^^
L-'iL.^
HD28
.M414
%
Time
as a Trigger for
Organizational Change
Marcie
J.
Tyre
Leslie Perlow
Nancy Staudcnmayer
Christina Wasson
August 1996
Marcie
Sloan
J.
Tyre
Leslie Perlow
University of Michigan
Ann Arbor, Michigan
MT.T. Sloan School of
Management
Cambridge, Massachusetts
617-253-6679
mtyre@mit.edu
WP #3920
313-764-4712^
Leslie Perlow@ccmail.bus.umich.edu
Nancy Staudenmayer
Christina
Wasson
M.I.T. Sloan School of
Yale University, Department of
Management
Anthropology
and E-Lab, Inc.
203-776-7843
Christina, wasson @yale.edu
Cambridge, Massachusetts
617-258-7269
nstanden@mit.edu
Presented
at the
Annual Meeting of the Academy of Management
Cincinnati, Ohio, August 13, 1996.
.^ i
^S
81998
LIBHAFrttiS
ABSTRACT
In
this
we
paper,
processes
in
integrate
findings
from four
technology-intensive organizations.
field
studies of change
Our findings suggest
thai
temporal shifts
i.e., events that alter normal rhythms or that insert a
complete break in the action — can be powerful triggers for organizational
change.
By contrast, problems (even when they were well-recognized)
noticeably failed to generate active adaptation.
In each of the cases
studied, change efforts did not begin unless and until some event occurred
that altered the rhythm of daily activities.
We find that such breaks in the
action forced people or groups to change their pace of work and the
temporal focus of their work, and that only once this occurred could people
devote attention to outstanding problems.
Specifically, temporal shifts
--
altered time
in
four important ways to enable change.
(1)
Time
as a
temporal shifts woke people up from their routines and refocused
attention on possibilities for change; (2) Time as a resource: temporal shifts
provided actual or perceived "time-outs" from regular, productive
activities to undertake reflection and change; (3) Time as a coordinating
mechanism: temporal shifts enabled all members of a group or
organization to focus collectively on a change agenda; and (4) Time as
symbol: temporal shifts stood as a powerful symbol of management's
commitment to the need for change.
trigger:
A
fundamental message
while routines, scripts,
constrain
adaptation,
from routines
to
eruption
problem can
work
other
that tend
theme.
this
and
examines
that
get noticed
to
away
1958; Cyert and March,
we have
Thus,
the
sorts
a considerable
Kiesler and Sproull,
(e.g.,
of problems or discrepancies
1992, Feldman,
how and why
of research explaining
between expected and actual outcomes lead
search and adaptation (March and Simon,
Lant 1992)— or
to
attention
shift
Louis and Sutton, Sitkin
(e.g.,
other streams
still
discrepancies
that
is
an enormous amount of the organizational literature can
on the difficulty of noticing problems
literature
1981),
of a
change (March and Simon,
be seen as discussions on
1981),
on organizations
literature
and repertoires that develop over time tend
the
elicit
Indeed,
1963).
the
in
how
this
March 1963,
1958; Cyert and
process can go astray
(e.g.,
organizational
to
March and Olsen
1975; Levinthal and March, Lant and Mezias).
In
yielded enormous
search
has
certain
other
adaptation.
we argue
article
this
phenomena
insights,
help
that
adaptation occurs but
problems or other discrepancies.
here,
unless
and
groups
and
to
that
change
their
important ways
to
or needs for change;
actually
not
Indeed,
that
fact
triggered
in
in
in
change or
many
cases,
any direct way by
four organizations
the
failed
to
some event occurred
generate
that
altered
active
the
rhythm
find that such breaks in the action forced people or
pace of work and the temporal focus of their work,
perceived
temporal shifts altered time
Specifically,
enable change.
woke people up from
or
organizational
only once this occurred could people devote attention to
outstanding problems.
actual
is
They noticeably
until
We
of daily activities.
the
obscures
systematically
also
problems were known and they significantly
underlying
predated change efforts.
adaptation
it
trigger
to
Our argument stems from
organizational
described
while the focus on problem centered
that
their
(2)
(1)
routines
Time
"time-outs"
as
a
Time
as
a
trigger:
in
four
temporal shifts
and refocused attention on possibilities
resource:
temporal shifts provided
from regular, productive
activities
to
undertake reflection and change; (3) Time as a coordinating
mechanism: temporal
shifts
enabled
all
members of
a
group or
organization to focus collectively on a change agenda; and (4)
Time
as
temporal
symbol:
commitment
We
begin this paper by outlining some of the vast literature on
problem centered search and
enormous
the
effort
specifically,
studies
separate
into
developing and refining
in
organizations.
appears
that
We
interruptions
problems
that
we
studies,
breaks in
change
served
search
"triggers"
as
both
action
We
activities.
subsequent
(and
focused
that
attention
shifts
the
also note significant differences across our cases.
the
in
alteration
Based on both
cases,
on
change and shaped or supported
triggered
horizon of ongoing activities.
to
adaptation)
By juxtaposing these four
commonalities in the ways that temporal
routine activities, while in others
limited
critical
a
Rather,
se.
example, in some cases the temporal break took the form of a
fundamental
—
had previously been ignored.
discover significant
the
model,
this
plays
we develop
the
a
shift
the
in
We
rhythm.
organization's core beliefs,
stop to
rhythm or the time
in
other cases
technologies or operating parameters.
similarities
and the differences among our four
thinking about organizational change that
close by comparing our
new model
We
the
to
because
It
enriches
helps to explain not only
it
conditions allow change to happen
approach
that
understand
broad,
using
why change
in
accepted
a
particular
notions
complements
groups and organizations.
some organizational phenomena
to
it
occurs, but also
focuses on temporal events also enables
not local
more widely
of problem-driven
theories
problem area
of problem-driven
us
to
search
when
An
better
such as search that
---
shifts
argue that our model
does not compete with the existing view, but rather that
approaches.
was
it
of specific
accepted notion of problem centered search.
traditional
For
Also, while "change" sometimes consisted of
new way of
a
meant
it
full
emphasizes the key role of temporal events such as interruptions or
in
note
then analyze data from four
were not triggered by the appearance of problems per
temporal
—
temporal breakpoints
other
it
We
organizational adaptation.
or
change
where
in
smaller set of work suggesting that time
far
interruptions
instigating
in
role
its
has gone
that
and we also highlight a
role
symbol of management's
very powerful
a
need for change.
'D^
the
to
stood as
shifts
that
search.
are
is
not well explained
LITERATURE
/.
have contributed
that
the
A number
Problem-Centered Search and Adaptation.
an
to
organizational
experience of failure
adaptation.
In
is
central
to
search and
organizational
view, organizational success
this
which suggests
perspective
learning
leads
of authors
repetition
to
Organizational search and adaptation are costly
routinization of activities.
and often aversive (Nelson and Winter 1982; Cyert and March
Kiesler and Sproull,
Failure
between
realization
engenders willingness
adaptation
routines
existing
to
insufficient,
Thus, the
1988).
failure
triggers
this
realization
and
a
time and other resources in search and
invest
This
between failure experiences and
relationship
the
organizational adaptation
such.
one:
March and
1963,
March
of
activities.
However,
as
are
logical
a
is
new ways
search for
Levitt and
1970,
and adaptation
failure
that
to
undertake change (Cyert and March
to
Simon 1958, Argyris and Schon
link
however, especially sudden and
experiences,
can induce organizations
failure,
doing things and
1963;
1982), and thus they will be avoided in the absence of
powerful motivators.
unexpected
and
is
simple.
not
the
First,
problematic given the tendency
is
must be noticed
"failure"
to
ignore disconfirming or
very discrepant information by both individuals (Darley and Gross,
1981) and organizations (Kiesler and Sproull,
Feldman.
failures,
far
1983;
Major
1982).
from encouraging adaptive responses, often lead
to
greater
and therefore persistence with existing action programs (Staw,
Small failures, or those which carry minor
Sandelands, and Dutton 1982).
rigidity
consequences for the person(s) involved, are also
(Louis and Sutton,
1981:
Van de Ven. 1980;
likely
to
be ignored
Thus,
1990.).
Isabella,
it
appears that only certain kinds of failures (those which are surprising but
not too
discrepant,
implications
but
Further,
are
since
but
significant
not overly
failure
is
not
overwhelming, carry personal
threatening)
essentially
a
will
be
noticed.
gap between expectations and
outcome, failure can be defined away by altering expectations
outcomes.
Research suggests that change
suppresses
adaptation
behavior
Another major problem
interpreting
in
(Lant
in
aspiration
levels
to
fit
actual
frequently
1992).
learning from
failure
is
the
difficulty
of
what happened and why, given uncertainty, ambiguity, and
constrained action in organizations.
learning
the
underlying
and
undertaken only
generated
processes
that
Starbuck,
1976;
(March and Olsen 1975
be overly narrow, so that change
procedures
those
Argyris and Schon
sum, the bulk of the
In
the
to
is
failure,
limited
rationality
events
they
experience,
routines
organizational
motivation
suggests
despite
that,
organizational
and
actors
the
goals,
adequate for achieving
be
to
While
salient.
must be adapted. This provides the
routines
engage
to
the
in
major gap between expectations and outcomes helps
a
that
of
difficulties
ambiguity
a
(unmet expectations) triggers
failure
normally assumed
are
realize
to
on organizational learning as
change by making problems obvious and
organizational
existing
among
Nystrom and
(Hedberg,
1978).
literature
organizational change
for
actors
also
exist
fact
examining the deeper organizational norms, assumptions and
without
basis
may
in
procedures immediately related
the
in
do not
ihal
Interpretations
etc.).
indeed induce change, yet
change may be superstitious, assuming
that
action-outcome linkages
may
Failure
in
information search and adaptation of routines or
even rules, despite the very
real
costs
and risks involved
in
such change
measures.
