|
|
Knowledge Communities as Socially Translucent Systems
An Adventurous Research Project
This is the original proposal for our AR project, written in August, 1998
by Tom Erickson, Wendy Kellogg, Mark Laff, and John Richards. The proposal
was subsequently funded and was carried out -- changing along the way,
of course -- from 1999 through the end of 2002. For results, see our publications
on the topics of "Babble," "Loops," "Social Translucence,"
and "Knowledge Communities."
Introduction
Technology is always deployed in the context of social systems. Our starting
premise is that digital systems can be (but rarely are) designed to both
support and take advantage of this fact. We offer a vision of what we
call socially translucent systems (STS), and suggest that viewing digital
systems from this perspective can drastically shift the computer’s
role in organizations. We then apply the STS perspective to the design
of a knowledge community, a conversationally- based system that supports
the creation, management, and re-use of knowledge in a social context.
Socially Translucent Systems
A socially translucent system is a system that permits the transmission
of social information, and thus provides traction for our social intelligence.
Consider these real world examples of STS:
- In another town on business, you and a few colleagues
are looking for a place to have dinner. You come to a small restaurant:
and through its window you see a cozy room with candle-lit tables, and
waiters bustling about; you can hear the faint murmur of conversation,
and the clink of glasses and cutlery. In you go.
- You’ve arrived at the opening reception for
a convention. You look around for someone to chat with and notice someone
you recognize gesturing excitedly as others listen intently. Curious,
you wander over.
- You’re shopping for a red wine for dinner. You
come across an almost empty bin amid the otherwise well-stocked racks.
It’s a medium priced Cabernet, and so without more ado you grab
a bottle for yourself.
These examples are unremarkable. Every day we make countless decisions
based on the activity of those around us. As social creatures we’ve
spent hundreds of thousands of years evolving an exquisite sensitivity
to the actions and interactions of others. Whether it’s wrapping
up a presentation when the audience starts fidgeting, or homing in on
a crowded booth at a trade show, social information like this provides
the basis for inferences, planning, and coordination of activity.
In all of this the only thing that is remarkable is
how different things are when we move from the ordinary world into the
world of digital systems. Digital systems are almost completely opaque
to social information. Most of our knowledge about people, most of our
sensitivity to their interactions, most of our ability to make use of
what others know, goes unused. In the digital world we are socially blind.
An STS Approach to Knowledge Management
What might it mean to have social translucence in a digital system? To
answer this question, let’s look at knowledge management from an
STS perspective.
Knowledge management is often seen as a problem of capturing, organizing,
and retrieving information. Given this perspective, it isn’t surprising
that when we think of knowledge management, we think of data mining and
text clustering and databases and documents. This is not wrong, but it
is only part of the picture.
For example, one of us once interviewed accountants about how they would
use a (proposed) database of their company’s internal documents.
A rather startling theme emerged: the accountants said that they’d
love to access the documents — so that they could find out who wrote
them. As one 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!’
How curious: the accountants wanted to use a data access system to access
the people who produced the data. It was only through the people —
and the social networks they were part of — that the accountants
could get the knowledge and social resources they really needed.
This sort of situation is not the exception; it is the rule. Knowledge
— whether it is of bugs in the Java Virtual Machine or how to begin
negotiations with an executive from another culture — is discovered,
shared, and used in a social context. Having to depend on a database isn’t
as useful as we would hope, unless it also provides an entree into the
social networks that produced the data.
Imagine a knowledge management system that was designed from a social
perspective, a system predicated on the assumption that knowledge is distributed
throughout a network of people and that only a small proportion of it
is captured in concrete form. As the above vignette suggests, such a system
would, along with its data and documents, also provide a rich set of connections
back to the social network of people who produced the information. But,
if we think in terms of STS, additional possibilities suggest themselves.
Imagine that the knowledge management system provided access not only
to authors, but to people who were accessing and using the knowledge.
Suppose that — just as we look for busy restaurants, notice crowded
trade show booths, or are drawn to engaging conversations — we could
see similar traces of those using information in a knowledge management
system. We could notice popular knowledge sources, encounter other users
with similar interests, and perhaps get glimpses of how knowledge was
being re-purposed. That is, because users often must do a lot of work
to adapt knowledge to their own ends, they develop an understanding of
its shortcomings and particularities (as well as building on it) which
would be very valuable to others engaged in similar efforts. Such a system
would not be just a database from which workers retrieved knowledge, it
would be a knowledge community, a place in which people discover, use,
and manipulate knowledge, and encounter and interact with others who are
doing likewise.
