The information society
Ewan Sutherland
The computer industry
Industry sectors
- hardware
- software
- services
World market shares by sectors
Collision with telecommunications and entertainment.
The infobahn
Hardware
- supercomptuers
- mainframe computers
- minicomputers
- workstations
- personal computers
System/360
- launched in 1964
- 360 degrees - to encompass all applications
- US $ 750,000,000 engineering development costs
- US $ 4.5 billion in five new factories including one chip
factory
- brought in US$ 16,000,000,000
- redefined the concept of business computing
Barriers to entry in 1960s
- R&D costs of hardware
- development costs of operating system and application software
- need for economies of scale
- need for experienced sales force
Seymour Cray
- Control Data Corporation built Cray a laboratory in Chippewa
Falls, Wisconsin
- effectively a recluse
- launched the CDC 6600 in 1963
- meanwhile CDC had diversified into more commerical machines
Cray Research
- founded Cray Research in 1972 and left CDC in 1973
- made the world's most powerful computers
- used for processor intensive applications
- later left Cray Research to found new company, Cray Computer,
which has yet to deliver a single machine
Digital Equipment Corporation (DEC or Digital)
- Ken Olsen founded DEC in 1957 with his brother and Harlan
Anderson
- Olsen had worked in the Massachussetts Institute of Technology
(MIT) computer laboratories he had also worked for IBM on behalf
of MIT
- the "non-computer"
- launched the PDP-1 in 1961
- niche market below IBM
- the minicomputer
- major company in mid-1960s
- launched VAX range in 1979
- featured clustering and networking
- VMS operating system (competing against Unix)
- failed in the PC marketplace
- now trying the Alpha RISC chip
Graph of DEC profit and loss
Wang Laboratories
- founded by Dr An Wang
- sold his patent for magnetic core memory to IBM
- manufactured electronic calculators
- minicomputers (from 1964)
- standalone word processors
- office automation
- personal computers
- entered "Chapter 11" bankruptcy in 1992
- emerged again in 1994 without manufacturing capacity
Apple Computer
- Steve Wozniak and Steve Jobs
- started in a garage in Silicon Valley
- Apple I (1976)
- Apple II (1977)
- brought in John Sculley from PepsiCo
- LISA failed in 1983
- Steve Jobs left to form Next Computer Inc
The Apple Macintosh "borrowed" the user interface from Xerox
- windows
- icons
- menus
- pointers
Macintosh launched ten years ago
- head-to-head with IBM
- legendary "Big Brother" advert
- niche market for desktop publishing
Xerox Corporation
A major photocopier company, founder of the industry. Establised
a "blue skies" research centre in Palo Alto California,
where the staff developed:
- WIMPs
- laser printer
- desk top publishing
Then failed to exploit them...
The IBM PC
- "standard product"
- launched in August 1981 after one year Project Chess high-speed
development
- Apple Computer welcomed IBM to the market
- DOS, processor, etc made from off-the-shelf components
The clone wars
- rivals tried variations
- then realised it was a "standard" and copied it
- companies such as COMPAQ which produced the portable version
- hundreds of companies tried to make "clones"
- no longer premium prices, but "brands" are still
significant
The empire fights back
- IBM lost market share
- IBM lost design control
- IBM went proprietary: Personal System/2 and Operating System/2
- IBM misjudged the market
The fall of IBM is not the conventional American story of underinvestment,
slipshod standards, and lagging technology. Even with all the
recent defections, it still has outstandingly talented employees.
Its rate of investment through the 1970s and 1980s was massive;
R&D spending in the last half of the 1980s averaged almost
$6 billion a year. No other company in the history of the world
has spent that much. But still the loss of market share has been
precipitate in every major product category.
Ferguson and Morris (1993) p. 84.
IBM Market shares in 1983
Changes in market shares between 1986 and 1991
RISC computing
- Reduced Instruction Set Computing (RISC)
- simpler and faster
- conceived at IBM in late 1970s
- 1983 launched by Ridge and Pyramid Corps
- MIPS
- IBM RT workstation (failed) in 1987
- also in 1987 Hewlett-Packard and Sun SPARC
- re-launched by IBM as RS6000
Underlying causes of IBM's failure:
- bureaucracy
- fear of anti-trust action
- failure to understand implications and applications of new
technologies
- flexibility of rivals in Silicon Valley
- AT&T is finally a rival through vertical integration
Ramsey Clark, the attorney general of the United States during
the last waning days of the Johnson administration, on January
17, 1969, signed a complaint charging IBM with unlawful monopolization
of the computer industry and requested that the federal courts
dismember the company.
Ferguson and Morris (1993), p 10.
An extensive review of the IBM case was an early priority of the
Reagan administration Justice Department. The suit was dropped
in June 1982 with a curt four-sentence appraisal by the solicitor
general that the case was "without merit."
Ferguson and Morris (1993) page 11.
The rise of the Japanese
First in cars, motorbikes and then in televisions, video
cassette recorders.
Japanese exports of VCRs to the USA
Finally Japan has been exporting:
- computers
- memory chips
- flat screens
- lap-tops
Shares of the world market for microprocessors
Selling chips
- how do you keep up with developments
- defence by law suits
- how do you sell a product the customer cannot see
- cross-licencing (dual source and to increase availability)
Long-term product planning is dangerous ... Three years is long-term.
Even two years may be. Five years is laughable.
Ed McCracken, CEO Silicon Graphics in Harvard Business Review
We don't pay a lot of attention to whether or not a product is
still gaining market share when we decide to launch a new product.
This approach is the only way to ensure that no one gets to the
market first with a faster, better, cheaper product. When someone
beats us to the market, which has happened only a very few times,
we are embarassed.
Ed McCracken, CEO Silicon Graphics in Harvard Business Review
Software
- operating systems
- networking
- generic applications
- bespoke applications
Key factors in company formation
- technological knowledge
- entrepreneurial flair
- availability of venture capital
- availability of managerial and marketing expertise
- individual and corporate role models
- a 'supportive' and conducive environment
Silicon Valley
- Santa Clara County, California
- hot bed of new technology
- thousands of start-ups and spin-offs
- close links to world-class universities
- continuing success
Japanese Keiretsu
- massive combines
- complete range of manufacturing and products
- few discernible individuals
- inexorable advances
- successfully copied in Korea, Singapore and Malayasia
Conclusion
- fiercely competitive
- rapidly advancing technologiies
- merger of sectors (previously distinct)
- no defensive walls
- shift from hardware to software and services
- two models (Silicon Valley and Japan)
Readings
- Campbell-Kelly (1990) "ICL" OUP, London
- Wallace and Erickson (1992) "Hard Drive" Wiley,
New York
- Sculley, John (1987) "Odyssey" Fontana, Glasgow
- Bauer et al. (1992) "The Silverlake Project"
OUP, New York
- Ferguson, C H and Morris, C R (1993) "Computer Wars"
Times Books, New York
URLs
Copyright © Ewan Sutherland, 1995.