The Role of Time
2.
Organizational Change.
in
The idea
that
temporal
breaks can serve as a trigger for search and adaptation has not received
much
in
the
organizational
various
aspects
The most
literature.
of this concept have
explicit
surfaced
treatments of this
work on temporal milestones (1988, 1989,
Gersick's
are
Nonetheless,
attention.
1994),
idea
Weick's
(1990) discussion of the impact of interruptions. Tyre and Orlikowski's
(1994) research on
(1988) description of
Barley's
social
windows of opportunity
order
hospital
in
Gersick (1988,
milestones,
mark
to
in
new
that
to
shift
the
1994) has suggested that clock-based events or
1989,
midpoint of a time-limited exercise, or the five-year
venture, can
milestones
deadlines,
change, and
departments.
turn
the
focus of activity
Here, the focus
is
from normal routine
not on events that alter the
rhythm of work, but rather on temporal markers.
important
its
technical
how temporal changes helped
radiology
reevaluation and change.
actual
on
a
the
like
for
and
that
often
this
focus
serves
a
as
group's
a
or
Gersick suggests
organization's
attention
problem-signal thai initiates
change.
other words, organizational
In
how
for deciding
Weick
with
new
more closely with events
deals
(Lewin via Marrow,
are
that
create
a
events that
stochastic
break
1969; Mandlcr and Watson,
1966),
the
in
work
interrupt
Drawing on experimental research
technologies.
interruptions
time as a heuristic
"use
long to remain on the same track" (1989:303).
He focuses on unexpected
action.
members
psychology
in
Weick argues
that
key because they create an alarm signal which evokes
high emotion (in Abelson's (1963) words, "hot cognition").
Since the
breadth of attention varies with the level of emotion, Weick suggests that
adaptive effort
is
some maximum
likely
to
increase with increased emotion,
point, after
up
least
at
to
which arousal can actually disrupt adaptive
activity.
Tyre and Orlikowski (1994) build on Weick (1990) and others
focus on the particular power of new beginnings.
They show
to
that
technology users do actively adapt their process technologies, but that
these
changes occur
discrete
in
following users'
spurts
first
experience
with a new technology, or following some disruptive event that renews
users'
relationships
with
their
tools.
Dutton (1993) has also argued
that
understood as a problem or an opportunity
demarking
shift
sharp break in the
a
people's
routines
--
to
future,
value of investing time
thus
now
making
to
it
effect
--
--
whether
can create change partly by
--
She argues
of time.
from the present
attention
the
How
unexpected event
an
everyday
such events
that
and
pressures
possible for people to see the
improvements.
future
Barley (1988) uses data from his ethnography of technical change
hospital
radiology departments to show
"socio-temporal order" of a workplace.
CAT
scan equipment entered hospitals,
radiologists'
work (from
xray departments
groups),
to
less
thus paving the
departments'
social
hectic,
how new
is
he shows that
Specifically,
it
altered
short-attention
the
span
work
longer attention span
way
larger changes
for
in
alter
the
when
rhythm of
frenetic,
the
in
in
traditional
CAT
scan
radiology
order.
Other researchers have made the point
buffer time
technologies can
in
needed
in
that
some
slack
time or
order to consider changes to routine and to effect
March and Simon (1958) and Scribner and Gutek (1987) note the
importance of some organizational slack or buffer time for making changes;
them.
Hedberg (1981) suggests substantive change
is
some
without
unlikely
Lounamaa and March (1987) add that while
change can proceed without a pause, it may fail to lead to performance
pause
routine
in
activities.
improvements unless actors take the time needed
While many of
scholarly
papers cited above have received considerable
the
perspective
how temporal breaks
understanding
for
action can create opportunities for search and
organizational
balance of
the
In
learning
theory
integrative
to
this
explain
paper,
the
we
the
time for enabling
role
attempt to provide a more
of temporal breaks
order to do
In
from four different organizations undergoing
illustrate
opening up the change agenda.
In
we draw on
this,
in
enabling and
in
distinct
importance of a temporal break
kinds
the
of change,
action for
each case, problems existed and were
recognized by organizational members, but they noticeably failed
search and adaptation.
the
and change.
encouraging organizational adaptation.
all
in
in
Each deals with
adaptation.
power of time and breaks
different aspect of the
that
change
past
none has attempted or achieved an integrated
attention,
theoretical
data
assess
and their impacts.
actions
a
to
A
temporal
shift
to
trigger
or break in the action not only
enabled change by creating some slack time, but actually triggered the
change process by shifting the focus of attention from routine
activities
to
improvements.
potential
Nature
the
of
The
results
Data
reported
in
this
paper come from four separate studies,
each examining a different context and each focusing on a different kind of
"break
in
the
papers could
various
types
Each of
action".
the
Together, they show more powerfully than separate
relationship
between
of organizational
the
studies
temporal break
a
action
and
change.
included here
is
a
field
study examining one or
more successful examples of organizational change
company.
the
in
inside
a
particular
Thus, while different kinds of change were achieved and while
the
precursors of change were different
across the cases covered here,
each case there was evidence of significant, documentablc change
organization following the event
each of the studies
we
label a
addition,
In
acknowledged leader
innovativeness.
each of the organizations studied
in
and
may
for
long-term
represent one of the
survival
These cases were selected
for inclusion
theme, they collectively cover
useful
a
from long-
recuperate
spectrum
in
the
type of organizational change that resulted.
Table
1,
below, provides
an overview of the dimensions along which our cases vary.
we
them
describe the methods used
results
existing literature,
theoretical
in
from each study regarding
significant change.
to
terms of both
involved, the nature of the "break in the action" observed, and
settings
to
in
because, while focusing on a
the
led
ways
problems.
ingrained
main
may
Thus, while the actions and reactions described here
which companies bound
below,
are
it
an
is
terms of market share, quality,
not be responsible for that success, they
common
Also,
an in-depth case study, so that details about the
is
nature of the change and insights about the causality behind
available.
the
in
"break in the action".
in
Finally,
and
understanding of
organizational
change.
we
each study.
the
way
integrate
that
these
the section
In
Next we present the
a
break
findings,
in
the
action
compare
use them as the basis for developing a
how temporal
shifts
can help
to
bring
about
10
Table
I
:
Four Case Studies of
Ori^aniz.ational
Change
11
heavily contingent on the accuracy of up-front planning activities.
approach tends
produce robust and reliable products, but
to
long and inflexible development cycle.
relatively
was experimenting with several
Microsoft
approaches and practices
sacrificing
"buffer"
inserting
time
During the
periods
initial
development
alternative
The present research analyzes
reliability.
the
in
impact of
the
development schedule.
phase of investigation, two senior researchers
Selby,
1995).
company.
focused on 7 projects currently
across
the
The
projects.
product areas
(e.g.,
components).
The
site
research associate
a
During
Microsoft.
period she
this
development and interviewed 34 people
in
projects
at
develop
to
and development process across
history
During the second phase of research,
spent two and one-half months on
Microsoft
at
The goal of these interviews was
knowledge of Microsoft's
detailed
the
&
without
flexibility
conducted 40 interviews with team members and managers
(Cusumano
cost of a
the
at
At the time of the study,
would provide greater
that
This
were selected
represent the major
to
desktop application, operating system, tools and
projects also varied in terms of their size and
complexity (the amount of code or number of components
in
software
the
product as well as team size and structure), legacy factor and performance.
The second round of interviews therefore permitted
examination of actual development practices
complemented
detailed
selected
in
broader perspective taken in the
the
a
first
which
projects
phase of the
investigation.
Several points speak to the validity and reliability of the data
collected.
First,
researchers
74 interviews were conducted by three different
yielding
over
These interviews were distributed across functional areas
transcribed).
Over
and levels of experience.
projects
throughout
documentation.
a
also
approximately 60 different
represented
are
in
this
study.
granted access to various forms of
For example, postmortem project reports yielded a rich
of data on
over time.
data on
all,
history
Microsoft's
The researchers were
history
200 hours of taped conversations (ultimately
how development
These postmortems
are
practices
written by
at
team members
project and candidly discuss both the successful
aspects of the project.
They
also
usually
Microsoft had changed
and
less
at
the
end of
successful
include detailed performance data
such as development cycle time and testing
results.
These reports
12
served
therefore
and yielded some rich examples of actual experiences on
interviews
11.
(For further details, see
projects.
different
and augment evidence collected during the
corroborate
to
Study of technologies-in-use
Selby (1995)).
BBA
at
The second study was undertaken
Design and Sampling.
pseudonym),
&
Cusumano
at
BBA
(a
European-based manufacturer of precision metal
a
components.
This study investigated problem solving around new process
technologies.
Thus,
made by
of the technologies
users
mechanisms or
organizational
the
altered
machine
in
new ways of organizing production, new
(e.g..
technologies).
BBA
high-volume,
the
improvement
changes that may have been triggered by experience with
At the same time.
because
(e.g.,
procedures for using the machines) as well as larger
the
new technology
practices,
focused on the incidence of small-scale changes
it
was
a desirable
high-precision,
for the research,
setting
automated nature of
relatively
production placed significant time pressures on technology users.
made
it
challenging for users of
new technologies
resources for ongoing
adaptation.
