Ethernet at 40: Its daddy reveals its turbulent youth
Bob Metcalfe: How Token Ring and 'IBM's arrogance' nearly sank Big Blue
Feature When Bob Metcalfe, the prime mover behind the invention of Ethernet, recently visited the site of that invention, Xerox's Palo Alto Research Center (PARC), The Reg had the opportunity to sit down with him to discuss the history of Ethernet, its advantages over Token Ring, and IBM's perfidy.
"As you know," Metcalfe told us, "something like Ethernet is not actually invented in a single day. But if there were a day, it would be May 22nd, ."
The Reg asked what had happend on that day. "I wrote a memo," he said, "and in that memo named it and outlined its general operation." Ethernet, however, was not the original name of what has now become the world's networking standard.
"Prior to that we'd been calling it the 'Alto ALOHA Network,' Alto being the PC that we were building, and the ALOHA Network being a packet radio network at the University of Hawaii whose design we admired," Metcalfe said.
"But then on that day we realized that we were going to network things other than Altos," he said, "and Ethernet had by then progressed so that it was barely resembling ALOHA. So we changed it to Ethernet."
The term ALOHA, by the way, was a rather contrived acronym that stood for Abramson's Logic of Hiring Access, named after its University of Hawaii developer, Norm Abramson.
Ethernet daddy Bob Metcalfe
(source: University of Texas at Austin)
Metcalfe remains more than a little sensitive about just how much Ethernet owes to ALOHA. "Over the years people have said that Ethernet is just ALOHA network," he told us. "I wrote a preface to my own book in which I put a paragraph explaining how Ethernet was just like ALOHA network except – and then there was like 20 'excepts,' one of which was that the ALOHA network was not a LAN; it was a WAN. It was to connect the Hawaiian Islands over a distance of 200 miles, and it ran at 4800 bits per second." Ethernet first ran at 2.94 megabits per second.
There were other differences, as well. Ethernet, for example, operated within a mile, so its propagation time was roughly five microseconds; ALOHA's 200-mile range, of course, operated at much longer time scales. "The parametric values all changed, and therefore the operation of the network changed," Metcalfe told us – and he said that he provided more than enough examples in his preface to prove his point.
"When I reread that paragraph," he said of his protestations, "I can see whoever wrote that paragraph was fed up with that accusation."
But his fervent attempt to separate Ethernet from ALOHA wasn't entirely successful – even at PARC. During our interview in the center's cafeteria, Metcalfe caught sight of a man at a corner table whom he identified as being PARC's "artificial-intelligence guru," Danny Bobrow, whose office was next to Metcalfe's in the early 70s. Metcalfe, who last visted PARC in 1975, told Bobrow about Ethernet's 40th birthday, and the AI guru said, "And it all came out of ALOHA, eh?"
Metcalfe turned to us; "See? See?" he said, then turned back to Babrow. "I'll have to send you my book," he laughed.
But he's still a bit miffed. "Norm Abramson, my good friend from the University of Hawaii," he said, "I've seen his presentations and he takes credit for Ethernet, Wi-Fi, and everything else."
That said, "There's plenty of credit to go around," he admitted.
Metcalfe had worked on the ARPANET both in graduate school and at PARC, and when he arrived at PARC, researcher Charles Simonyi was working to create a LAN within the buiding called SIGNET. "His idea was to mimic the ARPANET," Metcalfe said, "and put packet switches all over the building."
Since Metcalfe was "the real networking guy," he told us, Simony was given the boot and he was given the SIGNET project. "By the way, SIGNET is short for Simonyi's Infinitely Glorious Network, named after himself," Metcalfe informed us.
Don't feel sorry for Simonyi being removed from his LAN project, however. He and PARC colleague Butler Lampson went on to develop a WYSIWYG text editor for the Alto called Bravo. It was transformed into Microsoft Word and Simonyi got rich.
"He now has his own three-engine jet and his own helicopter, and he's a billionaire," Metcalfe said. "Maybe I should have written Bravo and he should have done the network."
We asked Metcalfe what the main usage-model requirements that his work on Ethernet were intended to solve. "No one wrote a requirements document," he said, "this is a research center."
What PARC wanted to do was create a LAN to link the Altos that it planned to put on its researchers' desks to replace the 300-bits-per-second Texas Instruments Silent 700s they had at the time, a "revolutionary" change, according to Metcalfe.
"There was a lot of argument if it made any sense to put a computer on every desk," he recalled. "You'd have to have an MIS manager for every desk. What if they broke? Someone would have to come to your office and fix it."
Those objections, he said, "melted away over time." A fast LAN would allow PARC to not only put an Alto on each desk and preserve access to the ARPANET gateway, but it would also solve another problem: providing access to the laser printer that PARC was developing.
