Content uploaded by Thomas Erickson
Author content
All content in this area was uploaded by Thomas Erickson on Dec 31, 2013
Content may be subject to copyright.
Erickson and Kellogg 1 Knowledge Communities
Knowledge Communities:
Online Environments for Supporting
Knowledge Management and its Social
Context
Thomas Erickson and Wendy A. Kellogg
IBM. T.J. Watson Research Center
snowfall@acm.org; wkellogg@us.ibm.com
1. Introduction
The issue of how to support the re-use of knowledge — under rubrics such as
organizational memory, knowledge management and expertise management —
has received increasing attention over the last decade. In this chapter we take a
strongly social approach to the issue, arguing that knowledge (and expertise) is
created, used, and disseminated in ways that are inextricably entwined with the
social milieu, and therefore that systems which attempt to support these processes
must take social factors into account.
Our approach to managing knowledge or expertise is to do it on-line, via multi-
user networked environments that support group communication and
collaboration. That is, we are interested in designing on-line environments within
which users can engage socially with one another, and, in the process, discover,
develop, evolve, and explicate knowledge relevant to shared projects and goals.
We refer to online multi-user environments used in these ways as “knowledge
communities.”
2
This chapter consists of two sections. In the first section we make the case for a
deeply social approach to knowledge management. We begin with an example
that depicts a number of ways in which the production and use of knowledge is
fundamentally entwined with social phenomena. We note that this socially
situated view of knowledge is supported by research in a number of disciplines,
and also has made its way into the business discourse that surrounds knowledge
management. This view raises a challenge for those designing technology:
knowledge management systems must take into account, either explicitly or
implicitly, the social context within which knowledge is produced and consumed.
In the second part of the chapter we argue that one way of addressing this
challenge is via the sorts of online multi-user systems that we call knowledge
communities. We describe some examples of systems that currently function as
knowledge communities and then turn to our own work on designing
infrastructures for knowledge communities. Our general approach is to design
online environments that, by making users and their activities visible to one
another, can enable a variety of social phenomena that support social and work-
oriented interaction. We describe a system called "Babble,” which we have
designed, implemented, and deployed to about twenty workgroups over the last
four years. We report on our experience with Babble, and conclude by discussing
some of the general issues we see for designing online environments that support
a socially-oriented approach to the management of knowledge and expertise.
2. Knowledge Work as Social Work
Knowledge management is often seen as an information problem: how to capture,
organize, and retrieve information. Given this perspective, it isn’t surprising that
knowledge management evokes notions of data mining and text clustering and
databases and documents. This is not wrong, but it is only part of the picture. We
suggest that knowledge management is not just an information problem, but that it
is, as well, a social problem.
2.1 An Example
One of us once interviewed accountants at a large accounting and consulting firm
about their information usage practices. The goal was to find out how they
thought they would use a proposed database of their company’s internal
documents. In the course of the investigation, an unexpected theme emerged: the
accountants said that one of the ways in which they wanted to use the documents
was as a means of locating people. The accountants’ claim — that they wanted to
use a document retrieval system to find people — was, at the time, quite
surprising. However, in the course of further interviews, it came to make sense: It
was only through the people that the accountants could get some of the knowledge
3
they needed. As one accountant explained, ‘Well, if I’m putting together a
proposal for Exxon, I really want to talk to people who’ve already worked with
them: they’ll know the politics and the history, and they can introduce me to their
contacts. None of that gets into reports!’
For our purposes, there are five important points here. First, as the accountants
observed, some types of knowledge tend not to get written down. Sometimes it
may be that the knowledge is too politically sensitive: people shy away from
recording gossip and innuendo, even though knowledge of it may be very helpful
to someone about to do business in that environment. Sometimes knowledge —
in the form of comments, opinions, or conjectures — may not be written down
because the resulting records can be potentially be subpoenaed. And sometimes
knowledge that may seem too trivial to be recorded when first encountered — that
the CEO is a teetotaler or a Scotch fancier — can prove quite valuable in the
process of establishing a relationship. Because this knowledge is often quite
useful for social purposes, we will refer to it by the rubric “social knowledge.”
The second point is that the accountants were not just tapping into social
knowledge; they were also getting access to social resources such as contacts and
referrals. One accountant explained that the worst way to approach a company
with a proposal was by making a “cold call”. It is much better if the accountant,
let us call him Charles, can begin a call to a new contact by saying ‘My colleague,
Jil Smith, suggested I chat with you.’ Being able to say that one has been referred
by a mutual acquaintance is a frequent and powerful facilitator for interpersonal
interaction — and this is true even if the relationship is only a few hours old.
Charles, by virtue of having permission to assert a relationship with Jil, can draw
on — to some extent — Jil’s reputation and standing with the person with whom
he is trying to open negotiations. Notice, by the way, that social resources can’t be
extracted from a person and embedded in a database: opening the conversation by
saying ‘I found your name in the corporate knowledge base’ isn’t the same as
saying ‘Jil Smith said I should call.’
The third point we take from this example is that people don’t necessarily need
access to an “expert.” It may be that Jil Smith has had only one previous
engagement with Exxon, and that, in terms of facts, she may have far less
expertise than an outside consultant. Nevertheless, Jil’s experience may be
sufficient to provide Charles with the social knowledge and social resources
necessary to gain entry into the Exxon environment. In fact, it may be preferable
for Charles to talk with Jil, because, as a colleague who shares the same work
context, she will understand more about what he needs to know, the situations in
which he will use the knowledge, and how he is likely to go about using it, than
someone traditionally construed as an expert. That is, Jil has social and contextual
expertise, in contrast to an outside consultant’s factual expertise.
The fourth point we take from this example is implicit in the previous ones:
networks of personal relationships, which are created and reinforced through
4
interpersonal conversation, are critical in supporting knowledge sharing. Let us
return to the example of Jil and Charles. Assuming that Jil's assistance was
helpful, Charles has now accrued a small debt or obligation to Jil. When Jil needs
assistance, she is likely, in turn, to come to Charles with questions or requests for
social knowledge that falls within his domain. Even if the required information is
outside of his domain, she may seek to obtain access to his social resources — a
referral to one of his contacts, for example. Thus are professional relationships
established, and thus do social networks grow. In the long run, if not the short, it
may be more valuable for an enterprise if its members seek knowledge and social
resources from one another — thus building a web of mutual knowledge and
trusted relationships — than if, for instance, employees are given instant access to
a top-notch external domain expert.
