S
f
UBSASIES)
g;
-tp
DCV^
GROUPWARE
IN
PRACTICE:
AN INTERPRETATION OF
WORK EXPERIENCE
Christine V. Builen
John
L
Bennett
March 1990
CISR
Sloan
WP
WP
No. 205
No. 3146-90
Center for Information Systems Research
Massachusetts
Institute of
Sloan School of
Technology
Management
77 Massachusetts Avenue
Cambridge, Massochusetts, 02139
GROUPWARE
IN
PRACTICE:
AN INTERPRETATION OF
WORK EXPERIENCE
Christine V. Bullen
John
L
Bennett
March 1990
CISR
Sloan
•1990
WP
WP
No. 205
No. 3146-90
C.V. Bullen, J.L Bennett
Center for Information Systems Research
Sloan School of Management
Massachusetts Institute of Technology
.^
Abstract
Observers have identified a potential for major improvements in organizational
productivity made possible through the use of personal computers serving as a means to
people into task-oriented teams. The study we conducted, given in overview form
here, offers an early examination of how people are using personal computers for such
electronic exchanges via networking. Our interviews of 223 people who were using several
"groupware" systems in a sample of 25 enterprises indicate how they employ these software
tools to support their group work.
link
We
conclude that complex interactions of social and technical factors affect the
We outline issues which both the developers
use of groupware systems in organizations.
of systems and the managers implementing groupware systems must understand in order
to facilitate the design, introduction,
and use of these systems.
acknowledge the many individuals who agreed
to be interviewed, and interacted with us in the course of this research. Without their
cooperation and genuine interest in the topic, it would have been very difficult to learn
about the experience of groupware use in organizations. We wish to thank David L.
Anderson who assisted us in the fieldwork and provided support for the ideas presented
here.
We also acknowledge the following for their valuable roles in reviewing drafts:
John Henderson, J. Debra Hofman, Bob Johansen, Wendy Kellogg, Bob Mack, Tom
Malone, Wanda Orlikowski, Judith A. Quillard, John Richards, and JoAnne Yates.
Acknowledgements:
The authors wish
to
Groupware
I.
An
In Practice:
Interpretation of
Work
Experiences
Introduction
The
fact that personal
an astounding rate
computer (PC)
being heralded from
is
availability in the
work place
is
growing
at
many comers:
"We're going from 11 million to 34 million PCs by 1994. PCs now make
up about half the electronic keyboards in use but will account for 70% of
the total in 1994." (Dalton, 1988)
"About 7 million PCs and 300,000 multiuser systems were installed in the
United States by early 1985; and these numbers are still growing at about
15% annually. By 1985, there were terminals or microcomputers for at
least 10 million people, or 20% of the United States' white collar workforce."
(Kling and lacono, 1988)
The
usual assumption that the use of personal computers contributes to increased
worker productivity
economy
turning into an
is
open question: Loveman reports
that at a national
researchers have failed to establish a significant relationship between
level,
information technology investments and increased productivity. (Loveman, 1988).
What
How
aspects of
PC
use could contribute to measurable increased productivity?
do we get beyond the extravagant claims often associated with PCs
reality of their
systems
value?
needed
understanding
to
Are changes
bring
how PC's
the
in the organizational
potential
are being used
into
actual
workplace or
realization?
now can be important
to discover the
in the design of
We
believe
that
for understanding the value
that PC's could potentially bring to the office environment.
One
area where personal computers are being brought into use
is
to support the
work of business teams.
way of life in many organizations...
many as a wave of the future." (Bullen and
"Business teams are becoming a
Business teams are seen by
Johansen, 1988)
'Traditional departments will serve as guardians of standards, as centers for
and the assignment of specialists; they won't be where the work gets
done. That will happen largely in task-focused teams." (Drucker, 1988, see
training,
also Reich, 1987)
Our
particular focus
on looking
is
at
people
We
through personal computer workstations.
We
technology brings to the office environment.
who work
in
teams and are networked
seek to understand the value that the
believe the quality of the results of using
information technology to support group work has the potential to far exceed what
achieved
today
through
use
the
of
PC's
by
within
individuals
isolated
relatively
is
organizations.
If
indeed team work
so in the future
-
is
an important form of
then investigating
how teams
office
work now
-
and
will
be more
of people work and studying the use of
personal computers in a team environment should be valuable for understanding
information
technology
organizations.
is
and
used
how
The MIT Commission on
this
is
affecting
productivity
in
how
today's
Industrial Productivity found:
'The third recurring weakness of the U.S. production system that emerged
from our industry studies is a widespread failure of cooperation within and
among companies." "... most thriving firms in the U.S.... have learned to
integrate technology in their ... strategies and to link [their strategies] to
organizational changes that promote teamwork, training and continuous
learning." (Berger, et
The presence
A
1989)
of PC's networked together in communication paths provides the
physical infrastructure.
processes?
al.,
But
is
a
new kind
of software needed to provide tools for team
term which has become popular during the
a term applied to software that
is
intended to be used
in
last
few years
is
"groupware",
support of interpersonal work
within an organization:
"Groupware
is
a generic term for specialized
for the use of collaborative
work groups.
computer
aids that are designed
Typically, these groups are small,
project-oriented teams that have important tasks and tight deadlines...
Sometimes, groupware is used by permanent groups or departments...Group
interactions may be formal or informal, spontaneous or planned, structured
or unstructured." (Johansen, 1988; see also Engelbart, 1963, 1968; Hiltz and
Turoff, 1978; Stevens, 1981; Hiltz and Kerr, 1981; Kerr and Hiltz, 1982; Rice
1984)
Our
questions about the use of PC's, the role of PC's
when teams
are linked
through communications networks, the underlying issue of productivity, and the role of
specialized software for use on PC's and workstations,
began
status of
group work
employed
insight
on
and
to observe
of group work.
factors that should
served as background as we
used a case study methodology to investigate the current
in organizations
in the facilitation
that can help guide
II.
We
research project.
this
all
how computer-based
Our purposes
in this
tools
were being
research are to develop
be influencing software design, and to report experiences
managers who put group support systems
into practice.
Research Design
An
While the outline
interview framework served as a focus for data gathering.
provided for
initial
emerge from our
distinctions
interviews.
we knew would be
of interest,
we
let
other distinctions
This work illustrates a research methodology often used by
anthropologists and titled in a variety of ways, including "exploratory observation" (Malone,
1983) and "contextual inquiry" (Bennett, Holtzblatt, Jones, and Wixon, 1990).
of study
is
not intended to be a controlled experiment or a large sample survey.
technique focuses on interacting with people
The
goal of data gathering
interaction.
This type
The challenge of
to accurately report
and
the examples studied.
is
in their
to obtain insights
this
methodology
is
interpret, while allowing
own
The
contexts as they do actual work.
through observation, interviews, and
that
it
relies
on the
skill
of the observer
unexpected phenomena to emerge from
This approach often results in uncovering research questions which
can be investigated through controlled experiments or additional contextual inquiry.
conclusions present such opportunities for further research.
Our
Table
this
1
paper we
Therefore,
We
people
framework
outlines the topics that served as a
summarize,
we do
synthesize,
not present data
to guide the interview. In
and present points
interpret,
to
us.
each category shown on the interview outline.
in
used the interview outline as we spoke with two hundred and twenty-three
in twenty-five organizations,
represented at thirty-one
sites (see
Table 2 for
on companies, number of interviewees, and groupware systems available
interview lasted a
minimum
in each).
in the individual's office
We
chose organizations
in
as our
In
and
size of
which groupware systems were available, and those
systems helped to define the set of groupware systems that
names are coded,
Each
or work area.
twenty-five organizations represented a wide range of industries
companies.
details
of one hour, with the longest interview lasting two hours.
almost every case, the interviews were carried out
The
salient
we
studied.
agreement with those interviewed guaranteed
Organization
confidentiality.
We
consulted with each organization to choose groups for our interviews that met the
following criteria:
cohesive
conditions
business
teams,
which
would
challenging
facing
emphasize
the
environmental
importance
of
coordination for achieving their goals and objectives;
»
teams that had some form of information technology available
to support the
The
size of
work of
the group.
our work groups ranged from seven people to thirty-Sve people.
interviewed included individuals at
all
levels of
management
within the target
Those
work group
and, where appropriate, support personnel (administrative assistants and secretaries).
most organizations, the managers
to
whom
as part of the case study to help establish
the
work group reported were
some of
In
also included
the contextual information.
'
«
;
!
K.
'
!
Tg^ .V
,..:-.J.J.
GROUPWARE
'
;
:
!gg
Case Study
Interview Outline
General background information on the organization, the work group, and the
individual being interviewed;
Detailed information on the work group or project:
o
o
o
Members
Description
Mode
of operation:
meeting frequency
forms of communication (face-to-face, phone, electronic, video)
levels of stress
leadership
boundary management (relationship to world outside
Description of
how
project);
tasks are accomplished;
Determination of information technology (I/T) tools that are used to
accomplishment with detailed description of use;
Determination of general sense of satisfaction with existing
mode
facilitate task
of operation;
Suggestions for change;
Probing of interviewee's sense of the future:
Types of group work that will take place;
o Changes anticipated for organization as a whole;
o Needs for different lyT tools.
o
CISR Grpwr CVB/m*
Table
1
We
did not choose our groups for study
We
sample.
contacted potential research
knowledge of
it
a wide
includes
basis of referrals
resulting
sample
and our own
drawn from
is
a
range of organizational sizes and
While these characteristics do not guarantee that the
geographic dispersion.
random
basis of a statistically
on the
sites
However, the
their use of technology.
wide variety of industries, and
on the
can
results
be generalized, they do suggest that we are not seeing isolated and unusual instances of
groupware
tool use.
