DEC Essay, Research Paper
Digital Equipment Corporation (DEC)
Overview
Digital Equipment Corporation (DEC) is
one of the most successful computer manufacturers in the world. Throughout
its nearly 40-year history, it has struggled through the ups and downs that
have plagued virtually every company in the computer industry. Unlike many
others, however, DEC has managed to reemerge time and again as a top player
in the industry.
The Early Years
The first 35 years of DEC’s history
can be tied directly to the history of its founder, Ken
Olsen, who was once hailed by Fortune
magazine as "America’s Most Successful Entrepreneur." In an
industry that regularly chewed up and spit out most of its founders and
entrepreneurs, it’s amazing that he managed to stay at the helm of his
company for 35 years. Equally amazing is that DEC has managed to stay
intact, as well as successful, for four decades.
In 1957, Ken Olsen and Harlan Anderson were engineers at the Massachusetts
Institute of Technology (MIT) Lincoln
Laboratory. Both had spent several years working with and on big mainframe
computers. As part of his duties, Olsen had served as liaison between
MIT and IBM at the IBM plant. Two things
were clear to Olsen when he finished this assignment – he wanted to build
computers and he believed he could do it better than IBM. In addition,
Olsen believed computers should be brought out from behind the protective
glass walls built by IBM and into the main offices where multiple individuals
could interact with them.
Olsen and Anderson knew engineering, but not much about running a business.
Since they needed $70,000 to start up their business, they sought help
from American Research & Development (ARD), a venture capital firm.
ARD agreed to the loan with the provision that it retain a 70 percent
interest in the company and the head of ARD, General Georges Doriot, would
be an advisor in all business matters. For the next 30 years, Doriot became
Ken Olsen’s mentor, counselor, and guide and a key part of DEC’s success.
One of Doriot’s first pieces of advice was for the small company not
to draw the attention of companies like IBM until they had established
themselves. To this point, Doriot suggested that the company start out
making just printed circuit modules, not full blown computers, and change
the company’s name from ‘Digital Computer Corporation’ to ‘Digital Equipment
Corporation’. When the company opened in August 1957 in the tiny mill
town of Maynard, Massachusetts, it was as an electronics company.
Conservative by nature, Olsen insisted on one extravagance for the new
company’s products. His modules would be built using the newly developed
and expensive transistors. In early 1958,
DEC shipped its first products – Digital Laboratory Modules and Digital
Systems Modules. After one year in business, the company had sold $94,000
worth of logic modules for memory testing. Once Olsen had proven himself
and his company, he again approached Doriot and ARD about building computers.
This time they gave their blessing.
The Origin of the Minicomputer
Olsen realized that many of the tasks being performed by the huge mainframe
computers could easily be performed by smaller "fundamentalist"
computers, and that’s what he wanted DEC to produce. To seem less threatening
to the computer giants, DEC’s first computer came out in 1959 under the
innocuous name Programmed Data Processor or PDP-1.
Although it appeared to be just another advanced logic module, it really
was a solid-state, general-purpose computer, complete with its own keyboard
and a cathode ray tube (CRT) that allowed the users to see what was being
entered into and received back from the central processing unit. It was
the first commercial machine to use transistors instead of vacuum
tubes, which allowed it to fit into a space the size of a refrigerator.
Designed for individual interaction, it provided more power than most
users expected or needed, and at a price they could afford.
Initially, sales of the PDP-1 were slow, primarily because each unit
had to be customized for the specialized function it would perform. The
professors and scientists who used it were sufficiently computer literate
to develop some of their own software and had enough company support or
grant money to afford the modest $125,000 to $150,000 price tag. Then
in late 1962, DEC received an order from International Telephone and Telegraph
(ITT) for 15 PDP-1 units to control its message switching systems. This
large order, and several that followed from ITT, firmly set DEC in the
marketplace as a computer company. Meanwhile, Olsen donated a PDP-1 to
MIT with the sole purpose of getting engineering students to interact
with a computer as soon as possible. The students were ecstatic and eventually
DEC computers became common in universities and schools around the world.
After five years in business, DEC reported sales of $6.5 million and
net profits of $807,000. But as the company grew, the management style
that worked in 1957 was beginning to flounder in 1963. Mismanagement led
to a drop in sales and the steady growth of the first five years came
to a standstill. The two founders, Olsen and Anderson, were at odds on
how to solve the company’s problems, which eventually led to Anderson
leaving the company.