This brings us to our final point, which has to do with the centrality and
importance of conversation in knowledge sharing (see Fitzpatrick, this volume).
It is no coincidence that both social knowledge and social resources are best
shared through talk. It is the time spent discussing apparently trivial social
knowledge that suggests that a relationship goes beyond the purely professional
— that there is more in play than just a purely instrumental professional exchange.
It is the disclosure of politically sensitive information that indicates a degree of
trust between two people. It is the ability of one person to take generic
information and apply it — on the fly — to the other's problem that increases the
reputation of the giver and creates an obligation for the receiver. This sort of talk
— and the exchange of knowledge and social resources it involves — both
requires and strengthens networks of personal relationships in workplace.
2.2 The Social Construction of Knowledge
This sort of situation is not the exception, it is the rule. A wide variety of research
programs — for instance, ethnographies of workplaces, social studies of science,
critical theory, organizational memory, the sociology of knowledge — point to the
deep connections between knowledge management and social context.
For example, ethnographic studies of workplaces reveal a wide array of social
practices implicated in the production and dissemination of knowledge. Lave and
Wenger have developed the notion of a community of practice. They note that one
way in which people come to master a body of knowledge is through a sort of
apprenticeship or "legitimate peripheral participation" in the activities of a group
of practitioners (Lave and Wenger, 1991). Wenger (1998) describes the daily
work in an insurance claims processing office, and shows how it is entwined with
social relationships and processes. Similarly, in an ethnography of copier service
technicians, Orr (1996) reveals that technical knowledge is socially distributed
across a network of technicians, and that it is tapped into and disseminated
through oral processes such as storytelling.
5
A similar sense of the social nature of the production and dissemination of
knowledge comes from the field of social studies of science — see Latour and
Woolgar (1979) and Latour (1987). For example, Traweek's ethnography of
particle physicists (1988) examines some of the social phenomena that structure
the practice of high energy physics. She notes the impact of social relationships on
the placement of graduate students, the evaluation of experiments, and access to
equipment and facilities. Her comments on the role of conversation are
particularly interesting:
"...talk accomplishes diverse tasks for physicists: it creates, defines, and
maintains the boundaries of this dispersed but close-knit community; it is a
device for establishing, expressing, and manipulating relationships in networks; it
determines the fluctuating reputations of physicists, data, detectors, and ideas; it
articulates and affirms the shared moral code about the proper way to conduct
scientific inquiry. Acquiring the capacity to gossip, and to gain access to gossip
about physicists, data, detectors, and ideas is the final and necessary stage in the
training of a high energy physicist." (Traweek, page 122)
At a more general level, Brown and Duguid (1995) note that even documents,
which appear to be fixed, immutable public entities whose very purpose is to
transcend social contexts, "play an important role, bringing people from different
groups together to negotiate and coordinate common practices." Documents, they
suggest, in their production, use, and distribution, have their own social life, and
function as mediators of and catalysts for social activity.
2.3 Social Capitalism
An awareness of ways in which work is bound up with social factors has assumed
a prominent place in business discourse regarding knowledge management. Often
referred to as organizational learning in these contexts, knowledge management in
the organization is seen as a collective process in which teams create and share
knowledge (e.g. Senge, 1990; Nonacka and Takeuchi, 1995; Cohen and Prusak,
2001; Boone, 2001). While proponents typically invoke a systems perspective in
thinking about organizational processes, they also emphasize social factors —
such as relationships, trust, reputation, and commitment — in their descriptions of
how such processes play out. As a Vice President of Strategy puts it:
Expertise location is a big issue in companies today. The goal is not only to
provide access to information, but to provide access to people who have the
information. ... I don't want raw data, I don't even want information, I want the
judgments of people I trust. (Boone, page 22)
Recently the concept of social capital — the "features of social organizations
such as networks, norms, and social trust that facilitate coordination and
cooperation for mutual benefit" (Putnam, 2000) — and the possible role it may
play in the networked organization, has come to the fore. Cohen and Prusak
6
(2001) explain the connection:
Social capital makes an organization, or any cooperative group, more than a
collection of individuals intent on achieving their own private purposes. Social
capital bridges the space between people. Its characteristic elements and
indicators include high levels of trust, robust personal networks and vibrant
communities, shared understandings, and a sense of equitable participation in a
joint enterprise—all things that draw individuals together into a group. This kind
of connection supports collaboration, commitment, ready access to knowledge
and talent, and coherent organizational behavior. (Cohen and Prusak, page 4)
Elaborating on the connection between social capital and knowledge sharing,
Cohen and Prusak point out that exchanging knowledge depends on a social
connection — "without some degree of mutuality and trust, the knowledge
conversations will not get started; without some degree of shared understanding,
they will not go very far" (Cohen & Prusak, 2001, p. 86). They also note that the
knowledge exchanged in spontaneous conversations "is often social knowledge —
shared aims and interests discovered, signals and stories shared that build
confidence, trust, and connection—rather than technical or business knowledge
that can be directly applied to a product or problem" (Cohen and Prusak, pp. 86-
87).
2.4 The Challenge
Thus far we have argued that knowledge management is not just an information
problem, but that it is a social problem. That is, we’ve suggested that effective
knowledge management involves networks of people, relationships, and social
factors like trust, obligation, and commitment. One can’t isolate knowledge from
its social context without denaturing it, without stripping it of the social resources
and social knowledge that contribute to its utility.
Taking the social nature of knowledge seriously raises a considerable challenge
for those interested in designing knowledge management systems. We suggest
that the place to start is to stop thinking in terms of knowledge management, and
start thinking in terms of supporting the larger social context in which knowledge
management is embedded. Our response to this challenge is to explore the role of
online multiuser environments. In particular, we are interested in environments
within which users can engage socially with one another, and, in the process,
discover, develop, evolve, and explicate knowledge. We refer to online multi-user
environments used in these ways as “knowledge communities.” In what follows
we discuss current environments that function as knowledge communities, and
then turn to our own work on the topic.
7
3. Knowledge Communities
Knowledge communities have a long history, albeit not by that name. The idea
that networks of computers might provide a medium within which individuals
might come together to share knowledge and expertise dates back to at least 1960.
