Bob Metcalfe, going with the flow
ON A FRIGID February morning, Bob Metcalfe, the inventor of Ethernet technology, stands in his barn on a picturesque Maine hilltop amidst a couple of dozen rare and pregnant Cotswold sheep. He's just finished recounting that among his flock are some sheep that, until recently, were found nowhere else on earth but some Scottish Hebridean island. The gene pool there has remained sequestered for so long that "genetically, seen under a microscope, these sheep don't even identify as sheep." Then he pauses and says, "Well, you've got to admit it, that's pretty Darwinian...."
Metcalfe presides over Kelmscott Farm (its website is www.kelmscott.org) with his wife, Robyn -- a dedicated preservationist of rare and endangered livestock breeds. Recently he's spent much more time on the farm near Camden, Maine, than in his Boston town house. The inventor-turned-technology-pundit wants to spend more time with his family, and he can write his weekly newspaper column anywhere -- so why not here? (Full disclosure: Metcalfe is vice president of technology for International Data Group -- CIO and Darwin magazine's parent company.)
The farm, a nonprofit venture, is open year-round to the public. But on the day of my visit, with a tart wind whipping the snow horizontal, nobody else swings by. The lanolin-rich coats of the sheep are frozen stiff.
The house itself is a Sears and Roebuck prefab, ordered from the venerable catalog (precursor to Amazon.com et al.) in the early years of the recently lapsed century. Connected to a smaller, older barn than the one the sheep inhabit, it rambles comfortably and in a decidedly nonprefab way -- with lateral jogs, narrow, twisting stairs and adjunct alcoves -- as though plotted out by meandering livestock looking for ever-tastier bits of roughage.
Metcalfe, who appears to be somewhat obsessed with the obscure but all-important particulars of various kinds of insurance coverage, relates why barns catch fire so often -- and, thus, why insurance premiums are quite a bit higher for houses that adjoin them: "If hay is harvested wet, the moisture ferments inside the bales and creates a hazard of combustion. But when you harvest it dry, that risk goes away. Hence," he says, "the old saying, Make hay while the sun shines. That's where that comes from."
In this and other eclectic anecdotes and scraps of arcana, Bob Metcalfe gives the impression of someone who eagerly vacuums up new data of almost any kind. It all seems so interesting to him.
PERHAPS EVEN MORE DARWINIAN than the sheep, goats and pigs of Kelmscott Farm is the massive round of technology mutations that Metcalfe helped set in motion. In fact, you could argue that if it weren't for Metcalfe -- and a merry band of equally gifted compatriots -- you probably wouldn't be reading this magazine.
Ironically, Metcalfe, who airs his libertarian-conservative viewpoints both loudly and proudly, is one of the charter members of the revolutionary cadre who gathered in Palo Alto, Calif., in the early 1970s to overthrow the established computing order. "It was a crusade," he emphasizes vehemently, looking as though he might pound his fist on something. He'd pound it on a centralized computing model, if it were handy.
Metcalfe's baby, Ethernet, is one of the seminal enablers of letting the information genie out of its well-guarded bottle. As a networking technology, Ethernet let information roam. Without being able to roam, it would have remained in the iron-fisted grip of those custodial eminences in the information services function, unavailable to you and your kind except through a process of laborious petition -- during which you'd be told to bloody well wait your turn (frequently in the hope that you'd eventually give up waiting and just go away).
As the brainiac denizens of Xerox Corp.'s Palo Alto Research Center (PARC), Metcalfe and his antiestablishment cohorts were inventing personal computing -- doing the basic spadework that would trigger the broad evolutionary trend to distribute computing power: Out of the white rooms and into the streets! If, back then, IBM's model was to confine all authority over, and access to, computerized information within the IT priesthood, what Metcalfe and the others at PARC were doing was laboring to set it free, to democratize access and governance. After all, the information belonged to the enterprise, not to the archdruids who ran the mysterious machines (machines so persnickety that they needed the pampered, air-conditioned sanctuary of special rooms that no mere mortal dared enter).
The creation of the personal computer was arguably the Manhattan Project of IT. What the "controlled" triggering of nuclear fission had done for hitherto stable matter, the computer scientists of PARC would do for ones and zeros: make them grow, gallop, frolic, rampantly multiply -- and wreak havoc on the usual way of thinking about and applying technology.
Now the crusade is over. Metcalfe and his colleagues at PARC won hands down. IBM has, of course, long since capitulated and broadened its vision and portfolio. Mainframes know their place and keep it. (Modern mainframes are for the most part fat, happy bins of stored information waiting to be summoned by whomever needs it.) Metcalfe went on to found 3Com Corp., the company that first commercialized Ethernet (3Com is now a $6 billion networking powerhouse). Data flows everywhere like water, swift and mostly unimpeded, across fast Ethernet cabling. Nobody even thinks twice about it -- or about the fact that a key religious belief of the early days of computing was that the users couldn't be trusted with the stuff.
But Metcalfe never doubted the outcome. "It was inevitable," he says. "It couldn't have been stopped." His own contribution to victory was the pipe that would carry the freed information away from its white-room Alcatraz. And it isn't ever coming back.
