ÐÓ°ÉÔ­´´

MOST OF us learn the hard way that research and development do not always
succeed. Most doomed concepts die young, the victims of insufficient funding,
or of clear signs of technical or economic impracticality. Others stagger
onward, but soon lose the race for funding to better and brighter ideas.
A few reach a critical mass, and survive to blunder on through bureaucratic
jungles until economic reality or, more often, new management arrives to
put them to death.

Americans call such doomed programmes ‘boondoggles’, a word coined by
an American scoutmaster, a Mr R. H. Link, to describe futile and unnecessary
projects. The typical boondoggle begins as a bright idea that strikes a
resonant chord with management, funding agencies or politicians. It may
promise to solve some problem, often a military one, or to yield great profits.
Frequently, the early stages of the programme demonstrate some progress,
or at least produce reports claiming to have made some progress. The people
who control the purse strings hear what they want to hear, and gleefully
throw money at the programme. The staff grows rapidly, and the people who
originated the idea, or who seized upon it early, find themselves at the
top of a budding bureaucratic empire that has sprouted virtually overnight.

Unfortunately, empire growth has no correlation with project feasibility.
Typically the science works but the engineering doesn’t, and this starts
a long-running series of delays to the programme. A weapon system may sound
deadly on paper; but the prototype misses its target by a mile. Enthusiasts
and hangers-on quickly reassure the boondoggle’s sponsors that only a little
more development is called for. Perhaps all that is needed is a new material,
albeit one that is twice as strong and half as heavy as any known. But if
by some miracle the engineer can make such stuff, it is bound to be impossible
to cut with ordinary tools. That, in turn, sets off other research efforts.
Nothing is impossible – and nothing works correctly. Timetables continue
to slip and costs to soar, but progress reports always see an imminent breakthrough.

The longer a boondoggle survives, the harder management tries to keep
it alive. After all, no one wants to admit to having wasted careers and
a lot of government or corporate money. A boondoggle nurtured by government
bureaucracy can often survive until a change of government or an attack
of the dreaded ‘reform’ virus. Barring changes in top management or total
exhaustion of corporate cash reserves, a private boondoggle may plod all
the way to the marketplace, where potential customers greet its hype with
a long, loud yawn.

Take, for instance, the nuclear-powered bomber, my very own favourite
boondoggle: bred and nurtured by the Pentagon, it represented the American
effort to build a long-range bomber powered by a nuclear reactor. Military
programmes are fertile ground for boondoggles, with everything going for
them. Governments have vast financial resources; national security is a
convenient rationale; and secrecy can conceal misguided programmes.

The concept of the bomber sounds bizarre today. Radiation shielding
would be needed to protect the crew, but it would make the plane too heavy
to get off the ground. A crash could contaminate a large area with radioactive
debris. It is only 20-20 hindsight that makes those problems so obvious
now. The US Air Force began work on nuclear-powered bombers soon after the
Second World War, when reactor design was in its infancy. Military planners
knew that a small mass of uranium or plutonium could generate tremendous
amounts of energy, and they got very excited. Futurists were talking abut
nuclear-powered automobiles. Compact power sources could greatly extend
the ranges of ships, planes and submarines. This was important in the early
years of the Cold War because of the vast distances between the United States
and the Soviet Union. Nuclear power seemed worth studying.

Nuclear reactors did prove feasible for submarines, which could house
the bulky reactors and tolerate their heavy shielding; airplanes were quite
another matter. Many nuclear scientists, including J. Robert Oppenheimer,
a pioneer of the atomic bomb, were sceptical at the outset. Early study
groups warned the task would be difficult, but no one could produce conclusive
evidence of potential folly, so the Air Force and the Atomic Energy Commission
pushed ahead. The effort lasted 15 years, and consumed $1 billion before
the Kennedy administration cancelled it in March 1961, two months after
taking office.

The nuclear-powered aircraft programme did have some modest success,
however. Engineers built a high-temperature reactor, and in early 1956 they
used it to power a jet engine, albeit one bolted firmly to the ground. Other
technologies were advancing so rapidly that the programme never produced
more than a blueprint for an airplane. Aeronautical engineers extended ranges
of conventional aircraft, and intercontinental ballistic missiles brought
the Soviet Union into the range of American nuclear weapons.

By the end of the 1950s, the Air Force’s main reason for building nuclear
aircraft seemed to be one of pique: the Navy had nuclear submarines. Unnamed
government officials told Aviation Week that the Soviet Union was flight-testing
a nuclear bomber, though no one else saw the plane outside the pages of
the magazine. Years later, one former Kennedy adviser called the programme’s
cancellation ‘a sanitary engineering problem . . . like coming downstairs
in the morning and finding a dead walrus in your living room’.

Military agencies do not have a monopoly on boondoggles, however. When
it was part of Britain’s Post Office, British Telecom developed another
of my favourite boondoggles, called viewdata in Britain and videotex in
the US. The idea was for telephone lines to carry information from central
computers to televisions in homes. This all made sense in the early 1970s,
when computer power was growing steadily but the general public had no easy
access to it. Videotex would take advantage of communications equipment
already in homes, and supply only the missing pieces, such as a central
computer with data files and an interface between the telephone and the
television. Visionaries failed to forecast the coming of the personal computer,
which gives users much more power than videotex ever could.

The early development of videotex was deceptively easy. By 1974, Sam
Fedida, a British engineer and one of the pioneers, demonstrated a working
model, and British Telecom began a public trial in 1976. A commercial service,
Prestel, followed in 1979, which I saw ‘live from London’ at the International
Conference on Communications in Boston. Like many others at the meeting,
I was fascinated by the potential of electronic information services. They
seemed to have a great future.