Conducting a study of work groups
to define inclusion in a
defined
as
"identifiable
work group.
raises
What
are
groups of information workers" (Bikson, et
bounded whole
We
at
its
interesting questions about
bounds?
Work
how
groups have been
and bounded subsystems of a whole organization [with
recognized purpose, which unifies the people and
a
some
"collaborating
activities." (Trist, 1981);
and "multiple individuals acting
1989);
al,
a]
as
order to get something done" (Rousseau, 1983)
in
found a variety of organizational forms constituting work groups. For example,
CableCo
work group coincides with
the
the organizational department, although
it
is
spread geographically across the continental U.S. At SmallCons, the "work group" consists
And
of the entire firm.
times
it
at
SoapCo, the work group
entity (e.g., all
it
consists of the
marketing groups), and under
a subset of entities
(e.g.,
a flexible concept such that at
one location
involves an organizational entity at
group), while at other times
is
(e.g.,
the Boston marketing
worldwide instances of the organizational
still
other circumstances, the work group
Boston, Chicago, and San Francisco marketing groups).
In the world of electronic communications, the composition of
showing important changes from those observed
provided by communication technology of
identify a formal organizational unit as a
an
office area
separated
all
in
kinds,
the
it
is
work group. That
do not necessarily work together
may form
is
past.
Because of
becoming more
is,
as a group,
a team focused on achieving a
work groups
some people
is
flexibility
difficult
to
co-located in
and people geographically
common work
result.
Through the
power of electronic media, these groups are dynamic and
and informal organizational
fluid; this
is
tnie for both formal
units.
"The traditional concept of an 'organization' is no longer useful to managers
or students of organizations. It is dominated by models of structure and
physical identity at a time when telecommunications has eroded the
boundaries between firms and changed the nature of coordination across
geographic location" (Keen, 1988)
Given our research
interest
designed for use by teams,
we had
This term often
"groupware."
in
the
factors
develop a working definition of what constitutes
to
used to indicate computer software that
is
support interpersonal work within an organization.
clear connection
important for software specifically
between the software
tool
is
intended to
Early notions of groupware saw a
However
and the group processes.
current
manifestations of groupware tools appear to focus on the technical qualities of the
functionality
We
and may,
in effect,
ignore the dynamics of group use.
have employed a broad definition
in
our research
In time, the term
the evolving nature of this field.
groupware
to include only those tools specifically designed to support
present,
it is
useful to include
all
and calendars on a system designed
working definition of groupware
groups to
facilitate the
is:
to
(e.g.,
order to accommodate
will
probably be narrowed
group work.
tools being used to support
represent user adaptation of an existing technology
in
group work, even
group agreement
keep such functionality
private).
under
this
if
at the
the tools
to share files
Therefore, our
computer-based tools that can be used by work
exchange and sharing of information.
There are a large number of systems with a large variety of
fall
However,
groupware umbrella. Figure
1 illustrates
a
framework
functionality,
which
for organizing these
systems using the dimensions of time and place to create four domains which describe
circumstances of interpersonal work (Bullen and Johansen, 1988):
>
same
same
*
different time,
»
different time, different place.
time,
same place
time, different place
same place
While each of these domains
we decided
is
to investigate those
important and the four are interdependent, for
computer systems which can be used
this study
to facilitate
work
in
the different time, different place domain.
III.
Information Technology Tools Studied
In the
course of the case studies,
we
focused on the use of eight different
information technology systems at various times and various locations.
specific systems
participate.
that can
make
was influenced by
their
The choice
of the
presence at the organizations that agreed to
Within the various systems a number of functions can be considered as tools
be used for support of groups.
In order to describe these systems broadly,
we
All of the systems studied provide the following
the following generalizations.
functionality (i^ummarized in Table 3):
Construction/Editing Facilities:
creation and editing
import graphics.
and editing
as
facility.
One
Some
All systems provide at least a rudimentary text
include elaborate editors and provide function to
special purpose system
(ForComment) focuses on
an aspect of group work.
Electronic Exchange of Text:
This includes electronic mail and/or conferencing,
gateways to other systems (both internal and external,
document
joint authorship
transfer.
As
e.g.,
facsimile
a result of the text being captured in a computer-
transfer),
and
^
LU
LU
GC
LU
UL^
Q
0)
UL
LU
LU
<
GROUPWARE
Functionality Provided
In
Groupware Tools
1
Construction/Editing Facilities
A
Electronic
Exchange
of Text
Directory
^
Time Making/Time Keeping
^
General Tools
CISR Qrpwr CVB/mfb 005
Table 3
11
readable form, the content can be reused, edited, re-sent,
to
exchange
of
environments.
images,
graphics,
Some
and
spread
of the tools provide ways to
data
sheet
Capabilities with respect
etc.
manage
differ
in
different
tool
the exchange of text through
folders or through automatically linking related messages.
Directory:
address
file
a traditional
to support data
"little
minimum
This functionality at a
provides a
exchange by users of the system.
Some
name and
electronic
of the tools provide
black book" functionality, where extensive data on mailing addresses,
multiple telephone numbers, and notes about individuals
(e.g., secretary's
name) can be
stored.
Time Marking and Time Keeping:
provide a
facility for
ranges from
lists,
this
The
capability
example, recording repetitive events, reminders, "to do"
data to other system functions.
General Tools:
financial
one (ForComment)
recording events scheduled for a particular date.
this basic task to for
and linking
Ail the tools except
Some
of the systems provide tools for the support of
and budgeting areas, and because of the ways
in
work
in
which the tools can be used,
support some project tracking capability.
Integration across the functionality provided
is
an interesting concept
in
groupware
research for two specific reasons:
We
found people hindered in the use of tools because of problems
associated with the degree of integration;
1.
2.
This term
is
used imprecisely by both vendors and users as a measure
of quality in describing groupware systems.
in
our interviews and we
exploring
its
there
is
Therefore
it
showed up often
value in defining integration and
application to these systems.
The concept of
relative term.
feel
One
"integration of function"
needed
for support of
group work
is
a
aspect of integration can be measured by examining the process
12
required for each user to
move
of system-dictated steps
(e.g.,
Another aspect
freely
between functions, and by looking
log off
from one function, log on
one function to another without requiring special transformations.
we
to another function).
move data from
the extent to which the software enables the user to
is
for the presence
Thus, integration, as
use the term, refers to the flow of control during work by an individual or by team
members, and
to the flow of data during the process of individual interaction with the
software or during interaction
significantly
and across the functional categories
within
Integration
among
among team members.
the various tools.
As
a result
some
above
listed
differed
of the systems resembled a group
of functions rather than a cohesive package.
Brief descriptions of the individual systems appear in Appendix
a
list
each.
Table 4 shows
I.
of the systems and which of the general categories of functionality are provided
We
operation
do not provide here
*"or
each
ordering rules,
this
The range
of capability
is
on
the details of
wide
in
terms of search mechanisms,
etc.
rv. Observations,
In
tool.
specific information
Key
Issues,
section
and Conclusions
we summarize
the
results
of our
study.
It
is
important to
understand, that because of the complexity of intervening factors, the observations
report here do not have simple explanations.
Markus and Forman (1989) supports
in
performing research
in
We
if
we need
to take multiple perspectives
have found the framework suggested by
Table 5) to be particularly useful to us as
factors influencing the adoption of technology.
From
we
sorted out
a tool perspective, technical solutions
the innovative benefits of the function would overshadow any
13
we
Research by lacono and Kling (1988) and
the notion that
in organizations.
lacono and Kling (shown
are offered as
in
S Ui
<
Z ^
Ui Ul
a a
lit
o
and
historical, political,
might be present
social factors that
in the
environment.
Instead,
lacono and Kling find that the environments into which computer-based tools are
introduced should be viewed as institutions. As institutions, the environments exhibit
barriers to adoption that have httle or nothing to
Consideration of
innovative tools.
based support
"We
do with the technical merits of
historical, political,
them
those developing tools and those introducing
many
and
into the
social factors
the
can forewarn
environment where computer-
provided:
is
conceptualize these patterns as the social organization of computing.
We
define 'social organization of computing' as the choices about computing
(both social and technical) which
become embedded
in
and which are experienced by the users as part of the
their everyday work world." (lacono and Kling, 1988)
work environments
social practices in
Other researchers have stressed the importance of understanding the balance between a
technology perspective and an organizational one.
"...group
members
technology,
the
Bikson, et
al.