Perhaps the first vision of this nature was offered by Simon Ramo (Ramo, 1961),
who wrote of “many millions of human minds ... connected together.” Ramo
offered a number of scenarios, including one of an attorney consulting an on-line
database that contained more than data:
“Even on the nonroutine legal processes, the attorney, in the coming intellectronic age, will be
able to consult with the equivalent of a host of informed fellow attorneys. His request to the
system for similar cases will yield an immediate response from the central store, together with
questions and advice filed by other attorneys on those similar cases -- even as he will add his
facts and guidance into the system for future use by all.” page 10.
— Simon Ramo, The Scientific Extension of the Human Intellect. Computers and Automation,
Vol 10., No. 2, pp 9--12. February 1961. (Based on a talk given in 1960).
Over the ensuing decades the idea spread and evolved. From its beginning as a
vague if exciting vision, it took concrete form in the special purpose DELPHI and
EMISARI systems pioneered by Murray Turoff in the early ‘70s (Turoff, 1972;
Hiltz and Turoff, 1993) and in the PLATO Notes system in the mid ‘70s (Wooley,
1993). In the late ‘70s and early ‘80s the idea took off, spreading and evolving,
under pressures from application domains such as education and gaming, into a
variety of genres of software ranging from bulletin board systems to MOOs.
3.1 Some Examples of Knowledge Communities
A complete account of the systems which are used to enable online groups to
share knowledge among themselves is well beyond the scope of this chapter.
Instead, we will take the tack of looking at some representative examples to give
an idea of both the types of systems and the forms of use that are used in
managing online knowledge. It is important to note that we are not just interested
in the software; we are interested in the combination of the software and the way
in which it is put to use by its users — we refer to this comvination of technology
and usage as a knowledge community.
One genre of software that supports knowledge communities is the MOO.
MOOs, originally developed as multi-user text-based gaming environments, have
been applied to a number of pedagogical and business ends. Examples include
MOOSE Crossing, an educationally-oriented environment for children from eight
to thirteen (Bruckman, 1997); Pueblo, a school-centered MOO in Phoenix,
Arizona (O'Day, et al, 1996); Tapped In, a distributed community of teachers
(Schlager, et al., 1998; Schlager et al., in press); and a MUD used by employees at
Argonne National Labs for work-related talk (Churchill and Bly, 1999).
8
Another genre of system that can support knowledge communities is the
electronic mailing list, or listserv. While mailing lists are used for a variety of
purposes, the existence of mailing lists used to share knowledge among cohesive,
long-lasting communities is well documented. In one case, a community of about
a thousand professional journalists have used a mailing list to help one another
with technical problems and to find story-specific information sources for over six
years (Millen and Dray, 1999; Millen 2000). Another example, the use of a
mailing list to support discourse amongst a scholarly community, is described by
Ekeblad (1999). And a third example, the use of a listserv by a community of soap
opera fans, to share knowledge ranging from plot summaries to character
background information, is described by Baym (Baym 1995; Baym 1997).
In addition to the genres of mailing lists and MOOs, which can be turned to a
variety of ends other than knowledge sharing, quite a few systems have been
designed with their principal aim being the support of a knowledge community.
One example is Answer Garden (Ackerman, 1998), a blending of bulletin board
and email systems that makes a network of questions and answers available to its
users, and uses email to automatically route new questions to appropriate experts
whose answers are then incorporated into the network. The Zephyr Help Instance
(Ackerman and Palen, 1996) has a similar purpose — providing online help
information — but uses a synchronous chat-like mechanism to broadcast
questions and answers to the user community. Another genre of knowledge
community system is the collaboratory. Collaboratories are aimed at the needs of
the scientific community, and provide real time access to scientific instruments
along with synchronous communication channels ranging from textual chat to real
time audio and video (Olson and Olson, 2000). Collaboratories are a highly
successful class of applications, with many in existence that have supported
dozens to hundreds of users for periods of years.
If one examines these systems and the ways in which they’re used to share
knowledge, an interesting commonality emerges: Virtually all of these systems
exhibit a rich array of social phenomena, in spite of the fact that most provide
only textual communication mechanisms, typically synchronous chat,
asynchronous email, or both (as in MOOs). (Even collaboratories, which are
increasingly support ing various forms of high bandwidth synchronous
interaction, functioned well when chat was their dominant communication
channel.) Examples of the social phenomena found in most knowledge
communities range from interpersonal phenomena such as the negotiation of
status and reputation or the development of trust, to the emergence of group
norms and conventions. While these systems bear eloquent testimony to the
ingenuity of their users in using textual representations to support a rich array of
social phenomena, we suspect that we can do better.
This brings us to the question which informs our own work. What would it
mean to design an infrastructure for a knowledge community from the ground up?
9
That is, if we take seriously the charge that knowledge management is a social
problem as well as an information problem, one response is to ask how we can
better support social interaction. How do we go about designing a system which
supports not just information sharing, but that supports the exchange of social
knowledge and resources, the creation and growth of interpersonal networks and
accompanying social phenomena such as trust, obligation, commitment and
accountability?
To address this question, we’ve developed a system called “Babble” which
we’ve used as a testbed for exploring these issues over the last four years. We
begin by discussing the rationale that underlies Babble's design: the notion that
increasing the visibility of the presence and activity of participants in an online
environment can provide a foundation for a variety of social processes and
activity. Next we describe the system that we’ve implemented, and discuss the
ways in which we’ve come to use it as part of our daily work practice. Finally, we
discuss our experiences in deploying Babble to other work groups.
3.2 Supporting Online Social Interaction
In the building where our group works there is a door that opens from the
stairwell into the hallway. This door has a design problem: opened quickly, it is
likely to slam into anyone who is about to enter from the other direction. In an
attempt to fix this problem, a small sign was placed on the door: it reads, “Please
Open Slowly.” As you might guess, the sign is not a particularly effective
solution.
Let’s contrast this solution with one of a different sort: putting a glass window
in the door. The glass window approach means that the sign is no longer required.
As people approach the door they see whether anyone is on the other side and, if
so, they modulate their actions appropriately. This is a very simple example of
what we call a socially translucent system.
While it is obvious why this solution works, it is useful to examine the reasons
behind it carefully. We see three reasons for the effectiveness of the glass
window: First, the glass window makes socially significant information visible.
That is, as humans, we are perceptually attuned to movement and human faces
and figures: we notice and react to them more readily than we notice and interpret
a printed sign. Second, the glass window supports awareness: I don’t open the
door quickly because I know that you’re on the other side. This awareness brings
our social rules into play to govern our actions: we have been raised in a culture in
which slamming doors into other people is not sanctioned.
