This article is from the March 20, 2001, issue of Red Herring magazine.

The Dutch, more than most, understand the perils of water. Nearly 50 percent of their country is below sea level, and without its labyrinth of dikes and drainage systems much of the Netherlands would disappear into the unforgiving North Sea. The Dutch also understand the power of water. The Netherlands' ability to control encroaching bodies of water elevated the country to global maritime dominance in the 17th century. Now, one Dutch startup hopes to exploit that same liquid asset to bring the Netherlands into a new golden age.

This time around, Dutch entrepreneurs aren't reclaiming land or diverting rivers. Instead, they believe that they can use existing infrastructure and a highly proprietary data-compression technique to deliver Internet traffic through the one remaining "network" that communications providers have yet to utilize -- water pipes. "In most developed countries, more people have plumbing than have telephones," explains Dennis de Boer, the cofounder of WaterNet, which grew out of research at the Dutch Retrograde Internet Project (DRIP). "After more than five years in the lab, we're ready to make this technology commercially available. It will be cheaper and capable of carrying more data than any existing network." The first fully functional network, or "WaterNet," isn't expected until mid-2002. Mr. de Boer and his team are confident, however, that all current conduits for Internet traffic -- copper wire, cable, satellite, wireless, and even fiber optics -- will approach capacity in the near future.

Though some might dispute the prospect of imminent limitations, demand for bandwidth is clearly on the rise. A recent report by research firm IDC indicates that while many communications providers still have unused bandwidth in their networks, bandwidth trading exchanges are now facilitating the sale of most excess capacity. And the executives of most leading bandwidth providers seem to agree with Mr. de Boer's sentiment that capacity will always be in demand. For example, Global Crossing's president and chief operating officer, Gary Cohen, recently told an audience at the National Communications Forum 2000, "There may be small pockets of the planet that will have more bandwidth than they need, but for the providers that plan to the demands of their customers, no amount of bandwidth is excessive."

According to Mr. de Boer, "As exchanges become more efficient and bandwidth becomes more scarce, the DRIP water solution will just be coming in to its own." In fact, tests are already under way in the Belgian cities of LiŠge, Nieuwpoort, and Verviers, and the fledgling company is rumored to be striking deals with two Dutch water authorities and at least one U.S.-based telecommunications provider. Mr. de Boer hopes these tests will pave WaterNet's way into a global broadband market which, according to Comsys, a U.K.-based telecommunications consultancy, will be worth $580 billion by 2010.

In the meantime, despite an effort to avoid publicity, Mr. de Boer has attracted inevitable attention as he begins to seek private funding for DRIP's commercial rollout. After raising $2 million in seed funding from individuals in January 1999, Mr. de Boer is now rumored to be raising $240 million in a first round led by New Worlds Associates, a New Orleans-based venture capital firm.

Those who have seen the blueprint for the new network express mixed opinions. Noted venture capitalist "Surreal" J. Neil Weintraut was approached by Mr. de Boer for funding last November. Though unwilling to discuss details of the meeting, Mr. Weintraut's guarded comments reveal a certain degree of skepticism: "I'm not saying they can't pull it off. I'm no water expert. But sending Internet data through water pipes? The regulatory issues concern me. The environmental issues concern me. We passed on the investment."

But according to Juan Man of The Man Group, a research consultancy in San Francisco, the water-based bandwidth solution could revolutionize the market for broadband content. "This could be the Holy Grail for providers of rich data over the Web," Mr. Man says. "I've seen a demo and I'm not giving anything away by telling you that we downloaded a feature-length Dutch adult film in roughly five minutes with zero packet loss." Mr. Man added that this feat was easily accomplished by attaching a special nozzle to a faucet in the kitchenette of his offices.

AQUAMAN Mr. de Boer, a lifelong academic, knows his water. As a young graduate student in the early '80s, he developed an affinity for water when he worked on the final stages of the Netherland's Delta Project -- a massive program that included the construction of a 3,200-meter storm-surge barrier along the Eastern Scheldt basin and the creation of inland lakes. When the project reached completion in 1986, Mr. de Boer -- then loosely affiliated with the Eindhoven University of Technology -- began nearly ten years of independent research on the properties of water.

At the end of that period, Mr. de Boer published his findings in a two-volume series, The Mystery of Water: Properties, Uses, and Potential (Utrecht Press, 1996). Within the academic research community, he recalls, "I was treated like a visionary by some; a lunatic by most."

According to Milton MacGregor, editor in chief of Water Today, an industry trade journal, "In the sedate world of advanced water research, de Boer's book was considered a startling document, a real wake-up call." Though many researchers chided Mr. de Boer for his audacious claims (for example, Mr. de Boer devotes two chapters to his thinly supported nanotechnology findings, suggesting that by shuffling atoms he would literally be able to turn water into wine within the next 15 years), others believed his core research about water and the Internet was groundbreaking.

