OK Computer

In the summer of 1996, not long after people began to get excited
about the pace of the new economy, two of Yale’s highest officers
met to discuss how the University would keep up. For Provost
Alison Richard and Associate Vice President Joseph Mullinix,
there was reason to worry: Yale’s mainframes and administrative
systems were becoming obsolete, and business practices lagged
years behind the latest technology. Most antiquated was the
University’s 15-year-old accounting software, which had been
designed when Yale was a significantly less prosperous place
than it is today. On top of questions of mere obsolescence was
the almost religious fervor surrounding Y2K, the modern twist
to the familiar apocalypse story (which still turned out less
bang than whimper). Even more dire than the state of aging buildings
and facilities around campus, it seemed, was the delicate status
of Yale’s digital infrastructure.
Richard and Mullinix quickly called for a committee and working
groups to address the problem. Over 60 individuals from all levels
of the University would assist in the effort and report back
to the administration with their findings. They were asked to
submit a preliminary budget, evaluate potential contractors,
and develop an overall program strategy. They eventually proposed
what turned out to be the largest, costliest, and most controversial
restructuring program in Yale’s history-the ominous, if unimpressively
named, Project X. Project X promised to change almost every
aspect of the way Yale conducted business, from making office
life paperless to automating tasks that once involved an intricate
hierarchical chain of people and processes. It heralded nothing
less than a new work environment, free from so-called "redundancy,"
guided by the new (and yet old) economic virtues of efficiency,
rapidity, and automation. Four years, $100 million, and the
retraining of several hundred employees later, nearly everyone
at Yale agrees that Project X has done all these things. They
only disagree as to whether this constitutes a success, or a
minor catastrophe.

Nobody seems to know what the X in Project X stands for.
Certainly it’s better than the official name (the Yale "Finance
and Human Resources Modernization Project") that was somehow
approved. Still, almost every affected Yale employee, at one
time or another, has probably felt its unofficial name signified
something unsavory: One woman I spoke with said she believed
it stood for "X-cruciating"; "X-cessive"
was another guess I heard; a third, "X-crement." Feelings
of prolonged frustration are also expressed by the Project’s
most vocal supporters and by people who were instrumental in
bringing it about. Ask any Yale employee to talk about the early
stages of Project X, and the response you get will inevitably
be clinical. (The word "painful" seems especially
popular; one person made an extended comparison to surgery.)
"We survived," said Dean Plummer, one of the Project’s
architects. "We got through it. It was painful, but we
made it."
Before Yale had to worry about such things as Y2K compliance,
the routine functions of office work-check requests, payroll,
account maintenance-were largely a matter of directing paper
forms to specific offices. Requests to purchase anything from
a stapler to a new computer required handling by the Purchasing
Office. There were Payroll people, Grants Management people,
Personnel people, Accounts Payable people, and so on. In order
to be reimbursed for the simplest personal expense, for example,
an employee would have to fill out one form, have that form signed
by the office business manager, then send it up the street to
the first floor of 155 Whitney, Accounts Payable, where another
pale face would look it over, sign it, and then send it back
down the street. Even then, you could never really be sure the
envelope would complete its journey.
No one much doubted that the Project X operation would be at
least mildly painful; what came as a surprise was the extent
of the trauma. Back in 1996, when the Yale Steering Committee
first began investigating the available upgrade options, it had
the experience of peer institutions to instruct it. The University
of Pennsylvania had beaten everyone to the line, going live with
a limited installation of Oracle database software in July 1996.
Penn had asked Oracle simply to update certain of its financial
systems and to patch up others for Y2K compliance in a gradual
rollout over the course of several years. Princeton and Stanford
had similar plans. Harvard, on the other hand, announced that
it was overhauling nearly all of its administrative systems at
once. In so doing, it kick-started the kind of back-and-forth
race that characterizes most policy changes in universities today.
The imperative was clear: Yale had to meet its competitors’
challenge. Would Old Blue merely hold pace with the new economy-or
would it pull, so to speak, into the lead?
The working groups reported back with their findings in a lengthy
study, which discussed the many shortcomings of Yale’s existing
systems and proposed a bold corrective. They found, in their
own words, that obsolete software and antiquated business practices
were often "impeding the best efforts of faculty and staff."
They argued that the entire University community was frustrated
and distracted by "complex, time-consuming, and costly administrative
processes in the course of their academic work." The study
maintained that while Yale could, conceivably, follow the more
tentative course taken by Penn-patching some systems and upgrading
only those in the most critical condition-an alternative if more
challenging option was open to them. Rather than go halfway,
Yale could go the better way-the way of Harvard-and even a step
beyond. In one giant leap forward, Yale could "reengineer"
not only its technology, but also the equally obsolete routines
of its users.
The University agreed; and in September, with Pentium-like speed,
the Yale Corporation approved funding for a large-scale tech
upgrade. Accompanying the ongoing facilities renovation project
would be an equally intensive renovation of information technology.
The lucrative contract, initially projected at $30 million,
went to Silicon Valley database giant Oracle, whose ceo, Larry
Ellison, is the second richest person in America. With Oracle
promising that "the deluge of papers and forms that can
overwhelm even the most efficient Yale staff members will soon
become a thing of the past," the University signed a four-year
agreement with the company, making Yale a test site for some
of its newest database and financial systems. Project X was officially
born.

