in Ireland," she adds, "the yield is about 35 percent on land." Therefore, wind farms in Ireland produce more electricity than those on the Continent.

Moreover, according to Ms. Layden, offshore wind farms typically have higher yields than their landlocked equivalents because there are no physical obstructions to seaborne wind currents. If the Arklow Banks yield runs as high as 40 percent, it will earn $125,900 per MW per year. But other costs must also be factored in: maintenance runs roughly $9,000 per year, and site rental is 2.5 percent of project revenue. All things considered, it will take about ten years for Airtricity to see a profit from the Arklow Banks. However, after the recuperation period, wind farms typically begin to yield significant profits, according to Andrew Ennis, an analyst with the Dublin-based financial services firm NCB Stockbrokers, who is compiling a report on the energy industry for the Irish government.

GAEL FORCE WIND

Wind energy isn't new to Ireland: records of windmills date back to the 12th century. And in the 19th century, more than 2,000 windmills dotted the landscape. In 2001, however, renewable energy sources, like hydroelectric and wind energy, accounted for only 2 percent (about 1 percent each) of the country's 4,800-MW power consumption.

Currently, Ireland is a very small player in the field. Denmark, the per-capita wind-energy capital of the world, has an installed capacity of 2,400 MW of wind-generated power; Ireland has a total installed capacity of only 125 MW. But if all the available wind power in Ireland were harnessed, 20 times that amount could be delivered, according to Sustainable Energy Ireland, a government-funded agency that researches and develops new energy efficiency policies. "We have enough wind energy available in the North Atlantic to provide power for the whole of Europe," says Mr. O'Connor.

A few years ago this would have merely sounded like the rhetoric of a fervent environmentalist. But Mr. O'Connor has proven wind power's profitability. After starting with just $449,000 in private seed capital, Airtricity's last share placement in January (albeit a private one) valued the company at $65.7 million.

Airtricity's success is due partly to Ireland's newly deregulated electricity industry, which in 2000, gave renewable-energy companies free rein to sell power directly to the consumer. The impetus for this policy change came from the EU. After many of the world's governments agreed to the environmental guidelines of the Kyoto Protocol, the EU pressured the Irish government to become less dependent on imported fuels.

Ireland's expected increase in energy consumption, a function of the country's well-publicized economic boom in the '90s, is likely to force the country to exceed the Kyoto limits on greenhouse-gas emissions by more than 30 percent prior to the protocol's 2010 deadline, according to a survey conducted by Ireland's Department of the Environment and Local Government. In fact, the country is on track to displace Luxembourg as the largest per capita emitter of greenhouse gases in Europe. That will mean heavy punitive fines. Currently, Ireland imports more than 86 percent of its fuel. "This is all good news for the renewable-energy sector," says Mr. Ennis.

That isn't to say Airtricity will have an easy time. The company is taking a massive risk to build its Arklow Banks project, which will be 13 times bigger than the world's largest offshore wind farm, in Middelgrunden, Denmark, which belts out 40 MW of electricity. The first phase of Arklow Banks, due to begin construction in early 2003, will generate 58 MW.

PRIME THE PUMP

Unlike such European neighbors as Great Britain, which is pumping $144 million into wind energy this year, the Irish government has done little to encourage renewable-energy suppliers. Rather than offer subsidies, Ireland has given 21 approved wind providers 15-year guaranteed fixed-purchasing contracts. These contracts guarantee a fixed income for the operator, but if energy prices increase, potential profits will drop. To make matters worse, the Irish government has also subsidized the cost and operating expense (up to one cent more per unit) of an ecologically unfriendly peat-fired plant.

Mr. O'Connor's long-term objective is to cover the costs of connecting Arklow Banks to Ireland's national electricity grid--a tab typically footed by the government--thus saving the government the cost of building and managing new power plants.

Whether or not he meets this objective, Mr. O'Connor says his project will proceed. Pessimistic visionary or Don Quixote, Mr. O'Connor probably fits somewhere in that category of impractical idealists who champion hopeless causes.

The application is one of several personal information management services offered by the Tokyo-based mobile application service provider 104.com. Another, used for tracking weight gain or loss, displays a pig.

Each morning thousands of Japanese women record their body temperature and enter it into software that, after a month or so of use, can calculate a woman's average temperature and determine if she is ovulating. Though 104.com will not say how many woman use its service, it claims to have the largest ovulation database in the world.

The company provides mobile content services for the cell phone providers NTT DoCoMo, KDDI, and J-Phone in Japan. "We're not interested in providing data services in other markets," says Arjen van Blokland, 104.com's vice president of international business development.

The underlying technology will not penetrate other regions unless cellular providers abroad buy the software. Then couples in the United States and Europe could enjoy foreplay with Koala bears and cell phones, too.

