September 01, 2016
Digital Man/Digital World
I'm a big fan of documentaries about the early years of the personal computer industry. Digital Man/Digital World provides a much-needed look at the often overlooked DEC (it can be watched online at the above link).
Before Intel, before Microsoft, before Apple, before the IBM PC and Compaq Computer (the company that later acquired it), there was Digital Equipment Corporation. The new reality that a computer could be "small enough to be stolen" (based on an actual incident) began not in Silicon Valley but in Maynard, Massachusetts.
Digital Equipment Corporation (DEC) was founded by Ken Olsen in 1957.
Ken Olsen had served in the Navy during WWII as a radar technician. After the war, he earned a degree in electrical engineering at MIT, where he worked on the Whirlwind project. The Whirlwind computers powered the prototypes of the Distant Early Warning (DEW) Line during the early years of the Cold War.
Olsen had acted as a liaison with IBM during the Whirlwind project (IBM build the computers that MIT designed), and recalled that IBM was "like going to a communist state." He brought that attitude with him when he founded DEC.
Although a conservative Christian who wore a tie, banned alcohol from company parties, and always ate dinner with his family (before going back to work), Olsen was one of the first "hippie" CEOs. He championed the flat corporate hierarchy, the employee-friendly workplace, and the customer-oriented business.
He drove a sedan and vacationed in the Canadian wilderness. Abandoning the organizational chart, he managed by walking around, and was on a first-name basis with his employees. He actively recruited women and minorities (this was in the 1960s and 1970s) and didn't lay off a single person until 1988.
In committee meetings, managers were expected to own their ideas, defend them, and fight things out: "You could get in somebody's face as long as you didn't stab them in the back."
DEC was the first high-tech company funded by venture capital (American Research and Development Corporation) and produced a crop of multi-millionaires when it went public in 1966. After Ken Olsen retired (involuntarily), he gave away most of his accumulated wealth.
DEC's truly disruptive innovation was the minicomputer. Instead of IBM's room-sized mainframes, the PDP-1 was a filing-cabinet sized time-sharing computer that cost less than $120,000, a bargain back in the early 1960s. The later PDP-8, introduced in 1965, shaved more than $100,000 off that price.
The 16-bit PDP-11 was the first computer to tie internal communications together on a shared bus, a feature later adopted by the Altair, the Apple II and the IBM PC. The 32-bit, network-ready VAX debuted in 1977 and became its most popular minicomputer, a mainstay of university engineering labs.
By the early 1980s, Digital had become the second-largest computer company in the world. In one of the great ironies that typify the last half-century of the tech industry, while disrupting the staid mainframe business and making computers truly affordable, DEC sowed the seeds of its own downfall.
This trend in affordability accelerated in the 1970s with the advent of a slew of inexpensive 8-bit CPUs that powered the Altair, the Apple II, and the Commodore, with the Intel and Zilog varieties running the soon ubiquitous CP/M operating system.
IBM responded to the PC threat with the 16-bit IBM PC, engineered in its freewheeling Boca Raton division using low-cost OEM components and a second-hand operating system from an upstart software company called Microsoft. It produced a smash hit product that eventually sowed the seeds of its own downfall too.
DEC went in the opposite direction, sinking resources into the VAX 9000 supercomputer, high-end multiprocessor microcomputers, and the Alpha 64-bit RISC processor. This was a decade before 64-bit computing would arrive in an affordable PC package. Everybody loved the Alpha but nobody knew what to do with it.
Powered by Intel's inexpensive x86 chips, the PC was growing so fast that "good enough" quickly became more than enough to do the job. Before long, personal computers were easily matching the power of DEC's previous minicomputers. The PC had turned into a minicomputer.
In a complete turnaround (just as Clayton Christensen would have predicted), DEC found itself defending the high end of the market and getting disrupted from below.
DEC's premium hardware failed to find a market, resulting in a $2.8 billion loss in 1992. That year, Olsen was ousted as CEO. Compaq acquired Digital in June 1998, only to merge with HP four years later. DEC all but disappeared amidst the corporate reorganization rubble.
Though largely forgotten, its influence lives on.
When Paul Allen and Bill Gates developed a BASIC interpreter for the brand-new Altair in 1975, they didn't have an Altair computer. Instead, they used an Intel 8008 emulator Allen had written for the DEC PDP-10 in Harvard's Aiken Lab. Amazingly, the program ran the first time it was installed on an actual Altair.
BASIC was Microsoft's founding product, its first best-selling product, and would later be adopted by both IBM and Apple. And it was born on a DEC minicomputer.
Then in 1988, Microsoft hired Dave Cutler, architect of the VMS operating system for the DEC VAX, to design a preemptive multitasking OS. Since the release of Windows XP, all desktop and server versions of the Microsoft OS (plus Windows Phone since version 8) have been built on Dave Cutler's NT kernel.
Our modern technological world was in no small part created by Ken Olsen and Digital Equipment Corporation.
Before Intel, before Microsoft, before Apple, before the IBM PC and Compaq Computer (the company that later acquired it), there was Digital Equipment Corporation. The new reality that a computer could be "small enough to be stolen" (based on an actual incident) began not in Silicon Valley but in Maynard, Massachusetts.
Digital Equipment Corporation (DEC) was founded by Ken Olsen in 1957.
Ken Olsen had served in the Navy during WWII as a radar technician. After the war, he earned a degree in electrical engineering at MIT, where he worked on the Whirlwind project. The Whirlwind computers powered the prototypes of the Distant Early Warning (DEW) Line during the early years of the Cold War.
