Vance's
CALL resources page | esl_home index
Return to Papyrus News Archive Main Page
*******************************************************************
May 21, 2000: This message was distributed by Papyrus News, a free e-mail distribution list on the global impact of information technology on language, literacy, and education. Feel free to forward this message to others, preferably with this introduction. For information on Papyrus News, including how to (un)subscribe or access archives, see <http://www.lll.hawaii.edu/web/faculty/markw/papyrus-news.html>.
*******************************************************************
[This was distributed by Phil Agre's Red Rock Eater News Service (RRE).
For info about RRE, see <http://dlis.gseis.ucla.edu/people/pagre/rre.html>]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Date: Sat, 20 May 2000 11:22:20 -0700
From: Christine Borgman <cborgman@ucla.edu>
From Gutenberg to the Global Information Infrastructure: Access to
Information in the Networked World
Christine L. Borgman
MIT Press, March 2000
http://mitpress.mit.edu/book-home.tcl?isbn=026202473X
Table of Contents
1 The Premise and the Promise of a
Global Information Infrastructure
2 Is It Digital or Is It a Library?
Digital Libraries and Information Infrastructure
3 Access to Information
4 Books, Bytes, and Behavior
5 Why Are Digital Libraries Hard to
Use?
6 Making Digital Libraries Easier to
Use
7 Whither, or Wither, Libraries?
8 Acting Locally, Thinking Globally
9 Toward a Global Digital Library:
Progress and Prospects
The Premise and the Promise of a Global Information Infrastructure
Let us build a global community in which the people of neighboring countries
view each other not as potential enemies, but as potential partners, as members
of the same family in the vast, increasingly interconnected human family. -- Vice-President Al Gore (1994a)
The information society has the potential to improve the quality of life of
Europe's citizens, the efficiency of our social and economic organization and
to reinforce cohesion. -- Bangemann
Report (1994)
The premise of a global information infrastructure is that governments,
businesses, communities, and individuals can cooperate to link the world's
telecommunication and computer networks together into a vast constellation
capable of carrying digital and analog signals in support of every conceivable
information and communication application.
The promise is that this constellation of networks will promote an
information society that benefits all: peace, friendship, and cooperation
through improved interpersonal communications; empowerment through access to
information for education, business, and social good; more productive labor
through technology-enriched work environments; and stronger economies through
open competition in global markets.
The promise is exciting and the premise appears rational. Information technologies are advancing at a
rapid pace and becoming ever more ubiquitous.
Many scholars, policy makers, technologists, business people, and
pundits contend that changes wrought by these new technologies are
revolutionary and will result in profound transformations of society. Physical location will cease to matter.
More and more human activities in working, learning, conducting commerce,
and communicating will take place via information technologies. Online access to information resources will
provide a depth and breadth of resources never before possible. Most print publication will cease;
electronic publication and distribution will become the norm. Libraries, archives, museums, publishers, bookstores,
schools, universities, and other institutions that rely on artifacts in
physical form will be transformed radically or will cease to exist. Fundamental changes are predicted in the
relationships between these institutions, with authors less dependent on
publishers, information seekers less dependent on libraries, and universities
less dependent on traditional models of publication to evaluate scholarship. Networks
will grease the wheels of commerce, improve education, increase the amount of
interpersonal communication, provide unprecedented access to information
resources and to human expertise, and lead to greater economic equity.
In contrast, others argue that we are in the process of evolutionary, not revolutionary, social change toward an information-oriented society. People make social choices which lead to the development of desired technologies. Computer networks are continuations of earlier communication technologies such as the telegraph and telephone, radio and television, and similar devices that rely on networked infrastructures. All are dependent on institutions, and these evolve much more slowly than do technologies. Digital and digitized media are extensions of earlier media, and the institutions that manage them will adapt them to their practices as they have adapted many media before them. Electronic publishing will become ever more important, but only for certain materials that serve certain purposes. Print publishing will co-exist with other forms of distribution. Although relationships between institutions will evolve, publishers, libraries, and universities serve gatekeeping functions that will continue to be essential in the future. More activities will be conducted online, with the result that face-to-face relationships will become ever more valued and precious. Telecommuting, distance-independent learning, and electronic commerce will supplement, but not supplant, physical workplaces, classrooms, and shopping malls. Communication technologies often increase, rather than decrease, inequities, and we should be wary of the economic promises of a global information infrastructure.
Which of these scenarios is more likely to occur? Proponents of each offer historical precedent and argue
rationally for their cases. Many other
scenarios exist, some between those presented above and some at the far ends of
the spectrum. The extremes include
science-fiction- like scenarios in which technology controls all aspects of
daily life, resulting in a police state where every activity is monitored, and
survivalist scenarios in which some catastrophe destroys all technology, with
the result that new societies are reinvented without it. The science fiction and survivalist
scenarios are easily discounted because checks and balances are in place to
prevent them. Choosing between the revolutionary, discontinuity scenario and
the evolutionary, continuity scenario described above is more problematic. Each
has merit and each is the subject of scholarly inquiry and informed public
debate.
In view of the undisputed magnitude of some of these developments, it is
reasonable to speak of a new world emerging.
It is not reasonable, however, to conclude that these changes are absolute,
that they will affect all people equally, or that no prior practices or
institutions will carry over to a new world.