With the news that cybercrime is on the rise comes the realization that there's only so much an organization can do to protect itself. Security software can make a hacker's job more difficult, but there are no guarantees against intrusion and damage to a company's network systems. There is, however, insurance.

On the surface, insuring corporate networks against hacking is rather straightforward: insurers sell multimillion-dollar policies to cover damages in the event a company is hacked, basing their premiums on the security of the company's information and the policies it has in place. Yet, because of the relative newness of the crime and lack of hard data, cyberinsurance quickly becomes complicated when it comes to quantifying both the effectiveness of security measures and the cost of damages.

Cyberinsurance is a bit like employers' liability insurance, says Laura Rippy, CEO of Handango, an online seller of handheld computers and software: "You do everything you can to make sure that you never have to claim, but it's nice to feel that if worst comes to worst, you have all your bases covered."

SECRET SERVICE Handango, however, is one of only a few companies that have bothered to purchase a hacker insurance policy. Despite the increase in cybercrime, less than 10 percent of all e-commerce companies have such policies, according to an FBI official who asked not to be named. And those companies that do have insurance don't like to admit to it, because they fear it will make them seem vulnerable.

"Our clients fear that if the word gets out that they have cyberextortion insurance they will become a target," says Ty Sagalow, chief operating officer of the e-business risk solutions division at the American International Group (NYSE: AIG), an insurance company. This insurer and Lloyd's of London are two of the larger companies providing hacker insurance and are the only ones that go so far as to promise to produce ransoms in cyberextortion cases. However, neither company will say if it has ever had to do such a thing.

There are only a handful of insurance companies prepared to underwrite the crime. Policies can be hard to come by, because cybercrime is too new a phenomenon for the insurance statisticians, or actuaries, to assess risk. Actuaries calculate the likelihood of an insurable event taking place, like a death in the case of life insurance, by looking at the statistics and pricing premiums accordingly. Yet so far there is little reliable data available on how many companies get hacked and the cost of covering the resulting damage.

It's also difficult for insurance companies to assess the security of a company's digital information, making it harder to spot potential risks. Emily Freeman, practice leader of e-business risk solutions for the insurance broker Marsh, says the company uses two methods to gauge a computer system's security: "Applicants fill out an extensive form detailing their security systems, and we do electronic penetration testing." The latter entails using an off-the-shelf software tool that scans the applicant's Web site and systems for weaknesses. The average policy to cover such a setup would pay out up to $5 million in the case of an attack.

SECURITY CLEARANCE Insurers do more vigorous security assessments for policies that cover up to $25 million in the event of a break-in. In these instances, the insurance companies contract a security consultant to test the applicant's systems extensively. The American International Group, for example, has strategic relationships with the security consultancy Securify and the systems integrator Unisys (NYSE: UIS) to assess the security of a potential client's systems.

Another hurdle to cyberinsurance adoption is the fact that it's hard to assess its effectiveness, so most companies simply don't buy it, says Marci Glazer, an analyst with the market research firm Jupiter Research. But that may change, she says. The increasing number of high-profile break-ins and denial-of-service attacks have worked to undermine the public's confidence in the Internet. Certainly it will not take the litigious American public long to figure out that a company without the means to protect its customers' data is a company ripe to be sued.

Niall McKay is a freelance writer based in the San Francisco area. Write to letters@redherring.com.

Picture this: a tourist is on vacation, motoring down the Mall in London. To his left is St. James's Park; ahead is a large, stately building. He wonders what the building is, but he's too busy negotiating the perils of "wrong-way" driving to check his pocket guide of Britain.

Soon tourists won't have to risk a wreck to learn that building's name, because "augmented reality" (AR) headsets will be able to transform everyday street scenes into living guidebooks. AR headsets work with a visual-data display, like those projected inside the helmets of fighter pilots. One look at the stately London building would signal the headset to display a pointer and a label that reads "Buckingham Palace." A voice command would pull up another layer of data -- perhaps the fact that Buckingham is home to Queen Elizabeth II.

AR is a marriage of virtual reality and, well, reality. The notion was first popularized by William Gibson in his 1994 cyberpunk novel Virtual Light (Bantam Spectra). In the book, a bicycle messenger finds a pair of AR goggles and stumbles upon a plot by gangsters to completely remodel a futuristic San Francisco.

The real-life eyeglasses, first developed by HRL Laboratories in Malibu, California, during the mid-'90s, supplement the wearer's view of the physical world with a virtual data overlay. HRL Labs, now owned by Raytheon (NYSE: RTN.B), Boeing (NYSE: BA), and General Motors (NYSE: GM), came up with the concept for the Defense Advanced Research Projects Agency's military applications. Recently, HRL Labs spun off a venture capital incubator called X-Laboratories to commercialize the AR headsets and other such technology.

