On April 7, 1946, the year after the end of World War II, a baby boy was born into an ordinary family in Brooklyn, New York.
The boy’s father was a technician at a gyroscope factory, and his mother was a homemaker. For this family, having just come through the upheavals of war, the birth of a son was a joyful occasion.
The couple placed great hopes on the child, wishing that he would grow up to be successful and become a top engineer—a highly respected profession at the time.
Indeed, this baby boy, named Metcalfe, did not disappoint. He showed remarkable academic talent as he grew. In 1964, after graduating from high school, he was admitted to the Massachusetts Institute of Technology (MIT), one of the top universities in the United States and a cradle of elite engineers.
Metcalfe’s parents could hardly have imagined that their son would not only become a top engineer, but would also go on to found a Fortune 500 company. One of his groundbreaking inventions changed countless lives and had a profound impact on the IT industry.
Yes, this Metcalfe is Robert Melancton Metcalfe—the inventor of Ethernet, founder of the famous tech company 3Com, and the proposer of Metcalfe’s Law.
01
█ A Rookie Faces Rejection: His Doctoral Thesis Is Turned Down
In 1969, 23-year-old Metcalfe graduated from MIT with degrees in electrical engineering and business administration. A year later, he earned a master’s degree in computer science from Harvard University and continued pursuing a PhD there.
During his doctoral studies, Metcalfe took a job with the MAC project group at MIT, which specialized in operating systems, computational theory, and artificial intelligence research, and later became very well-known.
In 1969, the U.S. Department of Defense launched ARPANET—the predecessor to the modern internet—connecting mainframe computers at four major universities.
Metcalfe recognized the potential and strongly advocated for Harvard and MIT to connect their systems to ARPANET. (He was a Harvard graduate student and an MIT researcher.) Arrogant Harvard refused, but MIT agreed.
Metcalfe soon completed the network interface and connected MIT’s mainframe to ARPANET. Based on his design and research—in which he created a high-speed network interface and protocol software between the ARPANET IMP and the PDP-10 time-sharing mini-computer—he wrote a doctoral thesis and submitted it to Harvard’s degree committee.
In June 1972, Metcalfe failed his thesis defense. The committee found his work lacking in “mathematical rigor” and “theoretical depth.”
The setback wasn’t just from Harvard. While working on ARPANET, Metcalfe once led a group of 10 AT&T officials through a demonstration. Unfortunately, the system crashed during the demo.
Metcalfe later wrote: “I looked up in agony and saw them laughing at the unreliability of packet switching… I will never forget it. To them, it confirmed that circuit switching—the technology used by traditional telephony—was here to stay, and packet switching was an unreliable toy that would never have commercial relevance.”
Repeated setbacks left Metcalfe disheartened. Not long after, Bob Taylor—the head of Xerox’s Palo Alto Research Center (PARC) and one of ARPANET’s founders—extended a warm invitation for Metcalfe to join PARC and complete his thesis. Metcalfe accepted.
PARC was known for revolutionary inventions such as the laser printer, mouse, graphical user interface (GUI), and bitmap graphics—many of which later inspired Apple’s innovations.
02
█ Inspired by a Peer, Project Success Is Achieved
At PARC, Metcalfe quickly started his work. The lab aimed to build the world’s first personal computer—the now-famous Alto. Metcalfe’s job was to design a network interface to connect these machines.
The biggest challenge in building a multi-user terminal computer network was managing how each computer accessed the network.
In the early 1960s, computer scientist Leonard Kleinrock proposed using queuing theory from mathematics—by simulating traffic jams and waiting lines—to manage data flow. ARPANET used this and proved it effective.
In 1971, University of Hawaii professor Norman Abramson created a wireless network called ALOHAnet, which used a more “radical” method. ALOHAnet transmitted data in small packets and did not try to avoid packet collisions. If a message was lost due to a collision, the user would simply retry after a random time interval.
It was like two people talking: if both speak at the same time, they stop, wait, and try again. Eventually, one side pauses long enough for the other to speak successfully.
However, ALOHAnet’s method had a clear flaw: it wasted resources. It worked well with low traffic, but in crowded networks, collisions became frequent and transmission efficiency plummeted.
After reading Abramson’s paper, Metcalfe was deeply inspired. He improved the ALOHAnet model and proposed a new one.
In Metcalfe’s model, computers would independently adjust their retransmission wait times based on collision frequency. If collisions were rare, they would retry quickly; if the network was congested, they would back off to preserve overall efficiency.
This new model addressed the shortcomings in his thesis. In May 1973, he finally passed his Harvard defense and earned his PhD. (Interestingly, Harvard never published his dissertation—MIT did. Metcalfe held a grudge about this.)
He also applied the new model to his research project. On May 22, 1973, Metcalfe distributed a memo titled “Alto Ethernet,” officially introducing the Ethernet concept.
In the memo, he sketched out Ethernet’s operating principle: “Participating stations, like those in ALOHAnet or ARPANET, inject their packets at megabit speeds; there will be collisions, retransmissions, and backoffs.”
His Ethernet concept incorporated Abramson’s random retransmission mechanism, his own clock timing adjustments, and other ALOHAnet improvements to minimize collisions.
While many of these ideas were developed by other researchers, Metcalfe was the first to integrate them into a practical network design.