"I wrote the operating system for that printer," he told us, "and then we plugged in an Ethernet card. Its requirement was 500 dots per inch, a page per second, 8½ by 11. Five hundred times 500 times 8½ times 11 is about 23 megabits per second. And so the current methods of networking were not going to keep that printer busy."
When we jokingly asked, "Well, what are buffers for?" he responded, "Well, speaking of buffers, each of these Altos has a removable-disk cartridge that could carry two megabytes, which was huge in the day. And so sneakernet was born."
To print, the researchers would put their documents on one of those removable cartridges, remove the cartridge, walk over to the Alto to which the laser printer was connected, plug the cartridge into that Alto, and print it. Sneakernet.
"But I argued," he said, "'Wouldn't it be easier just to hit P?' And they said, 'Well, you'd still have to walk down to get the printout!' So I said, 'Okay, you've got me there'."
But eventually Metcalfe did get the laser printer hooked up to his new LAN, and thanks to the printer's resident fonts and compression schemes, Ethernet's throughput was plenty fast enough to accept print jobs.
That was the genesis of Ethernet: connecting those "revolutionary" desktop PCs, providing access to the ARPANET, and allowing researchers to simply hit P to print. "We were not planning to do the web, or audio, or video," Metcalfe told us. Those came later. Much later.
Fighting IBM in the Token Ring wars
The Reg asked Metcalfe what it was that boosted Ethernet past its early competitors such as Token Ring and ARCnet, and he told us that despite what some said at the time, "They all worked. Various proponents claimed that the other thing didn't work, 'And therefore you should use mine.' That wasn't at all true. All these things worked."
Some of Ethernet's early detractors, he said, should hang their heads in shame. "In the dark days," he reminisced, "IBM was paying professors to write papers showing that Ethernet didn't really work – when it did work – and that Token Ring was better. Those professors should be ashamed of themselves."
Calling the papers that IBM had paid the professors to produce "clearly hatchet jobs," Metcalfe said that "it was really disheartening to watch IBM and these professors stooping to that shit."
Should you, aging Reg reader, recall one of the papers to which he was referring, Metcalfe has an explanation of what he claims the profs were up to. "What they would do was develop a model," he said, "a mathematical model ... They would then move the model into regions of absurd parameters. They would say, 'There's a million users, and they're sending full-time, all the time.'"
Those "absurd parameters" would cause the model to show that Ethernet would fail under those conditions. The papers would then conclude, "'See, that doesn't work here,'" Metcalfe said, adding, "But no system would ever be used 'here'. Duh."
IBM didn't play fair in other ways during the war between Ethernet and Token Ring, Metcalfe contends. "IBM always had 90 per cent market share in Token Ring, and some of it was cheating – they'd sprinkle SNA dust over their implementation," he said.
He also argues that IBM made it difficult for other vendors to get in on the Token Ring action. "3Com, my company, shipped Token Ring before IBM did," he said, shipping cards based on Texas Instruments chips. "That was because my board [of directors] would not allow me to 'risk the company' on this 'obsession' with Ethernet, and so we had to sell Token Ring."
That effort ran into problems. According to Metcalfe, when customers tried to plug their 3Com cards into IBM systems, they wouldn't work "because IBM had put some little thing in there that was particular to SNA," he said. "And eventually we'd figure it out, but we were always behind."
Metcalfe explained that IBM, being by far the dominant industry player in those far-away days, exploited its power. "IBM's arrogance was that they could make standards," he said. He recalled an old joke he he said he liked to tell that added a third type of standard beyond the familiar de facto – a standard by market acceptance – and de jure – one laid down by a standards-setting organization such as the IEEE.
"There was a third kind of standard in those days which I called 'de IBMo', which was a standard that was not in the market, and was not made by a legitimate standards body," he said. "IBM announced Token Ring – but they weren't shipping it, so it wasn't de facto, and it wasn't [a de jure] standard yet until they joined IEEE, so it was a de IBMo standard."
But IBM lost the power to create industry standards simply by announcing them during the mid-80s, he asserts – and from his point of view, Ethernet was the nail in the coffin of Big Blue's dominance. When we asked what exactly that nail was, he told us that "You can invent 20 or 30 answers to this question, including the superior technology of Ethernet," he laughed.
More to the point, however, was that Ethernet, as he put it, "understood its place," while the competition did not.
"Ethernet was developed in the context of the internet with its seven levels of the ISO reference model," he said. "So we had a job to do at level one and two, and we didn't burden Ethernet with all the other stuff that we knew would be above us. We didn't do security, we didn't do encryption, we didn't even do acknowledgements."
It's not that Ethernet was incapable of handling acknowledgements, Metcalfe said, it's just that he and his cohorts wanted to keep things simple. "Ethernet carried packets. Now, they could be acknowledgement packets, or not, whereas the other methods built some sort of elaborate acknowledgement scheme to boost reliability and so on."
As a result, "By virtue of knowing our place, we built something that turned out to be faster and cheaper."