An AR headset uses tiny displays embedded in its glass lenses. Images are sent from a handheld computer -- say, a Palm -- by means of a radio frequency or a short-range wireless-networking protocol, like Bluetooth. The AR system also includes a global positioning system chip so the headset knows where it is, and a gyroscope so it knows in what direction its wearer is looking.

X-Laboratories develops and integrates the operating and database systems that provide the data overlay; the company plans to buy the headsets themselves from specialty providers and to use off-the-shelf components to build most of the remainder of each system.

But X-Laboratories isn't the only AR headset game in town, and guidebook glasses are only one potential use. Companies like Microvision (Nasdaq: MVIS) in Bothell, Washington, and Telesensory in Sunnyvale, California, are developing AR headsets and systems to assist surgeons while operating. The headsets can furnish everything from photos of similar operations to a patient's medical records.

Current AR technology, however, faces some glitches: headsets don't work well in sunlight, the displays use a lot of power, and the imaging needs to be sharper.

Researchers at the University of Washington claim to have solved many of these problems with a technology they've developed called retinal scanning display (RSD). It uses a laser to write an image on the retina at the back of the eye. The image is sharp, shows up in bright light, and consumes significantly less power than does projecting text onto an external lens. RSD systems will be commercialized by Microvision, which plans to launch displays this summer.

X-Laboratories' first application of its AR headset will be in the real estate market within a year. The company already has its first customer, an as-yet-unnamed construction firm that will issue headsets to its architects, engineers, and draftspeople so they can visualize new developments before breaking ground.

"This is one of six products we're working on currently," says Gerard Casale, CEO of X-Laboratories. "We'll develop the first product for the commercial market, because each headset will cost in the region of $10,000."

Mr. Casale says AR headsets will next be marketed to the construction company's real estate salespeople, to show buyers what their houses will look like in undeveloped lots.

By the time the sixth product is released, in three to five years, Mr. Casale says, X-Laboratories should have costs low enough to sell AR headsets in the consumer electronics market.

This article originally appeared on RedHerring.com as "Lab Rat: Virtual light." Write to letters@redherring.com.

Picture this: a tourist on vacation, motoring down the Mall in London. To his left is St. James's Park; ahead is a large, stately building. He wonders what it is, but he's far too busy negotiating the perils of wrong-side driving to check his pocket guide to Great Britain.

Soon such tourists won't have to risk a wreck to learn that building's name. "Augmented reality" headsets will transform everyday street scenes into living guidebooks. AR headsets work with a visual-data display like those projected inside the helmets of jet fighter pilots. One look at the stately London building would signal the headset to pull up a pointer and a label that says "Buckingham Palace." A voice command would pull up another layer of data, such as the fact that it's the home of Queen Elizabeth II.

Augmented reality is a marriage of virtual reality and, well, reality. The eyeglasses, first developed by Hughes Research Laboratories in Los Angeles in the mid-'90s, supplement the wearer's view of the real world with a virtual data overlay. Hughes Labs, now owned by Raytheon (NYSE: RTN.B), Boeing (NYSE: BA), and General Motors (NYSE: GM), came up with the concept for the Defense Advanced Research Projects Agency's military applications. Recently, Hughes Labs spun off a venture capital incubator called X-Laboratories to commercialize the AR headsets and other such technology.

AR YOU READY? An AR headset uses tiny displays embedded in its glass lenses. Images are sent from a handheld computer -- a Palm, say -- via a radio frequency or short-range wireless-networking protocol like Bluetooth. The AR system also includes a GPS chip so that the headset will know where it is, and a gyroscope so it knows which direction it's looking. X-Laboratories plans to buy the headsets from specialist providers; it can use off-the-shelf components for most of the rest of the system. What X-Labs brings to the table is the integration and development of the operating and database systems that provide the data overlay.

But X-Laboratories is not the only AR headset game in town, and guidebook glasses are only one potential use. Companies like Microvision (Nasdaq: MVIS) in Bothell, Washington, and Telesensory in Sunnyvale, California, are developing AR headsets and systems to assist surgeons in the operating room. The AR headsets can furnish everything from photos of similar operations to a patient's medical records.

There are some glitches in AR technology, however. Current AR headsets don't work very well in sunlight, and the displays soak up a lot of power; the imaging also needs work. Researchers at the University of Washington claim to have solved many of these problems with a technology they've developed called retinal scanning display (RSD). It uses a laser to write an image on the retina at the back of the eye. The image is sharp, shows up in bright light, and consumes significantly less power than projecting text onto an external lens. RSD systems will be commercialized by Microvision, which plans to launch displays this summer.

