Web Masters Episode #55: Bob Metcalfe


3Com , download 3Com :: Vector Logos, Brand logo, Company logo

Bob Metcalfe:

My father was a driver scope technician, and he had lots of electrical engineering type doodads in his little shop in the basement. And he and I built a train set on a 4×8 piece of plywood. We did that with switches and relays and neon lights and so on. In eighth grade, in order to do a project for my eighth grade science teacher, Mr. Gosher, I took those same switches and lights and Masonite, and I built a computer, or I built what he called a computer. It’s a box about this big, like a microwave oven, and it would add any number between one, two, and three to any other number between one, two, and three, and display the result by turning on a light, two, three, four, five, or six. And the grade I got on this project, I got an A+++H superior, was my grade on this project. And it was in that moment I decided to be a computer engineer.

Aaron Dinin:

Trust me when I tell you, we all owe Mr. Gosher, the science teacher from the story. A huge thanks for giving his student such a good grade and encouraging him to become an engineer. That student was Bob Metcalfe, and he’d go on to do some pretty important things in history.

Bob Metcalfe:

I’m famous for having invented Ethernet and founded a company called 3Com Corporation.

Aaron Dinin:

Yup. Invented Ethernet, as in the backbone of modern computer networking, that was created by Bob Metcalfe. And if the name is somewhat familiar to you but you’re not entirely sure why, it’s probably because you’ve heard the phrase Metcalfe’s law. Sure you might not know what it is without Googling, but you’ve definitely heard it before. If you don’t know what Metcalfe’s law is, don’t worry, we’ll get to that in a few minutes. All you need to know right now is when you’ve got a law named after you, well, that’s usually a pretty good thing. At least it is in this case. Are you ready to hear this story? Let’s get dialed in.

[INTRO]

Aaron Dinin:

Welcome to Web Masters, I’m your host, Aaron Dinin. I’m a serial entrepreneur. I teach entrepreneurship at Duke university, and this is the podcast that helps me teach about entrepreneurship by sharing my conversations with some of the web’s most impactful innovators. And by the way, the phrase most impactful innovators might even be an understatement for this episode’s guest, Bob Metcalfe, who invented Ethernet. In other words, he literally created a critical part of the infrastructure we all rely on, every day, to communicate with people across the internet. Here, I’ll let him explain just how significant a change that was.

Bob Metcalfe:

Up until Ethernet showed up, I had an RS-232 terminal in my office. A Texas Instruments machine, 730 characters per second. And it sent bits from my office to the wiring closet and the time sharing machine at 300 bits per second. Then we put Ethernet in. I had a wire in my office that ran at 2.94 megabits per second. And to save you the arithmetic, that’s a speed up of, not a hundred percent, it’s a speed up of 10,000. So with Ethernet, we accomplished three things. Bandwidth abundance by a factor of 10,000 times faster. That’s a lot faster. It delivered packets to your desktop. In the previous mode, the dumb terminal didn’t deal in packets. The dumb terminal dealt in characters. And so the Ethernet brought packets to the desktop so that software running in your PC could process packets and not characters. And the third thing it did is we later decided to make Ethernet an industry standard. So Ethernet joined the TCP/IP internet standard effort as the standard plumbing for the internet.

Aaron Dinin:

Yes, Ethernet is like the plumbing of the internet. And well, we all know how important plumbing is, even if it’s not something we think about often. So we’re going to spend an episode thinking about it right now. And we’re going to start by thinking about all the businesses and all the entrepreneurs that rely on Ethernet every single day, as I tell you about our sponsor, a company that supports internet businesses.

Web Masters is being brought to you thanks in part to the generous support of our sponsor, Latona’s. Latona’s is a boutique mergers and acquisitions broker that helps people buy and sell cashflow positive internet businesses and digital assets. All the types of companies that rely on Ethernet pretty much all the time, including ad supported content websites, SAS apps, e-commerce stores, Amazon FBAs, domain portfolios, and basically any other online work from anywhere business you can think of. If you’re running a business like that and thinking of selling it, take a few minutes to connect with the team at Latona’s. They can help you understand the process and when you’re ready, they can help you get your business sold for a great price. And if you’re hoping to buy an internet business, Latona’s is a great place to start too. Head on over to the Latona’s website where you’ll see listings for all the businesses they currently helping sell. That website is latonas.com, L-A-T-O-N-A-S.com.