(1989) comment:
on
are interdependent not only on one another but also
and
technical
and
organizational
issues
are
closely
The more advanced the information-handling tools provided to
group, the more critical it becomes to give equivalent and concurrent
interrelated.
the
attention to the social processes through which these tools are deployed,
and
mutual adaptation rather than maximization of either the social or
(See also Mumford and Ward, 1968)
to seek a
technical system in isolation." (p.89)
In
making our observations and drawing conclusions we were struck by
importance of understanding the complex interplay of factors which influenced the
organizations
conclusions
1.
designers
we
studied.
we have
From
sorted
However,
them
into
a Design Perspective
ought
to
groupware systems.
take
Our
into
-
in
order to
simplify
the
presentation
the
specific
of our
two categories:
In this category
consideration
when
we
discuss those conclusions which
conceptualizing
functionality
conclusions suggest that while designers must be concerned
15
for
about the technical solutions, they need to go beyond the "tool perspective" to include
an appreciation for organizational factors that
From an
2.
conclusions which
may
influence the use of groupware systems.
Perspective
Organizational
management ought
-
In
this
These conclusions suggest ways
could
factors
organizational
which
implementation process and subsequent use.
might
discuss
when planning
to take into consideration
implementing groupware systems.
anticipate
we
category
in
for
and
which managers
the
influence
those
groupware
However, managers must also be careful
to not fall into a pure "institutional perspective" to the exclusion of
concern about the
quality of the technical solutions.
As with any
There
will
categorization, this dichotomy of conclusions
considerable overlap
is
in the
groupings.
We
is
do not intend
an oversimplification.
to
imply that designers
not find useful information in the second grouping, nor that managers will not benefit
from the conclusions
As
in the first category.
interviewers, observers,
and interpreters of these work experiences, we have
been conscious of these perspectives.
conclusions
Some
we support
of the
In the following discussion of key issues
and
the ideas using observations and quotations from our field work.
researchable questions which emerged are suggested in
each section.
However, others can be formed from the work reported here.
IV. 1.
IV.l.l.
From a Design Perspective
Electronic message communication
The
functionality for sending
the products
we
studied,
was
far
is
the primary tool.
and receiving electronic messages, available
in all of
and away the one aspect of groupware that was heavily
used and universally stated as valuable.
"I
love this tool.
I
can reach people and they can reach
anyplace and discuss anything." (SnackCo)
17
me
anytime,
The
desire to have support for communication within a
primary motivation for acquiring the
tool.
work group was
People quickly grasped the idea and procedures
associated with electronic messaging in contrast with their failure to use
functions available in these systems.
does not
in
any way
available functions
capability,
mean
it
The simple presence
be used.
will
of
was used
its
user
interface
extensively.
we
However
ease
design,
many
of the other
of function in a groupware tool
In each system
were ignored by the using groups.
regardless
sophistication
that
usually the
or
studied, several of the
the electronic messaging
or
difficulty,
In several instances interface features
level
of
were either
ignored or adapted by the people to accomplish a simphfied process of communicating
electronically.
For example, the interface provided by The Coordinator (Version
language related to the underlying theory of speech acts (Searle, 1969).
is
I)
contains
This language
intended to lead the users to think about what they are doing and to characterize the
communication as one of several choices,
e.g.,
a request or promise, etc.
software provides a choice for simple e-mail (called "free form"),
While the
we found people
Not
consistently sending each other "requests" regardless of the content of the message.
surprisingly, "request"
people
we
a message.
is
the
first
menu
choice and where the cursor
falls
interviewed reported that they ignored the choices and just
However, they had high praise
by default.
Many
"hit enter" to
send
for the tool:
"The Coordinator opens
'The Coordinator gets the information out!"
communication lines." (CableCo)
'The Coordinator's major advantage:
audience." (SnackCo)
instant
communication with
a large
In another example, the user interface for Higgins electronic mail walks the user
through a series of steps to answer questions about
dates, etc.
in the
A
how
to assign key words, follow-up
person using Higgins to help manage information or for sharing messages
groupware sense, would answer these questions.
skipped through
this
menu
However, the users we observed
leaving blanks in most categories.
18
People reported that they
understood the value of the categories, and knew that they "should" be
but that they were
One
in a
possible response to this observation
immediately suggests that
a
the blanks,
hurry and just wanted to get the message sent.
is
that the message-flow needs of
groups are so great that they overshadow the other system-based
the basis of
filling
we need
This
activities.
a definition for this kind of communication which, on
what we observed, could be termed sending and receiving messages.
number of
work
Any
of
the tools provided as part of a groupware system can be also thought of as
entering into "communication" in the sense that they can be used to support transfer of
information in some way, for example:
calendar entries
expense reports
reports of project status
telephone directory
spread sheets
budget reports
tickler files/to
It is
do
lists
interesting to note that these particular functions tend to
electronic messaging in the information technology tools themselves.
this distinction as "correct".
that they acted as
spoke
in
if
this
be separated from
We
do not support
However, as we observed and interviewed people, we found
separation of functionality was real for them, and they generally
terms of "e-mail and the other things the system can do."
This raises the interesting point of
how
to distinguish
between what people say
they do with information technology tools (or want to do) versus what people think they
should or should not do because they have been influenced by the tools at hand.
cannot answer
this
question definitively.
We
can, however, say that given the choices
existing in information technology tools today, the
calling
"message functions" almost exclusively.
19
We
people
we
studied used what
we
are
Our
interviewees frequently stated that they chose groupware systems because of
the mix of functionality offered.
Given
an important question
that,
is:
why do
they not
use most of the functions.
For example, users say they want groupware tools that provide calendaring
Yet
functions.
in the
majority of organizations
People gave a variety of reasons;
being used.
the desired function was present, this did not in
"We
studied the calendar tools were not
the net result
and of
itself
the fact that although
is
mean
that
it
was used.
thought having the calendaring function available on Profs would be
useful in organizing meetings, but
it."
we
no one has taken the time
to learn
about
(InsurCo)
"One of
the reasons
functionality
it
we chose Higgins was because
provides.
However, most people
of the wide range of
just
use
the
e-mail."
(RBOC)
If
is
developers commit resources to provide function,
it
is
important to understand what
seen by users as a barrier between "offered" (by the system) and "used" (by the people).
Other
factors, as
we
shall report,
were important
in
determining whether a function was
used.
In conclusion, our field observations
electronic messaging.
If
we can assume
show
that the tool people use the
that these busy people
however,
A
we
is
electronic messaging support.
is
would use the portions
of the systems that they need the most to accomplish their work,
what they need the most
most
we may conclude
that
In the following sections,
discuss other factors that increase the complexity of forming this conclusion.
researchable question here revolves around gaining a better understanding of the value
to the
user of the electronic messaging function as
functionality.
Another research topic raised by
this
compared
conclusion
to the value
is
understanding the
barriers to effective use of the functionality other than electronic messaging.
20
of other
Message linking
rV.1.2.
One
improvement provided by electronic communications.
a key
aspect of electronic message communication that was emphasized in our
was the
interviews
is
distribution
ability
messages concerned with one subject area or a
link
to
This functionality
list.
Coordinator) and
it
is
provided
(VAX
Notes) and
Computer conferencing organizes
Therefore when a message
is
is
"look in one place for
is
I.
electronic messages according to topic areas.
all
sent,
it
is
addressed to a topic. The concept
in
all
complained about the
(i.e. files
individual, etc.
list,
our interviews that they gained
much
difficulties
PROFS
of tracking
for grouping
down
value by being able to
XYZ."
discussion pertaining to project
observation, users of e-mail systems like
•
and The
not limited in any way, and can in fact represent a project, general
People reported
Message
tools (Higgins
computer conferencing, which
In-House System
in
composed and
discussion, software release, distribution
mail folders
two of the
also inherent in the concept of
is
available in All-in-One
of a "topic"
in
In contrast to this
and All-in-One (without Vax Notes),
related messages
and managing
their
messages by categories).
linking provides four primary values:
collection of notes in
one place
chronological record
ability for
•
In
latecomers to view an entire record of interaction
knowledge of the
The Coordinator (Version
to the underlying theory:
of
this,
"right" place to
I) this
put
new messages.
functionality
the "conversation"
is
is
embodied
in a
the primary unit of interaction.
each time someone replies to a message, the reply
is
Because
automatically linked to
previous messages in the stream and becomes part of the "conversation."
21
concept basic
all
"I
use the traceback function a
lot to find
out
how we
arrived at a particular
point in the conversation." (SnackCo)
receive an answer to
"If I
because
it
messes up the
my
request that
ability to follow
is
not linked,
get annoyed
I
back through the conversation."
(CableCo)
This feature of The Coordinator came out
our interviews as one of the most valuable
in
aspects of the tool for these users.
Our
showed
interviews
clearly that
messages that are related by subject.
available for
1988).
more than
people value the
It
group and
link
This "computer conferencing" capability has been
fifteen years in
electronic
However, general understanding of how
has been limited.
to
ability
it
remains interesting to us that
communication systems (Johansen,
works and general
this functionality
availability to users
should be singled out
by many users as a key benefit of groupware.
Historically,
knowledge workers have always sought ways to organize the volume
The
of information they have manage.
the vertical
file
system (the
file
first
cabinet,
major innovation to affect
in
this
process was
1892) which facilitated the grouping of
correspondence by subject:
"[Vertical files]
had several advantages over [other forms of
filing].
Most
importantly, the folders allowed related papers to be grouped together and
removed from the files for use." 'The new equipment alone was not
enough to make storage and retrieval efficient. In a textbook on filing
published by the Library Bureau, vertical filing of correspondence was
easily
defined as including the organization of papers
filing
apparatus
itself:
'The
definition
in the files, as well as the
of vertical
filing
as
applied
to
correspondence is -- the bringing together, in one place, all correspondence
to, from or about an individual, firm, place or subject, filed on edge, usually
in folders and behind guides, making for speed, accuracy and accessibility."'