There is a third,
somewhat subtler reason for the efficacy of the glass window.
Suppose that I don’t
care whether I hurt others: nevertheless, I’ll open the door slowly because I
believe that you know that I know you’re there, and therefore I will be held
accountable for my actions. (This distinction is useful because, while
10
accountability and awareness usually co-occur in the physical world, they are not
necessarily coupled in the digital realm.) It is through such individual feelings of
accountability that norms, rules, and customs become effective mechanisms for
social control.
We call systems that exhibit these properties – of perceptual salience,
awareness, and accountability – socially translucent systems. But there is one
other aspect of social translucence that deserves mention. Why is it that we speak
of socially translucent systems rather than socially transparent systems? Because
there is a vital tension between privacy and visibility. What we say and do with
another person depends on who, and how many, are watching. Note that privacy is
neither good nor bad on its own—it simply supports certain types of behavior and
inhibits others. For example, the perceived validity of an election depends
crucially on keeping certain of its aspects very private, and other aspects very
public. As before, what we are seeing is the impact of awareness and
accountability: in the election, it is desirable that the voters not be accountable to
others for their votes, but that those who count the votes be accountable to all.
We see these three properties of socially translucent systems — visibility,
awareness, and accountability — as critical building blocks of social interaction.
Notice that social translucence is not just about people acting in accordance with
social rules (see Erickson & Kellogg, 2000). In socially translucent systems we
believe it will be easier for users to carry on coherent discussions; to observe and
imitate others’ actions; to engage in peer pressure; and to create, notice, and
conform to social conventions. We see social translucence as a requirement for
supporting online communication and collaboration in general, and knowledge
communities in particular.
This brings us to the question of how to support social translucence in online
environments. How can we provide the cues that allow our socially based
processes to operate — and which are so ubiquitous and lightweight in the
physical world — in online systems? Two obvious approaches are to use video or
3D virtual environments. However, these have several drawbacks for our
purposes. First, they don't scale well: we would like to support conversations
among fairly large numbers of people. Second, both approaches are best suited for
supporting synchronous interactions, whereas we would like to support both
synchronous and asynchronous interaction. Third, they are both relatively
demanding in terms of processing power, bandwidth, and display space and
characteristics: we would like to be able to support mobile employees working
over sub-56K connections and using devices with smaller displays.
As a consequence, we have taken a more abstract approach to supporting social
translucence. The abstract approach involves portraying social information in
ways that are not closely tied to its physical analogs. Exemplars of the abstract
approach include the Out to Lunch system (Cohen, 1994), which uses abstract
sonic cues to indicate socially salient activity, and Chat Circles (Viegas, et al.
11
1999), which uses abstract visual representations. This approach also includes the
use of text to portray social information; as we have already noted, text has
proved surprisingly powerful as a means for conveying social information in
knowledge communities.
3.3 The Babble System
Babble (Erickson, et al., 1999) is an online environment intended to support both
synchronous and asynchronous text-based conversations within small to medium
sized groups. The principle goal of Babble has been to serve as a platform for
exploring ideas about the social effects of supporting mutual awareness among
online groups. However, to do this effectively, we needed to be able to observe
'real' workgroups using it as part of the daily work process. As a consequence,
Babble needed to be sufficiently robust and lightweight to be usable by groups
who don't care about the technology itself.
In terms of infrastructure, Babble is a client-server system with both
components written in SmallTalk. Babble stores all data, except for user specific
preferences and state (e.g., the user's last location, last items read, and so forth) on
the server and broadcasts it as needed. Babble clients request the data they need
from the server (e.g., when a user switches to a new conversation the client
requests the content), and also notify the server of events that it will broadcast to
other clients. As this architecture suggests, Babble only works when on a network;
when disconnected it has no cache of conversation text. The Babble server runs
on a variety of server-class machines; the principle client runs on PCs, though we
have had, for varying durations, clients that ran on the Macintosh (in Java) and on
the Palm Pilot. Here we discuss only the PC client, since that comprises the vast
majority of our experience.
In terms of functionality, Babble resembles a multi-channel, text-based chat
system in that many users can connect to it, and select one of a variety of
conversations to participate in (or create their own). However, Babble differs from
conventional chat in two ways, both of which stem from our interest in supporting
knowledge communities. First, the textual conversation that occurs in Babble is
persistent: that is, unlike conventional chat where newly arriving users only see
what has transpired since they've joined a channel, Babble users can see
everything ever typed in any existing conversation. These traces give the system
the potential to function as a knowledge store, or what we prefer to call a
“discourse base.” Second, Babble makes the presence and activity of the
participants visible by a variety of means, but principally through what we call the
social proxy.
12
Figure 1 shows the Babble user interface. In the upper left hand corner is a list
of the names of users currently connected to Babble. In the middle upper pane is
the social proxy, which we will describe shortly. In the upper right pane is a
hierarchical list of the Babble conversation topics (grouped in categories and
subcategories). And the pane that occupies the lower half of the window contains
the text of the current conversation (whose topic name is highlighted in the topics
list); within the pane, each 'comment' is prefaced with the name of the user, and
date and time of its creation (recall that Babble conversations need not be
synchronous; indeed, some are asynchronous, with hours, days or weeks
separating comments). Babble provides a variety of other types of functionality
via the menu bar, context-sensitive menus accessed via right clicks, and keyboard
shortcuts. These include functions for creating messages, creating, changing, and
deleting topics and categories, conducting private, ephemeral chats, and so forth.
The social proxy, in the upper middle part of the window, represents the
current conversation as a large circle, and the participants as colored dots, referred
to, hereafter, as marbles. Marbles within the circle are involved in the
conversation being viewed; marbles outside the circle represent those who are
logged on but are viewing other conversations. What makes the social proxy
interesting has to do with the position of the marbles in the circle. When a user
becomes active, either 'speaking' (i.e., typing) or ‘listening’ (i.e., interacting with
the conversation window by clicking or scrolling), the user's marble moves
rapidly to the center ring of the circle. If the user stops interacting, the marble
Figure 1. The Babble Interface. From the upper left, clockwise: the list of all users
logged on; the social proxy; the topics list; the conversation pane.
13
gradually drifts out to the inner periphery of the circle over the course of about
twenty minutes. Thus, when there is a lot of activity in the conversation, there is a
tight cluster of marbles around the center of the circle. The social proxy shown in
Figure 1 depicts a situation in which five people have been recently active (i.e.,
speaking or listening) in the current conversation, and two others have been idle
for a while (and an eighth person is off viewing another conversation).