We already know that Bob Metcalfe got interested in engineering at an early age, thanks to a dad who was an engineer, a train set, and an encouraging teacher. That interest led him to inventing a world changing technology and a billion dollar company. So yeah, on the surface, this episode is going to be one of those dull, boy plays with trains, then revolutionizes the world stories. [inaudible 00:05:56], who cares, right? In fact, when I spoke with Bob, he did a nice job of summarizing the entire thing in, basically, a few sentences. So yeah, let’s listen to that explanation.

Bob Metcalfe:

I went to MIT and digital electronics was just then emerging. Time-shared computers were just then emerging. And so I took various courses and then I was fully employed as a student. I worked from midnight to 8:00 every day, for four years. And those jobs were related to computer programming and building a hardware. I graduated and then I got a job at MIT, as a Harvard grad student, building a connection between MIT and the then internet called the ARPANET. So I built a high-speed network interface between MIT’s computer and the ARPANET. And then I went to Xerox Corporation, did it again. I built a connection between Xerox and the internet. And then I left Xerox and started my own company to make lots of these connections, connecting computers to the internet. And we used a technology that I led the invention of, called Ethernet, a local area networking technology. And my company 3Com started selling Ethernet, industry standard Ethernet products, and it eventually became a multi-billion dollar company. And I retired from there in 1990.

Aaron Dinin:

Pretty straightforward story when told that way, I suppose, but there’s more to it, I promise. And that’s what we’re going to get into, starting with a question I had for Bob which is how does someone become curious about high-speed networking in the first place?

Bob Metcalfe:

Well, let me tell you how I got into it. My senior project at MIT was to build a memory for a computer. And the technology that students used then was an acoustic delay line. We used store the bits by torquing a cable and the torque propagates at the speed of sound, around and round and round, and finally comes back, and then you keep cycling it. And in the course of building this device, I learned how to send bits, one at a time, down a long wire. So then when MIT needed to be connected to the internet, it needed somebody who knew how to send bits one at a time, down a long wire, so I got the job. And then at Xerox, when Xerox needed someone to connect all these personal computers that we were going to put on every desk, the lab looked around for who was good at sending bits one at a time, down a long wire. So that’s how I got into the business is an early specialization, blossomed into a multi-billion dollar company.

Aaron Dinin:

Could you give us some context into what networking was like in the early days of the internet?

Bob Metcalfe:

The networking part was the ARPANET starting at 1970. And there, the goal was to network together these time-sharing computers because ARPA wanted its research scientists, not to each have their own time-shared computer, but to use this network to connect up to somebody else’s, thereby saving ARPA the expense of buying a multi-million dollar machine for each of its sites. And that application, the first app of the internet was called resource sharing. So you would just use the network to connect all these machines. In fact, I was asked to edit a book in 1972 on scenarios for using the ARPANET. And so I collected the 20 or so applications running on these various time-shared machines. And in the book, explained how you could see them and try them out. And then we put that all on display at a conference in 1972, when the internet, still called the ARPANET, made its debut.

Aaron Dinin:

So hold on, I always heard the ARPANET was built by the US government to help distribute digital infrastructure around the country, and that was important because it would make the country less vulnerable in the event of a nuclear attack. Is that not a true story?

Bob Metcalfe:

Yes. The original purpose of the internet, ARPANET, was resource sharing.

Aaron Dinin:

So the nuclear war thing is just an apocryphal story? That’s kind of a bummer.