(Yates, 1989)
In effect nothing has changed:
groupware message linking or conferencing allows people
to carry out this task for electronic correspondence!
22
The need represented
linking),
is
in
our interviews,
(i.e.,
focus on the value of message
one which should be carefully investigated by both designers and implementers
People use message linking
of groupware.
to:
manage communications
»
keep records
develop "group memory"
exchange documents.
This conclusion
may be
telling us a great deal
more than what
is
at first obvious:
rather than looking at "fancy," innovative functions for groupware systems, designers should
be focusing on how to better solve the basic need of
office workers,
i.e.,
to
manage
large
There may well be ways other than those we see today
volumes of information.
for
designers to address these needs and better serve the groupware user.
IV.l J.
What
It
functionality
became very
functionality in
included and
is
is
is
of the
There are two
which we observed.
included.
the calendaring function.
interviews focused on one fact:
but
tool.
some
of the best examples of functionality requested by "the marketplace" but not
used effectively
traditional
offered are important factors.
groupware systems because of the design of the
What functionality
One
it is
clear through our interviews that people did not use
characteristics of design quality
IV.1.3.1.
how
paper ones.
The
we can summarize some
The explanations we were
electronic calendars in their current
topic of electronic calendars
of the key problems.
23
would
given in our
form can not replace
justify a
separate paper,
Traditional calendars are not simply places where you record times for events
to take place
on
though electronic calendars are usually limited to such a
dates,
Traditional calendars have notes on them, contain telephone
simple function.
numbers, are often color coded, and have other papers attached (such as yellow
sticky
notes or paper-clipped
traditional calendars
actually
is
memos,
letters,
The non-homogeneity
etc.).
of
an asset for finding important information (there
are parallels here with Malone's (1983) findings on desk organization).
"I
honestly tried to use the calendar in Higgins, but
have
notes, colors, and paper
frustrating to not
my
familiar
book with
all
I
found it
messy
its
(RBOC)
clips."
Electronic calendars are not portable, and paper copies of the information
contained
in the
computer are inadequate
substitutes.
Notes made on the paper
copies often do not get keyed back into the computer-based version.
need a portable calendar for traveling. My secretary makes
a copy of the computer one for trips, but it is ugly and
hard to read. Then I have to make notes on it and do not
put them in the computer, and everything gets out of sync for
a while." (HealthCo)
"I
me
The group calendaring value of
way
group
that they support
others (not their secretaries)
"We
is
lost unless
everyone
People do not have an incentive to maintain their calendars
cooperates.
a
electronic calendaring
use.
may
in
such
In addition, people object to the notion that
schedule their time.
and found that several
guys weren't keeping their calendars up to date. So almost as
many phone calls get made and it takes just as long."
tried to use
it
to schedule meetings
(ConstrucCo)
The process
times
when
of setting up meetings
negotiation
"When we
set
is
many
locations.
not always a mechanical one.
required to secure the presence of
up meetings we have
negotiation since the Board
well in advance
is
is
to
made up
go through
of
all
lots
desired parties.
of
members from
Dates for regular meetings get established
and put on everyone's calendar. But setting
24
There are
up
meetings
special
requires
of
lots
personal
contact.
(TransDist)
Very often those who take time
to input the information never gain the value
of group calendaring because others (usually secretaries) do the group scheduling.
Therefore people see a basic economic imbalance of input effort to output value
(see also Grudin, 1988).
"It
seems
benefits!"
IV.1.3.2.
How
do all the work and Harry's secretary gets
(BigU)
I
the functionality
The second aspect of
from the functional
is
tool effectively
way
it is
offered to users.
mentioned above, calendaring was not used
the systems because people found the process of use
indicate recurring events).
the
offered.
functionality relates to the
limitations
all
awkward
(e.g.,
Aside
in several of
no easy way
to
In other examples people reported that they could not use a
because they could not remember how to access a particular function and
could find no effective help on-line or in the written manuals.
Aspects of the user interface design were also important factors
people had to The Coordinator (Version
I),
(see also Bair
in the reaction
While
and Gale, 1988).
people in fact used the package, and stated that the product was valuable to them:
great for communication breakdowns since you can back track the
"It's
conversations and find out what went wrong."
they also
"I
commented on
am
Two
interfaces,
(SnackCo)
the terminology of the interface:
not enchanted with the verbiage." (ServBuro)
other products,
ForComment and
even though some other aspects of
"ForComment
is
a joy to use;
its
were consistently praised
designs were criticized.
Higgins,
their
so easy to understand the
without a manual or checking 'Help'." (SoapCo)
25
menu
for their
items
menus
"Higgins
are
self-evident,
and the use of color
is
really
nice."
(RBOC)
It
has long been recognized that user interface design
a critical element in the
Therefore
successful use of a software product (Martin, 1973).
it
is
it
is
not surprising that
continues to be an important element in the case of groupware tools.
be that because people
considered
in
work group use these
a
For example,
in interface design.
designers must be concerned about
it
may
additional factors must be
tools,
product, like the spreadsheet,
in a single-user
how each
However,
menus and
user interprets
takes action.
In
a groupware tool the designer must be concerned about the individual user, and in
addition,
must address the
individually
and
as a group.
the group which
"It is
may
how what
issue of
that user does
Additionally, the individual
influence
how
the tool
is
is
others,
used and interpreted:
interpersonal one, although both take place between
group
many
acting as a representative of
is
important to note that an intergroup transaction
member
interpreted by
is
not the same as an
individuals.
A
group
involved in intergroup transactions acts as a representative of the
in
accordance with the group's expectations. The
on an individual agenda." (in Ancona, 1987)
member
is
not
acting solely
For example, one person may choose
indicate "informality" in notes.
This style
users interpret this as not having
judgment of
this user's
to use
may be
detrimental in communication
enough concern
behavior can then be
an all-lower-case character
made
to
compose a note
layers of complexity into the interface design process,
analysis
of our
interview
data
rather
than
technological merits of design.
26
an
if
other
The
properly.
not only against that individual, but
against the group as being, for example, "unprofessional."
more
style to
analysis
Such
issues introduce
and they emerge from
which looks purely
many
a social
at
the
In conclusion,
it
is
clear
from our interviews that the quality of design, both
terms of functionality provided and access to that functionality,
how and whether people
suggested by
this
use groupware tools.
is
an important factor
The researchable questions which
conclusion focus on gaining a better understanding of 1)
design in software that serves a team of users, and 2)
individuals acting as
members
in
in
are
interface
the actual tasks carried out by
of groups.
Isolated tools hinder productive use of groupware systems.
rv.1.4.
The
tools are considered isolated with respect to the flow of user control during
work and with respect
to the flow of data
among
tools (as discussed earlier).
cases the process of accessing the function of a second tool
flow of control) requires an
transfer of data
from one
awkward sequence of user
tool to another
(i.e.,
when
actions.
flow of data).
some
In
using one tool
(i.e.,
Other cases require the
(See also Nielsen et
al.,
1986)
Transfer of User Control
-
In
several of the
organizations
we
necessary for the people to go through a series of steps in order to
groupware
tool they
were using
for their business
computer e-mail system
like
it
move from
was
the
group/team to other tools they were
required to use for tasks relating to the firm as a whole.
a personal
studied,
For example, some groups used
those available on Higgins or
The Coordinator
within their departments, but they had to change to a mainframe-based tool like Profs or
All-In-I to use e-mail in other parts of their
to
companies.
This was universally considered
be an aggravation and a waste of time, regardless of the ease or
difficulty associated
with the switch.
"I
I
want to log on to The Coordinator and Excel at the same time because
need to bounce back and forth from moment to moment." (SmallChem)
"Because Forecasting uses Metaphor to analyze sales data from last year,
pricing people are using HP3000's, and I have my analysis programs on a
Compaq 386, I have a heck of a time moving between functions I need to
access
when we
set
up a promotional campaign." (BeerCo)
27
such a pain,
"It's
Is this
modern
to
From
or difficulty.
desk and change plugs.
Tools that were not completely integrated required that the
-
from one task be consciously moved
perform additional
my
technology??!!" (SnackCo)
Tratujer of Data
result
actually have to crawl behind
I
Most
tasks.
users
into the
environment of another tool
were annoyed by
in
order
this step, irrespective of its
the examples given above to illustrate flow of control problems,
clear that transfer of data
problem
also a
is
at
some of
ease
it
is
the sites.
know it is hard to believe, but we actually have to print the output from
HP3000 programs and key the data in to the Compaq 386 because we
haven't found a more cost-effective way to move the data across directly."
"I
the
(BeerCo)
The ForComment
system, highly praised in most respects, was singled out here with
created elsewhere.
of the
Although
this
would prefer that the
that they
love
I
am
ForComment, but
a straightforward step, users consistently
to
some groupware
management
I
wish
it
were part of our word processing package.
will get lost in the transfer,
(i.e.,
the
tools.