THE FLOW OF INFORMATION, in lively gusts and eddies, is more or less continuous around Metcalfe. This truth dovetails neatly with his work. In fact, everything satisfying about life seems relatable to the burgeoning happy outcomes of simply letting information roam unfettered. Free-range information will inevitably produce benefits that are both logical and predictable as well as serendipitous and unexpected.
For instance, in high school Metcalfe loved the classic all-girl Motown group the Shirelles. Despairing of ever again dropping a diamond needle on their stirring rendition of "Will You Still Love Me Tomorrow?" he recently stumbled on their greatest hits 33 rpm vinyl LP on Amazon.com's auction site. "There was no hope that I was ever going to gget this album. In fact, I wasn't even looking for [it]," he says. "I just happened to be futzing around in Amazon.com, and it was like, 'Oh, there are people selling albums!' And now I have it. It cost me $15."
This matter of online auctions is one of the interesting trends Metcalfe thinks businesses should take note of: customers creating their own pricing on the fly. "We learned in elementary economics about supply and demand. In reality, it is very hard to find supply and demand curves that are dynamic. These auctions let companies optimize along supply and demand curves on a minute-by-minute basis. And that is good for economics. It is good for businesses, it is good for everybody. It is a win-win-win-win-win."
In its fluidly optimized state, information can be subversive. Customers can instantly interact with and choose from among a wider array of competing alternatives. And justice thereby prevails. "Many businesses rely on being monopolies within narrow geographical or topical areas, because information flow, in the past, [was so limited]." But the gracious protection conferred by those geographical boundaries is vanishing. "It's hard to be a monopoly anymore because your customers have access to [information] that lies outside of your little monopoly. That is going to be very disruptive."
He obviously relishes the idea that a system of broadly distributed information can right competitive wrongs. Information has a curative, purifying role to play in the world, in life, in business, shedding a cleansing light into corners where shady dealings might otherwise erupt. "The fact that communication is increasing means good news for ethics," he says, "because you can't keep secrets like you [might have done] in the past. If you ship a junk product, the community is going to be on you like white on snow."
METCALFE HAS A LAW OF computing named after him. Metcalfe's Law states, basically, that the asset value of a computer network increases exponentially as each new node -- or individual user -- is added to it. Because every new user brings along a wealth of linkages and resources, the fund of networked value grows to be far richer than the mere sum of its parts. Metcalfe's Law is at the heart of the Internet's power -- it's the place where (despite occasional courageous, and mostly unavailing, efforts to charge a fee for it) information really wants to be free. And it is the most inspiring model of a decentralized communications network that is likely ever to exist outside of insect pheromone messages in nature.
Metcalfe first became acquainted with the Internet when it was a cozy little loop connecting relatively small numbers of government and academic research outposts. Its users were a closed, purposeful community that invented, out of necessity, the early protocols for online information sharing. Now, he believes, with the community burgeoning toward the billions of users, we are on the brink of having the network connections themselves become despecialized. Already, pagers and cell phones and Palm devices are connected to the Internet.
"PCs are passé," he says, getting at one of the canonical beliefs of the informational new age: intelligent, networked everything -- from refrigerators to blenders to garage-door openers to contact lenses.
At the recent high-tech industry event known as DEMO (a gig where venture capitalists come -- checkbooks open -- to make deals for promising technologies), one of the more revolutionary concepts unveiled was "chips that some geniuses have designed that allow you to run Ethernet on the power line in your house. This is a real step toward ubiquity, being able to put the Internett wherever anything gets plugged in."
But even with all of these great leaps forward so tangibly at hand, Metcalfe sees fundamental problems still waiting to be solved. One thing he would like to tackle next -- whenever next turns out to be -- involves software design. "I have a pipe dream of returning to research," he confesses. "There has been very little software progress in the last 25 years. All the progress we have enjoyed has been hardware progress."
He longs to delve into something he calls "anticiparallelism" -- the ability of software to anticipate what the user will want to do next, and then to do it in the background, invisibly, concurrent with, or parallel to, whatever task is visible to the user.
"One of my prime irritants is when I am looking at a webpage and I click on Back to go to the previous webpage, I have to sit there and wait. For what? Three seconds, two seconds, four seconds? I don't understand why it isn't back in a hundredth of a second. Because Back is a pretty frequent [action], the software should know that it is very likely that I am going to go back. It should anticipate and in parallel be getting ready for me to say Back. Then when I [do], there should be no delay at all.
"Generalize this notion to other less-obvious examples, where the software knows what is likely to be done in the future, or needs doing, and pursues it contingently in parallel, anticipating what is going to be done. All that unused computer power is sitting there, while you are reading your e-mail; there are a million things it could be doing."