BBA
in its
is
process
a
world leader
This
find time and
to
Yet significant adaptation did occur:
industry with a reputation for excellence in
technology.
The study was
West Germany, and
Forty-eight projects
out
carried
the
in
major geographic divisions— Italy,
three
United States— and involved eight different plants.
involving the introduction
and use of new process
technology (production machines and systems) were
missing data
in
studied;
seven cases, forty-one cases are included
in
due
the
to
current
sample.
Projects were
experiences
were
studied using three types of data.
obtained
from
detailed,
Description and
semi-structured
interviews.
Interviews lasted from one to four hours and occurred between
18
months
after
supplemented
Respondents
by
project
completion.
multi-participant
included
project
zero
and
These one-on-one interviews were
discussions
managers,
and plant and division managers. Project
whenever possible.
operating
activities
and
technical
personnel,
and their timing were
13
on
reported
written
interviewed
both
their responses.
in
before
In
company and
questionnaires
below).
(see
Participants
and after completing questionnaires
addition,
the
plant documents.
had access
researcher
(For further details,
were
clarify
to
to historical
see Tyre
data
and
Hauplman, (1992) and Tyre and Orlikowski. (1994j.)
and Measurement of Organizational Change
Definition
Adaptation activities
the
new technology
users"
BBA
were defined
or relevant
how
or procedures,
skills
include
at
actions
as
aspects of the
activities
Activities.
intended to modify
operating context (including
were organized,
Examples
etc.).
debugging machine software, designing new tooling, training
machine operators, or developing new maintenance procedures. Activities
were considered part of normal production when the new technology was
used with no effort to alter the hardware, software, or related context and
procedures.
As
level
of the written questionnaire
part
at
BBA. respondents
rated
of effort devoted to adaptive activities such as modification of
machine software or change
in
factory procedures.
Respondents also
out a project history in the form of a time-line, showing
when
were undertaken, and when unusual events
of additional
equipment) took place.
the
level
(e.g..
arrival
tilled
activities
new
For each activity mentioned, respondents noted
of adaptive activity during the period (rated as high=3.
medium=2. low=l. and not
information,
computed
level
the
as
the
sum
significant or none=0).
including
date
integrated
(i.e..
parameters
specific
fullv
to
note major project milestones,
installation,
considered "production worthy"
basis)
this
of the scores for activities noted during that period.
of equipment
consistent
Based on
of monthly adaptive activity in each project was
Respondents also used the time-line
the
the
(i.e..
when
date
it
when new equipment was
was producing
parts
on a
and date when the new equipment was considered fully
satisfactory
defined).
efficiency
and quality achieved; operating
During interviews, respondents also described
problems they had dealt with, and when they were resolved.
14
III.
Study of Time Use
Ditto.
at
This study took place
Inc.
Ditto (a pseudonym),
at
major manufacturer
a
of office equipment, and involved intensive study of a single group of
software engineers working on the creation of a new color laser printer.
The purpose of
work (and
at
how people
study was to better understand
the
home), and
see
to
if
time
use
at
could be improved such that people
this
could accomplish the same amount of work, but also have more time to
spend outside of work.
The researcher observed
the
date
the
until
product launched
week on
The researcher
site
member
in
observing engineers
other
15
in
all
members of
17
all
members of
the
it
approximately midway
intervention,
all
one another
through
17
at
established
("quiet
interact
group on multiple
the
division),
and "shadowed"
down
study
the
engineers
paper took place
this
in
period
During
(January-March).
agreed collectively
Whereas normally engineers could and did
intervention
aspects of their work.
occurred.
The exploratory intervention described
interrupt
September
of the software group for at least half a day, writing
each activity as
time.
in
The researcher spent an average of
June.
interviewed
also
occasions (as well as
the
months from
software group for nine
funding was committed to the development project
four days a
each
the
to
restructure
interact
their
with and
any time that a colleague was working, the
certain
blocks of time
when engineers could
not
time").
At the end of the intervention, the researcher administered a
questionnaire
to
developers
of their
to
assess
worktime.
the
perceived productivity
Furthermore,
researcher was
the
restructuring
site
throughout the intervention and for two months following
recorded
changes
all
engineers on
multiple
that
she
observed.
the
see
occasions before, during and
Perlow (1995)).
it.
on
She
Moreover, she interviewed
after
about their work patterns and any changes they observed.
details
impact of
the
intervention
(For further
15
IV.
Shift
in
Work
Practices and Attitudes
Motorola
at
This study examined Motorola's Arcade plant
in
New
Western
York, a
30 year old plant producing automotive components such as alternators
The research focused on
and regulators.
work
attitudes
Specifically,
how
and
did
that
weeks
to
in
Elma
total
narrative of the
gathered
additional
(also
at
the
in
Western
Elma
New
facility
move from Arcade
plant's
1992 and
during
Elma
move from
The researcher spent seven
York).
to
the
after
is
1993.
The
resulting
based on oral histories
from 40 employees (supplemented by information from over 200
employees), as well as on archival materials such as reports,
memos, and
whom
the
researcher
The researcher interviewed
the
plant
The Elma employees with
presentations.
spoke covered a wide spectrum.
manager, most members of the plant
engineers,
another location.
occur?
it
Data were gathered retrospectively,
Arcade
to
what motivated the change,
question,
the
work practices and
in
move
occurred prior to the plant's
researcher asked
the
changes
the
production workers,
functions such
as
human
the
staff,
manager
for
empowerment,
team leaders, and members of
support
The researcher
resources and finance.
also
talked
number of former Motorola employees and with three employees who
had moved from Arcade or Elma to other Motorola facilities.
to
a
While collecting
the
facility,
this
data,
the researcher
including attendance during
spent
shifts.
all
full
meetings of
all
kinds
--
the
The researcher
meetings of production teams, the weekly
meetings of committees, the weekly off-site meeting of plant
forth.
The researcher
outside working hours.
also participated
(See
also
The researcher consistently attended
lines.
daily
at
Early on, the researcher
received plant and line tours from key team leaders.
worked on several production
working hours
Wasson
in
(x)
a
for
number of
social
further details.)
staff,
and so
activities
16
RESULTS
we
section,
this
In
describe the overall sequence of events at
first
Then we analyze
each of the organizations studied.
to
trigger
or
1.
how,
identify
setting
at
temporal breaks
all,
Shifts
Microsoft.
A
opportunities
efficiency.
for
In
action
problem
in
and
reevaluation
response
to
this
inserting
in
is
without
revision
challenge,
temporal
some teams
"space"
in
was intended
of buffer time
software production
to
at
their
in
Microsoft began
product
the
The
schedule.
enable team members to
and instead turn
periodically
stop
attention
broader issues such as the need
to
production
sacrificing
development cycle by creating specific "buffer times"
insertion
helped to
many ways the most straightproduct development is how to create
The case of Microsoft
experimenting with
the
Four Organizations
in
central
in
data from each
change.
enable organizational
Temporal
forward.
if
the
to
their
collective
product strategy,
revise
schedule or design.
tj'
To do
broken down
this,
into
implementation phase of
the
segments.
three
a
project
was
typically
Each segment consisted of several
weeks of regular development time (during which planned coding and
of features occurred),
testing
The exact
buffer time.
ratio
followed by one to two weeks of unallocated
of implementation to buffer time varied
depending on the complexity and degree of uncertainty of the product.
Project
buffer
managers deliberately
time
was limited
unspecified,
in
the
fast
left
insisting
the
tasks
that
their
changing world of
to
be
ability
PC
performed during
to
software.
forecast
this
future
the
As one team
leader
explained,
been in situations ... where you present a schedule to [the
manager] with some buffer space and he says 'What, are you guys
just going to go fishing or something?
What are you going to do in
that buffer space?' ... Well, if
could tell you exactly what I was
going to do, I'd tell you what the tasks are... But trust mc
don't
know what (buffer time is] going to be needed for, but lime and lime
"I've
I
..
again,
I
know
it's
needed."
I
17
Analysis indicated
several
kinds
that,
of activities.
in
fact,
was typically used
buffer time
for
Buffers enabled the teams to capitalize on
unexpected competitive events
PC
fast-moving
the
in
new
industry,
customer demands or even new ideas they acquired by "learning by doing"
during production.
unanticipated
to
overcome
noted,
And
also
we
the 'oops
Teams
to
biases and limitations of
the
use.
things--
think of that'
didn't
experimented
that
their
inevitable
the
First,
with
is
analysis.