In
management
addition Ancona's
of the group's
I
take the time
She found
that, for
would imply
that
therefore, productivity.
to
boundary management
If
relations with
teams equally matched on group process
their
is
to their
(1987) research on boundary
environments and
an interesting point with respect
performance could be differentiated based on
ability
so
from either integration perspective was a barrier
individuals external to the group) raises
finding
by ForComment be available as part
both flow of control and flow of data those interviews showed very
clearly that a lack of integration
control.
commented
to create the text.
always afraid something
With respect
of
the person must import text
check the whole document." (BigU)
to
use
is
functionality provided
word processor they used
"I
ForComment,
In order to use
respect to data transfer.
to flow of
characteristics, their
boundary management
capability. This
a key aspect of team performance and,
teams are using groupware systems that interfere with
perform boundary management
(e.g.,
28
the
team e-mail system
is
their
not easily
may be
connected to the organizational e-mail system), their productivity
From
affected.
this
we conclude
that productive use of
groupware
is
adversely
reduced when the
tools are isolated.
IV.2.
rv.2.1.
FROM AN Organizational Perspective
People report most value from tools that parallel their non-electronic
Those we interviewed reported
analogous
to,
traceable, geography-
home, while
in
learning
was
as
an improvement over "the old way" because
traveling).
how
to
Therefore
communicate
people saw as needed
(e.g.,
not currently employing.
it
was easy
People saw
was
it
faster,
(e.g.,
people to see the benefits to them
for
from what
significantly
electronic calendars) or presented capabilities that they were
In
the
were not undertaking. Therefore,
to
was
electronically.
latter
category,
reminders, directories, and expense tracking
them
it
and time-independent, and accessible from almost any location
Other functions provided by the systems either differed
requiring
"easy" because
but better than, what they did without groupware tools.
computer messaging
at
that e-mail, for example,
activities.
all
functions such as
project tracking,
represent tasks that the people interviewed
to use the electronic version of these tools
expend resources
was
actually
for activities they did not normally carry out or carried
out only infrequently.
We
therefore
conclude
that
the
designers
of,
developers
of,
introduce groupware tools are presented with an interesting challenge:
going to
make
the transition to
new
Part of the answer to this question
remember
in
practices which
lies in
after long periods of non-use.
some of
how
are people
the functionality enables?
designing functionality that
is
easy to learn and
Another part of the answer, however,
the organizational considerations which address the
processes.
and those who
need
to
is
found
examine current work
rV.2.2.
Benefits gained need to balance or outweigh the invested resource.
The
from some of the functionality (other than that provided
benefits
messaging) were not clear, nor balanced
in
saw more
much
effort
on
their part for not as
our interviewees' minds.
in
maintaining an electronic calendar
as redundant (since most also wanted to have a portable calendar
who
In fact users often
gain.
For example, people saw the work involved
benefitting others in the organization
for
did meeting scheduling.
maintaining calendars and project information to "keypunching"
on paper), and as
They likened
activities.
their
Although they
agreed that their managers and groups would benefit, the benefit to them personally was
too far removed to motivate their behavior.
Messaging functions, however, had a beneficial impact on them
directly.
experienced the satisfaction of "getting the message out," "putting the ball
guy's court," assigning tasks to group
members,
etc.
On
in the
They
other
the receiving side, they had a
record of what they were expected to do, and through being on copy
of being in touch with what was going on in the group.
lists,
had a sense
They had no need
to
conceptualize anything beyond a personal benefit.
Other functions as mentioned previously
actually required additional effort
"I
should do expense tracking",
kept an electronic directory",
invariably they
were unwilling
necessary to employ
this
"I
project tracking, reminders, etc.)
on the part of the users
different way, independent of the technology.
said things like
(e.g.,
to learn to
do
things in a
While the people we interviewed often
"I
know
it
would be more
efficient
if I
could really benefit from the reminder function",
to adapt their behavior
kind of functionality.
and
invest the personal resources
They had not
identified benefits to using
the technology which equalled or exceeded their resource investment.
30
Therefore,
we can conclude
that unless there
is
a balance between the effort
required on the part of the user and the benefit delivered to that user, a person
likely to
employ the
management
functionality present in a tool.
Other forms of motivation
group agreement, education) can be important
directives,
(See also Grudin, 1988.)
perception of balance.
is
Research
not
(e.g.,
in influencing the
to investigate motivation
and
change management as part of the implementation of groupware technology could be
beneficial in understanding the dynamics here.
rv.23.
Groupware implementation
Our
simultaneously a social and technical intervention.
research observations support the ideas offered by Kling and lacono (1988)
that "computerization
is
this
(Kling and lacono, 1988, p.8).
kept
implicit,
In our research
it
complex intervention
Whether
exists
we saw
is
by merely training new users
that
it
is
and can have a
significant
in
way
the mechanics of the tools,
of the functionality present in the systems. That
we
that
functionality present in the
to streamline
procedures
we saw people
instruction
groupware systems, or 2) how to
went beyond mechanical
steps,
work
tasks,
how
to
creatively use
their work.
however, to include, for example, a
presentation on the concepts of groupware, or material on
functionality to accomplishing their
using a
people used what they were
is,
which they had learned to have an impact on the way they carried out
When
made
interviewed.
taught to use without any innovative thinking on their parts about either, 1)
employ other
is
impact on the organization.
the effects of strategies on the individuals
as a
of the
a "strategic intervention"
the strategy of technology introduction
For example, when a groupware system was introduced
minimum
One
simultaneously a social and technical intervention."
most important aspects of
explicit or
is
how
then people
to relate the
made
use
of,
groupware
and applied
creative thinking to using the functionality present in the tool.
When
the
a groupware system
people did not take
it
was introduced as a new technology
seriously
and did not look
31
for
to experiment with,
ways to augment
their
productivity.
When
enhancement
decision makers held high expectations for productivity
through groupware, yet gave no attention to examining the work process, people reported
that they felt
under pressure
clear understanding of
how
to
perform better while learning a new tool and without a
the tool would
make
a difference.
In
many
of these cases, the
introduction of the groupware system had a negative effect on productivity.
Organizational factors showed up as consistently important as
people
in the twenty-five firms.
We
we
interviewed
report our observations in the following four general
categories.
Management support
Champions.
1.
varied significantly in our sample.
from the top
some
groupware
organizations, support for the tool
tools
emanated
levels of the firm:
"When
the president wanted us to use this e-mail package without even
looking at any others,
[him] to try
At
In
for the introduction of
others, like
it."
we thought
was
it
strange, but
SnackCo, the management support was
"We thought
had enough
faith in
(CableCo)
this tool
was weird, but
if
departmental
at the
WW asked us
to try
it,
level:
we knew
it
was worth doing."
In
middle
some
instances, the support
management supporters who
came lower
felt
they
in
the organization in the form of
could
engineer successful
pilots
and
demonstrate the value of the tools to upper management:
"Through my own coaching and interpersonal skills I have been able to
teach people and win them over to the value of using The Coordinator.
Now management
is
paying attention." (ServBuro)
While these instances of managerial support represent very different
levels of
power
within each organization, they demonstrate the importance in general of a committed
leader in the introduction of a
new
technology.
32
Management
literature for
decades has
discussed the value of leadership and champions for the successful implementation of an
In the area of
innovation.
However,
groupware
tools, this
"common wisdom"
as the previous observations have shown,
continues to be valid.
and the next observations
managerial support cannot guarantee successful implementation by
Ejcpectations.
2.
We
observed two different work groups
which the same software had been introduced.
was
tool
described as a
originally
familiarize themselves with
In
will suggest,
itself.
in
one organization
in
one of these groups (Group A) the
new technology
the group
that
and see what they could use
it
for.
members should
In the other
group (Group
B) the tool was described as an important new technology that was going to be used to
improve communication throughout the organization. Five years
our interviews,
the
original
attitudes
influencing the use of the software.
taken seriously and was
tool
still
was described by people
It is
clear
were present
As a
result, in
in
later,
two groups and were
these
Group
A
when we conducted
the tool had never been
considered "an experiment" and informal.
In
Group B
the
as "critical to their jobs."
from our studies and those of others "that the kinds of expectations with
which an organization approaches new information technology do much
consequences that
will
be observed." (Carroll and Perin, 1988).
which new groupware tools are introduced into the work group
to define the
Therefore the way
will
in
influence the ways in
which they are used.
3.
Training.
Those interviewed generally described the
training that they
had
received in the use of their software as directed toward building procedural or mechanical
skills.
By
this
specific tasks.
Coordinator,
we mean
they reported basic instruction in what keys to push to accomplish
This was true for
we
did interview
all
some
the tools
users
we
"The
linguistic
However,
in the
case of
The
who had
received training that included an
A
subset of this group did report that
introduction to the theory underlying this product.
the ideas were too sophisticated for
studied.
them and
their colleagues to assimilate:
concept went over most peoples' heads." (SnackCo)
33
"The training
(CableCo).
left
However, some reported
to
me
that
but
cold,
we pursued
the
value
on our own."
knowledge of the theory helped them to use the tool and
implement the communication practices which the
tool supports:
"Knowledge of speech-act theory has really helped me use The Coordinator
to be more effective in my daily communication." (ServBuro)
'The workshops were
vocabulary
much
make
and
inspirational
easier."
using
The Coordinator
(SnackCo)
Given the previous observations that people are not using the functionality provided
by these
tools, the fact that they
have also received only basic, mechanical training, would
tend to indicate that the training
is
not adequate.
we
'Training was not very good, but
4.
users
we
Evolution.