SURREAL ESTATE It was William Gibson who first popularized the notion of augmented reality in his 1994 cyberpunk novel Virtual Light. In it, a bicycle messenger finds a pair of AR goggles and stumbles on a plot by gangsters to remodel completely a futuristic San Francisco.

As ever, Mr. Gibson's futurist lens has served him well -- the first application of the X-Laboratories AR headset will be in the real estate market. The company already has its first customer, an as-yet-unnamed construction firm that will issue headsets to its architects, engineers, and draftspeople so that they can visualize new developments before they break ground.

"This is one of six products we're working on currently," says Gerard Casale, CEO of X-Laboratories. "We'll develop the first product for the commercial market because each headset will cost in the region of $10,000."

Mr. Casale says AR headsets will next be marketed to the construction company's real estate salespeople to show buyers what their houses will look like in green-field sites.

By the time the sixth product is released, in three to five years, Mr. Casale says, X-Laboratories should have costs low enough to sell AR headsets directly to the consumer electronics market.

Editor's note: This will be Niall McKay's last Lab Rat; the column will continue under Red Herring's science team. Mr. McKay, we're sure, will continue to prowl the halls of labs everywhere. He may be reached at www.niall.org.

To get this column sent to your inbox, subscribe to the e-newsletter.

Microsoft (Nasdaq: MSFT), long known for its desktop software, is showing a more playful side these days, with Xbox, the game platform that wowed game enthusiasts and technologists alike, and new software tools that will provide consumers with the ability to create their own games and animations.

But the Redmond giant isn't stopping there. It wants to make the game experience even more personal, which is why the graphics group at Microsoft Research is working on Face Mapping, software that lets computer users scan their heads in 3D using a $100 webcam. The tools include Expression, a technology that creates facial animation; Verbs, which can make an animated character walk, run, jump, or jive; and Adverbs, which gives characters emotional expression.

Of course, the ability to animate with emotional expression or scan a cranium is not new. Any Hollywood director with a large wad of cash, some expensive scanning equipment, and half a dozen highly paid animators can achieve such wonders.

What's new here? Microsoft is making the creation of realistic animation easy and cost-effective. Simply put, Microsoft is doing what Microsoft does best: bringing complex computer programming tasks to the less-technical user. What's more, if successful, such tools may do for animation and games what the digital camera has done for desktop video by making the creation of animation as easy and as popular as editing a home video.

HEAD SHOT To make the economics of game development more tangible, let's take a look at head scanning. To do this now, a developer needs a very complex piece of equipment that uses lasers to gauge the shape of the head and cameras to read the hair color and skin tone. When all of those data are captured, the computations are done on a large computer so that the texture and shape can be mapped onto a computer-simulated replica.

With Microsoft's Face Mapping, all the game user or developer does is have someone sit in front of a cheap webcam and have the camera record their facial image, which is then used to assemble a model of the face, utilizing 50 predefined parameters. It's not that much different from the way a police artist builds the face of a suspect with an IdentiKit package. Face Mapping simply matches the subject's chin to one of its predefined chins, then the nose, ears, and so on. The finished product, it has to be said, is about as pretty as a police sketch and makes most subjects actually look like they have a criminal record.

Why bother, then? Well, it appears that gamers have a desire to see themselves and their friends take part in the game they are playing. Other applications could include using a 3D representation of one's self during online meetings or instant messaging conversations, according to Microsoft Research graphics group team leader Michael Cohen.

Mr. Cohen's team is also developing animation software that provides novice game users with simple point-and-click tools for creating a realistic image.

WALK THIS WAY But the quest for realism doesn't end at the neck. Game developers want to give their characters more personality; for example, a happy walk or a wounded walk. That isn't so easy. Typically, animators get an actor and put him in a sensor-embedded suit. The sensors' data are fed into the computer; the animator then must map the motion to the animated character.

That's where Microsoft's Verbs and Adverbs come in handy. The actor's motion still has to be recorded, but with the tool, the mapping is automatically generated, thus dispensing with the need to have highly paid animators hand-code the animation.

Some game observers even believe it's possible that not too long from now, actors' motions could be sold by companies like Viewpoint (a firm that provides an A to Z of 3D objects) as prerecorded data objects. At that stage, enthusiastic amateurs will be able to build their own animations without ... well, having to animate. Once they have added the motion from Verbs, they can add the emotional expression of that motion from Adverbs.