As one worker
'Oops,
we
think of
lengthen the schedule;
what the buffer
buffer
times
team leaders were able
it
Finally,
only appeared to do this
Microsoft improved
how
its
if
are
although there
success rate and popularity of the concept.
is
At BBA, the central problem was, how
ongoing improvement
to
technologies-in-use,
production workers and engineers
look
at
where
to
focus
given
to
the
firm
in
both
find opportunities for
the
pressures
on
on current production
Regular use of the technologies was not consistent with the
requirements.
kind of mental
and physical
(Tyre and Orlikowski.
left
little
or
effort
1994).
required
to
develop and implement new-
Production schedules and efficiency
no lime or attention
for
experimentation and
Routines became established and, even when
recognized as flawed or inefficient, they were protected
get the
a
variation across
the
adaptation.
take
to]
of this success, most projects have voluntarily adopted
buffer time practice,
requirements
As one manager
"[we come
projects use buffer time,
the
ideas
the
filling
clustering..."
result
BBA.
fact
without
projects
for
to
in
one based the
problem-solving process.
understand what are the risks on the project. ..and
a
meet customer
by monitoring what activities ended up
explained, by examining
As
several
schedule and predict
to
better
to
that.'
there for."
is
attributed
comparison on unrealistically optimistic estimates
[problems]
didn't
Second, teams found that adding buffer time did not
expectations.
buffers,
an
human
market launch dates more accurately, and thus
buffers.
as
Buffer times also provided a mechanism
or vacation.
illness
"there's
benefits
with internal events
deal
to
wreak havoc with pre-planned schedules, such
often
that
Buffers also enabled teams
work done.
if
they
were
they
served to
18
however, during brief and intensive spurts of
Adaptation did occur,
activity
that
were almost always triggered by some disruptive event.
events
Significantly,
disrupted
that
only occasionally problems
events which
interrupted
normal production schedule were
the
More
and of themselves.
in
they were
often,
normal flow of production by necessitating
the
a
new problem with the
These events included the introduction of new machines
technology itself.
or tools (14 cases, or 40% of all instances observed), the addition of new
product requirements (6 cases, or 17%) or permission from management to
temporary
take the
stoppage, but which did not signal a
line
machine temporarily off
events
breakdown
(2 cases,
6%).
ignored during normal production,
reported
thus,
12%
to
outstanding
but which
teams not only
problems
had been
that
had caused chronic
examples of new technology studied
adaptation
activity
occurred
following
at
such
BBA, 88%
of
disruptive
events;
of technical changes were accomplished
some kind of break
the
in
action proved beneficial
improvement
for
to
in
all
absence of
the
a
the
in
retrospect
that
specific
breaks
the
in
given project, often enabling a team to
technology.
make
However, not one engineer or
manager who was sampled questioned entrenched assumptions about
need for uninterrupted production.
situation
the
action.
Managers often recognized
significant
sudden machine
a
inconveniences.
or
the 41
only
as
In each of these cases project
breakdown), but they also attended
Of
Occasionally,
immediate problem (corrected the source of the
with the
inefficiencies
17%).
were themselves problems, such
disruptive
dealt
(6 cases,
line
Thus, no one saw
so that "breaks in the action"
a
way
to
the
change the
could be instituted as a legitimate
and ultimately efficient mechanism for enabling
technological
improvement.
Ditto.
their
in
At Ditto, the fundamental problem was
work done
early,
stay
in
late
a
were not
the
as
no one could get
"reasonable" amount of time; engineers had to
and work weekends
complete their own work.
affected
that
in
come
order to get sufficient time to
These efficiency problems also adversely
product development
process
itself,
although
apparent to the members of the organization.
these
effects
19
The researcher observed
"me
about their lack of uninter:_r
software
to
group,
research::
the
examine
collectively
when engineers could
to
work done.
get their
way
a
was
One engineer
and assumptions.
work
respect others'
Interruptions
occur,
still
doing before
are
agreed that the quiet
Managerial as>
i:
fr
"The value was
:.
that
I
,
"I
do not
I
am
me
:,
move from traditional work
:
workforce"?
From 1955 ur:
plant
it
me
to
to
a
to
and
its
management of
state
a
and
local
to
local
the
practices,
community were
more
.:er
ta-^k
v
constantly
my
do
One engineer
^:
said:
Managers
shoulder.
efficient
It
how does
study was.
that
plants.
a
an empowered, involved
However,
strong.
sold
are
:;s."
.
operated in
and technology, though
.Automotive Group decided
close the .Arcade plant.
in
:\g
:
presented an appearance of great stability to
company, and announced
manufactured
..:o
grour o-e.
1980s, the .Arcade
late
^''^
.
The software
in this
practices
....:..-,
incremental changes over time, were relatively stable, and
the plant
to
It
Employment, work
workforce.
to
manner.
traditional
do
._
ways of
:ir
.
.
The question investigated
Motorola:
a
a
_
_:...
engineer^
___-i
.:
constantlv looking over
not constantlv standing over
:•".,
change as well.
this
..
-e've<
..
n.
to
f'-^e"
to
in
shift
..;.j
t^.e
detlne a
to
ij-.::-.ed
The engineers noticed
feel like
._
individua.
were also
-ns
:^
"interaction
-^trained
..
:
..me
:
believe -^eo-^'e ^-.ve begu"
^
and then just give the engineers time
inspecting."
"I
"0'::et
j
.;.^...^^^,n
•-d
r.
-eorle take
b',::
.
:
:
.
wrote:
the dav
;.^
end of the intervention
the
evidence of a
inter..::.
tin:e
—
about productivity
said:
e-^
v..
The focus has movec
time.
...
^--
>;...„.„_.
..
.At
still
.:
.-
even once interruptions were
Yet,
lifted.
what they
...re"'
wav
r_. _:so as a
periods of the day. there was
team.
.; .^ineers
formal schedule dictating "quiet lime"
the
••
provide
to
recognize the value of interactions.
period,
...a
c
Block>
_„..:
c
Tv-^'-t^-
an
.-eate
-
.;.:.:_..
time" was created both as
continually
work.
to
....:,^d
their -.vo'V
not
engineers
that
to
-
some
a>>et>
lies
subject
between
in
1987. the
of
New York
and product lines
other product lines
One reason
all
its
for this
would be
was Motorola's
20
rebirth
as
low-tech products produced
relatively
was
purely high-lcch producer, and thus
a
months
Si.x
later,
however, the decision
was reversed, partly due
management decided
line
Also,
Arcade
the
the
facility
and increasingly unprofitable.
out of date,
old,
Arcade.
at
move away from
its
out of
pull
New York
lobbying efforts by major customers.
to
build
to
to
a
new
facility
and
nearby,
Instead,
manufacture a
to
They eventually selected
of high technology products there.
state
a
site
in
Elma, some 26 miles away from Arcade.
This second announcement marked the beginning of a specific time
(about
period
14
months) during which plant personnel would have
develop a new production system, and
As
productive team.
to
ready themselves to become a
of this process, the plant manager introduced a
part
massive "empowerment" or involvement program.
daily
rhythms were radically
workers'
production
additional
activities.
required
new
designing the layout of the
extensive training
were altered
example,
at
to
required
In
existing
all
review,
competing
for a
levels
large
were involved
number of meetings.
meetings.
For the
to
production schedules.
deal
with
temporal
activities.
For
safety,
workers
time,
their
among
time
demands.
during the time when the plant was readying for the move,
rigid
considerable
there
in
and each
first
allocate
to
After
was
all,
a
the
one hand,
for
the
many
squeezed out by
At the same time, however, employees had
time pressure
customers
constant
On
make time
meetings and training sessions which were previously
Thus,
e.g.,
etc.)
temporal rules and rhythms were discarded.
lime.
in
schedules
Third,
skills.
empowerment roadmap,
workers were encouraged and even required
their
production
Second, employees participated
make decisions about how
to
short,
at
facility.
available
announcement
the
limited
their
and organizational
frequent cross-functional
multiple,
was
employees'
result,
35 new teams were created (including,
peer
were required
these
technical
make time
least
housekeeping,
in
a
spend considerable time on three
to
employees
First,
As
Whereas before
altered.
responsibility
sole
now were
they
routines,
to
still
and
uncertainty
about
how
to
to
use
expected regular on-time deliveries.
and unprecedented tension between the
demands of production and "involvement"
or
empowerment
21
This shift
the
in
temporal rules and rhythms affected worker
plant's
behaviors and attitudes.
The
change was a new awareness among
initial
employees; they moved from taking the workplace and
granted,
what
serious
to
reflection
encourage long-term employees
Thus, the temporal
apparently
changes
led
adopt
new awareness appeared
workers'
in
for
shifts
attitudes
following announcement of a
toward
their
to
kind of work practices and
the
major behavioral changes
to
How Time
2.
to
This
which can generally only be introduced successfully
attitudes
sites.
routines
about the future of their employment and
would require from them.
it
its
in
work
practices
greenfield
at
new
plant
and
to
work.
Enabled Change
Time As A Trigger
Production
interruptions
in
forced people
stoppages
the
to
or
rhythm changes created distinct
midst of familiar work patterns.
focus on
underlying problems, partly because people
could not easily use perpetual busyness as a way
increasing
uncertainty
boundaries,
to
these
Such interruptions
to
ignore problems.
By
and by cutting people off from familiar temporal
interruptions
may
also
have helped
increase
to
openness
change.
For example,
team members tend
that
at
Microsoft, engineers agreed that without buffer time,
become
to
so focused on
meeting specific project goals
they forget or are unable to reflect about what they are doing and
what
is
happening
in
their
environment.
According
to
one,
"[With an] end-date motivated schedule. ..developers were so busy
meeting deadlines that there was no time for the reflection on 'where
we stand'. ..[We] did not stop and reassess problems... [just] plunged
ahead."
By
teams
to
contrast,
actively
the
insertion
"shift
of a buffer time within a project enabled
gears" during development
from
a
mode
of
22
implementation
one of reflection, awareness and analysis.
to
words
the
In
of one engineer,
"[Buffer times] are a time for stopping and insisting that developers
heads up from their work for a reality check on progress....
Too often, teams are too close to projects to see (and admit) where
they are falling behind."
pull
their
BBA
Similarly, at
breaks
and change, even though
this
was not an intended outcome of random
did not really focus on
[the
placer was brought in [and the line was shut
weeks]; that sort of focused our attention."
new
addition of the
the
speeds),
other,
previously-ignored operating issues.
also
engineer explained that bringing
inflection
the
in
"[When
need
(the
new
in
In
work on
the
run the line at
another project
tooling
was
a critical
at
BBA,
an
point of
project:
new
the
came
tools
and the
in
line
forced to take the time to do a great deal of
ideas suggested from
to
to
two
for
helped the project team stop and reassess several
faster
it
down
placer not only triggered
problem raised by the new placer
specific
that:
remaining] problems until the
new
Significantly,
of analysis
levels
For example, one project manager noted
production stoppages.