Bob Metcalfe:

If you look get the early ARPANET papers, they all refer to resource sharing as the app. But it’s true that prior to that, there was a 12 volume study produced by Paul Baran at the RAND corporation, and he wrote this report in the sixties. And in it, one of the goals of creating a packet network was to distribute it so that it couldn’t be knocked out. The joke was it couldn’t be knocked out by a simultaneous rise in temperature in 10 cities of 30 million degrees, without mentioning nuclear war. It was just a simultaneous increase in temperature by 30 million degrees. But the people who built the ARPANET were not doing that. Paul Baran said that was the goal of his report. And maybe there were some people in the Department of Defense who thought that that would be a reason to support the funding of it. And it was supported, so maybe there’s a string that goes back to that. But Larry Roberts, our fearless leader, was having us build a resource sharing network.

Aaron Dinin:

Okay then, I guess I learned something new. The ARPANET, the thing that came before the internet, was just a way for the government to save money, but that’s, I guess, an interesting origin story too, I suppose.

Bob Metcalfe:

You just danced over a big issue. You just said there was the ARPANET and that was before the internet. Whether the ARPANET is called the internet or not, is a big issue, especially among the people who claim to have built it. So I like to think of ARPANET as Internet 1.0. But there’s a bunch of people who believe the internet didn’t get started until TCP/IP protocol who were instituted in 1985 or so, and I guess it’s inside baseball. Al Gore, by the way, Al Gore famously was accused of claiming to have invented the internet. And the people who defend Al Gore, they’re the ones who think it wasn’t called the internet until 1990 when the Gore Act was approved. But since the internet, in my mind, started operating on October 29th, 1969, it’s rather rich to have Al Gore claiming to have invented it in 1990. So anyway, you just touched a third rail there. I believe the ARPANET was internet 1.0, and then we transitioned it and began calling it the internet in the mid ’70s.

Aaron Dinin:

As Bob mentions, we got into a bit of inside baseball during our conversation in the sense that debating whether the ARPANET was actually the internet or a precursor to the internet, is the kind of thing only people passionate about internet history would do. And sure, you can go ahead and call me a nerd for caring about that. But Bob cares not because he is a nerd, but because he was there and that, at least to me, is really cool. And it’s cool in part because, well, I’m fascinated by the stories of people with unique insights into the motivations and external factors that led to important innovations in history. In other words, not the stories people tell decades later, but the real reasons things are the way they are and ultimately evolve the way they do, like for example, why the internet is the way it is.

Bob Metcalfe:

One of the specs of the ARPANET was that you had to be able to send a character from your terminal to a remote computer across the country and back, in half a second. That was the spec. An eight bit character. Actually, they were seven bits in those days. Across the country and back in under a half second, that was the spec that all of us grad students were trying to meet. And there came this day where you’re writing, you’re building hardware and you’re writing programs, and then one of these days you hit the key and that character goes through your hardware and software across the country, getting switched and everything, gets turned around at the remote computer and gets sent all the way back and appears on your screen, on your paper. Actually, it was usually paper. And I can’t tell you how exhilarating that is the first time it happens, because you know that that letter that you just typed went across the United States of America and came back in under a half second. Holy smokes. Action at a distance, cheap thrills from making things happen far away, that’s an early motivator

Aaron Dinin:

As you’re sending these characters around the country in fractions of a second, are you thinking about the implications of what you’re building? Are you imagining the future of the digital revolution you’re about to catalyze?

Bob Metcalfe:

Well, a lot of my colleagues or a lot of my generation make up stuff about what they were thinking. And they know that you’re dying for them to say that they had some vision of the future, worldwide connectivity, promoting freedom and prosperity. But I’m telling you, I went to all those meetings and I didn’t hear a lot of that. What I heard was the cheap thrill. So I organized a workshop at MIT and all the grad students from all the ARPA universities came and we all demonstrated that we could log into each other’s computers, and if we couldn’t, we would fix it, right there at a workshop.

And there was no to talk about Facebook or Google. There was all the, “Wow, I can wiggle that computer over there in… Chicago is like thousands of miles away and yet I can log into it. Isn’t that cool?” So I think cheap thrills was the early motivator. I’m sure the people who funded it all must have had some grander scheme. Well, we know what their grand scheme was, resource sharing. They wanted to save money. That’s what funded the early ARPANET, and it didn’t last. It became a serendipitous outcome, a surprise killer app. And there’s been a series of them.