An army of programmers and consultants from Oracle and KPMG
were called in to begin implementing the new system. Soon, it
became clear that Project X would be a learning experience for
both the doctors and the patient. By 1997, Oracle had already
overseen complete system overhauls for businesses and governments
around the world, but never for a university. New technology
was to replace obsolete systems in any department or office that
maintained its own payroll, purchasing, general accounting, financial
planning, or grant funding-which is to say, every single office
at Yale. The everyday tasks that once required eight separate
forms sent to eight separate central departments would now be
handled by a single web-based system. The old mainframe was
to be completely replaced with two brand-new IBM servers, which
could be accessed by 2,000 screens simultaneously over a new
fiber-optic network. Beyond mere boldness, Yale could also claim
credit for being a trendsetter, the first university to commit
itself to such a massive undertaking. But being first counts
more in some races than in others, and the distinction of being
first with Project X seems to have been rather dubious.
A month after the Corporation approved the project, Richard and
Mullinix announced the new initiative, already going by the unofficial
moniker "Project X," in a letter printed in the October
7-14 issue of the Yale Bulletin & Calendar. An official
statement of goals, in no particular order, was soon drafted,
and would become nearly omnipresent in the Project X literature:

Also in October, the Oracle people came in and set up temporary
headquarters at 12 Prospect Place. Working with them to test
the software were 110 full-time workers from across the University-people
from the Arts and Sciences faculty, Athletics, the School of
Forestry, Yale College, and Animal Care. Over the next four
months, team members would learn the software and then propose
how it could be tailored to best meet Yale’s needs. They developed
product specifications and suggested design changes along the
way. Meanwhile, Oracle planned Yale’s future database system-projected
at 1.2 terabytes worth of code, or the equivalent of around 50
million volumes of the Encyclopedia Britannica.
Once the pilot testing of the Oracle software was completed in
March, Yale began the more arduous task of phasing out its existing
systems while adjusting employees to the new ones. That same
month, Project X sent out some 500 invitations to business managers,
department heads, and other staff to take part in a training
program advertised as "the most extensive ever offered by
the University." Sessions covering everything from software
usage to "Preparing for Change" would help employees
with the difficult learning process. The handsome new office
housing Project X on the ninth floor of a downtown building was
equipped with its own training room and a fleet of new computers
for group sessions. A separate learning center, similarly appointed,
soon went up at 220 Whitney. Suddenly, it seemed that not a
day went by when Yale employees weren’t being invited to avail
themselves of focus groups, full-day workshops, business redesign
workshops, management issues forums, hands-on training, implementation
planning, in-depth briefing sessions, and telephone support.
No topic was considered self-explanatory, and training sessions
always restricted class sizes, of course, so as not to discourage
the group from asking questions.
The planners, it must be said, were certainly mindful of the
people uncomfortable with Project X. Steve Sunderland, the Project
manager, told the Bulletin & Calendar (in the first of its
series of articles devoted to the subject) that he was "fully
aware" that "for some people on campus who have become
accustomed to doing their jobs the same way for many years, the
modernization might make them anxious about how the changes will
affect them." Sunderland stressed that everyone-everyone-would
have the opportunity to be trained. "There’s no doubt that
such a major change can be scary," acknowledged Margaret
Plympton, another director of Project X. For that reason, she
said, her team was working hard to make people feel at ease about
the transition, since, naturally, "we’ll all be going through
the change-over together."