No doubt game developers will be the first to take advantage of such tools, but it may not be long before enthusiastic amateurs follow suit and use them to create their own games and animations. So why is it that Microsoft, a company that, after all, has been focused on the business market, has taken an interest in games and animations? Well, it's just another take on Bill Gates's famous maxim: content is king. Only this time, the corporation is trying to get users to create their own content. Who knew playing games would become such a serious business?

Niall McKay is a contributing editor to Red Herring magazine and can be contacted at www.niall.org.

To get this column sent to your inbox, subscribe to the e-newsletter.

Several years ago, Dr. Sagman decided that it was time to stop treating the symptoms and start finding cures. After some research, he stumbled across buckyballs, also known as fullerenes (named after R. Buckminster Fuller, designer of the geodesic dome). These are tiny geodesic-dome-shaped carbon molecules that could revolutionize the drug delivery business. They are so tiny, in fact, it would take 1 million of them to make up the diameter of a human hair. Unroll them and you have a material not unlike miniature chicken wire. This property, together with their size, strength, and nontoxicity (they are a simple carbon molecule), makes it possible for scientists to treat them as miniature pincushions, sticking various drug compounds into the chicken wire-like structure and putting them into the human body for deployment.

There was just one problem: the science is still in its infancy, and that means university research scientists are doing most of the pioneering work.

Dr. Sagman saw his opportunity. He joined forces with Stephen Wilson, professor of chemistry at New York University, and started C Sixty, a company that raises venture capital and invests it in university research that furthers the modification of the fullerene molecule. It then takes the technology, helps commercialize it, and sells it to pharmaceutical companies. So far, the company has been able to sign up several prestigious universities: the University of California at Los Angeles, Columbia University, Rice University, Dartmouth College, and the University of Taiwan.

KILLER APPOnly a few years ago Dr. Sagman's pitch would have been a much tougher sell because buckyballs (which were discovered in 1985 by Dr. Richard Smalley of Rice University) were considered near the fringe of science in the emerging discipline of nanotechnology. But times have changed. Just last year it was discovered that coating the tiny structures in sugar or other water-soluble compounds enables them to be discharged with the rest of the waste in the human body through the renal system or kidneys. And the outlook for buckyballs is very bright.

"There is no doubt that the killer application for buckyballs and their by-product, carbon nanotubes [sheets of buckyballs rolled into a cylindrical shape], will be in the life sciences industry," says Deepak Srivestava, senior scientist at NASA's AMES Research Facility in Silicon Valley, California, who is working on nanotube-based biosensors.

More recently, companies like IBM (NYSE: IBM) and Xerox (NYSE: XRX) have been investigating the use of buckyballs as electronic components or for energy storage. Indeed, next month Mitsubishi will announce a new division, called Mitsubishi Corporation-Fullerenes, dedicated to manufacturing and exploiting the properties of buckyballs.

SINKING PRICES, RISING HOPESThree years ago, the cost of the molecules was prohibitive, but recently prices have dropped dramatically from about $600 per gram to about $30 per gram. That price is expected to fall again to about $10 per gram by 2002, according to Dr. Lon Wilson, chemistry professor at Rice University.

What's all this mean for potential treatments? Dr. Sagman believes that the first commercial application for buckyballs is likely to be for the treatment of HIV. Buckyballs fit perfectly into the socket in the HIV protease enzyme, which is responsible for the growth of the disease in the body.

"Think of it as a stopper fitting perfectly into a drain," says Dr. Wilson. "This is an extremely effective way to deliver the HIV drugs, and it's even effective against the many drug-resistant strains of HIV that are appearing today."

With HIV making a serious comeback and nearly 35 million people infected with the disease worldwide in 2000, we need new and stronger drugs and drug-delivery methods.

So far C Sixty has completed successful animal testing of the drug-delivery method and hopes that it will enter the Food and Drug Administration approval process as early as next year.

But HIV is not the only disease that can be treated with fullerene-based drugs. The carbon compound is also thought to be effective against cancer. Buckyballs can be, for example, filled with a radioactive heavy metal and used in radiation therapy safely because the radiation does not leak. Moreover, they are powerful antioxidants and can be used to fight tumors, says Dr. Sagman.

Perhaps Dr. Sagman will find his cure for cancer after all.

To get this column sent to your inbox, subscribe to the email newsletter.

Dr. Steven Goldsmith, a former Star Wars research scientist at Sandia National Laboratories, believes he has the solution. It involves using a team of autonomous software agents -- software routines that act without human intervention -- that roam the network looking for intruders, isolating systems that have been compromised, and spotting new viruses by their behavior rather than checking them against a database of old ones.