When people leave the current conversation their marbles move outside the
circle; when they enter the conversation, their marbles move into the circle. When
a person logs onto the system, it creates a virtual wedge for their marble, adjusting
the position of all the marbles in the social proxy; when they depart, the wedges
are destroyed, and the remaining marbles adjust to uniformly occupy the space.
All marble movements are shown with animation, thus making arrivals,
movements, and departures visually salient. Although simple, this social proxy
gives a sense of the size of the audience, the degree to which the audience is
actively listening or contributing, as well as indicating whether people are
gathering or dispersing, and who it is that is coming and going.
In addition to the social proxy (which we refer to as 'the cookie'), Babble uses
additional mechanisms to reveal the presence and activity of users. In the topic
list, to the left of the topic names, are 'mini-cookies', thumbnails of the social
proxy for each topic with a participant in it. So, in Figure 1, we can see that there
is a single person in the second topic, "Amusing Wendy." Babble also highlights
information that the user hasn't yet seen: the names of topics with new material in
them are shown in red (e.g., "Amusing Wendy" in Figure 1), and comments that
have been added to the current conversation since the user last 'touched' Babble
are shown in reverse highlighting.
One of the shortcomings of the cookie is that it only works for synchronous
interactions — that is, it shows only the presence and activities of people who are
currently logged on to Babble. This is a considerable drawback because the
majority of the conversations carried on in Babble are asynchronous, with just a
few comments per day (or per week, or per month). As a consequence, we
designed a second, asynchronous social proxy for Babble: the Timeline (Figure 2).
The basic goal of the Timeline was to provide a way for a ‘speaker’ to see that
people were ‘listening’ (or not), even when the listening was offset in time. The
Timeline proxy works as follows: each user is represented by a row in the
Timeline; when they are logged on to Babble, they leave a flat trace or line, and
when they ‘speak’ they leave a vertical mark or blip on the line. If the line/blip is
in color, it means that that user was present/speaking in the conversation currently
being viewed by the user of the timeline; if they were in a different conversation,
the line/blip is shown in gray (and the line becomes thinner). As the user mouses
over the Timeline, the name of the topic, the user, and the time being examined is
shown in the upper left corner; the user can scroll back through as much as one
week of activity. The Timeline also provides access to other functionality via a
14
menu accessed via a right-click on another user’s row (e.g., private chats).
For example, in Figure 2, we can see that nine people have logged onto Babble
(shown by the presence of lines), and that all of them have spent some time in the
current conversation (shown by the color/increased thickness of the lines), and
that many but not all have ‘spoken’ (shown by the blips). The line being indicated
by the cursor shows that the user ‘Peter’ logged on around 11am, made a couple
of comments in the “Commons Area” conversation, switched to another topic,
and then switched back to the Commons area about 1pm, and then logged off.
3.4 How a Babble is Used by a Group
While one must be wary about drawing conclusions concerning the usability of
software when it is used by its developers, our aim here is to simply provide a
sense for how Babble is actually used by a group. We’ll begin by describing the
group, and then move on to discuss how Babble is actually used. In the next
section we'll discuss our deployments of Babble to other groups and some of the
phenomenon that we’ve observed across different deployments.
Our group has used Babble for about four years. The group is centered around
the software development group (AKA “the lab”) that designed and implemented
the system, and includes a mix of computer scientists and social scientists
(including the authors). The size of the group has varied in number over the years
from four to nineteen users. Part of the variance is due to the ebb and flow of
people characteristic of groups in large organizations; and part is due to current
members of the lab inviting "associates" — colleagues with whom they had strong
social or professional ties — to join Babble.
Geographically, the group of Babble users is about half co-located in New
York, and half distributed. Most of the lab members are located in the same
building, although offices tend to be distributed around the building — so actual
adjacency is rare. Three members of the lab are telecommuters, and spend the
majority of their time tens to hundreds of miles away; other members of the lab
frequently work at home. Four of the six associated colleagues (i.e. those not
officially members of the lab, but users of Babble) are remotely located.
Socially, the lab is a cohesive group, with considerable camaraderie. In
addition to work-based collaboration, the lab members occasionally socialize,
although usually within business hours (e.g., going out to lunch) The associates
Figure 2. The Timeline (showing 3 hours of activity).
15
vary in the strength and number of their ties to the lab members, some known to
almost all lab members, and others known only to one or two lab members with
whom they have shared interests. Conversation in the Babble system moves
fluidly between work and social talk; it is always civil, frequently informal, and
joking, teasing, and other ludic behavior is not unusual.
Overall, the Babble system as used by the lab can be characterized as a core of
relatively synchronous activity surrounded by a constellation of asynchronous
conversations. At the center of activity is a topic called the "Commons Area," a
place where collocated and remote members greet one another, share news,
engage in banter, and ask general questions. Members of the lab tend 'hang out' in
the Commons Area, often remaining logged on for most of the work day.
Comments in the Commons Area tend to be short and informal, with relaxed
syntax and punctuation, use of paralinguistic expressions (“ummm”),
onomatopoeia, emoticons, and playful tropes (for example, the ‘tossing of
cookies’ to ‘a dog’ who usually ‘accompanies’ one of the participants — all done
via text, of course). The content of conversation in the Commons Area
ranges
from purely social talk (such as the custom of saying “good morning”), to the
posing of general questions, to reminding people of an impending meeting of
general interest, to more technical discussions about work projects. (In theory,
more topic-oriented discussion is ‘supposed’ to take place in specific topics; in
practice, work talk often grows out of social discussions, and the recognition that
a substantive conversation that ‘belongs somewhere else’ is taking place is often
not recognized until after the fact.) Because of the amount of talk that occurs in
the Commons, the content of the Commons Area is automatically archived once a
week.
In addition to the commons area, there are a variety of other topics or
conversation areas in Babble. These have ranged in number from a dozen or so in
the early days of Babble to several dozen, the growth being facilitated by the
addition of an expandable hierarchical topic list. These topics tend to have
asynchronous and mostly sporadic conversation, and they tend to be focused on
particular purposes, typically either project-oriented or person oriented. Examples
of topics include personal offices (e.g., "Tom's Office"), project-oriented topics
(e.g., "Babble Ethnographies—CB Babble"), and occasional non-work topics
(e.g., "Bad Jokes").