"We
action spurred unusual
the
in
go through
[all]
sides.
"what
a lot of
-ifs"
was stopped], we were
mental
testing of the
So instead of running
in
parts,
we had
our minds."
During these "mental simulations", project members went back through
project logs and
developing
a
outstanding
Shifts
own
their
new
successful
tooling package,
but
in
also
addressing some
temporal patterns and schedules, like those
a
trigger
accepted,
taken-for-granted
forced
entertain
the
for
rhythms could be altered.
possibility
intervention
at
Ditto
at
Ditto and
change because they showed
that
other
Thus,
aspects of their
could change, and thai they could effect change.
time"
they ended up not only
problems with the new technology.
Motorola, also created
to
recollections of events;
that
people
work
were
settings
For example, the "quiet
provided engineers with the blocks of
23
uninterrupted
not
time
they desired,
that
work could be done
all
exactly what
They
others.
could wait.
interactions
that
like
doing
others
namely, interrupting
them,
to
became more conscientious about
therefore
that
both the
realize
to
realize
to
recognize that they were doing to others
to
they didn't
fact
them
forced
also
They came
isolation.
in
importance of interactions, and the
Furthermore, they came
but
interaction
their
patterns.
sum, temporal breaks
In
activities,
or shifts in
the
to
accepted rhythms of work
--
because they forced people
for
full
banishing uncertainty
In each
organization
members
normal
because
triggered
change
to
points out, the most powerful
organizations
in
random or "self-imposed temporal boundaries
constraints."
normal productive
make uncomfortable decisions about how
to
As Barley (1988: 125)
use their time.
technique
whether
stops
--
is
treat
to
inviolable
as
even
external
of the four cases studied, temporal breaks forced
make
actively
to
temporal
boundaries
decisions about
and
how
were
constraints
to
use
time,
temporarily
suspended.
At Microsoft, for example, managers deliberately avoided specifying
activities
the
to
be
accomplished during buffer times, realizing
time could only function
Thus, teams had
the
team.
the
temporal space as
it
how
decisions about
if
to
buffer
that
use the time were
left
to
determine for themselves what to do with
to
At
occurred.
BBA,
production halts were not
planned; they had the effect of throwing people out of the extremely wellstructured world
of production
how
they had to decide
used
best to
At Motorola, employees
fill.
to
make
who had been
pace of work did not slow down
transition
complex
In
structure
time than
to
to
—
faster during
if
anything
had been previously
it
--
was
it
it
in
shift
rhythm
at
this
new
ways of working
exist;
happen whenever
a
structure
the
old
off attention
While the
plant.
the
to.
Ditto provided more, not less
team members' days because of
However,
trade
was also much more
manage than what workers had been used
one way, the
to
how and when
planning for the new
decisions about
production, to training, and to
time.
and into a temporal void that
relying on a rigid production schedule to structure their time
to
suddenly had
to
schedules,
the
imposition
forced people
assumption
(that
need arises) had been broken.
to
of rigid quiet
recognize that multiple
interruptions
can and must
Engineers also began
to
24
experience a feeling of empowerment for the
that
first
Managers learned
time.
engineers could be trusted to do their work and, as a result, managers
trusted
engineers to determine their
manager
said:
time to do
"Managers
it
are
"I
own usage
As
of time.
software
the
learned to define a task and then just give the engineers
One engineer explained:
without constantly inspecting"
not constantly
standing over
me
me
pulling
to
do other
things."
Table 2 provides some summary examples of
trigger for change at each site.
Table 2: Time as a Trigi^er for Change
Evidence
or
"[Buffer times] are a time for stopping and insisting
that
developers pull their heads up from
for a reality
BBA
time acted as a
Four Organization s
Representative Quotes
Organization
Microsoft
in
how
"We
the
line
work
check on progress."
did not really focus on
problems
their
until
the
new
was shut down];
[the
placer
remaining]
was brought
in
[and
that sort of focused our
attention."
Ditto
People were forced to reflect on their current ways
of working; "It made me think about making any
interruption
Motorola
first."
"People are having to take on more responsibilities
as part of their basic jobs..." This complexity in
schedules caused some pain, since people did not
know how to juggle meetings, experiments, and
regular production. But, they believed that this
"pain
drives
change."
25
Time as a Resource
One
action
the
in
This
of the most obvious but nevertheless valuable aspects of a break
most clear
is
BBA
at
adjustments
provides the time to undertake change activities.
it
Microsoft and
at
at
BBA, where temporal
the forward action of production.
a full break in
leader
that
is
was very
explicit about the
shifts
For instance,
involved
one project
importance of making time for
new technology:
to
"[During the time when the line was stopped] we were able to set
aside time and effort to a special job [involving adjustments to
tooling and procedures]. ..That gave a calm environment for doing
the work, and the people involved [could] work very hard."
At Motorola and Ditto, where there was no break
temporal shifts
provided pockets of time
still
For instance,
work.
hard to find the time to attend to diverse
make time
radically
get
in
the
activities,
even though
it
was
it
was possible
At Ditto the engineers were under immense pressure
altered.
product
that
They suspended
market.
to
to
because old temporal routines were already being
these
for
new
the action,
undertake new kinds of
to
Motorola there was a sense
at
in
their
to
normal ways of working
order to implement periods of "quiet time" because of their desire for
uninterrupted
Quiet time, however, not only provided such
time.
uninterrupted blocks of time to work on technical problems,
provided time
they
to
reflect
on interaction patterns, and
to
but
realize
also
the
impact
were having on others.
Besides simply providing a calm period
in
which
to
consider and
accomplish change, temporal breaks also provided resources for change
a
more conceptual
At
level.
all
four companies, managers and employees
were normally so focused on speed and time efficiency
conceiving of time only as
once they began
to
constraint or
we
did
However,
the
embedded rhythm of
members began
to
see
that
For example,
secret that
outcome measure.
they ended up
in
measure, and a variable that can be used
project.
that
experience breaks
organization
activities,
a
at
at
to
time
is
these
both an outcome
control what happens
in
a
Microsoft a project manager explained, "The big
when we
finally
started
shipping things on time
-
[is]
26
we
finally
put time
in
simply adding weeks
buffer"
—
the
to
As he explained,
schedule."
schedule
the
but rather by using
people up, and enabled them
time
do
to
was not done by
what he termed "lazy and stupid
--
novel and creative ways.
in
better job
a
this
in
This freed
same amount of
the
time.
Table 3 summarizes the evidence supporting the use of time as
resource
for
change.
Table 3: Time as a Resource
Organization
Microsoft
a
in
Four Organizations
Quotes or Evidence
Representative
"The big secret
is
that. ..we
finally put time
in
the
schedule..."
BBA
"...[During the time
were able
to
set
when
the
line
was stopped] we
aside time and effort to a special
job [involving adjustments to tooling and
procedures]. ..That gave a calm environment for
doing the work,."
Ditto
Managers and engineers set aside blocks of time
each week to try new ways of working and to
reflect
to
do
Motorola
on the
"...quiet
time has enabled
get rid of a lot of unproductive things
that
diverse
time
for
already
move.
I
used
me
to
were unnecessary."
Even though
to
effect,
it
new
these
being
was hard
to
activities,
it
find
the
time to attend
was possible
to
make
because old temporal routines were
radically altered by the impending
27
Time as a Coordinating Mechanism
Because schedules, deadlines, and temporal rhythms
means of organizing human
strong
structures
served
to
workgroup or even an
break
it
is
entire
these
activity,
shifts
the
shift
nonroutine
plant.
By
coordinate
to
are
in
such a
temporal
activities
across
a
without some temporal
contrast,
often hard to capture the attention of the critical mass of people
necessary to accomplish real change.
manager at BBA.
"From the time
According
to
one experienced project
equipment was delivered, we could see that we
needed to change the way that the machine was configured.
But
until we were forced to take the machine off line we could not get
[the necessary people] to go back and reexamine some of those basic
engineering decisions.
For all the high-level managers and technical
experts associated with this machine, we were not able to get them
to focus on the specific problem of internal configuration until that
So. we wasted a huge amount of time fixing things on the shop
point.
floor instead of attacking the real problem at the level of the
The whole process was very frustrating.. .we wasted a lot
technology.
the
of time."
At Microsoft, buffer times served as important "re-synching" points
during product implementation.
As engineers concentrate on coding and
individual pieces of a product, there
testing
is
a
tendency for things
"drift apart."
For example, a minor change made
communicated
to
others
design change
is
only noticed by one-half of the team
temporarily,
points,
but
who
collectively,
are
one area
impacted by that change.
stopping
team management ensured
the
in
that
to
the
reassess
(it
project
Or
a
fails
status
to
get
major
happens).
whole team was on
Table 4 provides evidence of time acting as
a
to
at
By
buffer
track.
coordinating devise.
28
Table 4: Time as a Coordinating Mechanism
Organization
Microsoft
Representative
you
"First
in
Four Organizations
Quotes or Evidence
finish
[individual
software
features]
100%. ..Then, [during buffer time], you reevaluate
where you are" as an integrated product.
BBA
managers and technical
experts associated with this machine, we were not
able to get them to focus on the specific
problem. ..until we were forced to take the machine
"For
off line.