Matrix Management
As Olsen agonized over the problems in manufacturing and the divisions
in management, he hit upon a concept of organizational structure that
would soon become DEC’s trademark. The structure, known as matrix management,
although not entirely unique to DEC, became Olsen’s baby and the solution
that would move his company from chaotic infighting to explosive growth.
Simply put, matrix management divided the company and individual responsibilities
between product lines. A senior executive took ownership of each product
line and was responsible for developing the product, marketing it, and
turning a profit. Resources from central functions, such as sales, manufacturing,
and marketing, were shared by line managers, thereby opening up communications
between all groups. This matrix structure encouraged such concepts as
corporate democracy, consensus decision-making, and creativity at all
levels. Under the new management scheme, profits again began to climb.
The PDP-8
In 1966, the first product to be released under the new management, the
PDP-8, was
very successful and became a mainstay of the company for years to come.
As with all DEC equipment, the PDP-8 included a high-quality video display
terminal (VDT), but it was also small enough to sit on a table top. And
it was cheap, only $18,000, which was far below any comparable machines
being sold in the industry. Thousands of PDP-8s were installed in small
businesses, universities, high schools, newspapers, and book publishers.
Many baby boomers got their first taste of computing on one of these machines.
When one of DEC’s salesmen took the new, small computers to London to
try to establish a presence there, he drew a connection between the current
craze of miniskirts seen all over that town and the new "mini"
computers he was hawking. The phrase caught on and industry publications
began referring to the new computers as minicomputers.
The PDP-8 was also responsible for opening up a whole new market and
new style of selling in the industry. Original equipment manufacturers
(OEMs), such as scientific instrumentation makers or typesetting companies,
bought PDP-8 units, attached their own hardware to it, wrote software
to support their particular application, and sold the resulting package
as their own product. Soon, OEM business accounted for 50 percent of all
DEC sales.
As DEC recorded more sales and achieved higher profits, other computer
manufacturers quickly jumped into the minicomputer business. By 1970,
nearly 70 companies were manufacturing minicomputers, but none with as
much success as DEC.
8-Bit Words
Just as things were starting to look up for the company, the next challenge
came. DEC was creating computers that used 12- and 18-bit words. The industry,
though, was moving toward the 8-bit word with its various multiples as
an industry standard. In 1967, a group of DEC engineers led by Ed de Castro
was assigned the task of designing a 16-bit computer. Their final plan
contained a basic 16-bit system that could be grown to 32-bits as well
as the development of a series of compatible products that would allow
users to upgrade their existing machines rather than replace them. But,
what de Castro’s group was suggesting amounted to scrapping the entire
DEC product line and replacing it with the new 16-bit machines.
DEC’s management soundly rejected de Castro’s computer and his long-range
plan. So, in April 1968, de Castro and two other engineers found their
own venture capital, left DEC, and started their own company, Data
General Corporation, to produce 32-bit computers. In 1969, Data General
was one of the hottest new companies in minicomputer manufacturing, tapping
a market that could have been DEC’s.
The PDP-11
When de Castro’s group left, DEC still didn’t have a 16-bit computer
product and no longer had an engineering team that could develop it. A
new engineering team was hired and development began again. It was a slow
process for the new engineers, but the result was the PDP-11, which would
prove to be the precursor of a new era in DEC computers. The final PDP-11
plan actually encompassed a whole family of minicomputers with larger
memories and more processing power than any small machine previously built
by DEC. In addition, the PDP-11 was simple to use, while covering a wide
range of computing performances.
Although it was late arriving on the market, the PDP-11 was a major success.
By 1971, DEC was selling 100 machines per month. DEC quickly brought out
two more models and by 1972, DEC was back on top of the minicomputer market.
Bad Decisions
But all of DEC’s decisions weren’t good ones. In the spring of 1972,
the PDP-11 group at DEC proposed a new concept they called the DEC Datacenter.
The Datacenter combined a PDP-11/20 with a VT05 terminal and a printer
on a desk. The intent was to bring a computer to all people, from scientists
and technicians to secretaries and bookkeepers. This "individual"
computer never made it to the product line and was eventually scrapped.
In a similar story, David Ahl, a DEC employee in the new products division
of the company, proposed a small computer that could be sold to individuals
such as doctors or engineers and to small businesses as well as to schools.
Although they were probably too expensive for most homes, Ahl proposed
that the Heath company create a kit version that would make them more
affordable. Olsen and the DEC management didn’t think there would be a
market for either product. For DEC, these were missed opportunities, but
they were also being missed by most of the major computer manufacturers
at the time.