Aaron Dinin:

Like what? What would you say were some of the surprise killer apps of the early internet?

Bob Metcalfe:

Very quickly, it became email and stayed email for the next 20 years. Email may still be the killer app of the internet. It’s hard to tell sometimes. I’ve been sending emails since the ’60s and over the internet, since the ’70s. Now, my company later got its start helping people share printers. For example, the Apple LaserWriter cost $7,000, so you would want to share that among a bunch of people. And if you put a bunch of PCs on an Ethernet, you could share a LaserWriter, or IBM came out with a 10 megabyte disc in 1982. And no one knew what to do with all that storage, just so much storage. It made sense to share it among PCs so the network could be used to share access to this 10 megabyte disc that we had.

So that was sort of another round of killer apps. That’s a feature of networking, is that you make progress along one of the several dimensions of interconnectivity. And that gives rise to unexpected applications, new application, serendipity. So one of the principles that we learned in the building of the internet is build it and they will come. So we’re now working on the gigabit internet. No one really knows what we need a gigabit internet for, but we’re pretty confident that when we build it, somebody’s going to come up with nine new Facebooks to make use of the upcoming version of the internet.

Aaron Dinin:

This kind of network growth Bob is describing and the importance of enabling network growth, even without necessarily knowing what you’re ultimately going to support, is the foundation of what would famously become known as Metcalfe’s law. Metcalfe’s law states that value of a communications network is proportional to the square of the number of connected users. So let’s unpack that concept for a minute. Basically, what Metcalfe’s law points out is for every new node being added to a network, you’re actually exponentially increasing the potential value of that network. Imagine the concept in terms of money. In the normal world, for every additional dollar you earn, you have one additional dollar of spending power, it’s linear growth. But think about what would happen if each additional dollar actually compound on itself. You only earn $1, but because of all the other dollars you already have, it actually provides $10 worth of additional value, or $100, or a million dollars, depending on your current spending power.

That’s the concept behind Metcalfe’s law. It’s that the bigger a network is and the more net nodes you have in it, the more value you get from adding new nodes, because then, all the currently existing nodes have more connections to make. An example of what that means might be easier to understand when you think about it in the context of, say, telephones. Two telephones can only make one connection to each other, right? But three telephones can make three unique connections. The original connection and then the two connections between the original two phones, and the one new phone. Add a fourth phone and now we’ve got six possible connections. Add a fifth phone and we’ve got 10 and so on. Now, imagine what happens when you’ve got a network with a billion telephones and you add a new telephone. That one new telephone creates a billion possible, unique, new connections between itself and all the currently existing telephones.

Whew, that’s a lot of additional potential value, right? And that’s what Bob Metcalfe articulated so elegantly. He correctly predicted the exponential growth of value we’d be able to derive by growing the number of users and machines on the internet. But here’s the thing, you’d think a concept like that would be the kind of theory some early digital evangelists would come up with while trying to as espouse the potential of the internet to make the world a better place. And to be fair, that’s how Metcalfe’s law has been used for decades. And people still use it like that today. Only that’s not the story behind Metcalfe’s law or where it came from. To understand that story, we have to go back to Bob’s work, commercializing Ethernet, that began while he was at the famed Xerox Palo Alto Research Center, otherwise known as Xerox PARC.

Bob Metcalfe:

The internet was started, as I’ve said now a couple times, to network these time-sharing machines, but right in the middle of doing that, Xerox research and others started doing things called personal computers, which were not time-sharing machines. So then it became how do we connect personal computers to the internet? And that’s what Ethernet was designed to do. So instead of there being one computer per building connected to the internet, there’d be hundreds of them connected to the internet. And how would you do that? And Ethernet was the answer to that question. You’d collect all the traffic. Imagine this, a computer on every desk. So we started doing that in 1973 at Xerox Research and I lucked out. The computer science lab needed somebody who knew how to send bits down a wire, one at a time, I got the job.