As for the name “Ethernet,” it’s worth explaining. Before electromagnetic waves were discovered, scientists believed in “ether”—an invisible medium that transmitted light. Later, ether was proven nonexistent.
Metcalfe chose the name “Ethernet” to symbolize it as a transmission medium. He even earned the nickname “Ether Daddy.”
In June 1973, Metcalfe was authorized to build a 100-node Ethernet prototype. To complete the complex tasks of logic design, circuit board construction, and microcode writing, he enlisted Stanford grad student David R. Boggs.
On November 11, 1973, thanks to their efforts, the world’s first Ethernet prototype system was born.
It achieved a transmission speed of 2.94 Mbps—roughly 10,000 times faster than earlier terminal networks.
03
█ Founding 3Com: Driving Ethernet’s Commercial Success
After Ethernet’s invention, Metcalfe urged Xerox to commercialize it. However, Xerox’s management was slow to act.
By 1979, after six years of waiting, Metcalfe left PARC. He decided to start his own company to push Ethernet into the market. The company he founded was 3Com.
The name 3Com stands for: Computer, Communication, Compatibility—reflecting Metcalfe’s vision of improving computer communication interoperability.
3Com boosted Ethernet’s commercial viability by selling networking software, Ethernet transceivers, and network interface cards (NICs) for minicomputers and workstations.
In 1980, thanks to Metcalfe’s coordination, DEC (then the world’s second-largest computer company), Intel, and Xerox formed a technical alliance to publish the DIX (from the companies’ initials) Ethernet standard.
In 1983, the IEEE formed a working group and released the IEEE 802.3 standard based on the DIX variant.
The first 802.3 version was 10BASE5, offering 10 Mbps throughput via thick coaxial cable and using CSMA/CD (Carrier Sense Multiple Access with Collision Detection), a method familiar to engineering students.
IBM and General Motors also released their own network standards. IBM’s Token Ring in particular became Ethernet’s major rival. After 20 years of competition, Ethernet prevailed, and Token Ring was phased out.
04
█ After Success: Personal Transformation
Throughout the 1980s, Metcalfe dedicated himself to promoting Ethernet. On March 21, 1984, 3Com went public.
In the mid-80s, Metcalfe proposed a key idea: “The value of a network is proportional to the square of the number of its nodes”—a concept now known as Metcalfe’s Law.
The law became a cornerstone for understanding network effects and the internet economy.
In 1990, Metcalfe left 3Com to become a commentator and tech columnist.
During this period, he made several misjudgments that drew media mockery. In 1995, he predicted the internet would suffer a “catastrophic collapse” the next year, saying he’d “eat his words” if wrong.
He was wrong. In 1997, at the 6th International World Wide Web Conference, Metcalfe blended a printout of his article into a slurry and ate it in front of the audience, admitting his mistake.
Other failed predictions included: Linux would be overtaken by Windows 2000; wireless networking would be abandoned in the mid-1990s; and by 2006, Windows and Linux would be unable to handle video services.
In 2001, Metcalfe left media and founded Polaris Venture Partners, becoming a venture capitalist. In 2011, he became a professor at the University of Texas at Austin.
In 2022, more than 50 years after his time at MIT, Metcalfe returned to its Computer Science and Artificial Intelligence Laboratory (formerly MAC, now CSAIL) as a researcher.
Meanwhile, 3Com faced ups and downs. In 1999, its revenue peaked at $5.7 billion. But soon after, the dot-com bubble burst, and 3Com’s value plummeted. In November 2009, Hewlett-Packard acquired 3Com for $2.7 billion in cash, ending the company’s independent existence.
In his later years, Metcalfe received many honors for his contributions to Ethernet.
In 1996, he received the IEEE Medal of Honor. In 2003, he was awarded the National Medal of Technology and the Marconi Prize. In 2007, he was inducted into the National Inventors Hall of Fame.
Most recently, on March 22, 2023, 76-year-old Metcalfe was awarded the 2022 ACM Turing Award, with a $1 million prize funded by Google.
Jeff Dean, Google’s Senior VP of Research and AI, said in ACM’s statement:
“Today, there are about 7 billion Ethernet ports globally. Ethernet is everywhere, and we take it for granted. But it’s easy to forget that without Bob Metcalfe’s invention and persistence—ensuring every computer must be networked—our connected world wouldn’t look like it does.”
05
█ Conclusion
Jeff Dean was right. Ethernet is the foundation of modern data communications. As the father of Ethernet, Metcalfe made an immense contribution.
Today, Ethernet remains the primary global standard for wired communication. Its data rates have evolved from 2.94 Mbps and 10 Mbps to 400 Gbps, 800 Gbps, and even 1.6 Tbps.
That familiar number—802.3—will stay with us for a long time, until the day it is finally replaced.
That’s all for today’s article. Thanks for reading. If you found it helpful, please share and like it. Thank you!
References:
- “Ethernet Inventor Bob Metcalfe Wins Turing Award,” The Paper
- “Turing Award Goes to Father of Ethernet,” Synced
- “Ethernet’s 50th Anniversary: Metcalfe Wins Turing Award, 2023 Roadmap Released,” SDNLAB
- Wikipedia
Disclaimer: This article is created by the original author. The content of the article represents their personal opinions. Our reposting is only for sharing and discussion purposes and does not imply our endorsement or agreement. If you have any objections, please get in touch with us through the provided channels.