One by one, Yale began to transform existing systems into
"modernized" Oracle equivalents. By the summer of
1998, Yale had successfully installed the first round of Oracle
human resources (hr) and accounting applications. Already, it
was becoming easy to see how Project X would change office life.
Redundancy could now be replaced by simple, efficient automation.
No longer would hr staffers spend their days photocopying and
filing the 12,500 resumes that Yale receives each year; now they
could simply be dropped into a scanner and filed electronically.
As Mullinix told Oracle’s company publication, the baldly titled
Profit magazine, "The principal idea is, ‘Do it right; do
it once.’"
But the Oracle transition team stood to learn something from
the ethic propounded by Mullinix. It is difficult to gauge precisely
how faithful Project X was to its self-imposed deadlines because
those dates were changed many times and hardly any of them was
actually met. Originally, Project X was to be completed in two
phases-the first phase, largely clerical and technical in nature,
in July 1998, and the second phase, largely administrative, in
April 1999. In reality, the final launch was delayed three times
because of, well, technical difficulties. As the April deadline
approached, test after test uncovered new problems to solve.
As April 1 became May 1 and then, just as quickly, July 1, the
Project architects went on the defensive. With uncommon prescience,
Dean Plummer contrasted the modernization process with an election:
"It’s not like an election when on a certain date you find
out who won and that’s the end of it," he told a Yale Daily
News reporter. "It’s more like a learning process where
you sign up for a lesson and come out much, much smarter."
Without a finished product to introduce to employees, Project
X was forced to continue offering training programs for software
that didn’t yet exist. And the delay only served to aggravate
further the hideous budget overrun that saw Project X go three
times over its original $30 million allocation. When the new
Oracle software finally went live on July 6, 1999, no one really
knew what to expect. On July 9, when the first batch of paychecks
was printed, they got an idea.
The checks showed up at the wrong location. Such was the inauspicious
opening, and more of the same was to come. Staff benefits were
calculated improperly; data entry errors, bugs, and glitches
kept reappearing in the system. Employees, upon opening the web-based
Oracle programs, seemed much less competent than the Project
directors had hoped. Some employees complained that there was
not enough qualified technical staff around to help the less
experienced, and some problems left even the technicians themselves
scratching their heads.
Almost nothing went well during the first year of Project X.
Clerical and technical staff, in particular-the 2,800 predominantly
middle-aged, female employees sitting at desks and keeping the
University operating-were finding it difficult to learn programs
they used only sporadically. This proved worrisome, because for
the first time, all Yale offices and academic departments were
to submit their budget proposals-formerly plotted on a simple
Excel spreadsheet, or even handwritten with pen and paper-in
the new Oracle format. In February 2000, officials announced
that the University budget for fiscal year 2001 would not be
completed until two months after its usual April submission deadline
because of lingering Project X growing pains.
Still, in hindsight it appears that Project X was far less painful
than it could have been. In March 2000, Ohio State and Cleveland
State, which had chosen to work with Oracle’s competitor, PeopleSoft,
reported that their system upgrades were proceeding miserably.
Ohio learned that its project, estimated at $53 million over
three years, would cost $90 million and take six years. Yale
also had the satisfaction of trumping its neighbor in Cambridge.
Like Yale, Harvard had begun planning a systems upgrade back
in 1996, and even gave the effort a similarly bland name, Project
adapt. Also like Yale, Harvard, working with Oracle, went far
over budget, faced delays, and-perhaps believing that one Game
a year wasn’t enough-also settled on a July 1 deadline. But
whereas the turnover at Yale is retrospectively deemed manageable,
Harvard offices are still struggling with their transition to
Oracle, and Project adapt is widely considered an embarrassment.

In December 1999, with all the new systems finally in place,
Mullinix told the ydn that it would be difficult to calculate
exactly how much Yale had spent on the Project: "The problem
is that the Project is many projects and so it all depends on
what you include in the cost estimate." Mullinix had a
personal stake in Project X as one of its original planners,
which may explain his weak explanation to the same reporter:
"It’s just such a mishmash of stuff that it is very hard
to come up with a number." Officially, the matter was handled
with more decorum. In Yale’s financial statement for fiscal
year 1999, the figure given for all the training and technology
was over $100 million-$33 million in 1998, when things were running
smoothly, and $70.5 million in 1999, when they were not. A staggering
amount of money-and yet, in these fat and plentiful economic
days, only a fraction of the $1.24 billion Yale spent to keep
the University running in 1999.
One year later, in February 2001, a near-unanimous consensus
deems Project X a success. Across the board, administrators
say that, while they wouldn’t want to revisit the early days
of the Project, it has had a positive effect on working life
at Yale. Chuck Paul, Yale’s Director of Total Compensation,
says that Project X has empowered individual employees and rid
the University of long trails of paper. According to Brian Tunney,
Director of Labor Relations for the University, the transition
to Project X has constituted a near non-issue for Yale’s workforce.
Tunney explains that Project X has not caused intimidation or
lost jobs; rather, Yale now employs over 200 more full-time workes
than it did prior to the Oracle overhaul. Of the University’s
upcoming contract renegotiations with Locals 34 and 35, Tunney
added that Project X would almost certainly not be an issue.