Perhaps the most significant result of the book's controversial acclaim was that it brought Mr. de Boer's work to the attention of an unnamed private benefactor -- reportedly a Dutch shipping magnate. This private benefactor funded the initial research that would ultimately lead to the WaterNet project.

With ample resources, Mr. de Boer was also able to assemble a small group of like-minded researchers. Most notably, Mr. de Boer attracted Guy Spofford, an expert in hydrodynamic modeling who previously served at Britain's Defence Evaluation and Research Agency, and Henri Savard from IBM's Next Generation Internet projects group. The team, consisting of eight researchers, has been cloistered in a research facility outside Eindhoven since late 1996 and has now been joined by nearly 60 other employees for the Belgian trials.

RUNNY NOZZLE Based on Mr. de Boer's thumbnail descriptions and according to sources close to the project, here's how a WaterNet might function: the network architecture is similar to that of any standard telephone network. However, instead of central offices (which act as important nodes in a telecom network), the WaterNet will use the existing network of municipal water authorities as information hubs. By installing a "highly cost-effective, router-like device," says Mr. de Boer, water utilities will be transformed into communications providers.

At the consumer end, rather than a telephone jack, a simple device called the client-side nozzle (CSN) will serve as the connection to the Internet. Reportedly, the CSN will attach to any standard water faucet and will provide a connection to the Internet for as long as the faucet is on. Observers point out that conservationists needn't fear -- the CSN is water-tight and in no way increases water usage.

When asked about sending information "upstream" -- back out across the Internet -- Mr. de Boer is evasive, yet confident. For instance, when challenged with the observation that the Internet is a global network and that not all water "networks" are connected, he suggests the WaterNet "will benefit from the very technologies with which it competes." Sources familiar with the project believe disparate WaterNets will be stitched together with wireless technologies -- most likely satellites.

Though adroitly vague on DRIP's technical specifics, Mr. de Boer let slip the following claim: "If we can reach critical mass with the first several WaterNets, the cost to communications providers will ultimately be cut in half, and free bandwidth will become a reality for consumers." For example, he says, the current $19.95 Internet access model in the United States will be eradicated, with access speeds easily eclipsing cable and DSL services. In some cases, data delivery could exceed T3 speeds.

UP THE CREEK Of course, projects of this scope breed suspicion -- and detractors. "Yeah...and I'm the Pope," scoffs Phyllis Coles, a water management specialist at the U.S. Department of Energy. She believes that even if the technology did exist, the regulatory hurdles would be overwhelming.

And according to Truman Schuman, a project manager at The Water Tank, a Seattle-based think tank devoted to water preservation issues, environmentalists will surely move to block the deployment of a network that uses drinking water as a "pack mule" for Internet traffic. "We don't have any evidence of an environmental threat yet," he remarks, "but we're going to be all over these guys like a cheap suit to ensure that they adhere to the law."

But back in the outskirts of Eindhoven, which give way to the bucolic landscapes made famous by the paintings of Jacob van Ruisdael, Mr. de Boer is as serene as the surrounding countryside. As a man who has faced skepticism -- and even mockery -- before, he remains impervious to it. He points out that the Belgian trials are ahead of schedule and that New Worlds Associates has entered into a tentative agreement to lead the next round of financing. Clarence Madison, the venture firm's managing partner, would not comment on the pending investment other than to say, presumably of its potential returns, "This investment could be one tall glass of water."

Indeed, but it could also be as arid as the Sahara desert. Until Mr. de Boer can deliver the WaterNet on a grand scale, his detractors will continue to rant. And should his endeavor fail, he may well be remembered as a blindly ambitious researcher with a little too much water on the brain.

WATERNET AT A GLANCECHAIRMAN AND CEO Dennis de BoerLOCATION Eindhoven, the NetherlandsURL www.dutchwater.comOWNERSHIP PrivateFOUNDED 1996EMPLOYEES 68PRODUCT High-speed data network which uses existing waterways and a highly proprietary data-compression techniquePARTNERS Two undisclosed Dutch water utilitiesCOMPETITORS Most major communications service providersREVENUE LAST 12 MONTHS PrerevenueTHE HERRING TAKE Not only does the company's ambitious project have technical hurdles to overcome, but it must also meet the stringent requirements of regulators and environmentalists. If the beta network is not flawless, the company may face insurmountable challenges.

Former Red Herring editor Andrew P. Madden is a San Francisco-based freelance writer. Write to letters@redherring.com.