Parallel Program
The Global Electronic Community:
Modeling the Sociology of Self-Organization
Joan Roca
Memorial Library
Mankato State University
Mankato, MN 56002-8400, USA
Bill Ammentorp
University of Minnesota
Minneapolis, MN USA
Tom Morgan
Augsburg College
Minneapolis, MN USA
Abstract
The revolution in telecommunications of the past decade has brought about fundamental
changes in the patterns of professional communication. Internet and other electronic
networks make it possible for geographically-dispersed individuals to engage in meaningful
dialog concerning common problems. These exchanges result in a Global Electronic
Community which is unique in human experience. This is a community that is truly self-
organizing in that it can take on new forms momentarily. Consequently, it is a dynamic
community with continually-changing membership and structure; a society populated with
both human and electronic agents. It is, as a result, more complex in the roles its members
can assume. The sociology of this community is both novel and open to inquiry in depth -
in ways never before imagined by students of human society.
This paper explores the sociology of the Internet as a self-organizing system. The
Internet is represented as a “electronic landscape" where users and their agents seek
collections of relevant information. The "landscape" is defined by the accessibility,
connectedness and relevance of data files. As users navigate across this "landscape", they
add both connectedness and relevance. This creates collections of useful information and
welds users into Electronic Communities held together by shared models and paradigms.
The authors present a System Dynamics model of the prototypic Global Electronic
Community. Using data drawn from Internet file servers, the authors identify the model
and show how its structure and behavior conform to the principles of self-organization.
Simulation results are used to outline the sociology of such Communities and the
implications for professional contributions to sustainable development.
Internet: An Information Landscape
As the Internet grows in size and complexity, it is apparent that users experience an ever-
greater difficulty in finding relevant information. This is a particularly vexing problem for
the inexperienced user. Consider the following observation resulting from an extensive
review of Internet resources.
806
System Dynamics '95 — Volume II
As the number of resources grow, it is increasingly difficult for a user to identify and
locate potentially relevant resources to satisfy a need for information in a specific area,
particularly if the user is not a specialist in that area. (Lynch and Preston 1990, 282).
The challenge posed by the explosive growth of networked information utilities is one of
organization. As Gregorian (1993) states,
We must rise above the obsession with quantity of information and speed of
transmission, and recognize that the key issue for us is to organize this information
once it has been amassed - to assimilate it, find meaning in it, and assure its survival for
use by generations to come (5).
To date, networked information systems have been largely driven by
telecommunications and computing technology. Ever larger data bases have been made
accessible on a global scale and endless links have been created to enable users to "surf" the
sea of information. However, there is no theoretical foundation which describes how users
are navigating this sea - nor are there general principles whereby relevant information might
be made readily accessible (Brett 1993).
The Infoscape Model:
The authors have attempted to address this shortcoming by proposing (Ammentorp, et. al.
1994) and testing (Roca, 1995) a model of the information landscape (Infoscape) which
locates data bases in a three dimensional space (Figure 1).
Figure 1. The Infoscape Model.
Accessibility
Connectedness
807
Parallel Program
Each dimension of the Infoscape describes a critical feature of electronic information.
Accessibility refers to the extent to which the "common speech" of users can lead them
to media of interest. Connectedness refers to links among media which enable users to
collect information from a variety of sources. Relevance is a measure of the value of
information to a community of users. As these measures are applied to a particular set of
electronic information, the result is a grouping of media in "islands" on the landscape.
Two of these dimensions, accessibility and connectedness, are drawn from
bibliometric studies of scholarly communications. Accessibility is measured by the link
between materials in a particular database and the keywords of the thesaurus used as an
index. This is an instance of the "common speech" of a particular field or profession in that
those who assign keywords to materials have a similar understanding of the thesaurus, as
do those who use it to access the database. Connectedness is determined by the extent to
which data bases "point" to one another - thereby assisting the user in navigating across the
landscape.
Roca (1995) has attempted to assess the construct validity of this model by positioning
library information servers in a three-dimensional space. Accessibility of a given server
was measured by the number of servers that pointed to it. Connectedness was defined
as the number of outgoing pointers emanating from a particular server. Relevance was
determined by the total number of user accesses in a given time period. When the 92
servers in this study are plotted on these dimensions, we find a landscape that is
remarkably like that proposed in the Infoscape theory (Figure 2).
Figure 2. Library Gophers on Infoscape.