Olsen had acted as a liaison with IBM during the Whirlwind project (IBM build the computers that MIT designed), and recalled that IBM was "like going to a communist state." He brought that attitude with him when he founded DEC.
Although a conservative Christian who wore a tie, banned alcohol from company parties, and always ate dinner with his family (before going back to work), Olsen was one of the first "hippie" CEOs. He championed the flat corporate hierarchy, the employee-friendly workplace, and the customer-oriented business.
He drove a sedan and vacationed in the Canadian wilderness. Abandoning the organizational chart, he managed by walking around, and was on a first-name basis with his employees. He actively recruited women and minorities (this was in the 1960s and 1970s) and didn't lay off a single person until 1988.
In committee meetings, managers were expected to own their ideas, defend them, and fight things out: "You could get in somebody's face as long as you didn't stab them in the back."
DEC was the first high-tech company funded by venture capital (American Research and Development Corporation) and produced a crop of multi-millionaires when it went public in 1966. After Ken Olsen retired (involuntarily), he gave away most of his accumulated wealth.
DEC's truly disruptive innovation was the minicomputer. Instead of IBM's room-sized mainframes, the PDP-1 was a filing-cabinet sized time-sharing computer that cost less than $120,000, a bargain back in the early 1960s. The later PDP-8, introduced in 1965, shaved more than $100,000 off that price.
The 16-bit PDP-11 was the first computer to tie internal communications together on a shared bus, a feature later adopted by the Altair, the Apple II and the IBM PC. The 32-bit, network-ready VAX debuted in 1977 and became its most popular minicomputer, a mainstay of university engineering labs.
The PDP-11 control console (top) looks like a "computer." Built for computer geeks, the Altair front panel resembled the PDP-11 console. |
By the early 1980s, Digital had become the second-largest computer company in the world. In one of the great ironies that typify the last half-century of the tech industry, while disrupting the staid mainframe business and making computers truly affordable, DEC sowed the seeds of its own downfall.
This trend in affordability accelerated in the 1970s with the advent of a slew of inexpensive 8-bit CPUs that powered the Altair, the Apple II, and the Commodore, with the Intel and Zilog varieties running the soon ubiquitous CP/M operating system.
IBM responded to the PC threat with the 16-bit IBM PC, engineered in its freewheeling Boca Raton division using low-cost OEM components and a second-hand operating system from an upstart software company called Microsoft. It produced a smash hit product that eventually sowed the seeds of its own downfall too.
By comparison, the IBM PC looks like an appliance. |
DEC went in the opposite direction, sinking resources into the VAX 9000 supercomputer, high-end multiprocessor microcomputers, and the Alpha 64-bit RISC processor. This was a decade before 64-bit computing would arrive in an affordable PC package. Everybody loved the Alpha but nobody knew what to do with it.
Powered by Intel's inexpensive x86 chips, the PC was growing so fast that "good enough" quickly became more than enough to do the job. Before long, personal computers were easily matching the power of DEC's previous minicomputers. The PC had turned into a minicomputer.
In a complete turnaround (just as Clayton Christensen would have predicted), DEC found itself defending the high end of the market and getting disrupted from below.
DEC's premium hardware failed to find a market, resulting in a $2.8 billion loss in 1992. That year, Olsen was ousted as CEO. Compaq acquired Digital in June 1998, only to merge with HP four years later. DEC all but disappeared amidst the corporate reorganization rubble.
Though largely forgotten, its influence lives on.
When Paul Allen and Bill Gates developed a BASIC interpreter for the brand-new Altair in 1975, they didn't have an Altair computer. Instead, they used an Intel 8008 emulator Allen had written for the DEC PDP-10 in Harvard's Aiken Lab. Amazingly, the program ran the first time it was installed on an actual Altair.
BASIC was Microsoft's founding product, its first best-selling product, and would later be adopted by both IBM and Apple. And it was born on a DEC minicomputer.
Then in 1988, Microsoft hired Dave Cutler, architect of the VMS operating system for the DEC VAX, to design a preemptive multitasking OS. Since the release of Windows XP, all desktop and server versions of the Microsoft OS (plus Windows Phone since version 8) have been built on Dave Cutler's NT kernel.
Our modern technological world was in no small part created by Ken Olsen and Digital Equipment Corporation.
Related posts
The accidental standard
MS-DOS at 30
The grandfathers of DOS
The future that wasn't
Something Ventured
Labels: computers, tech history, technology
Comments
DEC lived on in another way. The architect for Windows NT (what eventually became Windows XP/7/8/10) was an ex-DEC [kernel] architect, Dave Cutler. His influence is now seen in ALL major operating systems, not just Windows.
BTW, VMS was a marvelous operating system and so much better than anything IBM ever produced. I'd also suggest that what did them in the most wasn't disruption, but good old competition from IBM and the AS/400. For big data, main/mini-frames were still the only real solution in the 90s. (For super-big, high transaction, secure data, that's still true today.) It didn't help any that Ken and DEC didn't rid themselves of all their dinosaur managers in the early 90s and the non-dinosaurs bailed (see Xerox for another example.
BTW, VMS was a marvelous operating system and so much better than anything IBM ever produced. I'd also suggest that what did them in the most wasn't disruption, but good old competition from IBM and the AS/400. For big data, main/mini-frames were still the only real solution in the 90s. (For super-big, high transaction, secure data, that's still true today.) It didn't help any that Ken and DEC didn't rid themselves of all their dinosaur managers in the early 90s and the non-dinosaurs bailed (see Xerox for another example.