Professional problems, personal crises, interpersonal spats—whatever the issue—Taylor's people came to Taylor. "Bob's office was like
the hub of a wheel. Through the day each of them would come in
there. The question could be anything. 'Can I go to the bathroom?' 'Is
it time for me to go to lunch?'"
One day Spencer witnessed an incident that seemed to sum it all up.
He was sitting and chatting in Taylor's office when a member of the
CSL staff poked his head in to mention that he was going out for a
game of tennis. As Spencer recalled, "Taylor said, 'I see you've got a
new can of balls there. You're not good enough yet to play with new
balls. Here's a can of old ones. Use these, they'll be better for you.'
"Now, here's a thirty-five or forty-year-old man, a Ph.D., and he needs
Bob Taylor to tell him how he's going to play tennis!"
After Spencer became PARC director the troublesome consequences of Taylor's all-encompassing paternalism got driven home.
One of the most serious problems he faced was CSL's inability to
resolve its members' disputes with the other labs, no matter how trivial. As he settled into his new office, Spencer started to bridle at the
amount of time he spent mediating quarrels that should have been settled further down the line. "Whenever a CSL guy had a problem with
anyone else there was no one he could take it to but Taylor, and Taylor
would never deal with it. So every time there was a problem it ended
up in my office. And I said, 'This is not a workable situation. I will not
solve the problems of fifty individuals in that lab.'"
He also recognized more strongly than ever the other deficiencies in
Taylor's management style. Taylor's vaunted "flat" management structure, which meant that every researcher in the Computer Science Lab
reported to him rather than to an intermediate level of management,
had outlived its usefulness. In the old days it had served the purpose of
ensuring that all the researchers pulled together in pursuit of a coherent vision (with the exception of renegades like Shoup and Bobrow).
But
it
had also stifled creativity and—more important from the point
of view of PARC management—prevented the center from spotting
and nurturing management talent among the CSL staff. If any of the
researchers possessed the skills to be the next Bob Taylor, no one
would ever find out.
Meanwhile Taylor, still viewing Spencer as a potential ally, strived to
draw him into his own vision of PARC, which was as a research center
devoted entirely to personal computing, free of the dead weight of a
physics lab. Shortly after Spencer's promotion to head of PARC, Taylor
asked him along to CSL's annual Greybeard offsite at the Bear Hollow
resort north of San Francisco.
"We invited Spencer so we could tell him how the resource allocation
should
be reorganized," Taylor said
—in
other words, to repeat the case
he had made to the ill-fated Bob Spinrad. "Everybody there complained
to
him
about the investment imbalance."
The
wary Spencer, however,
thought the plan sounded like
a blueprint for
giving Taylor control of
both
CSL
and
SSL,
something
he was not about
to do.
He
sat through
the discussion noncommittallv.
Later, after they
had all returned
home,
Taylor
talked as though he
thought the group
had made its case to the
new director. Ed McCreight
disabused him
of the notion at once.
"Spencer listened," he cautioned,
"but don't get
it into your head that he
agreed."
On
the contrary, relations
between these two
strong-willed individu
als were
destined to build
rapidly toward a
crisis.
Twenty years of experience
as a laboratory
manager had given Spencer
very
firm ideas about the obligation
of researchers
to help
their
employers turn their work into product
—"technology
transfer," as it was known
in industry.
On
this task he judged
that much of PARC,
and particularly
CSL,
deserved a failing grade.
Henceforth, he
decreed, fifty percent of
the lab bosses' annual evaluations
would be
based on how well they
worked
with their "customers"—
i.e., the
development and manufacturing units of
Xerox.
Taylor's
response was,
perversely, to step up
his demand for new
resources.
If
Spencer wanted this
sort
of additional contribution out of
CSL,
he said, he would need
more "billets"—
in other words, more
staff openings. Spencer was astonished.