And so Ethernet was the answer to the question, how are we going to network these PCs together? And the killer app for Ethernet was laser printing. At the same time we were building the Alto personal computer, arguably the first modern personal computer… That’s a whole other argument you can get into. We were building, arguably, the first laser printer, page per second, 500 dots per inch, and they gave me the job to develop the little operating system and the network connections for this printer. And I arranged so that the only way you could print on this printer was through the Ethernet. And this printer printed a page per second at 500 dots per inch. It was a miracle. Everybody wanted to print their documents and multiple fonts on this printer, and the only way they can do it was to put Ethernet card in their new brand new PC. So you can see how printing was the killer app for Ethernet. Everybody had to have Ethernet because they all wanted to print on this printer.

Aaron Dinin:

So you did all that work originally developing Ethernet at Xerox PARC. How and why did you end up creating your own company, 3Com, to help supply the world with Ethernet?

Bob Metcalfe:

Well, I had lucked out by getting accepted to MIT in 1964 and went to Boston, and Boston was, at that time, Silicon Valley. The startups were coming from what was called Route 128, around Boston. And that was the hot bed of venture capital, which had been invented at the Harvard Business School, roughly speaking. So there I was, immersed, surrounded by people starting companies like Amar Bose started Bose, Ken Olsen started Digital Equipment Corporation, and so on. So I started three companies, consulting companies, as an undergraduate to just pay tuition and stuff. But I got to name the companies and I got to choose the stationary, and I got to find customers, and I got to collect revenue, and I got to pay people. And each of those three companies lasted a year or two. But then I moved, 1972, moved from Route 128 Boston to Silicon Valley in its early days, 1972.

Then again, I was surrounded by entrepreneurs, Bob Noyce of Intel, Bill and Dave, founders of Hewlett-Packard. So it was in the water that you would start companies. So after eight years at Xerox… I left Xerox with a general intent of starting a company, some company, any company. I didn’t have one in mind. It took me five months to incorporate after leaving Xerox. And the idea was to use the machinery of free enterprise. Use that to have impact with this technology that had been developed in the labs. So I’ve used startups as a method of having impact. And so we began collecting people and resources and technologies around the ideas of the internet and began selling them. Founded in June 4th, 1979, 3Com. We grew rapidly, so in 1984 we went public. And by the way, for 10 nanoseconds in 1999, the company was worth, inflation adjusted, $52 billion. And that’s a fake number, but for a nanosecond, it was worth that. That was during the internet bubble. I started the company with $27,000, that was my net worth. And that 27 turned into $52 billion, some years later.

Aaron Dinin:

And what was it like building that company? I mean, I realized that’s a big question and kind of asking for a short answer in this context, but could you talk briefly about 3Com, which in itself was such an important company?

Bob Metcalfe:

I spent 13 years personally building 3Com. And every day of every year was high adventure and some of it pretty ugly, but it eventually succeeded. 3Com is an epitome of a Silicon Valley company. Think of Xerox PARC as a lab of Stanford University. We were all professors at Stanford while we were working at Xerox. And the father of the internet was Vint Cerf at Stanford, so we attended the same seminars together. So think of it as university research supported by venture capitalists, eventually becoming a large company, going public, and then eventually being acquired, acquired by Hewlett-Packard in 2010. It’s the epitome of the Silicon Valley company. And I fell into it. So I was an engineer, but I fell into this ecosystem, the startup ecosystem, that was equipped to surround me with all the stuff I needed to commercialize my technology. And that all worked out. So Ethernet is now, roughly speaking, the plumbing of the internet.

Aaron Dinin:

And how’d you manage to achieve that kind of enormous success and adoption of the technology?