For Dr. Goldsmith, security is no academic exercise. He is responsible for finding a way to protect the computer systems at Sandia's nuclear weapons lab in New Mexico from both physical and virtual intrusion. So the artificial intelligence expert dreamt up a network of robot swarms communicating over a peer-to-peer network. Cyber-bots, or intelligent agents, would roam the virtual world, and nickel-sized robots would roam the physical world, making sure that nobody had run an unauthorized network node or patch cable. 'Both run on the same basic software,' says Dr. Goldsmith. 'They detect the state of the world around them, they organize search patterns, and they learn. They are just instantiated with different input information.'

RED TEAM TEST The project was so successful that Dr. Goldsmith invited Sandia's Red Team -- a special force of professional government-paid hackers -- to put the agents through their paces and see if, indeed, they could protect the lab's computer systems. The team's purpose in life is to break open and expose weak government computer systems before malicious hackers get to them.

The Red Team got two opportunities to crack the system. The first was a two-day attack; they failed to gain entry. The second was a more prolonged seven-day attack, and they eventually prevailed. 'I was surprised at how devious these guys were,' says Dr. Goldsmith. 'But we learned from the experience, and now our agents are a lot smarter.'

So smart, in fact, that Sandia decided to let Dr. Goldsmith take the core software technology and start up a security company called Primitus that will commercialize these software agents. The nickel-sized robots, meanwhile, have been kept in-house and given a new tour of duty as battlefield sensors and land mine excavators.

LITTLE BLACK BOX Primitus is now developing a new security product that will go into beta release next September. It will consist of a black box about the size of a modem that will run a Macintosh G4 chip. The agents will reside on the box and will be programmed to read all incoming and outgoing packets. In this way, not only will the agents be able to prevent a virus from getting into a computer, they will also be programmed to prevent valuable information from being accessed or sent to the wrong person. 'This product is designed for government, businesses, and electronic commerce companies that have valuable information and armies of laptop and home workers,' says Dr. Goldsmith.

However, the agent box will have other important functions, like the ability to detect and prevent prolonged intrusion attacks and encrypt messages. 'This system will recognize slow attacks that take place over a number of weeks or months and that are co”rdinated over a number of machines,' he says. 'Typically, that's how smart hackers work. They will map out the network using a series of flow-port scans over a long period of time.'

The agents will have both a long memory and the ability to communicate with other agents on the network and carry out pattern analyses on each attack.

Primitus will also be able to encrypt and decrypt messages sent from other computers using Primitus hardware, which raises the question of how it is different from a firewall. 'Well,' says Dr. Goldsmith, 'firewalls will not do this detailed, contents-based analysis. Primitus will examine every packet and co”perate with other agents to recognize larger global attack patterns.'

There seems to be a consensus among observers that there's a real need for security to fend off costly attacks. Last year, the denial-of-service attacks on Yahoo (Nasdaq: YHOO), CNN, and others cost about $100 million in lost revenues, and the so-called Love Letter virus last spring may have cost as much as $2.6 billion, according to Boston's The Yankee Group. We obviously need all the security we can get.

Niall McKay is a contributing editor to Red Herring magazine and can be contacted at www.niall.org.

To get this column sent to your inbox, subscribe to the email newsletter.

This column also appears in the March 6, 2001, issue of Red Herring magazine.

Michael Dertouzos does not pull punches when he talks about the future of desktop computing -- and for good reason. As the director of the Massachusetts Institute of Technology's Laboratory for Computer Science, he spends considerable time every day thinking and working on future desktop projects. His assessment is clear: the current computing infrastructure is inadequate. He says this stems from the fact that the average person gets about 20 emails a day and spends three minutes dealing with each one. More messages means less time to deal with them -- especially as email traffic increases tenfold over the next few years. (You do the math.)

It boils down to what he calls human servitude. "We have been taken hostage by $500 PCs that force us to feed them data, nurse them, and obey their stupid commands," Mr. Dertouzos says.

He's issued a call to arms asking people to kill the PC and move to a better computing architecture. To him, that's one that operates as easily as a car, collects and organizes data on the fly, and doesn't crash and trash six months of hard work.

With that mantra, Mr. Dertouzos and his colleagues at MIT embarked on Project Oxygen. Over the next four years, MIT, Philips Electronics (NYSE : PHG), Nokia (NYSE : NOK), Hewlett-Packard (NYSE : HWP), Acer America, Delta Electronics, and NTT (NYSE: NTT) will throw a total of $50 million and 250 research scientists at developing a new platform that uses a constellation of devices that hear, see, and respond to our every need.

The idea is to augment the keyboard and mouse with speech-comprehension and vision technology, so that inputting data is as easy as saying, "Computer, capture this meeting and send it to my boss." In essence, the goal is to make computing invisible to the end user. Mr. Dertouzos calls it "humancentric" computing. Although hardly a new idea, this is the first time these components -- now finished and available -- have been bundled into one complete system.