In general, uses of Babble can be grouped into three general categories:
social/ludic; informative; and instrumental. Social/ludic activities are those
engaged in for social and entertainment purposes such as the custom of
exchanging morning greetings, and the topic devoted to jokes. Informative
activities have to do with actions on the system that are addressed to the group as
a whole, or to no one in particular, and generally are done without expectation of
a reply or responsive action. These activities include posting announcements and
other news believed to be of general interest, commenting on project activity, and
16
keeping on-line notebooks or offices. The third type of activity is instrumental,
that is, activities engaged in with a particular end in mind. These include starting
or participating in focused discussions, posting bug reports, holding on-line
meetings, and asking questions. These activities are often, though not always,
addressed to a particular participant or group of participants.
3.5 Adoption and Social Phenomena across Babble Deployments
Over the last four years we've deployed Babbles to about twenty groups. We’ve
studied the deployments using techniques ranging from ethnographic studies —
see Bradner, et al. (1999) for a study of six Babbles — to studies based on surveys
and analyses of log data and conversation archives.
We have had mixed experiences with the adoption of Babble. Sometimes
groups try Babble out, but fail to adopt it (typically about six weeks pass before it
is evident whether or not the Babble is going to be adopted by the group). Other
times groups use Babble for a period of months, and then cease (either because it
was for a particular event or period that has ended, or because the composition or
needs of the group change). It isn’t clear how to operationally define a successful
deployment of Babble: the group uses it for its entire existence? the group uses
Babble actively for six months? the group uses Babble to carry out a particular
activity? If we take, as a rule of thumb, that a Babble is successful when it is used
on a more or less daily basis by several people for more than six weeks, we can
say that about half of our Babble deployments have met with success. As of this
writing, we have five Babbles running, all of which are well past the six week
mark, and all exhibiting robust daily activity.
When a Babble is adopted by a group, it usually supports a variety of
communicative purposes and practices (often similar to those described in the
previous section). Here, we describe four social phenomena that we’ve observed
in a number (though not all) of successfully adopted deployments that are most
relevant to knowledge communities.
One phenomenon is waylay, in which a user watches for a particular person to
become active on Babble (signaled by the movement of their marble into the
center of the social proxy), and then initiates a conversation (either publicly
within Babble, via Babble’s private chat mechanism or by some external means
such as the telephone). Because the movement of the marble occurs when the user
has just begun an episode of typing or mousing, it indicates a opportune moment
for contact (since the user’s attention has just shifted to communication with the
group). Waylay is used for purposes ranging from asking questions to initiating
casual social chat. In general, forms of opportunistic interaction such as waylay
permit the same sorts of requests for assistance and transfers of social resources
that we’ve observed in face to face knowledge sharing situations, with the
accompanying effect of strengthening of interpersonal ties.
17
Babble also supports the maintenance of group awareness through the
exchange of social knowledge. For example, when members of a Babble travel,
many report reading through conversations that occurred in their absence to ‘find
out what happened.’ For someone who is a member of the group and understands
the context, seemingly trivial comments can convey considerable information
about what’s going on at the individual, group, and organizational levels. Thus, a
sign off — “I have to go to the [project] meeting now” — reveals that one
participant is still involved in a particular project, and a question — “Does anyone
know how to do a screen capture” — indicates that someone is beginning to write
a paper. Babble also supports group awareness through the Timeline proxy.
Babble participants have reported uses such as: looking to see who has visited a
topic in which they had posted questions; looking to see whether a colleague who
hadn’t posted recently had been online; and using the Timeline to get a sense for
the activity of the community as a whole.
Another phenomenon that can be observed across Babbles is the development
of social norms. That is, one participant may develop a particular way of doing
something, and others will imitate it. Examples of this include what users include
in their online nickname (e.g., in some Babbles users append “@mylocation” after
their name), the types of online conversations created (e.g., some Babbles have
categories for "personal places" or "offices"), and naming conventions (e.g., one
Babble uses the term "chit-chat" to signal that a topic is intended for casual
conversation. Babble groups also evolve various interactive customs, the most
common being to say 'hello' upon logging in (even when no one else is present).
Again, the existence of these norms supports social interaction by providing
expectations about how to behave.
Finally, we’ve observed that Babbles are typically regarded as semi-private,
“trusted” places. This became apparent when ‘strangers’ appeared in various
Babble systems. Sometimes the strangers were unannounced new members,
sometimes they were visitors provided access by an unreflective manager, and, in
one case, the stranger was actually an unannounced conversational software agent.
But in all cases, the arrival and presence of the stranger (reflected in the social
proxies along with the presence of the regulars) was greeted with considerable
consternation. In each case, the appearance of strangers provoked concern about
how unguarded conversations might be interpreted by those from different
contexts, and led to the creation of visitor and membership policies. We suggest
that this concern reflects the success of Babble as an online space that is rich in
social context.
One issue that is not clear, so far, is the degree to which Babble’s social
proxies contribute to these phenomena. Analytically, it is difficult to isolate the
effects of the social proxies, from the effects of purely textual cues. Certainly,
there are a number of social practices (such as waylay) which require (or are at
least greatly facilitated by) the proxies. It is clear that the participants, in general,
18
like the proxies and want them retained as a feature of the system. One user,
responding to a question in Babble, writes:
“Ah, the cookie… we love the cookie…the cookie is good – our colored dots
circulate around to ‘make room’ when someone new joins the conversation –
that’s fun. And when someone’s connection dies, they rather dissemble into the
ether, angelic like. Which is sort of fun to watch. … Also, when I’m wondering
whether my comments have fallen on deaf ears, I can tell when a response may in
fact be on its way when someone’s dot moves back to the center (happens as soon
as someone starts typing). So, yes, we like the cookie – it makes me feel like
there are actually people in a room with me…”
It is also clear that users are able to ‘read’ Babble proxies, using them to draw
inferences about the presence of individuals and the activity of the community as
a whole. Another user, commenting on the Timeline proxy, remarks:
“It’s a little like reading an electrocardiogram, the heartbeat of the community. I
noticed that I missed Sandy by an hour on Monday morning.... Pat comes in every
so often as a blip. Lynn jumps from space to space....”
Nevertheless, although we have compelling anecdotes and a large fund of positive
comments by Babble users, analytically separating social benefits conveyed by
proxies from those produced by text remains as a challenge for the future.