Left
Ditto
high-level
the
all
"
engineers
alone,
interrupted
each
other
without even being aware of the effect they had on
each other.
Shifting to established quiet times
changed this because it forced
engineers to change
their patterns of behavior simultaneously.
If
individuals had tried to set their own quiet time,
one would be left with business as usual or with no
way
at
to
all
get their
work done.
Given the more complex set of activities to be
completed before the move, the entire group of
employees who were moving from Arcade to Elma
had to coordinate their activities closely.
Coordination efforts enabled smooth completion of
tasks, and it also created a social bond among
Motorola
employees.
Time as a Symbol
In
each of the companies that
symbolic value.
meeting
tight
we
studied,
time had enormous
There was an emphasis on maximizing speed
production schedules.
to
market or
Thus, anything that broke the smooth
flow of time was also symbolically important.
This was especially true
at
Ditto and
Microsoft, where time
considered the most valuable and scarce resource of
in
the
all
was
and where
breaks
normal rhythm of work were intentionally created, and not just
unintended accidents.
In
the
software industry,
the
notion
of deliberately
29
inserting
unallocated
lime
-'bulTer"
into
the
development schedule runs
counter to current industry practices.
directly
focus on compressing the development cycle.
resource, usually allocated
"waste"
is
down
Most software companies
Time
is
scarce and precious
a
the task level, and no excess
to
"slack" or
As one Microsoft engineer explained,
tolerated.
what happens is when you schedule in buffer time
everyone goes crazy! They say 'you bozo, you idiot, how could you
schedule [that]?
That pushes our ship date way past the point we
want it, get rid of that!'"
"[Traditionally,]
Thus, by legitimizing breaks
and
mentality
At
notice.
the
normal schedule despite
busy schedules,
very
Ditto,
in
for example,
the
fact
management ensured
that quiet
...
crisis
that
people
took
time could be executed
during a high-pressure period of product development was a strong
symbol of management's commitment
to
addressing
problem
the
that
individuals could not get their work done during the normal (9-5) day.
At
both Microsoft and Ditto, by explicitly and deliberately slowing or changing
the
pace of work, managers were signaling their belief
and ultimately the
utility
time further reinforced
of reflection and change.
its
role
as
trigger,
as
in
the
importance
This symbolic role of
resource,
and as coordinating
mechanism.
BBA
At
and Motorola, the events we studied were driven by
exogenous changes
"do"
anything
place.
need
Still,
to
in
in
technology or competition; they were not intended
particular to
the
technologies or work practices already in
temporal shifts carried symbolic weight.
In
both settings the
meet production goals was considered overriding. Thus, the normal
flow of production symbolized "business as usual".
that
flow
signaled
a
shift
Anything
that
broke
from entrenched assumptions and embedded
routines.
Table 5 summarizes the evidence for time acting as a symbol of
change.
to
30
Table 5: Time as a Symbol
Organization
Microsoft
Four Organizations
in
Quotes or Evidence
Representative
Managers use buffer time as a way
to team members that while short
development cycles are important, people also
need to look for and sometimes respond to
"Once you admit you don't
the unexpected.
know everything about
the
alcoholic that says, 'Yes,
BBA
In
but no time
we
get
...
1
future,
it's
for trying
into production
I
can't
new
signal
an
like
have a problem.'
typical project there
a
to
"
is
solutions.
"plenty of
money
Once
do any more
improve the technology], only try to
Thus, when
attend to everyday issues".
[to
a
break in the action occurs, it stands out as a
highly unusual opportunity.
Ditto
The fact that quiet time could even be
implemented was a strong symbol of
management's commitment to addressing the
problem that individuals could not get their
work done during the normal work day.
Employees
Motorola
is
a
move
south of the border
make time
employee involvement
move.
stone...".
that
feel
I
if
that
they can't
Thus,
it
was
management was
for training
in
"our business
that
very competitive"..."
squeeze water from a
remarkable to people
willing to
felt
and
planning for the
there will be
31
A New Model
Discussion:
The data above suggest
about change
established
Change
of
a
model of how temporal
organizational settings.
in
theory,
in
It
is
notable that, in contrast to
Problems were obviously relevant; they
were generally well-recognized and openly discussed
by people
--
provoke change.
that
was no way
there
to
change
Temporal breaks or
ways.
a
First,
assumed
to
be
failed
to
Thus, the source of any organizational inertia was not
break
and acted as
felt
to
break
a
if
enable change in several
made time — which was normally
unchangeable, and unrelenting
When
of
sort
situations.
appeared
the action
in
inviolable,
trigger for change.
their
shifts
was some
Rather, until there
organizational rhythms, the people involved
in
sometimes
--
somehow
every level, but they
at
problems were unrecognized.
break
help to bring
none of our cases was organizational search and
change primarily problem driven.
incessantly
shifts
the
in
--
become
a
people suddenly
action occurred,
found themselves cut loose from regular routines or assumptions.
Even
where these events were not themselves problematic, they served
to
people out of an everyday mental frame that said, "problems arc
just
there", and into a frame that said,
Second, breaks
reasons
that
in
"it's
time for a change!".
the action provided time as a resource.
why well-recognized problems went unresolved
people
felt
for
they were simply too busy to attack them.
normal action provided the slack time or rerlective time
such time
create
tasks.
necessary
temporal breaks and
Indeed,
time"
--
to
assess
shifts
without actually adding
to
work done
break
during
in
the
action
in
a
at
the
for
But
—
long
a
or the
the
was
break
in
ability
to
resolve them.
experience of "more
accomplishing certain
Ditto and at Microsoft found that
product development.
operated as a coordinating mechanism.
event,
time
the
to
so
One of
given time period once they inserted a
Third, breaks or rhythm changes
collective
created
elapsed
Thus, for instance, developers
they could get more
problems and
often
the
jolt
in
the
regular tlow of activities
Change
in
organizations
is
a
and pacing, rhythm, and temporal milestones arc among
most powerful forces
for organizing
and coordinating
a
group of people
32
(Zerubavel,
By
1981).
group disrupt
all
same token, breaks
the
group members
the
at
same
in
the
normal rhythm of the
This facilitated change
time.
by ensuring that readiness for change was synchronized among
who needed
to
Finally,
a
be
the
break
in
role
its
For example,
at
"slowing" the pace of work
At
frequently carried enormous
action
the
as
trigger,
BBA
in
these
all.
settings,
importance and ultimately the
the
as
symbolic
resource, and as coordinating
both Microsoft and Ditto, time was considered
most valuable and scarce resource of
belief in
of those
involved.
power which reinforced
mechanism.
all
and Motorola, pauses
in
By
explicitly
and deliberately
managers were signaling
utility
of reflection
their
and change.
production were symbolically important
because they so directly contradicted the iron rule of focusing only on
maintaining
the
Figure
cases
studied
1
production
schedule.
summarizes how temporal
here.
shifts
led
to
change
in
the
four
33
Figure
J:
Temporal Triune rs for Organizational Chang e
BREAK
BREAK IS
INTENDED
AS TIMEOUTto
PRODUCTION
IN
Microsoft
Trigger Signaled people
made people
look up from their
work;
assess
work.
production
the
:
Coordination
common
together:
a
Symbol
Acceptance by
:
admission
is
inevitable.
Result:
to
:
Quiet time affects
team members;
all
as
:
and
schedules
problem".
"time
the
changes
Significant
works because
it
members follow the "rules".
Symbol Management acceptance
signaled commitment to solving
all
change
that
lime
find
reflection.
for
focus.
management was seen
an
about
people
time;
less
in
Buffer time
team
entire
schedules
daily
in
twice
think
impacts of "routine" interactions.
Resource: More work accomplished
status.
Freed from
:
Coordination
brought
Ditto
Trigger: Change
:
Resource
ALTER WORK RHYTHM S
vs
Result
Imposed
:
behavior changes during;
project
Mixed results
in changing how projects
managed and structured.
parameters.
intervention
lead
longer-lived
to
changes and new-
attitude
patterns
interaction
within the team.
BREAK IS NOT
INTENDED AS
Trigger
TIMEOUT
production,
Motorola
BBA
Forced
:
Resource
a
short
halts
Result:
how
out
Resource:
lime.
lime
:
Production
away from
:
Disrupts
affected
Symbol
together
to
Production
urgency
attracts
attention.
Significant
changes to
technology and
use
and
halt
novelty;
procedures,
use.
no
real
a
all
work
must
new
routines.
Management's
:
shift
make lime
in
thinking.
Result:
change
in attitudes and
routines for technology
but
Symbol
develop
willingness to
signaled
taking
production.
of the whole group.
:
activities.
routine
employees;
signals
figure
to
Legitimizes
Coordination
members
forced
"involvement"
with
new
for
juggle routine work
to
work,
Coordination
idled
workers
plant,
time.
Freed from
:
production
for
more
enter
unstructured
Trigger: To prepare
slop
to
Behavioral change
in
the
form of "involvement
adopted.
practices"
Also,
employee attitudes
and commitment to the firm.
changes
in
34
not proposing that a time-
are
model of organizational change should replace existing theories of
driven
problem-driven search and change.
to
we
important to point out that
is
It
existing
theory
and because
latter,
because
both
--
Rather,
phenomena
theory to
adaptation
may
is
that
show
power of
search.
the
well
not
are
that
understood using accepted notions of problem-driven
existing
complementary
is
enriches the explanatory
it
explains certain
it
our model
We
build on
one of the reasons why problems can provoke
that
problems often create temporal breaks.