THE ANTI-PC

The Xerox Palo Alto Research Center has been working on the concept of ubiquitous computing since the '80s. Last year, IBM (NYSE : IBM) spent $180 million on its pervasive computing program code-named Planet Blue (see Lab Rat, October 19). Other research institutions are running similar programs. Carnegie Mellon University's Project Aura, the University of Washington's Portolano project, and the University of California at Berkeley's Endeavor Expedition have each received funding from the Defense Department's Defense Advanced Research Projects Agency to develop pervasive, ubiquitous, invisible, or humancentric computing architectures.

For Project Oxygen, MIT is working on three prototypes: a handheld device called the Handy 21, a workstation-like computer called the Enviro 21, and a networking architecture called Network 21. The Handy 21 will be similar to a cell phone and will include a microphone and speaker, a view screen, a small video camera, and antennae for broadband and narrowband communications. However, 90 percent of the device's circuitry will be digital, allowing the user to reprogram it to carry out different functions by downloading software from the network. The Enviro 21 will provide users with a powerful computing environment in the home and office and will be designed to interoperate with the Handy 21 and with actuators and sensors dispersed in its environment. Network 21 will bind these devices together.

In some respects, the tech industry is well on its way to implementing Mr. Dertouzos's vision: initiatives like Jini from Sun Microsystems (Nasdaq : SUNW) aim to provide a platform for this constellation of smart devices, and chip designers like Parthus (Nasdaq : PRTH) and ARM (Nasdaq : ARMHY) are building cell phone and handheld device chip sets with Java Virtual Machine interpreters to permit software downloads on the fly.

Still, Mr. Dertouzos's mantra, which he hammers home in his new book, The Unfinished Revolution (HarperBusiness, 2001), is that "information technology should help people more by doing less."

This all sounds great until you fail to get a signal on your cell phone. "When computers vanish from sight ... we'll know that the IT revolution is finished," Mr. Dertouzos says. "But let's face it, we've got a long way to go."

IBM (NYSE : IBM)'s Thomas J. Watson research facility is famous for pitting computers against humans and having the computers win. For example, the company's Deep Blue supercomputer trashed chess world champion Garry Kasparov at his own game. Now Big Blue itself sets its silicon wits against commodities traders.

I recently took the train from New York's Grand Central station through the snowy suburbs to the research facility in Westchester County, about 50 miles away. I thought I was going to report on IBM's research, but as it turned out, I ended up becoming part of the story. I engaged in a friendly battle of wits against IBM's intelligent agent technology to trade commodities. The results weren't pretty.

I don't think IBM research scientists Robert Baseman and Jeff Kephart deliberately set out to shame the media. The object of the exercise was to test their intelligent agent's buying and selling powers against live humans. (An intelligent agent is essentially a software routine that assists people and acts on their behalf.) The result was that the technology performed 8.8 percent better than the other humans in the experiment and 21.8 percent better than this reporter. The research is part of IBM's Experimental Economics project -- a relatively new area of research (popularized by Dr. Vernon Smith of the University of Arizona in 1979) that tests the largely theoretical field of economics in a controlled environment. In this case, IBM wanted to test intelligent agent behavior in an electronic double auction.

Here's how it works. Take between six and ten humans -- typically NYU students -- and set them against an equal number of intelligent agents. Both agents and humans are given a task either to buy or sell commodities (it can be any commodity) at a profit. Buyers are given the price they can resell the commodity, the number of units available, and the current market price. Sellers are given the cost of producing the item, the number of units available, and the current market price. The buyer's goal is to buy as much of the commodity at as low a price as possible, and the seller's goal is to sell as much of the commodity at as high a price as possible. The trials were split into nine three-minute trading periods.

The biggest challenge that Mr. Kephart and crew had to overcome wasn't how to get the agents to trade efficiently -- indeed, that technology has been around for some time -- but how to give humans an incentive to trade efficiently. If paid a flat fee, they'd treat it like a game, sometimes winning and sometimes losing. So they devised a scheme to pay them an appearance fee, about $25 to $50, and also, they organized the auction so that successful traders could walk away with between $125 to $150. Not a king's ransom, mind you, but more like two or three shifts at Starbucks. Nevertheless, the scheme seemed incentive enough to students for a couple hours of effort. I, however, may have been better off taking the shifts at Starbucks.

UNFAIR ADVANTAGE

Unfortunately, the little digital sods seemed to have three distinct advantages over biological technology such as ourselves. First, they don't make stupid mistakes. When I started the trial, I noticed that I had to stop trading after several minutes and sit the rest of the session out unless I wanted to sell at a loss, so I did, just to keep doing something.