And more scalable, he believes. With Token Ring, he said, "you do have to wait for that token to zip around, and of course the more places it's gotta visit, the probability of its breaking goes higher."
Another Token Ring failing, he said, was that "They never modeled what happens when you lose the token; they always assumed that the token was going to be there."
When we asked if there wasn't some method to store, check, and re-forward the token, he told us that there was. "There were a lot of mechanisms. They needed a mechanism to detect that the token hadn't been around for awhile. Then you need a mechanism for deciding who was in charge of putting the token back on the ring – and in that time the network's been down for a minute."
From Metcalfe's point of view, the Token Ring v Ethernet war began in the IEEE in 1980 and '81. One big break was when, in December of 1982, 19 different companies agreed to use a specific Ethernet spec, despite that it hadn't be approved by the IEEE and wouldn't be for few more years. "As soon as we shipped product – we shipped for the IBM PC using that standard – that was a big moment," he said. "But that wasn't the end of the war."
In those days Ethernet still required coax, albeit thin coax. "The twisted-pair came in the middle of the 80s – I'm not sure what the particular day was," Metcalfe said. "That was one of the advantages that Token Ring had, that it was twisted-pair, so coax was a negative. But as soon as we had twisted-pair," he laughed, "we were running faster at half the price.
That wasn't the end of the war, of course, but it marked the beginning of the end. "IBM struggled to prove that their 4 megabits per second was faster than our 10 megabits per second," he recalled. "They kept saying, 'Well, there are all these collisions.' No."
IBM then bumped Token Ring's speed up to 12 megabits per second when Ethernet was at 10 megabits per second, but doing so didn't halt Ethernet's adoption. "Twisted pair killed Token Ring," Metcalfe said, "plus the fact that it was expensive."
"I think it was Ethernet that finally brought IBM down in the mid-80s," he said.
Forty years down, more to go
Since defeating Token Ring, Ethernet has gone on to not only become the computing industry's default networking protocol, it has also been improved many times over.
The first Ethernet LAN created by Metcalfe and his colleagues at PARC ran at 2.94 megabits per second. In 1978, Xerox developed X-Wire, an Ethernet LAN that ran at 10Mbps, and Metcalfe's 3Com began shipping 10Mbps parts in 1981. In 1992, Grand Junction Networks started up to develop 100Mbps – aka "Fast" Ethernet – then Gigabit Ethernet came along in the late 1990s, and in June 2010 the IEEE approved its standard for 40/100 Gigabit Ethernet.
The reason for these improvements may seem obvious to most of us, but in Metcalfe's experience each inflection prompted some folks to say that enough is enough. "Every single time – at 2.94, at ten, at a hundred, at a thousand, at ten-thousand – there's always a chorus of people saying, 'We, we already have enough bandwidth, we don't need any more, it's too expensive, and the stuff we have we barely use, and blah, blah, blah'," he said.
Those people, Metcalfe asserts, don't see what usage models are coming down the pipe. To illustrate his point, he told us a story from back in the mists of time, when Ethernet was still in its infancy:
I was on a panel in 1976. There were three of us on the panel – two were commercially oriented marketing people and then there was me from research. I'm vaguely remembering this, so I may have my numbers wrong, but one of them was selling the dominant 2400-baud modem, and the other one had new technology running at 4800.
And the argument was this: 'At 2400, the characters go by on the screen faster than you can read. Why would you want 4800?' And then I was the third speaker – 2.94 megabits per second – so clearly we were going to carry a different kind of traffic, we were not going to carry scrolled ASCII characters.
So I think that's how this elasticity thing has been sustained over the decades. Eventually, I suppose, maybe that won't be true – but it's been true for 40 years.
And in May there'll be a party to celebrate those 40 years. In addition to inviting other contributors to the original Ethernet project – "who have been annoyed for decades that I get all the credit," he said – Metcalfe has also invited some of his adversaries in the war between Token Ring and Ethernet.
We asked why proponents of Token Ring would accept an invitation to an Ethernet celebration. "I don't know. Some of them still don't like me, and they want to come just to take their last shot," Metcalfe mused.
"They'll probably argue that Ethernet resembles what they were proposing more today than the original Ethernet did – and they might be right." ®
After our interview, Metcalfe and your humble Reg reporter were joined by PARC's CEO Steve Hoover, and the conversation turned to Ayn Rand and the Atlas Shrugged movies, which Hoover asked Metcalfe if he had seen. "I have, both one and two, and am now looking forward to three," Metcalfe said.
"The people who don't like Ayn Rand thought it was a terrible movie, and the people who think the world of Ayn Rand, like I do, think it was a great movie. They can't talk to each other," he said, adding, "I've read Atlas Shrugged like six times, a long time ago."
Wrapping up his comments on Rand's magnum opus, Metcalfe said "Let's hope that California does not succeed in killing Silicon Valley, as it is earnestly trying to do. Let's hope John Galt intercedes somehow."