A knowledge community of this sort has a formidable social problem to
overcome: Why should those who produce and use knowledge take the time
to engage in such interactions? Why should they want to? What personal
benefits would they gain for sharing their knowledge? This is one of the
deep and difficult research problems that we propose to investigate. We
suggest that the solution to the problem arises from social translucence:
because such a system would make knowledge work visible (thus allowing
people to observe and contact one another), it would also enable those
skilled at unearthing, applying, and adapting knowledge to receive credit
for what is all too often an invisible form of work. If knowledge work
is made visible it can be recognized and rewarded by the organization,
and it can shift from something that takes time away from ‘real work’
to being ‘real work’ in and of itself.
Having sketched the notion of a knowledge community, we need to discuss
one other issue before turning to the details of our proposal: Through
what mechanism can knowledge work be made visible? How can it leave traces
so that not only can others see it occurring, but that those who were
not present can gain value from it at a later time? We believe that the
answer to this lies in conversation.
Conversation
Conversation is essential. We use it as a medium for decision-making.
It is through conversation that we create, develop, validate, and share
knowledge. When computational or bureaucratic systems prove too rigid,
we figure out the work-arounds via conversation. And with all our advances
in information retrieval, the preferred method for obtaining information
is still to ask a colleague.
Why is this? We suggest that the power and ubiquity of conversation is
due to the fact that it is both a deeply interactive intellectual process
as well as a fundamentally social one. Consider conversation as an intellectual
process. As we talk we refer to a common ground of already established
understandings, shared experiences, and past history. As the conversation
unfolds, we continuously attempt to interpret what is said, verify that
we have been understood, and offer new contributions to extend the common
ground. Sometimes misunderstandings occur, and so we attempt to fix them
by rephrasing our words, or ‘debugging’ the previous conversation
to reveal that our shared understandings were not, in fact, shared. What
all this amounts to is that conversation is a superb intellectual tool
for eliciting, unpacking, articulating, applying, and recontextualizing
knowledge.
But conversation is not just an intellectual endeavor: it is also a fundamentally
social process. First, people speak to an audience. Speakers notice how
their audience is reacting and steer their remarks appropriately: nods
and eye contact convey one message; questions and furrowed brows another;
it is this social feedback that guides the elicitation and interpretation
of knowledge. Conversation is also social in that people portray themselves
through conversation. They advance their personal agendas, project their
personal style, take credit, share blame, and accomplish other social
ends through their talk, often with a great deal of subtlety. The social
nature of talk is not an undesirable side effect, but rather the heart
of it: personal motivations fuel conversation and provide the energy for
the considerable intellectual work that conversation requires.
Even though conversation is essential to the conduct of our business,
social, and personal lives, digital systems don’t support it well.
E-mail, mailing lists, discussion databases, etc., all have substantial
shortcomings. There are many problems: addressing; managing threads; bringing
others into the middle of a conversation; avoiding conversation drift;
knowing who (or whether anyone) is listening; getting people to respond
promptly; finding old messages with crucial information; etc. In short,
it is difficult to conduct a long-running, productive conversation through
the digital medium, especially if there are more than a few people involved.
We claim that these shortcomings are not inherent in digital systems,
and that understanding how to design a system that supports deep, coherent,
long-running conversations among groups of people — especially if
those conversations could later be ‘mined’ for knowledge —
would be of immeasurable value to our customers.
The Proposal: Conversation-Based Knowledge
Communities
Our basic premise is that it is possible to design digital systems that,
by making users and their activities visible to one another, can engage
our social intelligence. In particular we propose to design, implement,
and deploy the infrastructure for conversationally-based knowledge communities.
Once deployed, we will study the adoption and evolution of knowledge communities,
so that we may iterate on the design of the infrastructure, and also develop
a better understanding of the social and institutional work necessary
to support knowledge communities.
 |
Early Work
We are not beginning from a blank slate. For the last year our group has
been developing and using a working prototype called Babble that supports
text-based conversation — either synchronous or asynchronous —
and begins the exploration of applying social translucence to a digital
system by way of what we call a social proxy. Our first version of the
social proxy (figure 1), which shows the participants in a conversation
as colored dots, gives an idea of whether they have recently ‘spoken’
or ‘listened’ (dots drift to the periphery with inactivity),
and shows when people leave or join a conversation. This social proxy,
although simple, gives a sense of the size of the audience and the amount
of activity in a conversation at any point in time; in addition to supporting
the focused conversation, the social proxy, by providing awareness of
who is coming, going, or otherwise active, supports the digital equivalent
of impromptu hallway conversations (individuals can be contacted by clicking
on their dots).