After an
figured
out."
(MedCons)
introduction into the use of a groupware tool, the
initial
interviewed tended to "practice" only those procedures that they needed to
accomplish their most urgent business
As a
tasks.
trained to do but did not continue to do regularly
'Two weeks
to
it
my
file
program
after the training
directory."
I
result,
much
of what they were
initially
was forgotten.
could barely
remember how
to get
(LawCo)
In the use of any system, people will encounter special case needs for functions
time to time
a manual
it
in their
when
were unable
work.
Those interviewed did not regularly look up procedures
these situations arose.
to find
When
what they needed.
online help was available, most
Instead, the typical
form of help sought was
refuse to read manuals and documentation; they aren't even written in
English!"
(BigChem)
"The only copy of the manual
inappropriate to the version
I
am
I
was two years old and
(SoapCo)
could
using."
34
find
in
who used
to ask a colleague or subordinate.
"I
from
"On-line 'Help'
is
a bust.
It
never covers the exact problem I'm having and
is
written with lots of jargon."
"I
always ask Joe for help.
(FoodCo)
He
can
tell
me
in
two seconds where
I've
gone
wrong." (BigU)
"Sue explains
me
for
Some
to
my
mistakes
remember
in the
context of our work.
for next time."
That makes
it
easier
(TerminalCo)
of the organizations provided a person or group to serve as the designated
support source to which the users would turn for help.
These organizations appeared
to
understand the evolutionary nature of a person's use of software, and they supported that
evolution through a formal organizational entity.
once the
initial
going nature.
groups
training
sites
we
studied assumed that
had taken place, there was no formal corporate role of an on-
In these cases, de facto support
who became
Other
grew up
in
the form of individuals in
work
"local gurus."
"Dick has become the guy we
He's part of our
go to for help.
department and understands our questions best." (TerminalCo)
all
"The Infocenter has been wonderful in supporting the use of this tool. They
are always available and polite in telling you where you made your mistakes."
(OilCo)
We
observed what might be called a "plateau of competence"
in using a tool.
Without a timely and user-appropriate incentive to move beyond self-standardized
people tend to
settle into
standard operations (Rosson, 1985).
interaction serving as a constructive stimulus for individuals.
sent out a newsletter of hints and ideas that
In
We
observed close group
SnackCo a
central person
was found useful by some.
presence of new ideas was countered by pressure to "get the job done," so
found
little
time to try
new
things.
use,
However, the
many people
This suggests that such stimuli must be in the form
of easily tried procedures with immediately visible value so that they
carried out during a busy day.
35
fit
into the practices
We
evolution
conclude
each of the categories
that, in
-
champions, expectations, training,
the need for sensitivity to organizational issues
-
is
evident.
degree and timing of organizational intervention must be planned.
In addition the
Without attention to
the organizational impacts of introducing a technology, the risk of failure
because the complexity of factors influencing the use of the technology
increased
is
is
not being
considered.
rv.2.4.
Process redesign
We
may
be required to realize productivity improvement
have just suggested that organizations should consider the perspectives of
people when introducing technology.
Another
interesting question
to
is
what extent do
work
organizations need to alter their basic processes for accomplishing
order to
in
achieve higher levels of coordination and productivity.
may
This process redesign
occur on a variety of
levels.
For example,
traditional
process redesign looks at formal processes which have been established in an organization
in
order to achieve
example,
light of
in
business goals and objectives. These processes are reevaluated, for
its
changes
industry, or the impacts of
in
products or services, changes
new technology on
in
basic functions
the structure of the
(e.g.,
manufacturing or
distribution channels).
However, process redesign can be employed on a much more
For example, the process of work coordination
organization.
process of conducting meetings
may be
work we observed
improvement
to result
instances of
from the simple
department or the
(e.g.,
Whiteside and Wixon, 1988).
management expecting
act of putting a
substantial productivity
groupware system
into place.
these instances our interviews did not uncover any significant change in
approached
their jobs.
Some
and therefore made them
felt
an
areas in which productivity gains could be achieved
through rethinking and redesigning the traditional forms
In our field
in a
local level in
that the
how people
new technology created more work
less productive.
36
In
for
them
Whenever we observed
concurrent examination of
we saw
or evolve,
the implementation of groupware technology without a
how work procedures and
that these systems
had very
These observations lead us
of the groups.
little
coordination should perhaps change
impact on the perceived productivity
groupware systems are implemented, not enough attention
the basic processes of
is
when
cases
being placed on examining
work and how technology may enhance these
may be
process redesign
some
to the conclusion that in
processes.
TTierefore,
required to achieve productive benefits in using groupware
technology.
Creating productive teams
rv.2.5.
Managers
the
way work was
"paradigm
would
the
Does
carried out,
a challenge.
we
the organizations
moving
their
studied had explicit goals of changing
groups to new planes of performance, creating
shifts":
am
"I
some of
in
is
intrigued by the linguistic theory underlying
like to
way they
The Coordinator and
see everyone undergo a paradigm shift and significantly change
(SnackCo)
interact."
the current generation of groupware systems facilitate this process?
needed
to create productive
teams?
In
What
TeamWork (Larson and LaFasto,
is
truly
1989), the
authors present eight characteristics of high performing teams which they draw out of
interviews with publicly acclaimed high performing teams, including sports, science and
shown
technology, and industry.
The
appear
understood notions of good team work, and they are described
at a
to reflect generally
somewhat
elements
visible
They
abstract level.
raise
In the
facilitate the creation
many
in
Table
6.
Most
of these
However, they serve the purpose of making these
and keeping them
also
operationalized.
eight characteristics are
at
hand
questions
for
in
managers
terms
context of our research,
of these key elements?
37
of
to think
how
critical
about and use.
these
characteristics
how can technology be
are
applied to
The
partnership
Management must
organizational
of
out
carry
specific
and
process
tasks
bring
to
technology
All of these things can be accomplished without any technology.
characteristics.
in the fast-paced,
ability to carry
barriers to productive
its
use.
team
However,
out the appropriate tasks.
Examining people's attitudes appears to be an important step
immediate value
clear.
geographically-dispersed environment of today's corporation, groupware
technology could enhance the individual's
to
very
productive
these
forth
is
team work.
In
our interviews
we noted
in
understanding
when people saw
that
to themselves of using a function (e.g., messaging) they quickly
However when
was
it
in the interests
department or organization would benefit
In these situations
it
-
of the "higher good"
-
that
is
the
adapted
the team,
the incentive for the individual was missing.
took other motivators to cause people to use these
tools.
Some
of
the motivators included:
A
charismatic leader
Orders from higher management
Workshops on organizational issues
Obtaining group agreement and individual permission
»•
In other words, the technology alone, regardless of
ease of use, could not inspire people to use
into a
poorl^
operating work group
researchers have found that
its
introduction
brings
new
is
it.
may
potential value, attractiveness, or
In addition, introducing
unlikely to
technology,
its
improve
its
new technology
performance.
In fact
very well degrade performance because
more complexity and a
threat
to
the
people on the
team
(Henderson and Cooprider, 1988).
The well-known concept of
1952) seems applicable here.
The
"unfreezing" in organizational change theory (Lewin,
potential users of
groupware systems need to "open
up" or "unfreeze" to the possible value of learning to use the technology. That unfreezing
is
not a simple matter of mechanical training, but rather an organizational process that
includes training, education, and rethinking the goals of the team, and then considering
how work
results will
be accomplished from the new perspective.
38
GROUPWARE
Characteristics of
High Performing Teams
A
A Clear
^
Results-Driven Structure
A
Competent Team IWembers
A
Unified
^
Collaborative Climate
^
Standards of Excellence
Elevating Goal
Commitment
^ External Support
^
Source:
1
Larson
and Recognition
Principled Leadership
&
LaFasto, 1989
CISR Qrpwr CVB/mfb 006
'
Table 6
39
One
of the lessons that comes out of
TeamWork
is
that
much
teams depends on communication among team members on key
groupware's greatest values, according to our research,
electronic messaging.
interaction in
work
V.
Our
some of
conclusion here
the items
into the realm of high
shown
in
is
that
if
Table
is
of the success of
the support
groupware systems can
6,
it
One
issues.
it
provides for
team
facilitate
has the potential to
of
move group
performance teams.
Summary
Is
groupware too new to study conclusively?
research (Rogers, 1983) that
it
takes time for
new
We
have learned from innovation
ideas to be assimilated by people.
Although the technology for electronic mail and conferencing has been available for
years, the concept of technology to support
work groups has only been discussed
four years.-^
We may
their infancy
and before people have learned
therefore be observing the use of these
to think about
new
tools
them
when
fifteen
for about
they are
in
as essential tools for
effective office work.
Nonetheless, experiences gained from studying people as they learn to use
tools
new
can benefit the designers of the next tool generation, thereby helping to accelerate
the process of acceptance and use of these tools.
We
also believe that
managers can
learn to be sensitive to the complex balance that exists between the organization and the
technology.
Our
observations were consistent across a wide range of organizations.
groups at the largest organization we studied, BigChem with $30
^The pioneering work done by Doug Engelbart
the forerunner of the groupware concept.
However
public.