3.6. Babble as an Infrastructure for Knowledge Communities
Babble clearly succeeds as a multi-user online environment where sustained social
interaction takes place. But does it support knowledge communities? Is the social
interaction combined with the sharing of information, social knowledge and social
resources via personal social networks that, we suggest, is a crucial part of
knowledge management? This is indeed what we have observed. In the following,
we refer to examples
1
and survey results drawn from a Babble whose membership
is composed of a world wide cross-section of people in IBM and Lotus interested
in online communities.
Perhaps the first point to make is that participants do feel as though they are
part of a community. This is particularly important to those who are remote
teleworkers:
“I work remotely and can feel very isolated when I don’t travel regularly (as has
been the case for the past six months because of travel restrictions). Babble has
provided me with a way to feel connected with a group of people outside my
basement walls. It is my portal (so to speak) into IBM.”
Another says:
“As a home office worker, this is perhaps one of the things I miss the most – the
ongoing banter I can have with colleagues who are focused on a similar work
1
Identifiers have been changed to protect confidentiality, and comments edited for brevity
19
topic as I am.”
This is not simply a feeling of a vague belonging to a group; participants report
feeling as though they are hooked into social networks. One participant reports
that participation in Babble strengthened an existing network:
“Babble has helped me establish a tighter social and professional relationship
with all of them – we have much more regular contact with each other, much as
we would if we were collocated, via the Babble connection. This in turn has
built social capital among us which may be of use in the future.”
And these social networks are not just about talk, they can also be tapped for
assistance. The participant continues:
“I have also contacted Vera about getting her input and advice about setting up a
knowledge network, which is part of my ‘real work.’ I felt much more
comfortable about approaching her with this question as a result of our frequent
contacts via Babble than I would have otherwise.”
Another Babble member notes:
“I like the back and forth. …we have a lot of reflective talk about our own
experiences... In at least one case, e.g., a half-joking comment of mine, “anybody
want to fund this?” has led to e-mail, phone, and face-to-face meetings and now a
serious proposal for funding. I don’t know the final outcome yet,
2
but we have
found out something significant about another part of the business and have made
a serious attempt to propose [a] solution to their problems.”
These comments are prima facie evidence that knowledge sharing and
expertise management are deeply social processes — that people value informal
exchanges with colleagues, and may only venture a non-trivial request for
information or assistance after a social relationship has been established.
A danger in using the summary remarks of participants to what happens in
Babble is that it makes it sound a bit more straightforward and calculated than it
is. It is difficult to convey the way in which these effects emerge out of a rich
melange of social and work talk. For example, one instance of the transfer of
social resources occurred over the course of a multi-threaded, 30 utterance, 17-
minute Babble conversation on March 7, 2001. The conversation consisted of two
primary participants (‘scienceguy’ and ‘Patrick’), and was composed of four
distinct threads. Two threads were related to work topics (Patrick explaining that
he had referred some colleagues to scienceguy , and a discussion of the use of
patterns in knowledge management), and two were more social threads (one an
attempt to identify an earlier participant’s real name, another a request by
scienceguy for assistance in developing an Irish accent for an upcoming
storytelling performance). The two work related tasks were treated relatively
seriously, even as the two interleaved non-work threads were used as an excuse
for banter. Yet both the social and work threads developed and played off one
2
The project was funded.
20
other throughout the conversation, which concluded with Patrick revealing the
names of the colleagues whom he has referred to scienceguy, and scienceguy
indicating that he would be happy to talk with them.
(
The situation grows more
complex when one recognizes that Babble users are remote from one another, and
may be simultaneously carrying on other work on their computers, via the
telephone, or orally with co-located colleagues.)
5. Concluding Remarks
In this chapter we’ve argued that knowledge management is not just an
information problem, but is, as well, a social problem that involves people,
relationships, and social factors like trust, obligation, commitment, and
accountability. This view raises a considerable challenge for those interested in
designing systems to support knowledge management. Our approach has been to
explore the creation of infrastructures for knowledge communities: on-line
environments within which users can engage socially with one another, and, in the
process, discover, develop, evolve, and explicate knowledge.
In our work on Babble, we’ve begun exploring ways of creating infrastructures
that support rich forms of social interaction. We’ve found that social proxies are a
promising development, and continue to be impressed with the power of plain text
as a means of supporting interactions that are both complex and subtle. We
believe that one of the most important aspects of a knowledge community is that
it can be used as a place for unguarded discussion among people who know one
another, who share professional interests, and who understand the contexts within
which their remarks are being made.
The notion of a knowledge management environment as a ‘trusted place’ is an
interesting and challenging one. How — technically, socially, and
organizationally — can we balance the need for a safe and trusting place with the
organizational imperative to share information? One decision facing us as
designers is how and to what extent we “design in” norms and social conventions.
For example, if we build in technical mechanisms to provide privacy, in addition
to the usability impact, we also eliminate opportunities for participants to show
that they may be trusted, or to rely on others to respect their privacy. The Babble
prototype has no technical features for controlling access: anyone who has access
to the client could, in theory, enter any Babble space. But, because Babble makes
users visible, this results in groups noticing, commenting on, and ultimately
discussing how to deal with this issue. We believe that a greater understanding of
how to design systems that permit social mechanisms to come into play is of great
importance in designing future systems for knowledge management.
21
Acknowledgments
Thanks to David N. Smith for creating the Babble prototype, to Mark Laff , Peter
Malkin, and Amy Katriel for implementation work on the Babble server and
clients, and to Cal Swart for critical assistance in the deployment of a multitude of
Babbles. Thanks, as well to members of IBM's Social Computing and Pervasive
Applications groups for productive conversations, and to the many dozens of
'Babblers' who have shared their insights, responded patiently to surveys, and,
most importantly, used Babble in many of productive (and often surprising) ways.
References
Ackerman, M.S. (1998); Augmenting organizational memory: a field study of
Answer Garden, ACM Transactions on Information Systems, Vol.
16
, No. 3,
July 1998, p203-24.
Ackerman, M. S. and Palen, L. (1996) The Zephyr Help Instance: promoting
ongoing activity in a CSCW system. In Human Factors in Computing
Systems: The Proceedings of CHI '96. New York: ACM Press, pp. 268 - 275
Baym, N. K. (1995) The Emergence of Community in Computer-Mediated
Communication. Cybersociety: Computer-Mediated Communication and
Community (ed. S. Jones). Thousand Oaks, CA: Sage Publications, 1995. pp
138-163.