Second, they are faster than humans. It seemed that most of the trading took place in the first ten seconds of each trial. My real problem was that I was hampered by the fact that I don't think and act in milliseconds.

Third, and most important, the agents have an algorithm that enables them to quickly cast a virtual eye over previous sales, calculate the mean strike price, and immediately bid that price. I, on the other hand, was just bidding what I thought I could get away with.

Why does IBM bother? Well, such electronic double auctions using intelligent agents have many applications today. EBay uses intelligent agents to allow customers to automate the bidding process, but these are less sophisticated than the ones used in the trial. Indeed, experimental economics is just one way in which these electronic double auctions can be tested before they are foisted on the unsuspecting public.

In Los Angeles, the California Institute of Technology has used experimental economics to test the agent trading of carbon emissions. A factory, for example, that purchased a pollution emissions license could auction off its surplus while the plant is off-line, or it could buy extra surplus during a busy production time. Such markets can be used for more complex auctions where money is not the most important factor.

NASA'S AGENTS

The National Aeronautics and Space Administration tests electronic double auctions to allocate carrying capacity in the space shuttle. The Federal Communications Commission has been considering using such technology for the allocation of third-generation (3G) spectrum licenses for the cellular communications market.

However, the testing process is almost as important as the creation of the technology, according to Mr. Kephart. "We must be very careful about the introduction of agent technology," he says. "Agents still behave in a way that we don't completely understand."

For example, in an all-agent auction, prices tend to rise, reach a peak, and then suddenly dip dramatically before the same process begins again. Mr. Kephart and his colleagues must answer the question: does this behavior differ from the way humans trade?

Still, to the converted (like Mr. Kephart), intelligent agents eventually will transform our world, which means they may trade information, gather information, translate information, and perform all sorts of negotiations for us in the future. Given my performance, I'd be well-served using the technology to transform my life. Then again, that's what I thought when I opened an ETrade account -- but that's another story.

During Bill Clinton's third and final visit to Ireland as President of the United States in December, he told the Irish people that peace in their country depended on building upon its recent economic progress.

It seemed fitting, then, that he delivered his speech at the Guinness Storehouse, a newly refurbished section of the 300-year-old Guinness brewery in Dublin that will soon become home to the Massachusetts Institute of Technology's Media Lab Europe.

Media Lab Europe is being established by technology visionary Nicholas Negroponte with financial aid from the Irish government and is part of the government's $670 million push to bring Ireland to the next stage of its economic development.

Indeed, so far, the country's economic gamble on the high-tech industry has paid off handsomely. Since 1994, it has reduced unemployment from 20 percent to less than 6 percent; attracted nearly 600 U.S. companies to its shores; and spawned hundreds of indigenous high-tech companies, about a dozen of which are publicly traded on U.S. stock exchanges, such as Iona Technologies (Nasdaq : IONA), Trintech (Nasdaq : TTPA).

But where will Ireland go from here? It no longer offers a low-wage, highly educated work force. Salaries have crept up and now exceed the European average, companies can no longer find suitably qualified employees, and a three-bedroom house in Dublin now costs about the same as one in the Bay Area.

"We've got to move up the value chain," says Conor O'Carroll, senior policy analyst with Forfás, the Irish government's economic policy division.

So Forfás has proposed a plan to bring original scientific research to Ireland. "We already have plenty of product development that has helped create new Irish high-tech companies, but what we need to do now is build the nation's intellectual property," says Mr. O'Carroll.

SUPER MODELS

To achieve this, Forfás looked to the United States' National Science Foundation and the Defense Advanced Research Projects Agency as models. Both organizations have a good track record of funding long-term research that has benefited economic growth.

"What we need is the left-brain kind of thinking," says Mr. O'Carroll. "The type of research that looks a great deal further than developing next year's products."

"What we are looking for here are the people who can create and lead projects that will lead to new scientific discoveries," says Mr. O'Carroll.

NOBEL CANDIDATES WANTED

Actually, Forfás is looking for the sort of people who are most likely to be nominated for a Nobel Prize. So it is offering a five-year, $5 million research budget, plus a six-figure salary, to anybody with a track record and a really good idea. So far, it has had more than 80 serious responses from scientists in the United States, Germany, India, Russia, Israel, and Slovakia.

From the applicants, it will choose ten winners who will be attached to an Irish university but will not have to teach. That way the researchers can use the university's resources and recruit the cream of the post-graduate student crop. Forfás has not chosen the winners yet, so it will not discuss any of the projects, but about 50 percent will be in the life sciences and the rest from high-tech sectors.

Still, it may take 20 years to assess whether Forfás's strategy is successful. In the meantime, MIT's Media Lab Europe should play its part by putting Ireland on the scientific research map.