Thus, problems
be powerful agents of change not just because they force people to
weaknesses
notice
existing
in
routines
assumptions,
or
but
also
because
they insert a break in the action (whether a full stop or a restructuring of
time).
This provides both a trigger
critical
resource
—
Furthermore,
time
in
this
and adaptation
search
required
--
paper,
to
to
recognize the need for change, and a
undertake change efforts.
we have
highlighted
that
both
described.
some break
what degree change
rather
shifts,
in
the action.
to
Yet even
trigger
in
problems and temporal breaks contributed
As outlined
is
cases
in
Figure
instigated
2,
one might ask
in
change
these cases,
to
the
until
it
any particular case
is
at
work.
Trii^i^ers
as
Complementary Forces
Motorola
v (LOW)
^
BBA
V
^
Microsoft
V
Ditto
V
(LOW)^
CHANGE
IS
clear
by problem occurrence or by temporal
than assuming that only one of these
Temporal Triggers and Problem
is
changes we
Figure 2:
(HIGH)^
where
were clearly not primarily problem-driven;
problems were well-recognized, but they failed
events created
several
(HIGH)
CHANGE IS
TRIGGERED BY
TRIGGERED BY
PROBLEMS
TEMPORAL BREAK
to
35
Thus, we suggest
change
temporal breaks can be important triggers for
that
and of themselves, or a temporal break can be a powerful aspect
of problems that helps to initiate organizational adaptation.
For example,
in Meyer's (1982) paper on "jolts", it is conceivable that the doctor's strike
in
created opportunities
immediate problem
for
slowing the pace of work.
The
latter
to
effect
provided the temporal space
problemistic
March,
1963).
temporal
their
that
shifts
attention
to
broke
where search
we witnessed
situations
where, following
normal routines and constraints, people turned
had previously been
that
This finding points to what
and the more affective impact of temporal
problems tends
to
(Cyert and March,
focus
may
be a fundamental
1988) and
attention
be
to
narrowly on the issues
more
diffuse;
increase
arousal
not a narrowing of attentional
Our theory may
also
But the effects of
appear
to
and awareness,
may
focus (Weick,
decrease certainty
at
least
up
1990).
complement existing work dealing with
or the
five-year
Gersick highlights clock time
the
the
role
1994) has suggested
mark
in
new
a
venture) can also turn
focus of activity from normal routine to reevaluation and change.
a
some
clock-based events or milestones (like the midpoint of a time-
limited exercise,
as
to
lead to a broadening, and
For example, Gersick (1988;
of time in organizations.
hand
at
1958).
they
Thus, temporal shifts
point of inflection.
certain
rather
The appearance of new
shifts.
1963; March and Simon,
may
temporal shifts
trigger,
not
is
between the mainly rational process of problem-driven search,
difference
that
by
also
it.
wide spectrum of issues
a
suppressed or ignored.
(Barley,
but
both created a trigger for change, and
also help to account for cases
our studies,
In
describes,
sense of remaining local to the problem (Cyert and
the
in
Meyer
revenues, etc.) as
(lost
Our theory may
change not only by creating an
organizational
--
but not necessarily temporal
and, to a certain extent, as a symbol.
We
shifts
Thus,
as
--
For example,
organizations
often
create
temporal
progress, go on retreats, etc.
attention,
and
the
at
breaks;
a
might suggest that time
coordinating mechanism, and time as a resource, are also powerful
cases that she studies.
the
in
key milestone points teams or
they
stop
work
to
evaluate
Yet these forces have not received much
independent effects of temporal
breaks have not been teased apart.
markers and temporal
36
Another stream of research examines the fruitfulness of beginnings
supporting search and adaptation.
for
hiring of a
of a
new employee,
new technology,
adaptive
Tyre and Orlikowksi,
support
new questions and
According
1994).
assumptions are not yet
to
new
things;
new employee
subversive) questions,
but
also
does the experienced worker.
simply have the time
to
In
potential
action
the
may
preferences)
questions
difficult
move from
the
also
A
down from
background
We
search process.
make possible
the
job-seeker
to
into
hectic
A
converse than
may
make
periods.
Transitions such as
life.
change simply by
trigger
the
capabilities
foreground for discussion
1991).
a
But transition
change
be newly laid off, or
in
and
her current job
in
the
may slow
during
the
suggest that these sorts of changes in pace help to
reflectiveness
individuals and groups
in
and adaptability often observed
transition.
among
Indeed, without such a break in the
simply surfacing difficult questions would be unlikely
action,
set,
help to explain the adaptive
They require
involvement
full-speed
full-time,
beginning of a
during beginnings people
regarding
(e.g.,
experiences also have a temporal effect.
tempo or pace of work.
often
(and sometime
1978; Louis and Sutton,
and consideration (Schein,
An
1994).
think and
more
later,
changing jobs (for an individual) may help
that
technologies are
ask questions, and to
to
of transition points in organizational
necessitating
the
at
to
other words,
notice things,
in
and
routines,
and deadlines are not pressing.
more time
has
changes that get squeezed out during
Temporal breaks
that
is
has license to ask "dumb"
not only
beginnings
researchers,
norms,
1985;
schedules have not yet been
multiplied,
yet
Murningham,
1990; Tyre and Orlikowski,
relatively relaxed;
commitments have not
the
or the introduction
organizations and
overlooked aspect of beginnings, however,
is
project,
Therefore, people have the freedom to ask
unusually malleable (Weick,
project the pace
these
to
and change because
set.
try
new team,
Bettenhausen and
1990;
structuration
rapid
new
often an intensive flurry of formative or
is
(Weick,
activities
of a
start
formation of a
the
there
At the
change, since there would be no time or flexibility
to
to
lead
to
any
seriously consider the
issues.
Our findings
In
recent
years,
also
relate
organizational
to
research
theorists
on organizational entrainment.
have begun
to
take
seriously
the
notion
human
that
activity
rhythmic (McGrath, Kelly and
innately
is
Machatka (1984); Ancona and Chong (1996), Gersick, 1994).
In particular,
these authors have focused on entrainment processes, or the adjustment of
the pace or cycle of
stronger one.
In
one activity
activity out of step with
its
a given
We
show
of a break in the entrenched rhythm of daily activities can
insertion
that
we examine the power of moving
own accustomed pace or rhythm.
paper,
this
match with the pace of another,
to
be a powerful
way
of helping individuals and groups recognize and deal
with problems that are ordinarily carefully suppressed or ignored.
suggests
that just
as
an organization's dominant rhythms can,
metronome, serve
to
entrench
and procedures, so changes
stability
and enable new questions
also raises
different kinds of change.
organizational
shifts
rhythm.
people's
in
in
these
rhythms can undermine such
surface.
to
questions about the origins of
interesting
Across the four cases studied
of change
kinds
different
identified
In
attitudes
cases,
or
the
the
analysis
break
by
in
indicates
the
in
intention
the
itself
versus
production
make
to
from experiencing
a
a break
and assumptions, the break
taught them
that
time
not
is
difference
in
is
change the way
time
be available.
that
is
in
the
the
Specifically,
action.
action
in
people's basic
people experienced
that
they
(as
had
experienced and the amount that appears
at
Microsoft and
fact
their
to
prove
devote time
to
to
experiences
with
in
that
a
At Motorola, people learned that they could
"nonproductive" activities
be productive.
to
Ditto learned
at
by shifting their usage of time, they could "create" additional time
choose
in
People learned that they could actually
For example, engineers
product development project.
nor
structures),
what people learned
is
immutable and absolute
malleable.
assumed), but rather
Our
was intended versus
each of the settings where we observed significant change
attitudes
behavioral
differences?
these
temporal
shift
behind the break ("time-out"
What does seem
unintended).
about time
(stop
explains
to
cannot be neatly explained by the type of
they
that
action
What
circumstances.
specific
fundamental
to
organization approached
problems, whereas in other cases change was more limited
adaptation
the
in
temporal breaks led
way
we
paper,
this
in
from breaks
resulting
some
in
a
tempo and even the content of various
the
routines
Our analysis
like
This
But
time-outs
at
BBA,
—
and
that
this
could
in
people were not able to generalize
from production or
to
reframe
their
38
assumptions about time
action
complete needed technological
to
no evidence
break
While they benefited from breaks
itself.
was
there
activities,
they recognized such breaks as a tool they could use to
that
of entrenched
out
improvement
the
in
unproductive patterns
but
of behavior.
Conclusion
One
paper
of the most puzzling features of the findings
surprisingly,
is,
how
they arc.
sensible
In
new schedule with
they spend their time
—
the
intent
and they
that
people used breaks
hardly
—
environment
their
in
Indeed,
surprising.
the
in
in
way of making time
occurred.
this
action
everyday
"Hold on just
or shifts
is
not
and change.
In
its
apparent
this
understanding about
paper,
of
why
In
this
we
unique power
time,
we suggest
yet
when we want friends
first thing we are likely
this
or
to
if
role
this
included not only a sophisticated
they
can
trigger
organizational
change,
temporal breaks and shifts play
that
way, we might be able
to
this
in
better address questions
when
but
not just
change might occur.
In
think that the idea of time as a trigger for change has
to
explain cases where problems have existed
they have failed to generate search or change.
common, but is also highly
paper, wc propose only that
may have important
If
threatening to organizational
contributions
the
long
for
a
the
multitude
to
is
not
survival.