Baym, N. K., (1997) Interpreting Soap Operas and Creating Community: Inside
an Electronic Fan Culture, In Culture of the Internet (ed. S. Kiesler).
Mahwah, NJ: Lawrence Erlbaum Associates, pp. 103-120.
Boone, M. E. (2001). Managing Inter@ctivity. New York, NY: McGraw-Hill.
Bradner, E., Kellogg, W.A., and Erickson, T. (1999) The Adoption and use of
Babble: A field study of chat in the workplace. Proceedings of the European
Conference on Computer-Supported Cooperative Work (ECSCW '99), 139-
158, Kluwer Academic Publishers, 1999.
Brown, J.S. and Duguid, P. (1995) The Social Life of Documents. In Release 1.0
(ed. E. Dyson), October 1995. New York: EDventure Holdings Inc., pp. 1-18.
Also at: www.parc.xerox.com/ops/members/brown/papers/sociallife.html
22
Bruckman, A. (1997) MOOSE Goes to School: A Comparison of Three
Classrooms Using a CSCL Environment. In the Proceedings of CSCL 97,
Toronto, December 1997.
Churchill, E. F. and Bly, S. (1999) "Virtual Environments at Work: Ongoing use
of MUDs in the Workplace." The Proceedings of WACC '99 (San Francisco,
CA, 1999). New York: ACM Press.
Cohen, D. and Prusak, L. (2001) In good company: How social capital makes
organizations work. Boston, MA: Harvard Business School Press.
Cohen, J. (1994). Monitoring background activities. In G. Kramer (Ed.), Auditory
display. New York: Addison-Wesley, pp. 439-531.
Donath, J., Karahalios, K., & Viegas, F. (1999). Visualizing conversation. In J.F.
Nunamaker, Jr., & R.H. Sprague, Jr., (Eds.), Proceedings of the Thirty-
Second Hawai’i International Conference on Systems Science. Los Alamitos,
CA: IEEE Computer Society Press.
Ekeblad, E. (1999). The Emergence of Multilogue on a Scholarly Mailinglist,
Paper presented in the symposium: “Time and coordination in a virtual
community of learners,” at the 8th European Conference for Research on
Learning and Instruction (EARLI), Göteborg, Sweden, August 24-28 1999.
Available at: http://hem.fyristorg.com/evaek/writings/earli99/multdyn.html
Erickson, T., Smith, D.N., Kellogg, W.A., Laff, M.R., Richards, J.T., and
Bradner, E. (1999) Socially translucent systems: Social proxies, persistent
conversation, and the design of Babble. CHI 99 Conference Proceedings:
Human Factors in Computing System, 72-79, ACM Press, 1999.
Erickson, T. and Kellogg, W.A. (2000). Social translucence: An approach to
designing systems that mesh with social processes. ACM Transactions on
Computer-Human Interaction, Vol.
7
, No. 1, pp. 59-83.
Erickson, T. and Laff, M.R. (2001) The design of the ‘Babble’ timeline:
A social proxy for visualizing group activity over time. In Extended
Abstracts: The Proceedings of CHI 2001. New York: ACM Press.
Hiltz, S. R. and Turoff, M. (1993). The Network Nation (Revised edition),
Cambridge, MA: MIT Press.
Latour, B. (1987) Science in Action. Cambridge, MA: Harvard Universtiy Press.
23
Latour, B. and Woolgar, S. (1979) Laboratory Life: The Social Construction of
Scientific Facts. London: Sage. (2
nd
edition; Princeton: Princeton University
Press, 1986).
Lave, J. and Wenger, E. Situated Learning: Legitimate Peripheral Participation.
Cambridge: Cambridge University Press, 1991.
Millen, D. (2000) Community Portals and Collective Goods: Conversation
Archives as an Information Resource. Proceedings of HICSS - 33
rd
Annual
Hawaii International conference on Systems Sciences, January 4-7, 2000.
Millen, D. R., & Dray, S. (1999) Information Sharing in an Online Community of
Journalists. In Proceedings of Esprit i3 Workshop: Ethnographic Studies in
Real and Virtual Environments: Inhabited Information Spaces and Connected
Communities. Edinburgh, January 25-26.
Nonaka, I. and Takeuchi, H. (1995) The Knowledge-Creating Company : How
Japanese Companies Create the Dynamics of Innovation. New York: Oxford
University Press.
O'Day, V. L., Bobrow, D. G., and Shirley, M. (1996) "The Social-Technical
Design Circle. In the Proceedings of CSCW 96. (November 1996,
Cambridge, MA). New York: ACM Press.
Olson, G. and Olson, J. (2000). Distance matters. Human-Computer Interaction,
Volume
15
, Vol. 2-3.
Orr, J. E. Talking About Machines: An Ethnography of a Modern Job. Ithica, NY:
Cornell University Press, 1996.
Putnam, R. D. (2000) Bowling Alone: The Collapse and Revival of American
Community. New York: Simon and Schuster.
Ramo, S. (1961) The Scientific Extension of the Human Intellect. Computers and
Automation, Vol 10., No. 2, pp 9-12. February 1961. [Based on a talk given
in 1960].
Schlager, M., Fusco, J., & Schank, P. (1998). Cornerstones for an on-line
community of education professionals. IEEE Technology and Society
Magazine, 17(4), 15-21/40.
24
Schlager, M., Fusco, J., & Schank, P. (in press). "Evolution of an On-line
Education Community of Practice." To appear in K. A. Renninger and W.
Shumar (Eds.), Building virtual communities: Learning and change in
cyberspace. NY: Cambridge University Press.
Senge, P. M. (1990) The Fifth Discipline : The Art and Practice of the Learning
Organization. New York: Doubleday.
Traweek, S. Beamtimes and Lifetimes: The World of high energy physics.
Cambridge, MA: Harvard University Press, 1988.
Turoff, M. (1972) Delphi Conferencing: Computer-Based Conferencing with
Anonymity, Technological Forecasting and Social Change, Vol. 3, 159-204.
Viegas, F.B. and Donath, J. Chat Circles. CHI 99 Conference Proceedings:
Human Factors in Computing System, 9-16, ACM Press, 1999.
Wenger, E. (1998) Communities of practice: Learning, meaning, and identity.
Cambridge: Cambridge University Press.
Wooley, David. (1994). PLATO: The Emergence of Online Community. On -line
version at http://www.thinkofit.com/plato/dwplato.htm; an earlier version
appeared in Matrix News (ed. J. Quarterman), January 1994.