The VAX Strategy
In the spring of 1977, IBM finally came out with its version of the minicomputer,
the Series 1. Although IBM was a major competitor, DEC felt secure in
its established customer base. In addition, DEC had a new weapon in its
arsenal, VAX,
short for Virtual Address Extension. Since 1975, DEC had been developing
a 32- bit minicomputer designed to provide virtually unlimited address
space (memory). The name VAX-11 was chosen to ensure customers that the
VAX was compatible to their PDP-11 units. On October 25, 1977, the first
VAX prototype came off the production line. Along with the VAX architecture
came a proposal from the designers called the VAX Strategy. Basically,
the VAX Strategy revolved around compatible machines being tied together
to form a huge network. For years, DEC had been developing its own networking
setup, called DECnet. Networking was something IBM’s machines could not
do.
Although DECnet worked well enough internally, Ethernet,
a system developed by Bob Metcalfe at Xerox’s
Palo Alto Research Center (PARC),
was more encompassing. The designers of the VAX Strategy chose Intel
Corporation, a leading Silicon Valley chip maker, to manufacture the
semiconductor chip on which to base Ethernet. In May 1980, DEC, Xerox,
and Intel announced their Ethernet plans to the world.
The Personal Computer Problem
Also in 1980, virtually every product-line manager at DEC was pitching
some type of personal computer for DEC to begin manufacturing. DEC had
the beginnings of a personal computer in the Components Group’s PDT (Programmable
Data Terminal), which included a computer terminal with built-in intelligence,
a CRT screen, and a keyboard. Olsen finally agreed to develop a small,
individual computer. However, the new product was called an "applications
terminal and small system" – the label "personal computer"
was not allowed at DEC.
Unfortunately, Olsen and DEC still didn’t quite understand the PC concept.
Instead of constructing an affordable, quickly-produced machine that would
appeal to a large market of new users, DEC spent 18 months coming up with
an elegant computer terminal that would not only stand alone but could
also be networked with other DEC systems. To complicate things further,
DEC continued its corporate policy of manufacturing all the component
parts itself. In contrast, IBM was purchasing MS-DOS
from Microsoft and drives, monitors
and circuit boards from the Far East.
Then, DEC made another tactical error. Instead of throwing all of its
support behind one product, it hit the market with three products. In
May 1982, DEC unveiled the Professional (its original PC concept), the
Rainbow (a low-end backup product that used MS-DOS), and DECmate II (the
latest version of its dedicated word processor). Although the Pro was
a high quality product, its high price took it out of most users’ range
and it sold poorly. The Rainbow, though more reasonably priced, didn’t
offer anything not already available from IBM. Although the Rainbow had
decent sales, most were to DEC’s already established customer base.
Reorganization
By 1983, DEC was struggling. It was now a very large company with several
products and was having trouble managing itself. In response to all of
the internal problems, Olsen reorganized the company. Many upper and middle
managers left the company, and employees floundered as they tried to fit
into new jobs with new bosses and new procedures. Schedules weren’t met,
orders were incorrect, and billing wasn’t accurate. Earnings were dropping
significantly. But even DEC was taken by surprise when it realized that
its third quarter earnings for 1984 were down 72 percent from the year
before. The next day, October 18, DEC’s stock plunged 21 points. This
day became known as "Black Tuesday" in DEC’s history.
Although things were bleak, most of DEC’s customers had so much invested
in their computer systems that they couldn’t do much but stay with the
company. But new customers were leery and DEC’s already limited success
in the PC market quickly dissipated. DEC quietly pulled out of the competition
as IBM and Apple fought for domination
of that area.
However, there was one positive result. By moving away from PCs, DEC
again refocused its time, energy, and money on the VAX product line. In
May 1983, DEC had unveiled the VAX-cluster, a means of hooking several
VAX minicomputers together in a proprietary local-area network, in essence
allowing them to become the equivalent of a mainframe. In October 1984,
DEC finally released its flagship VAX product, the VAX8600. The 8600 marked
the beginning of the second generation of VAX machines, with four times
as much power as DEC’s top-of-the-line machines.
As DEC brought out more and more components for its VAX system, it kept
preaching the benefits of the Ethernet and quietly built links to other
vendors’ equipment, including IBM, which landed them important new big
accounts. By 1985, DEC’s star was on the rise again, but this time it
wasn’t with a single product like the minicomputer in the 1960s, but rather
a concept – networking. Corporations were beginning to see that computers
could do much more for them than just crunch numbers or write letters,