Bob Metcalfe:

In 1982, 3Com Corporation was spending its venture capital and slowly going out of business. And the board of directors replaced me as CEO and made me head of sales and marketing. So here’s an engineer who’s now in charge of sales and marketing. Now, sales and marketing are much more complicated than electrons. They’re really complicated stuff. And I was head of sales and marketing for two years, from ’82 to ’84, and I got us from zero sales to a million a month, and then we went public. And that is what I’m most proud of is that accomplishment. I, somehow, got good advice on how to be a good VP of sales and marketing, and then I just did it and did it hard for a long time. So getting from zero to a million a month, I’m most proud of that.

Aaron Dinin:

So what was the strategy? How’d you figure out how to sell Ethernet so effectively?

Bob Metcalfe:

We used decks to sell our products. 35 millimeter slides in carousels. Later, by the way, I was on the board of the company that developed PowerPoint and sold it to Microsoft in 1987 for $14 million. But that was 1987. I was having to sell products in 1982. So we used 35 millimeter slides and we were selling Ethernet cards to connect PCs in an Ethernet. And one of our problems were there weren’t any PCs. And that’s a big problem. So we tried to make it as easy as possible for the people who bought PCs to try our products, so we made a three node trial network. Three of our cards, you take three IBM PCs, you’d plug one of our cards into each of them, you would run a cable among them, and then you’d load our software, which came on a diskette, and loaded into each of those three machines and thereafter, you could share a printer, share a disc, or exchange email among those three PCs. And that trial worked.

You could share the printer, you could share the disc, and you could exchange emails. But after selling this for a while… And people bought it because $3,000 isn’t that much money to a business. This is business to business. The customer said, “This trial works exactly like you promised, but it’s not useful.” And so useful, that’s a problem. If you’re a head of sales and marketing and your customers are telling you the product is not useful, you have to do something. So one night, I went over to Stanford. Some of the Xerox equipment had been donated, printers and stuff. And I made a slide, a 35 millimeter slide. And on this slide, I argued with a diagram that the cost of your network is linear in the number of my cards that you buy. There’s the number of nodes you put on the network is linear.

But the number of possible connections goes up as the square since each one of those N nodes can talk to N minus one other nodes. The total number of connections is N times N minus one, which is approximately N squared. I made six slides, gave it to my sales force. They went out to their customers and they said the following. They said, “The reason your networks are not useful is that they’re not large enough.” That is this quadratic, which is a measure of the value of the network. The number of possible connections, I claimed was the value. The reason that you were finding your network not that useful is it wasn’t big enough. And what’s the remedy for that was to buy more of our products with a very self-serving slide. Well, they believed us and they did grow their networks. And in 1984, we went public as a profitable company.

Aaron Dinin:

Wait, so that’s where Metcalfe’s law comes from? It wasn’t some academic research or advocacy work about the future and potential of the internet? It was a sales pitch?

Bob Metcalfe:

A sales tool, a 35 millimeter slide in a deck of such slides by sales force of six people, north and south, east, central, west. And the guy who handled the central region, David Colson, he also handled international. So our sales force had six people in. And by the way, Dave Colson got international because he had an English accent. And Dave DePew got to move from Santa Clara to the East Coast because he was single and he could. So he became the head of sales on the East Coast, and they all succeeded wildly and I’m forever grateful.

Aaron Dinin:

That is crazy. I feel like my mind is almost blown. I always just assumed things like that were these carefully constructed and rigorously tested theories. And how did it get the name Metcalfe’s law? It wasn’t you who would’ve named it that, right? I mean, that would be incredibly narcissistic to name a law after yourself, I suppose.

Bob Metcalfe:

In 1993, much later, a man named George Gilder called it Metcalfe’s law in a magazine article and a book, a book called Telecosm, and in a magazine called Forbes. I gave him this slide and he called it Metcalfe’s law. And one of my hobbies these days is defending Metcalfe’s law, to this day, that a reasonable quantification of the value of a network is the square of the number of its attachments. In fact, the law has progressed. Originally, it was talking about machines attached. Then later, in the Facebook era, it was people who were attached, and now we’re moving on to data being attached, connecting data also produces value. So once again, that’s one of my hobbies is defending Metcalfe’s law.

Aaron Dinin:

What and who are you defending it against?