While Babble has served as a proof of concept, and resulted in the creation
of working prototypes of clients and a server on which we can build, it
has also made it clear that there are a considerable number of deep and
interesting problems to be addressed — hence this proposal.
Our Research Agenda
We see three areas of work. To some extent these need to be pursued in
parallel, though we expect that emphasis will shift from one to the other
over the course of the project.
1. Conversation Support
First, it is necessary to support the initiation and conduct of conversations
via digital media. In particular, we want to support long-running, deep,
reflective conversations, which can be steered by their participants.
To do this it is important to represent the conversation as a first class
object (rather than as a collection of messages which the users must individually
manage), and to represent the participants in the conversation and their
activities with respect to it so that social information can help support
and shape the conversation (thus, it should be possible to see who is
listening, make inferences about the level of interest, and notice when
a crowd is gathering or dispersing). The Babble prototype has partially
explored this area. We see the need for much more work on the social proxy
concept: for example, we would like to design and test social proxies
that are not strictly synchronous (so that you could see, for example,
that ‘a crowd had gathered’ in a topic, even though individual
visits were spread out over several hours or days). We also want to design
social proxies for larger groups: our current social proxies can accommodate
about two dozen simultaneous participants.
 |
| Figure 2. A simple visualization
of a textual conversation |
2. Reuse of Conversation
Second, the knowledge embedded in conversations should be re-useable.
That is, we want to move from today’s state, where conversation is
of value primarily as it occurs, to a state in which conversation is a
useful work product that can be browsed, mined, and restructured at a
later time. Babble does not currently support such re-use, and the frustration
of dealing with a year’s worth of conversation riddled with useful
but hidden veins of information bears eloquent testimony to its value.
Thus we will develop tools for searching, navigating, and visualizing
conversations (note that conversations have considerable structure that
can be exploited for these purposes — e.g., ‘find all dialogs
between John and Amy that lasted more than a week and contain “ASSR”’.
It is also important to provide tools that permit participants to add
structure to conversations, summarizing, glossing, highlighting, linking,
and otherwise annotating them — this adds value for later browsers,
and creates ‘social landmarks’ (e.g., ‘Show me comments
linked to by more than 5 people’). With the help of a summer intern,
we have designed some visual (non-functional) prototypes of such tools
(e.g., Figure 2).
3. Knowledge Communities and Organizational Knowledge Spaces
Our experience with Babble suggests that knowledge creation, use, and
re-purposing will proceed most easily in an informal, semi-private environment
where knowledge workers feel ‘safe’ enough to venture tentative
interpretations and conjectures. But this is at odds with the goal of
making the knowledge visible to the larger organization. This is another
deep problem that we intend to tackle: how do we modulate between the
need for privacy and the value of visibility? Notice that neither privacy
nor visibility is inherently good or bad: each supports and inhibits certain
types of behavior (for example, the perceived validity of elections depends
crucially on keeping certain aspects very private, and others very visible).
Here, as with the earlier question of how to get people to engage in the
considerable effort that knowledge work takes, is where the translucent
aspect of STS comes into play: we need to understand how to strike the
appropriate balance between privacy and visibility. In this case, a search
engine might be used to reveal the locus and extent of a particular concept
being discussed within a knowledge community without revealing the particulars
of the discussions. Given clues that useful knowledge is present, interested
parties could request summaries of the topic, petition for admission to
the community, or simply converse with some of the community members.
Notice how this strategy blends technical and social mechanisms: technology
is used to locate hot spots, social mechanisms are used to control access.
Implicit in this description is the notion that organizations would have
many ‘local’ knowledge communities situated in a larger organizational
space from which the communities, and some of their knowledge and activities,
would be visible. This raises the issue of how to visualize a community
or organization and its activity.
Concluding Remarks
We believe that social information is crucial to any group activity, and
that the STS perspective does not just apply to knowledge management,
but to any domain — e-commerce, distance learning, workflow, workgroup
collaboration — in which groups or organizations use digital systems.
The work proposed here lays the groundwork for more fundamental inquiries,
such as how might social translucence be built into the digital medium
itself (rather than being designed in at the application level), and what
principles might be used to guide the design of socially translucent digital
systems.
Additional Information
Publications regarding Knowledge Communities as Socially Translucent Systems
are on the Social Computing Group publications
page.
The Social Computing Group gratefully acknowledges its Advisory
Board consisting of the following members:
Mark Ackerman, University of Michigan
Amy Bruckman, Georgia Institute of Technology
Judith Donath, MIT Media Lab
William Gaver, Royal College of Art, London
Judith S. Olson, University of Michigan
Steve Poltrock, The Boeing Corporation
|
Other
projects in