At the same time, some of Yale’s workforce say they have yet
to see the wonderful advances promised by Project X. Eighteen
months after Oracle went live, some of Yale’s clerical and technical
workers are dissatisfied-a frivolous response, according to their
higher-ups. The reasons behind their dissatisfaction extend
back to the earliest announcements of Project X, but also, in
a sense, to before such things as web-based financial accounting
systems existed. Historically, Yale’s employees have been nothing
if not class-conscious in their relationships with the University
and each other. Today (maybe more than ever) there exists a
palpable and long-standing line separating the work interests
of Yale’s three main labor pools. The groups usually refer to
each other by abbreviations-management and professional (M&P),
clerical and technical (C&T), and service and maintenance
(S&M)-and at all times maintain a clear sense of where they
stand, status-wise, relative to the other two.

Responses to Project X usually fall in one of two unevenly
matched camps. On the one hand there are the articulate and
confident views of M&Ps in the central business offices.
There is not much sympathy on this side of the fence for people
who persist in believing Oracle programs are too difficult to
learn, and they may be right. Then again, these are people whose
jobs handling finance and personnel matters were given new capabilities
by the powerful Oracle software, people who use the Oracle software
everyday-or, more likely, who have assistants who use it. To
ask these people about the difficulties incurred by Project X,
you will hear that while some old-fashioned people were more
cautious than others, more or less everyone has accepted it as
something that is here to stay. To press the point and ask if
there isn’t really someone who is less than thrilled by the change,
you will hear a slightly less diplomatic opinion, which holds
that some people, sadly, will always refuse to learn.
It’s difficult to find someone in the other camp who will rebut
the m&p point of view, at least on the record. Not that
there aren’t people who disagree-it’s just that they have either
retired early, or continue working, but are not at liberty to
tell their employers what they really think of the arrangements
made for them. Just as the Project’s supporters can almost always
be found in the more spacious offices of m&ps, the dissenters
are inevitably middle-aged, female C&Ts who work in smaller
offices and believe they are doing more work, not less, since
the new system arrived.
These are the pesky women who, to M&Ps dismay, still complain
of having difficulty with the Oracle software. They work in residential
college master’s offices, in libraries, and in smaller academic
departments, where on top of everyday clerical tasks was suddenly
placed the burden of learning a difficult and unwieldy Oracle
system. These are the women-so goes the complaint-who were never
asked about Project X and never once asked for it; who were told
one day that other people were "reengineering" their
jobs, and that they would have to get with the program. These
are the women who answer phones and process student payroll,
keep track of hiring and maintain department budgets, for whom
life is still anything but paperless, and for whom Oracle’s 30-digit
processing numbers are no simpler than the forms they once had
to fill out.
"The pressure and frustration of learning the systems and
the programs when you don’t use them every day makes it very
painful," said one master’s assistant, who said she could
only talk while the college master was out of the office. "Sometimes
you just want to pick up the computer and throw it through the
window, it’s so painful." She says that Project X has added
more tasks to her life without eliminating her pre-Oracle responsibilities.
A friend of hers says she knows people who have been told "another
person would always be ready" to take over any work she
found too difficult, which of course is true. This friend says
that any further comment she makes about Project X has to be
approved by the director of her office, who has asked that she
refer all questions to him. A third Yale employee in the Benefits
Office knows co-workers who have retired because of Project X
and current employees who feel overly burdened by the new systems.
She promises to ask if anyone would be interested in talking,
then calls back a few days later and says all would rather not
revisit the issue.

Nearly everyone directly involved prospered handsomely as a result
of Project X. Oracle came away with millions and a successful
partnership to advertise in wooing other universities. Yale
installed one of the most sophisticated financial systems in
the world and retrained its administrative workforce in the process,
while reporting only minimal injuries. Individual actors prospered
as well: Sunderland, the Project director, soon left Yale to
work for Oracle; Daniel Updegrove, head of its during the Project,
also left shortly after completion to head the much larger it
services department for the University of Texas at Austin. Most
successful was probably Mullinix, who took a position as Senior
Vice President of Business and Finance of the California state
university system, despite telling Profit magazine that he looked
forward to using Oracle software to print out his Yale budget
reports.
But while Project X was a smash for the administration, as soon
as they left, it became merely 9 to 5 for the people in the offices-the
ones who carry out the bold initiatives of the so-called new
economy. The C&T women’s complaints would probably strike
most M&Ps as idle, even willfully archaic. After all, no
one but the most wistful Romantic really feels any sympathy for
the Luddites, who smashed their bosses’ weaving machines but
live on today mainly as quaint relics of an older age. Today’s
working world, it seems, is unique in that the hammers can be
ordered directly with Oracle Purchasing-only minutes before attending
a session advising how to "Prepare for Change."

 

Ronen Givony, a senior in Branford
College, is associate editor of
TNJ.

More Stories
The Nuclear Testament