000
, 000
000
1000
1000
000
000
Relevance (# of Accesses)
000
1000
R08
System Dynamics '95 — Volume II
The main feature of this figure is the extent to which highly accessible, connected
servers also have high relevance. It is also clear that the converse holds such that servers
with low relevance are only marginally integrated into the information network.
These findings suggest a model where the interactions among the three dimensions are
detailed (Figure 3).
Figure 3. Dynamic Infoscape Model
paradigm
consensus
connectedness
relevance
citations conchg
relchg
thesaurus "
power Seog contributions,” conversation
users
usrpool
accessibility
addchg
network attraction reject
In this model, the key Infoscape variables relevance, connectedness, and
accessibility, are shown as 'boxes'. We have added two additional 'boxes' to capture
the effect of users and the usrpool on model dynamics. Values in the 'boxes' may
increase or decrease through the variables relchg, conchg, addchg, adopt and reject which
act on the flow ‘pipes' connected to the ‘boxes’. The seven ‘circles’, network attraction,
citation, access, paradigm consensus, thesaurus power, conversation, and contributions
are "links" which tie the major factors into a coherent model. For example, paradigm
consensus is a function of contributions, citations and users. In turn, paradigm consensus
causes changes in relevance. At the same time, through access and contributions users
contribute to the connectedness of the data base and, ultimately, to paradigm consensus.
This is the fundamental dynamic of the model in that it uses the prevailing patterns of
professional thought to organize the Infoscape.
The system has the potential to be self-organizing due to the feedback loops shown
above. As users increase paradigm consensus through citations and contributions, their
numbers increase due to feedback of effect from relevance through the network attraction
variable. There is also a key feedback loop which strengthens accessibility as the result
of changes in relevance. This is "fed back" to further increase ligm conse! as
accessibility improves access.
809
Parallel Program
The system is open to outside influence through the introduction of new users. They
enter the system from the usrpool at the lower right-hand side of the above schematic.
Their access and contributions add an increment of effect into the feedback and, over time,
the system can increase in scope of use and users can be drawn into the ruling paradigm.
The model also shows that users can reject the network and return to the usrpool.
Model Foundation:
The Infoscape Model grows out of two major research traditions: informetrics and
innovation studies. Each of the variables in the Model is defined using conventions
developed in these lines of work and links among variables are derived from selected
research studies as follows.
Accessibility: In distributed information systems, access is determined by the
organization of data bases and the key words or terms used in indices. It is especially
critical in electronic information networks due to the "location independence" of resources
(Tomer 1992). [Measure: Number of nodes in the information network.]
Connectedness: Information resources do not stand alone in cyberspace. They are
connected through information managing utilities and pointers. These pointers provide a
road map for navigating the network and for finding distributed information (Kappe, et.al.
1993). [Measure: Number of links among network nodes.]
Relevance: As information becomes accessible and connected, it merges into clusters
where new ideas build on existing theory and research. Such clustering or "mapping"
increases the value of information and its relevance to users (McCain 1990). [Measure: A
product of paradigm consensus and connectedness.]
Users: Professional and scholarly groups are the reference points for the study of the
sociology of electronic communities. They are the "invisible colleges" where roles are
taken and symbolic social behavior is carried out (Crane 1972). [Measure: Actual count of
network members.]
Usrpool: The professional group under study is defined by an area of practice and/or
scholarship. (American Physical Society 1991) [Measure: Membership in a professional
society or other independent assessment of potential use.]
Access: Users are drawn into electronic communities by the perceived value of
information. As the currency of information increases in the society, the probability of
access in a given user community increases (Drucker 1993). [Measure: Accessibility /
users.]
Contributions: Electronic communities can be passively explored, or they can be
actively engaged. This variable is a measure of the probability that users will offer to
include their opinions or products in network data bases (Dordick and Wang 1993).
[Measure: Users * probability of contribution.]
Citations: Contributions do not stand alone; they are linked to other resources in the
data base through citations. Thus, citation is a measure of the extent to which the data base
is organized (Egghe and Rousseau 1990). [Measure: Contributions * Connectedness *
probability of citation.]
Conversation: By participating in informal exchanges, users clarify contributions and
relate them to their personal needs and positions (Mulkay 1991). [Measure: Users *
probability of Eistserve participation]
The orientation of a user to the electronic community is shaped
by the ruling poradiem of his/her field. As more users are drawn to the paradigm, it
becomes a more powerful organizer of information (Sterman 4289). [Measure: Number of
co-citations = citations * probability of co-citations.]
R10
System Dynamics '95 — Volume II
Thesaurus Power: The sway of the paradigm helps to organize information via the
language of the profession. This, in turn, defines the keywords and terms of the indexing
thesaurus (Sutton and Davis 1992). [Measure: Paradigm consensus * keyword generation
rate.]