In
most ordinary of human processes
make toward understanding and
even resolving questions about organizations'
change.
search
would be both
of studies on organizational inertia are any indication, this situation
only
say
power of temporal pauses
these theories
that
organizations change, but also of
particular,
fact
is
life,
ordinariness,
how problems
also took account of the
process.
the
incorporated into accepted theories of organizational
more robust
stronger and
BBA,
at
to
a minute...".
despite
Yet,
Even
respond
to
undertake needed changes
to
colleagues to revisit their assumptions, the
is
how
At Microsoft, management forced
did.
and
we witnessed
each case,
they explicitly consider
that
engineers to stop their normal work as a
changes
present in this
At Ditto, developers were
processes that strike us as terribly ordinary.
given a
we
ability
--
or inability
--
to
Acknowledgments:
We
thank the following for financial assistance: the
MIT
Leaders for
Manufacturing Program, the Ford Foundation, grant #910-1036, the MIT
International Center for Research on the Management of Technology.
We
would
also
like
collected during
the
Cusumano
thank Michael
to
for granting
phase of investigation
first
at
access to data
Microsoft.
REFERENCES
&
Ancona, D. G.,
in
pp.
Chong, C.
L. (1996). Entrainment:
Organizational Behavior. Research
in
Pace, Cycle and
Organizational
Rhythm
Behavior Vol. 18
,
.
251-284.
Argyris, C, & Schon, D.
Addison-Wesley.
(1970). Organizational
Learning Reading,
.
MA:
Barley, S. R. (1986). Technology as an Occasion for Structuring: Evidence
from Observation of CT Scanners and the Social Order of Radiology
Departments. Administrative Science Ouarterly Vol. 31 pp. 78-108.
,
Barley, S.R. (1988).
On Technology, Time, and
,
Social Order: Technically
Induced Change in the Temporal Organization of Radiological Work.
F.Dubinskas (Ed.) Making Time: Ethnographies of High Technology
Organizations
Philadelphia: Temple University Press.
In
.
&
Murninghan, K. J. (1985). The Emergence of Norms in
Competitive Decision-Making Groups. Administrative Science Ouarterly
Vol. 30 pp. 350-372.
Bettenhausen, K.,
,
,
Cusumano, M.A. and Selby, R. (1995) Microsoft Secrets: How the World's
Most Powerful Software Company Creates Technology, Shapes Markets, and
Manages People NY:
.
Cyert, R. M.,
Englewood
Darley,
J.
&
March,
Cliffs,
M.,
&
Free Press.
J.
G. (1963).
A
Behavioral Theory of the Firm
NJ: Prentice-Hall Inc.
Gross,
P.
H. (1983).
A
Hypothesis-Confirming Bias in
Psychology Vol. 44
Labelling Effects. Journal of Personality and Social
20-33.
.
,
,
pp.
40
Dutton,
The Making of Organizational Opportunities: An
to Organizational Change. In B. M. Staw & L.
(1993).
J.
Pathway
Interpretive
Cummings
(Ed.),'^(Eds.),
Research
Organizational
in
L.
Behavior Greenwich, CT:
JAI Press.
Cognitive Processes in
Feldman, J. M. (1981). Beyond Attribution Theory:
Performance Appraisal. Journal of Applied Psychology Vol. 66 pp. 127148.
,
G. (1988). Time and Transition
Model of Group Development. Academy of
Gersick, C.
ll(No.
1),
J.
.
Work Teams: Toward
Management Journal Vol.
in
a
New
.
pp. 9-31.
Predictable Transitions in Task
G. (1989). Marking Time:
Groups. Academy of Management Journal Vol. 32 (No. 2). pp. 274-309.
Gersick, C.
J.
.
Gersick, C.
G. (1991). Revolutionary Change Theories:
A Multilevel
Equilibrium Paradigm. Academy of
J.
Exploration of the Punctuated
Management
.
Vol.
16 (No.
1),
pp.
10-36.
G. (1994). Pacing Strategic Change: The Case of a NewAcademy of Management Journal Vol. 37 (No. 1). pp. 9-45.
Gersick, C.
Venture.
Review
J.
.
Hedberg, B. (1981). How Organizations Learn and Unlearn. In P. C. Nystrom
& W. Starbuck (Ed.),^(Eds.), Handbook of Organizational Design (pp. pp. 326). Oxford. England: Oxford University Press.
Nystrom, P., & Starbuck, W. (1976). Camping on Seesaws:
Prescriptions for a Self-Designing Organization. Administrative Science
Ouarterly Vol. 21 pp. 41-65.
Hedberg.
B.,
.
.
A. (1990). Evolving Interpretations As A Change Unfolds: How
Managers Construe Key Organizational Events. Academy of Management
Isabella,
L.
Journal Vol. 33 (No.
.
Kiesler, S.,
&
Environments:
Administrative
Lant, T.
1).
pp. 7-41.
Sproull, L. (1982). Managerial Response to
Sensing from Social Cognition.
Quarterly Vol. 27 pp. 548-570.
Perspectives on Problem
Science
,
.
K. (1992). Aspiration Level Adaptation:
Management
Lant, T. K..
Science
&
Changing
An Empirical
.
Mezias,
S. J.
(1992).
An
Organizational Learning Model of
Convergence and Reorientation. Organization Science Vol.
.
71.
Exploration.
3 (No.
1).
pp. 47-
41
Levinthal, D., & March, J. G. (1981). A Model of Adaptive Organizational
Search. Journal of Economic Behavior and Organization Vol. 2 pp. 307-333.
.
&
Levitt, B.,
March,
Sociology Vol. 14
.
.
G. (1988). Organizational Learning. Annual
J.
M. R., & Sutton, R. I. (1991). Switching Cognitive Gears:
Mind to Active Thinking. Human Relations Vol. 44 (No. 1),
Louis,
of
Reviev^ of
.
,
From Habits
pp. 55-76.
Lounamaa, P. H., & March, J. G. (1987). Adaptive Coordination of
Team. Management Science Vol. 33 (No. 1), pp. 107-123.
a Learning
,
Mandler, G. (1975). Mind and Emotion
&
Mandler, G.,
Watson, D.
L.
.
NY: John Wiley and Sons.
(1966). An.xiety and the Interruption of
Behavior. In C. D. Spielberger (Ed.),'^(Eds.), Anxiety
Academic
March,
and
Behavior NY:
Press.
&
The Uncertainty of the Past:
Organizational Learning Under Ambiguity. European Journal of
Research Vol. 3 pp. 147-171.
G.,
J.
Olsen,
,
March,
J.
P.
J.
(1975).
Political
,
&
G.,
Simon, H. (1958). Organizations NY: John Wiley and Sons.
.
Marrow, A.
J.
Lewin NY:
Basic Books.
.
The
(1969).
The
Practical Theorist:
Life and
Word
of Kurt
McGrath, J. E., Kelly, J. R., & Machatka, D. E. (Ed.).^(Eds.). (1984). The Social
Psychology of Time:
Entrainment of Behavior in Social and Organizational
Settings Beverly Hills, CA:
Sage Publications.
.
Meyer, A. D. (1982). Adapting to Environmental
Science Ouarterly Vol. 27 pp. 537-551.
,
&
Nelson, R. R.,
Change
.
Jolts.
Administrative
,
Winter,
S.
(1982).
An Evolutionary Theory
MA: The Belknap
Cambridge,
Press of Harvard
of
Economic
University Press.
Schein, E. H. (1978). Career Dynamics:
Organizational
Matching Individuals and
Addison Wesley.
Needs Reading, MA:
.
&
Gutek, B. A. (1987). Some Time Dimensions of Work:
Measurement of an Underlying Aspect of Organization Culture. Journal
Applied Psychology Vol. 72 (No. 4), pp. 642-650.
Schriber,
J.
B.,
oi'
.
Sitkin,
S.
Research
B.
in
(1992). Learning Through Failure:
Organizational Behavior Vol. 14
,
.
The Strategy of Small Losses.
p.
231-266.
42
&
Staw, B., Sandclands. L.
Organizational Behavior:
E.,
Ou'arterlv Vol. 26
501-524.
.
.
pp.
A
Dutton,
J.
E. (1981). Threat-Rigidity Effect in
Multi-level Analysis. Administrative
Science
(forthcoming 1996). Gaining Competitive Advantage From New
Technologies: Patterns of User Adaptation. In R. S. Rosenbloom & R. A.
Tyre,
M.
J.
Burgelman
and Policy
M.
(Ed.),^(Eds.),
Research on Technological
Innovation.
Management
&
Hauptman, O. (1992). Effectiveness of Organizational
Response Mechanisms to Technological Change in the Production Process.
Tyre,
J.,
Science Vol.
Organization
Tyre,
M.
J.,
.
&
Orlikowski,
Patterns of Technological
Vol. 5 (No.
1),
pp.
3),
W.
(1994b).
J.
Adaptation
pp.
in
301-320.
Windows
Temporal
of Opportunity:
Organizations. Organization
Science
(1980). Problem Solving, Planning and Innovation:
Tests of the Program Planning Model.
Human
Relations Vol. 33
.
.
pp.
Part
711-
740.
Weick. K. (1990). Technology as Equivoque. In P. S. Goodman. L. S. Sproull,
& Associates (Ed.),'^(Eds.), Technology and Organizations (pp. pp. 1-44). San
Francisco,
.
98^-118.
Van de Ven. A. H.
1-
3 (No.
CA: Jossey-Bass.
Zerubavel, E. (1982) Hidden Rhythms: Schedules and Calendars in Social
Life
Chicago: University of Chicago Press.
.
75
3 3
0/.:.6
Date Due
MIT LIBFIARIES
•
llif
-
3 9080 01444 4399