The
issue of staffing and budgets was exceptionally sensitive
at Xerox, given
that the company's
financially desperate situation
had already led
to layoffs in its
125,000
strong workforce. "The fact that the research budgets weren't
cut
was
amazing," Spencer said. "In 1981
Xerox
hit a wall and by 1982 it was in
serious trouble. That year there was no profit-sharing for employees
for the first time in company history. The place was being torn apart."
In
contrast to Pake, who dealt with Taylor largely through avoidance,
Spencer was not one to suffer defiance mutely or let an affront go
unremarked.
He
liked to give as good as he got: Upon his departure
from one job his peers had presented him with a giant mock hypodermic, a testament to his penchant for "needling" others. When Taylor
pushed, Spencer pushed back harder, dictating more and more explicitly how he expected Taylor to manage his lab. Taylor, the old expert at
defining his role in the hierarchy, had finally met someone insistent on
defining his role for him.
"Taylor is spiraling out of control," Harold Hall observed to his diary on
June 28 from the secure perch of a corporate staff job. The CSL staff
watched apprehensively as the initial friendliness between Taylor and
Spencer deteriorated into outright animosity—the tennis Saturdays were
a
thing of the past. Butler Lampson, who had become one of the world s
first networked "telecommuters" by moving to Philadelphia, where his
wife was working as an immunologist, viewed the situation gravely
enough to fly to Palo Alto to beg Pake to avert the impending cataclysm.
But Pake, toughened perhaps by having Spencer around as spear carrier,
proved unexpectedly determined.
"His position boiled down to that Taylor was too much a pain in the
butt," Lampson recalled. "And it's true that Bob was a pain in the butt,
absolutely no question about it. But he wasn't actually causing George
that much trouble. It's ridiculous to say the functioning of the whole
research center was being disrupted. What was really happening was
these guys just got on some kind of power trip. They had to have control over Bob."
This is too limited a view. Pake regarded Taylor's behavior not merely
as a personal tribulation, but as a roadblock preventing PARC from fulfilling its corporate destiny. As much as the laser printer represented a
commercially valuable technology, Pake thought that only by persuading
the company to link it with other PARC digital technologies could the lab
help Xerox claim a piece of the future of digital electronic copying.
"That meant I needed to foster good relations between PARC and the
development and engineering units of the copier division," he said later.
"Having a group of computer scientists, including their manager, who
scoffed at and derided the copier engineers in Rochester did not help
PARC develop such good relations." Under Bert Sutherland the System
Science Lab had forged a suitably collegial working alliance with the
copier division. Pake's alarm at Spinrad's five-year plan had sprouted in
part from the thought of how swiftly those friendly ties would be obliterated if Taylor were allowed to take over SSL.
Which is not to deny that Pake also had become deeply distrustful of
Taylor's powerful personal influence over
his
researchers' souls.
"I
saw
Taylor's relationship to his lab members
as
analogous to Jim Jones and
the Jonestown cult," was his remarkable recollection. This impression
must have been reinforced, if not inspired, by one senior researcher's
explanation for why he had turned down a chance to work with Taylor's
lab. "George," the researcher told Pake, "I wouldn't be willing to drink
the Kool-Aid."
"We understood very well Taylor's relationship to his lab members,"
Pake said later. "What we did not understand was what to do about it."
Meanwhile, it fell to Spencer to get everyone on Coyote Hill working
together in an atmosphere of mutual civility. That August, in a desperate
attempt to ease the rancor among his lab
chiefs,
he convened his own off-site at Pajaro Dunes. There, in the same setting where Alan
Kay's
group
had regularly repaired to contemplate the digital future, the senior lab
managers of Xerox PARC flayed each other in an emotional showdown.
Until two or three in the morning they vented their feelings like patients
in gestalt therapy. "People let their hair
down
to talk about what really
were the problems at PARC," Spencer recalled. The debate seemed
honest and heartfelt. As dawn was breaking over the dunes, he allowed
himself to think that Taylor—the target of most of their complaints
—
had
finally realized the error of his ways.