40
(in
billion
in
Work
revenues,
the early I960's) on high performing teams
work is not well known to the general
this
is
experienced the same challenges
in
organizing work and using information technology as
did those at the smallest organizations.
Work groups
"forward thinking", "networked," and participatory in
their
problems
nor
in
attempted
their
at
companies recognized as being
management
from work groups
solutions
characterized as "conservative", hierarchical, and traditional in
For example, CableCo ($1
highly
company with
successful
management of
who
technology
is
in
management
companies
style.
a very young, quickly growing,
management
participatory
not differ in
style.
related to effective
The work group
communication and
an accelerating, fast-paced, environment that spanned several
At SoapCo ($17
time zones.
people
tasks in
a
revenues)
who had concerns
people
of nine
consisted
billion in
style did
billion in
revenues) one work group consisted of fifteen
expressed exactly the same concerns and were attempting to use information
same way
in the
to support their
work group. SoapCo
is
a very old
company
with a long tradition of hierarchical, conservative management, and with long-standing
procedures
for
environments
in
accomplishing
tasks.
A
priori,
at
differing
these two ^'organizations, and these factors would have dic'ated different
approaches to solving the coordination problems.
nor
we might have assumed
other research
sites.
We
did not find this to be true here
Apparently today's business environment of global, 24-hour
marketplaces with the concurrent acceleration of information and coordination needs
brings the
We
same challenges
in
managing work groups
to diverse organizations.
have discussed major questions and conclusions about work groups and their
use of information technology (summarized in Table
possible
interplay of factors
groupware
more
is
a relatively
in
We
7).
For example, the
our major conclusions.
new technology may account
must not overlook the
for
fact
why people seemed
to
that
need
extensive training and introduction than the simple mechanical instruction that
accompanies these
tools.
their inherent value
to their use
more
In the future, as the tools are
may become more obvious
easily.
41
to people
more widely known and
and they therefore
will
used,
adapt
However, the organizational
inhibitors
that
we observed cannot be
Recognizing the long-lasting constraints of history and the power of
organization at the
may
result in
new
same time
as considering the
These contrast with
insights.
new
possibilities for
insights suggested
dismissed.
politics
the
in
technology support
when
using traditional
requirements analysis, often focused on individual users to the exclusion of organizational
Understanding the interplay
factors.
1)
economic balances
(i.e.,
of:
input resources versus output value) inherent in the
use of a tool,
2) the differential impacts
differ
from support
on organizational
roles
(i.e.,
managerial impacts
may
staff impacts) and,
3) the organizational readiness
(i.e.,
management
attention through planned change
or intervention)
may
lead
management toward
different technological paths than those discovered through
simple analysis.
We
have stated earlier that managing the volume of information has been
traditionally,
and
still
is,
the
As we have
major task facing knowledge workers.
we
interviewed, observed teams, and better understood the tasks they are undertaking,
have come to the conclusion that a groupware system
could have an effect on knowledge work.
That
is,
like
The Coordinator,
The Coordinator,
if it
for
example,
were used as
its
designers intended, could reduce the volume and complexity of information by compressing
it
through a focus on action-oriented conversations.
in focusing
this
on results-related content and meaning
This could support group
in their
idea further in a second paper that focuses on
communications.
The Coordinator and
members
We
its
explore
underlying
theory as an example (Bennett and Bullen, forthcoming).
Revolutionizing work
have
in organizations.
may be
the type of role groupware systems could
Most groupware systems of today are not designed
to
come
do
to
this.
Instead they attempt to provide electronic versions of the tasks people are believed to
42
GROUPWARE
Summary
of
Conclusions
From A Design Perspective
1.
Electronic message communication
2.
Message
3.
What
4.
Isolated tools hinder productive use of groupware systems
linking
is
functionality
a key
is
is
the primary tool
improvement provided by electronic communications
included and
how
it
is
offered are important factors
From An Organizational Perspective
2.
People report most value from tools that parallel their non-electronic
Benefits gained need to balance or outweigh the invested resource
3.
Groupware implementation
4.
Process redesign
5.
Creating productive teams
1.
may be
activities
simultaneously a social and technical intervention
is
required to realize productivity improvement
is
a challenge
CISR Qrpwr CVB
Table 7
43
carry out in performing knowledge
and business teams
is
work
groups.
If
indeed the concept of work groups
the organizational concept of the future,
understand the interaction of individuals
technology can play
in
in
in these
it
becomes
critical to better
groups, and the role that information
supporting or even enhancing the work of these groups.
44
Bibliography
Ancona, Deborah Gladstein. "Groups in Organizations," Group Processes and Intergroup
Relations, Editor Oyde Hendrick, Sage Publications, Newbury Park, CA, 1987, pp 207230.
James H. and Gale, Stephen, "An Investigation of the Coordinator as an Example
of Computer Supported Cooperative Work," Extended Abstract, submitted to the
Second Conference on Computer-Supported Cooperative Work, Portland, Oregon,
Bair,
September, 1988.
Bennett,
J.L., Holtzblatt, K.,
Jones,
Contextual Inquiry," Tutorial at
1990),
ACM, New
and Wixon,
S.
CHI
'90,
Using
D., "Usability Engineering:
Empowering People,
(Seattle,
WA,
April 1-5,
York, forthcoming.
Berger, Suzanne, Dertouzos, Michael
L.,
Robert M., and
American, Volume
Lester, Richard K., Solow,
Thurow, Lester C, 'Toward a New Industrial America,"
260, Number 6, June 1989, pp 39-47.
Scientific
Bikson, Tora K., Eveland, J.D., Gutek, Barbara, "Flexible Interactive Technologies for
Current Problems and Future Prospects,"
Multi-Person Tasks:
for Work Group
Collaboration, edited by Margrethe H. Olson,
Associates, Inc., Hillsdale,
New
Jersey, 1989,
in
Technological Support
Lawrence Erlbaum
pp 89-112.
Blomberg, Jeanette, 'The Variable Impact of Computer Tecnologies on the
Organization of Work Activities," in Computer-Supported Cooperative Work:
Readings, edited by Irene Greif,
Morgan Kaufmann
Publishers, Inc.,
A
Book of
San Mateo, CA,
1988.
Bullen, Christine V., and Johansen, Robert R., "Groupware:
Business Teams?,"
CISR Working Paper #169, Center
Research, MIT, Cambridge,
MA,
02139,
May
A
Key
to
Managing
for Information Systems
1988.
and Perin, Constance, "How Expectations About Microcomputers
Influence Their Organizational Consequences," Management in the 1990's Working
Paper 88-044, Sloan School of Management, MIT, Cambridge, MA, 02139, April 1988.
Carroll,
John
S.,
Dalton, Richard, Open Systems,
Drucker, Peter, "The
New
March
1988, p.7.
Organization," Harvard Business Review, January-February
1988.
Engelbart, Douglas
C, "A Conceptual Framework
for the
Augmentation of Man's
Intellect," in Vistas in
Information Handling, Vol.
1
(P.
Howerton,
Ed.), Spartan Books,
Washington, D.C., 1963, pp 1-29.
Engelbart, Douglas
Human
C, and
Intellect," originally
English, William K.,"A Research Center for
published
in
A Book of Readings,
San Mateo, CA, 1988.
Cooperative Work:
Publishers, Inc.,
Grudin, Jonathan,
"Why
CSCW
1968 and reprinted
in
edited by Irene Greif,
Augmenting
Computer-Supported
Morgan Kaufmann
Applications Fail: Problems in the Design and
Evaluation of Organizational Interfaces," Proceedings of the Conference on ComputerSupported Cooperative Work, September 26-28, 1988, Portland, Oregon, pp 85-93.
and Cooprider, J., "Dimensions of I/S Planning and Design
Technology," Working Paper #181, MIT Center for Information Systems Research,
02139, September 1988.
Cambridge,
Henderson,
J.C.
MA
Hiltz, S.R.
and Kerr, E.B., "Studies of Computer-Mediated Communication Systems:
A
Synthesis of the Findings," Final Report to the National Science Foundation, 1981.
Roxanne, and Turoff, Murray, The Network Nation: Human Communication
Computer, Addison-Wesley Publishing Company, Inc., Reading, Massachusetts, 1978.
Hiltz, Starr
via
lacono, Suzanne and Kling, Rob, "Computer Systems as Institutions:
Social Dimensions
of Computing in Organizations," Proceedings of the Ninth International Conference on
Information Systems, 11/30-12/3 1988, Minneapplis, Minnesota, pp 101-110.
Johansen, Robert, Groupware:
New
Computer Support for Business Teams, The Free
York, 1988.
Keen, Peter, G.W., "The 'Metabusiness' Evolution:
Advance, October 1988, Washington, D.C.
Challenging the Status Quo,"
Kerr, Elaine B., and Hiltz, Starr Roxanne, Computer-Mediated
Status
and Evaluation, Academic
Kling,
Rob and
February
6,
Press,
New
ICIT
Communication Systems:
York, 1982.
lacono, Suzanne, "Computerization as a Social and Technical
Intervention into
8b,
Press,
Work
1988,
Organization:
Department of
The Case of Desktop Computerization," Draft
Information and Computer Science and Public
Policy Research Organzation, University of California, Irvine, California 92717.
Larson, Carl E., and LaFasto, Frank M.J., TeamWork, Sage Publications,
CA,
1989.