But, reckons Metcalfe, that would require a new computer programming language that allows for reasonably reliable predictive allocations of processing power. To undertake such a project would mean more than retiring to his backstairs lair and hacking out a solution. It would take something on the order of a research institute with a team of zealous engineers parsing the project into manageable chunks. In Thomas Edison's Menlo Park, N.J., lab 120-plus years ago, dozens of assistants specialized in various subsets of raw materials, ransacking all of nature looking for the perfect lightbulb filament. To reform the design of software, Metcalfe would have to preside over something comparably intense. Whether he is interested enough in the challenge to do so remains to be seen.
THE LICENSE PLATE ON METCALFE'S Volvo station wagon reads ETHER -- though in the ancient garage of his Maine redoubt (two side-by-side spaces emptying out through one width of door), it is pinned so far into a post-blocked corner that getting it out and on its way requires the patience and finesse of a bomb defuser. In its tortuous extrication, the car is far from ethereal. Metcalfe is clearly amused, challenged and frustrated by the painstaking task (which dwarfs by vast time factors the longest imaginable wait for the previous webpage after hitting the Back button).
But he executes the needed maneuvers with disciplined skill, backing slowly among the obstructing timbers and finally out into the snow-blown sunlight -- all without once reversing direction. It becomes an interesting accomplishment. In the climate of natural directness at Kelmscott Farm, surrounded everywhere by animals whose survival as species is threatened, Metcalfe can appreciate the simple problem of getting the damn car out of the garage.
He readily cites automobiles as belonging among the technologies he does not understand. "I don't even open the hood anymore. For a while, like most males, I would open the hood and look at it as if I could possibly fix something under there. But now if it's not functioning properly, I get on the phone and call a pro. I do not know how to fix cars. I can change a tire."
Last summer, out in his lobster boat on Penobscot Bay, he "mixed up port and starboard," as he puts it. "I was running out of the empty tank instead of the full tank. I had fuel. All I had to do was switch tanks. But when you run out of fuel with a diesel, you have to bleed the air out.
"For somebody who does not open the hood of a car, this is a major problem. Here I am, moored to a buoy at dark, in the middle of the Penobscot Bay, on a cell phone whose battery is running down, finding a diesel expert who taught me over the phone -- in two minutes flat -- how to bleed [the fuel lines]. That is, take the injectors off each of the eight cylinders, crank the engine to start pumping the fuel, wait until the air gets out and [the injector] sprays diesel fuel. My kids were with me, watching -- they had diesel fuel all over their faces. Then tightening down the things. This is a major accomplishment. I am so proud of it! I am now the kind of guy who knows how to bleed a diesel engine!"
This last, newest attribute is asserted with a degree of pride perhaps equal to that he might take in being the kind of guy who knows how to invent a technology or start up a company. Or how to write an incendiary column or be a pundit bold enough to predict -- incorrectly, as it turned out -- the complete and utter collapse of the Internet because of overwhelming traffic on an underdeveloped infrastructure.
But he's never lacked the courage of his strong convictions, nor the zeal required to deal with the consequences (he literally ate his words in public -- his column liquefied in a blender -- when the Net-collapse prediction didn't pan out). And he has a way of choosing pursuits that require more than a bit of a stretch.
When he left PARC in 1979 to start up 3Com, he had just met his future wife Robyn. She counseled him not to leave Xerox, not to do this madcap thing of starting up a company (back then, the launching of upstart ventures was not nearly the bland clich it is now). "If I had listened to her, I would still be working at Xerox, because there was no rational way to explain why I should start 3Com -- except that I wanted to."
He continues to admire various forms of irrational gumption. He once sold Ethernet to a banker who had decided to invest in the bleeding-edge technology well ahead of any obvious need the bank had. "He decided on his own, and then I fed his enthusiasm," says Metcalfe. The banker had concluded that his career fortunes would "do better at the bank if he was aggressive on technology adoption. Years later he got in touch with me. He was now the head of the bank, [and] he remembered the meeting we had where he decided to take this risk.
"But then there are other examples where people decided to go too soon, got their faces eaten and lost their jobs. These are not obvious decisions. They are complicated. I have seen all the cases. I have sold people stuff they should not have bought. I didn't do it knowingly, but then I watched them crash and burn. And it was not really a risk they should have been taking, in retrospect."
In the life of every company there arise these critical moments of choice. In the early days of 3Com, when a newly hired senior manager invested in an advanced manufacturing planning system, Metcalfe plainly thought the guy was nuts. "I opposed the decision, I want you to know."
The system, he says, had "capabilities that were several orders of magnitude beyond what our little company needed to manage its manufacturing. But the company was growing at 300 percent a year, and [eventually] we reached a stretch where we could not have operated without this system. Our competitors had not made that forward investment. So our products were cheaper, reliable, delivered on time, at a cost we could make a profit at. And having that little systtem -- and we were experienced using it by the time we actually needed it -- we knocked out three or four competitors."
It's almost never easy to identify situations when too soon is just right. "Companies that make the right decisions flourish, and the ones that make the wrong decisions die."
Which is an increasingly harsh reality. And, unfortunately, there's no biodiversity farm dedicated to preserving the hapless victims of errant business decisions -- though Metcalfe would likely regard Cotswold sheep as more deserving of such protection.