Bob Metcalfe:

Well, the problem with having a law is that it attracts critics. And so a lot of the people I run into on that topic are disputing the truth of Metcalfe’s law. But none of them have really succeeded, like my famous professor, Odlyzko, the University of Minnesota, got a cover story in the IEEE magazine claiming that Metcalfe’s law was dangerous because it tend to over inflate the value of networked companies. And he proposed that instead of N squared, we would use N log N. And all that did was reduce how quickly the value grew, but it still went to infinity. And plus he had no data to support his law. Of course, I had no data to support my law either. So neither one of us had any data. So what we were arguing about was a hunch about the value of networks.

I’ve since written a paper for the IEEE demonstrating how Metcalfe’s law N squared fits the first 10 years of Facebook growth. That is taking revenue as a surrogate for value and monthly average users as a measure of the number of attached users, you can pretty much fit it to N squared. And that was my first attempt, in 2013, to bring some data to that debate, but the debate continues. So anyway, the downside of having a law named after you is you have to defend it. And every grad student and every professor who knows how to do arithmetic comes after you, proving that your law is foolish and fraudulent, dangerous on the cover of IEEE magazine, dangerous. And you have to smile and… But I don’t give an inch. My law is exactly true and has always been and always will be, as far as I’m concerned.

Aaron Dinin:

So Metcalfe’s law is true according to you. And so what do you think about how Metcalfe’s law has been applied to other types of networks like social networks, which came along after you developed it? Do you agree with the newer applications of Metcalfe’s law?

Bob Metcalfe:

So one of the things that has me nervous is that Metcalfe’s law has now begun to be referenced frequently in the crypto world. There’s actually a graph. It’s called The Metcalfe that shows what Bitcoin should be worth based on Metcalfe’s law. And I don’t know anything about that. I just know Metcalfe’s law is being used that way, which means when Bitcoin goes down, guess who they blame. I have another audience of hostile to me. So I’m worried about that. So there’s Metcalfe’s law being used to quantify the value of cryptocurrencies. And I don’t know anything about cryptocurrencies, so I’m a little nervous about that.

Aaron Dinin:

So you’re cautious about it, is that a fair assessment? You’re keeping an eye on it?

Bob Metcalfe:

So I’ve parleyed that into a interest in connectivity. So to me, connectivity is plastics. Connectivity is the cure all. And Metcalfe’s law is part of the math of connectivity. Connectivity as a thing has dimensions, and it has math, and laws, and it has disruptions, and it has pathologies. I remember the first pathology I ran into on the internet was porn. In the early days of the ARPANET, it was almost shut down because it was carrying porn. And that was a pathology of that version of connectivity. And we dealt with that. And then came advertising. The internet was not built to carry ads and all the intelligence, the internet intelligentsia were offended when the internet started carrying ads. Little did they know that the entire internet would soon rely on ads for its complete financial system. So then spam became the pathology and we’ve pretty much, not entirely, but we pretty much dealt with spam.

But then there are the modern pathologies. I’ll skip ahead, the worst one, of course, being fake news. So that’s a pathology of connectivity. And I think my general explanation for these pathologies is that we’ve become connected suddenly. As the internet is now only 50 years old, it started operation way before Al Gore. Actually, it started operation the year that Al Gore and I graduated from college, 1969. So it’s been going for 50 years, but in 50 years, more than half the human race is now on the internet, 4.7 billion people. And that connectivity is overwhelmed us. We don’t know how to deal with it. We’re still learning how to deal with it, hence, these pathologies. We’re learning how to deal with fake news and we will. And that some new pathology will crop up.

Aaron Dinin:

Is it reasonable to say that these pathologies are almost like a consequence of Metcalfe’s law?