Newbury
Park,
Lewin, Kurt, "Group Decision and Social Change," in G.E. Swanson, T.N. Newcomb,
and E.L. Hartley (Editors), Readings in Social Psychology, (revised edition), Holt, New
York, 1952.
Loveman, Gary W., "An Assessment of the Productivity Impact of Information
Technologies," MIT Management in the 1990s Working Paper 88-504, Massachusetts
Institute of
Technology, Cambridge,
MA,
July 1988.
Malone, Thomas W., "How Do People Organize Their Desks? Implications for the
Transactions on Office Information Systems,
Design of Office Information Systems,"
pp
'99-112.
Vol.1, No. 1, January 1983,
ACM
Markus, M. Lynne, and Janis Forman, "A Social Analysis of Group Technology Use,"
UCLA Information Systems Research Program Working Paper #2-90, July, 1989.
Martin, James, Design of
New
Man-Computer
Dialogues, Prentice-Hall,
Englewood
Cliffs,
Jersey, 1973.
Mumford,
E.,
and T. B. Ward, Computers: Planning for People, Batsford, London, 1968.
Mack,
and Grischkowsky, N., "Integrated Software
Usage in the Professional Work Environment: Evidence from Questionnaires and
Interviews," in Proceedings of CHI'Sd Human Factors in Computing Systems (Boston,
Nielsen,
J.,
April 13-17, 1986),
R., Bergendorff, K.,
ACM, New
York, pp 162-167.
Reich, Robert B., "Entrepreneurship Reconsidered:
The Team
as Hero,"
Hansard
Business Review, May-June, 1987, pp 77-83.
Rice, Ronald, and Associates, Vie
New
Media, Sage, Beverly
Rogers, Everett, Vie Diffusion of Innovation, Free Press,
Hills,
New
CA,
York,
1984.
NY,
Rosson, Mary Beth, "The Role of Experience in Editing," Proceedings of
Elsevier North-Holland, Amsterdam, 1985, pp 45-50.
1983.
INTERACT84,
Rousseau, D.M., 'Technology in Organizations: A Constructive Review and Analytic
Framework," in Assessing Organizational Changes: A Guide to Methods, Measures and
by Seashore,
Sons, New York, 1983
Practices,
Searle,
John
S.E., Lawler, E.E., Mirvis, P.H.,
R., Speech Acts,
Cambridge University
&
Press,
Caman, C.
Cambridge, England, 1969.
Stevens, Chandler Harrison, "Many-to-Many Communications,"
3
(Editors), Wiley &.
Working Paper #72,
MIT
Center for Information Systems Research, Cambridge,
MA,
02139, June 1981.
"The Sociotechnical Perspective," in Perspectives on Organization, Design and
Behavior, by Van de Ven, A.H. & Joyce, W.F. (Editors), John Wiley & Sons, 1981
Trist. E.L.,
Whiteside, John, and Wixon, Dennis, "Contextualism as a World View for the
Reformation of Meetings,"
in
Proceedings of the Conference on Computer-Supported
Cooperative Work, Association for Computing Machinery,
Yates, JoAnne, Control Through Communication:
Management, Johns Hopkins University
New
York, 1988.
The Rise of System
Press, Baltimore,
MD,
1989.
in
American
Appendix
I
All-In-l""^
All-In-I
environment.
means of
is
more
accurately described as a family of tools or even an office tool
This system resides on a centralized computer, with PCs often serving as
access (both locally and remote).
does not provide flow of data
integration, but does provide flow of control within its environment. The basic tool
provides a variety of functions ranging from electronic mail to a spread sheet package.
An organization can customize All-In-I by adding other commercial products under the
general All-In-I "umbrella." For example the popular word processing software. Word
a
It
be installed to operate under All-ln-I. The extent to which the
under All-In-I can be used as groupware depends on which
functions are used and on what agreement the people in the organizational unit reach
on how the functions will be used.
Perfect, can
functionality provided
The
logical
The
basic All-In-I functions described above are in use at three organizations
groupware use of this tool involves employing the electronic mail
function and Vax Notes (computer conferencing) for communication and the exchange
of documents. A calendar function can be used for scheduling group meetings.
in
our study.
ForComment'*^
ForComment
group authoring or
editing of documents. This system is available in single PC and local area network
configurations. ForComment "imports" text produced in most of the popular word
processing environments or ASCII and allows multiple authors or reviewers to rewrite
and/or comment on the document. Control over the final version of the document
always remains with one individual, the designated primary author. Proposed rewrites
and comments are noted through a symbol in the margin, and the actual text of the
revision is displayed in a second window on the viewer's screen. Each entry is
identified by reviewers' initials, color coding, and date of entry. The software
automatically merges entries from multiple reviewers so that the primary author reviews
is
a single-purpose system.
a single, aggregated version.
In this respect
It
assists users in
ForComment
provides flow of data
integration.
ForComment
uses
it
is
used by four organizations
unmodified as provided by the vendor.
in
our sample.
Each organization
Biggins'*^
Higgins
a personal computer system based on a local area network that
is
provides a variety of functionality including electronic mail, personal information
organization, project tracking, and project expense tracking.
The
electronic mail
function links messages and their associated replies, allowing users to trace the history
of communications leading to a current message.
All of the functions are integrated
on
For example, a user
Therefore the
all
name "project xyz" can be used to find electronic mail, "to do" entries, expense reports,
calendar entries, etc., that relate to that project by its code name. In this way Higgins
can be used both as a personal organization tool and as a groupware tool.
Higgins both with respect to flow of control and to flow of data.
can employ key words
Higgins
used
is
entries dealing with specific categories.
to find
in five
of ihe organizations in our sample, in each case
stand alone local area network
In-House System
One
(LAN) mode
in a
as provided by the vendor.
I
large organization in our
sample developed
its
own
global electronic mail,
document exchange system. This system resides on a mainframe
accessed by PCs acting as workstations (both locally and remote).
conferencing, and
computer and is
Both integration in terms of flow of data and flow of control exist to varying degrees
this system.
Development of increased integration in both areas is a current priority.
This system has been in worldwide use by a very large number of people at this
organization for
more than
In-House System
One
ten years.
II
small organization in our sample developed
electronic messaging tool.
its
own
relatively basic
This system resides on a centralized computer and
accessed by both local and remote PCs.
States (although
in
one European node
is
The system
in place)
is
used primarily
and has been
in
in the
is
United
use for approximately
eight years.
Metaphor"^^
Metaphor provides high-end
networked workstations and software to
in constructing complex queries agains.
multiple data bases. Users build queries by specifying data elements graphically and
then by linking sequences of operations on that data. These queries can be saved in
"capsules" for later use or for use by others. Data results can be passed to others on
the specialized local area network in the form of spread sheets, reports, and graphs.
specialized,
support professionals, managers, and executives
ii
The graphical
user interface
intended for easy and effective use by business
is
who need to review and aggregate data
Flow of control and flow of data integration exist
professionals (such as marketing analysts)
extracted from large data bases.
Metaphor environment.
within the
Metaphor
is
in
In one of those sites it is being
designed with a gateway into the
use at two sites in our sample.
used as a standalone system;
in
the other
it
is
corporate data network.
PROFS^^
PROFS
centralized
remote).
is
computer with PCs often serving
PROFS
as a
means
for access (both locally
folders for mail
management.
PROFS
Other than for the electronic mail component, the
can be used as groupware depends upon the agreements
an organization reach for allowing access to calendars and folders.
in
and
includes functionality for electronic mail, calendaring, reminders, and
extent to which
people
This system resides on a
a general purpose office system tool.
Flow of
control integration exists to a limited degree within the Profs environment.
PROFS
was studied
at
two of our
The Coordinator System"*^ (Version
The Coordinator System (TCS)
support people
The system
is
in effective action
sites.
I)
is
a groupware system which was designed to
during the course of their work in an organization.
generally available in two hardware configurations:
area network, or via dial-up
mode supported by
the vendor.
on a PC/local
from most of
based on an
either
TCS
differs
we examined in that the software implementation is
of human interaction. The theory suggests that the basic
the other products
underlying theory
interaction
is
a conversation, and that people use language (speech acts) to
requests, promise results, decline requests, declare
commitments completed,
software makes these distinctions visible and thereby
to
conduct their interactions
in a
is
unit of
make
etc.
The
designed to encourage people
way presumed (under the theory)
to
be more
effective.
The Coordinator Version
I is
available in seven of the organizations in our
remote mail, local area
network), but these differences do not play an important role in what the users see, or
how they tend to employ the tool. The fact that Version I is the tool we studied is,
however, important because the user interface of Version I differs significantly from
that of Version II. Version II became available in 1989, and it is currently being
marketed and used in a number of organizations.
sample.
Technical details of the implementations differ
Ill
(e.g.
The degree to which flow of control integration exists
and bridges
depends upon the nature of the implementation
in some of
exists
other systems. Flow of data integration
to
IV
926
7
'J
I
in a
TCS
that
have been established
the tools.
implementation
Date Due
APR 30 1991 JUN. 09 1985
MAY 23 1!J9|
^^i^
i 1
FEB 15
Sift.
10
1992
HAR.0219S3
OCT. 25^998
APR.
27
1§84
Lib-26-67
Mil
3
UBR4RIES
TOflO
DllPl
1
00563505
7