Unfortunately for Telecom, British consumers did not agree. By the end
of 1981, Prestel had only about 15 000 subscribers, many of them travel
agents who used the service to obtain information about transport schedules.
When I looked more closely, I understood why. The flaws one overlooks in
a demonstration haunted Prestel. The telephone-to-television interface units
were expensive. The information service conveyed precious little information.
Each screen showed only about 100 words of text, and the graphics, which
were intended to attract subscribers, were awkward. The service was very
slow over telephone lines, and the interface, which you had to prompt repeatedly
using a menu on screen, was so clumsy it soon became frustrating.

Prestel’s problems did not stop Americans from making the same mistakes,
as well as some of their own. Knight-Ridder, a national newspaper chain,
launched a service called Viewtron in the early 1980s. The graphics were
better than Prestel’s but the service was still slow, the information sparse,
and the terminals cost several hundred dollars each. It was clear that Knight-Ridder
did not understand the market. A Viewtron representative, who must have
gone to a boondoggle finishing school, summed up the difficulties by saying
that Viewtron wanted to be ‘the McDonald’s of videotex’ and cater to upmarket
consumers. I marvelled at the simile. Viewtron management did not understand
that an upmarket audience would not deign to dine on fast-food videotex.

The latest effort to interest Americans in videotex is Prodigy, a joint
venture of IBM and the giant retailer Sears Roebuck, which together have
invested more than $100 million. Introduced in 1988, Prodigy runs on personal
computers with standard modems. The service claimsto have more than 100
000 subscribers, but I am not convinced it will ever reach its goal of two
million. While it hasovercome some of the problems of videotex, Prodigy
isagonisingly slow and each screen conveys little information. Watching
Prodigy assemble itself on screen, I felt like a gourmet faced with a fast-food
menu.

The classic industrial boondoggle was the Edsel, the automobile that
nobody wanted. The vehicle was ugly and unreliable. It cost Ford, the manufacturer,
$500 million and much public embarrassment in the late 1950s. High technology’s
answer to the Edsel was another of my favourite boondoggles, the RCA SelectaVision
Videodisk player.

RCA had started in the 1960s with a good idea: consumers wanted more
choice over television programmes than commercial broadcasters gave them.
At the time, video players did not have sufficient electronic storage capacity
to meet the need and, besides, they were much too expensive for the average
consumer. RCA tried several technologies for producing a home video player,
such as holographic tape, before deciding to develop pre-recorded discs
that could be read electronically. Other companies chose different approaches.
The Dutch electronics giant Philips developed optical discs read with laser
beams, an approach later adapted for digital audio in compact disc players.
Japanese companies developed magnetic videotape recorders, an approach that
was ignored by RCA and other Americans.

The technology race gathered speed in the late 1970s, but I found it
hard to swallow the glittering market forecasts for pre-recorded discs.
I would gladly listen to my favourite music, again and again, but few people
are likely to want to watch even an exceptional movie many times. Those
doubts were strongly reinforced when several marketing and sales managers
for high-technology companies expressed no interest in buying videodisk
players: as well-paid executives deeply involved in electronics they were
prime market prospects.

They were not alone. Philips got to the market first, but its laser
video disc players sold slowly. RCA followed at a lower price in March 1981,
but sales were only a little better. RCA cut prices and advertised heavily,
to no avail. The public wanted video cassette recorders, and many potential
buyers turned away from video discs when they learnt the discs could not
record. By the time RCA abandoned its discs in April 1984, the company had
sold 550 000 players and lost about $580 million. Those losses weakened
RCA so much that it was later taken over by General Electric of the US.

The roster of boondoggles continues to grow. Many observers think the
list will soon include the Strategic Defense Initiative of the US, or Star
Wars. President Ronald Reagan launched the idea with great fanfare in 1983,
promising a leak-free shield against nuclear attack. It soon became the
single largest research programme of the Pentagon. Polls indicated public
support for nuclear defence.

A cold, hard look shows that the technology needed to build a leak-proofnuclear
shield is not available. Thatdidn’t bother Reagan or his advisers, whosimply
told Pentagon scientists to figureout how to do the job. Then the Reaganadministration
began throwing moneyat the problem. Soon military researcherswere claiming
assorted breakthroughs,mostly in areas ripe for advances becausethey had
received little previous attention. Nonetheless, it is hard to findmany
scientists, even among thoseworking on SDI contracts, who believeStar Wars
can work. Doubts have evenpenetrated the mind of vice presidentDan Quayle,
who was a staunch supporterof the programme while he was a senator.

Meanwhile, SDI officials have backed away from Reagan’s promises of
a nuclear umbrella. The incoming Bush administration quickly reduced the
budget request for the programme, and Congress, in turn, cut the request
still further. Star Wars now seems to be groping for direction, and influential
members of Congress warn that more concrete goals are needed to avoid harsher
budget cuts next year. Too many things have been lumped under SDI for the
programme to go away entirely, but Reagan’s original vision is receiving
a quiet burial.

Doubtless big science will continue to yield important results, like
our exploration of the Solar System. Yet I suspect that someday we will
sit and reminisce about boondoggles we knew and loved, the overgrown megaprojects
of the late 20th century. Big science can outstrip the capabilities of management,
leaving behind monuments like the $242 million Mirror Fusion Test Facility
at the Lawrence Livermore National Laboratory in California. This test reactor
was built to study an alternative approach to magnetically confined nuclear
fusion. The facility was shut down as soon as it was completed in 1986,
because the Department of Energy would not pay for its operation.

We need to continue ambitious research programmes, even if their success
is not guaranteed. And we also need to learn how to stop boondoggles in
their tracks, lest they consume the resources needed for more promising
efforts.