Bob Metcalfe:

Well, it’s related. There’s impact. Connectivity has impact. And one quantification of that is Metcalfe’s law. And I guess that contributes to the pathologies. The power of connectivity is manifest in certain pathologies like fake news. I think the overall impact of the internet has been fabulously positive in achieving our big goals, which are freedom and prosperity. There’s this great chart that shows the number of people living in desperate poverty versus the ones who don’t. And you look at that graph and subsequent to the arrival of the internet and particularly the worldwide web, the number of people living in extreme poverty plummets. Now, I know the difference between causality and correlation, but correlation is a prerequisite for causality. I mean, there’s a big hint that connecting everybody together, although it has its negative consequences, for example, the radicalization of jihadists. It generally has positive impacts. And the spread of democracy and the growing of our economic output is our evidence of the impact of greatly increased connectivity, thanks to the internet.

Aaron Dinin:

And if we’re going to talk about Metcalfe’s law and the internet more broadly, we have to talk about this. Yes, the internet creates problems or as Bob calls them, pathologies, but that’s not a flaw, that’s a standard byproduct of the innovative process. Every new innovation, no matter what it is, creates new problems or as an entrepreneur might frame it, new opportunities. In fact, there’s no such thing as a universal good. So by definition, as network scale in accordance to Metcalfe’s law and exponentially impact more people, some of those impacts are necessarily going to be bad. And they stink, and they’re terrible, and when possible, we should work to counteract them and avoid them, and minimize them, and eliminate them. But on the whole, if we zoom out and look at the bigger picture in terms of the benefits created by increased interconnectivity, well, you can probably guess what I think of them.

Let’s just say I’m a fan of what Bob Metcalfe helped build. I suspect you are too otherwise, you’re probably listening to the wrong podcast. But let’s assume you aren’t and in that case, I hope you’ve enjoyed this episode. I hope you’ll leave us a nice review on your podcasting app of choice and I hope you’ll consider sharing it with a friend. I want to thank Bob Metcalfe for taking the time to speak with me and share the story Ethernet, 3Com, and Metcalfe’s law. I also want to thank our audio engineer, Ryan Higgs for all his help pulling together this episode, and a thanks to our sponsor Latona’s.

If you’re interested in buying or selling an internet business, be sure to check out latonas.com. For any thoughts or feedback about the episode, you can find us on Twitter, we’re @WebMastersPod. I’m also on Twitter @AaronDinin, that’s A-A-R-O-N D-I-N-I-N. Or you can connect with me on my website, aarondinin.com, where you’ll also be able to access my articles, newsletter, and all sorts of other content about startups and entrepreneurship.

Thanks again for hanging out with me for another episode of Web Masters. If you haven’t already, be sure you’re subscribed so you get the next episode downloaded to your phone automatically, as soon as it’s released. I hope you’re as excited about it as I am. Until then, well, it’s time for me to sign off.

[OUTRO]

Aaron Dinin:

I’m just curious, why’d you call it Ethernet? Where’d that name come?

Bob Metcalfe:

So we, Dave Boggs and I, in this designing this first Ethernet had a choice of the medium. Would it be radio? Would it be coax? Would it be twisted pair? We chose coaxial cable. In fact, it was a mile long, half inch coaxial cable, but we knew that the Ethernet didn’t depend on the medium. You could use a variety of media for it. And we anticipated we would in the future. So instead of calling it coax net, we wanted to call it something that admitted future variations of media. And there was this physics phenomenon called the luminiferous ether. Luminiferous meaning light bearing. It was once theorized that the luminiferous ether was a medium that carried light from the sun to the earth. And in around 1900, they discovered there was no medium that carried light from the sun to the earth. Light propagated through a vacuum that didn’t need a medium.

So this luminiferous ether term was now debunked. So we adopted it. We called it Ethernet because it was a pervasive media for the propagation of electromagnetic waves, in particular, our data packets. So we used the word for this a similar purpose and called it Ethernet. Ethernet got attacked many times over the next 10 or 20 years. And one of the early attacks was from a physicist who didn’t like Ethernet at all. And then he said, “It was named after a debunked term anyway,” as if this was a critique of Ethernet that we had chosen to call it Ethernet. And since the ether had been debunked, this didn’t reflect positively on whether Ethernet worked or not.

Aaron Dinin:

Wow. That seems like a strange thing for someone to fix it on, but okay.