Network Attraction: The degree to which relevant information brings new users into
the network. This is a function of current levels of relevance plus an effect due to the
number of users already in the network (Rogers 1974). [Measure: Relevance *
attractiveness multiplier.]
The Sociology of Infoscape:
The behavior of users of information networks takes shape in the content and direction of
communication. Although the initial steps toward network involvement may be based on
an individual's membership in a non-electronic group, his/her involvement in electronic
communications is based on the benefits of such exchanges (Blau 1964). Interaction in the
Infoscape is governed by three principles which are, in turn, representative of feedback
loops in the Infoscape Model.
1) User participation in relevant exchanges leads to substantive contributions linked to
other users and resources. These are, at first, one-way with the user as receiver. Later,
they become two-way exchanges among users. And, finally, they are multi-way as the
user contributes to the data base. This is a modern (and simplified) version of the
processes described by Merton (1973).
2) Paradigm consensus develops through shared information and results in a “common
speech" in general use among users.
3) User populations will grow in size according to logistic innovation patterns in
information-relevant environments.
Propositions 2 and 3 make it possible for the model to be self-organizing as users build
relevance through paradigm consensus and, ultimately, draw new users into the network
(Prigogine 1993).
References:
American Physical Society. 1991. "Report of the APS Task Force on Electronic
Information Systems," Bulletin of the American Physical Society, 36 (4): 1119-1151.
Ammentorp, W., Morgan, T. and Michels, D. 1994. The Electronic Community Paper
presented at the annual meeting of the University Council for Educational Administration.
Philadelphia, PA.
Blau, P. 1964. Exchange and Power in Social Life. New York: J. Wiley.
Brett, G. 1992. "Networked Information Retrieval Tools in the Academic Environment:
Towards a Cybernetic Library,” Internet Research, 3(3) 26-36.
Crane, D. 1972. Invisible Colleges: Diffusion of Knowledge in Intellectual Space.
Chicago: University of Chicago Press.
Dordick, H. and Wang, G. 1993. The Information Society. Newbury Park, CA: Sage.
Drucker, P. 1993. "The Rise of the Knowledge Society," The Wilson Quarterly, 17 (2):
52-71.
811
Parallel Program
Egghe, L. and Rousseau, R. 1990. Introduction to Informetrics. New York: Elsevier.
Kappe, F., Pani, G. and Schnabel, F. 1993. "The Architecture of a Massively Distributed
Hypermedia System," Internet Research, 3(1): 10-24.
Kuhn, T. 1973. The Structure of Scientific Revolutions. Chicago: University of Chicago
Press.
Lynch, C. and Preston, C. 1990. “Internet Access to Information Resources," in M.
Williams (Ed.), Annual Review of Information Science and Technology, 25, 263-312.
McCain, K. 1990. "Mapping Authors in Intellectual Space," Journal of the American
Society for Information Science, 41(6): 433-443.
Merton, R. 1973. The Sociology of Science. Chicago, IL: University of Chicago Press.
Mulkay, M. 1991. Sociology of Science. Bloomington, IND: University of Indiana Press.
Prigogine, I. 1993. Time, Dynamics and Chaos. In J. Holte (Ed.) Chaos: The New
Science. St. Peter, MN: Gustavus Adolphus College: Lanham, MD: University Press of
America.
Roca, J. 1995. Road Signs in Infoscape: A Study of the Links Among Academic Library
Gophers. Unpublished doctoral dissertation, University of Minnesota, Minneapolis, MN.
Schultz, T. 1992. "A World Physics Information System," Serials Review, 18
(Spring/Summer): 45-48.
Shannon, C. and Weaver, W. 1949. The Mathematical Theory of Communication.
Champaign-Urbana, IL: University of Illinois Press.
Simsek, H. and Ammentorp, W. 1993. The Paradigm Shifts of the 1990's. Paper
presented at American Vocational Association annual meeting. Nashville, TN.
Sterman, J. 1985. "The Growth of Knowledge: Testing a Theory of Scientific
Revolutions," Technological Forecasting and Social Change, 28:93-122.
Sutton, B. and Davis, C. (Eds.) Networks, Open Access and Virtual Libraries. Graduate
School of Library and Information Science, University of Illinois at Champaign-Urbana.
Tomer, C. 1992. "Emerging Electronic Library Services and the Idea of Location
Independence," Journal of Computing in Higher Education, 4(1): 88-121.
812