Nor is it reasonable to assume that any individual institutions, whether
libraries, archives, museums, universities, schools, governments, or
businesses, will survive unscathed and unchanged into the next millennium. Strong claims in either direction are
dangerous and misleading, as well as lacking in intellectual rigor. The arguments for these scenarios, the
underlying assumptions, and the evidence offered must be examined. Upon close examination, it will often be
found that strong claims about the effects of information technologies on
society, and vice versa, are based on simplistic assumptions about technology,
behavior, organizations, and economics.
None of these factors exists in a vacuum; they interact in complex and
often unpredictable ways.
I argue throughout this book that the most likely future scenario lies
somewhere between the discontinuity and continuity scenarios. Information
technology makes possible all sorts of new activities and new ways of doing old
activities. But people do not discard
all their old habits and practices with the advent of each new technology. Nor
are new technologies created without some expectations of how they will be
employed. The probable scenario is
neither revolution nor evolution, but co-evolution of information technology,
human behavior, and organizations.
People select and implement technologies that are available and that suit
their practices and goals. As they use
them, they adapt them to suit their needs, often in ways not anticipated by
their designers. Designers develop new
technologies on the basis of technological advances, marketing data, available
standards, human factors studies, and educated guesses about what will
sell. Products evolve in parallel with
the uses for which they are employed.
To use a simplistic aphorism: Technology pushes, while demand pulls.
The central concern of this book is access to information in a networked
world. Information access is among the
primary arguments for constructing a global information infrastructure. Information resources are essential for all
manner of human affairs, including commerce, education, research, participatory
democracy, government policy, and leisure activities. Access to information for all these purposes is at the center of
the discontinuity-continuity debates. Some argue that computer networks,
digital libraries, electronic publishing, and similar developments will lead to
radically different models of information access. The technologies of creation, distribution, and preservation will
undergo dramatic transformation, as will information institutions such as
libraries, archives, museums, schools, and universities. Relationships among these and other stakeholders,
including authors, readers, users, and publishers, will evolve as well. Others argue that stakeholders,
relationships, and practices are so firmly entrenched that structural changes
will be slow and incremental because most new technologies are variations on those
that came before. My view is that some
degree of truth exists in each of these statements. These and other arguments are examined throughout the book.
Much has been written about technology, human behavior, and policy regarding
access to information. Most of the
writing, however, focuses on one of these three aspects with little attention
to the other two. In this book I
endeavor to bring all three together, drawing on themes, theories, results, and
practices from multiple disciplines and perspectives to illustrate the complex
challenges that we face in creating a global information infrastructure. Technical issues in digital libraries and
information retrieval systems are addressed, but not in the depth provided in
recent books by Lesk (1997a) and Korfhage (1997). Nor are design issues addressed to the degree covered by Winograd
et al. (1996). Information-related behavior
in electronic environments is covered, but in less depth than in Marchionini
1995. Institutional and organizational
issues are treated more fully in Bishop and Star 1996, Bowker et al. 1996, and Sproull
and Kiesler 1991. Policy issues of the
Internet are addressed in more depth in Branscomb and Kahin 1995, Kahin and
Abbate 1995, and Kahin and Keller 1995.
In this book I draw on these and many other resources to weave a rich
discussion of access to information in a networked world. In view of the early stages of these
developments, more questions are raised than yet can be answered. My hope is to provoke informed discussion
between the many interested parties around the world.
Converging Tasks and Technologies
People use computer networks for a vast array of activities, such as
communicating with other individuals and groups, performing tasks requiring
remote resources, exchanging resources, and entertainment (whether with
interactive games or passive media such as videos). Among the few common
threads in predictions of future technology (see, e.g., Next 50 Years 1997 and
Pontin 1998) is that we will see more convergence of information and
communication technologies, blurring the lines between tasks and activities and
between work and play. We will have "ubiquitous computing" (Pontin
1998) and "pervasive information systems" (Birnbaum 1997). We will become "intimate with our
technology" (Hillis 1997), and "information overload" (Berghel 1997a)
will be more of a problem than ever. An
underlying theme of such predictions is "digital convergence", indicating
that more and more information products will be created in digital form or will
be digitized, allowing applications to be blended more easily. Digital technologies will co-exist with
analog and other forms of information technologies yet to be invented. Analog technology is based on continuous
flows, rather than the discrete bits of digital technology. Computer and communication networks are an
example of the bridge between these technologies. The word "modem" was coined from "modulate"
and "demodulate", which describe the device's function in converting
digital data produced by computers into analog signals that could be sent over
telephone lines designed for voice communication and vice versa. Predictions of ubiquitous computing are
based on an increasing reliance on small communication devices and embedded
systems such as those that control heating and lighting in homes and
offices. Future computer networks are
expected to link these devices just as they now link personal computers, data storage,
printers, and other peripherals (Pontin 1998).
Modes of Communication
No matter what technologies gird the framework of the global information
infrastructure, human activities involving the network will be intertwined. As the editors of Wired magazine (1997, p.
14) put it,
... broader and deeper new
interfaces for electronic media are
being born. ...
What they share are ways to move seamlessly
between media you steer
(interactive) and media that steer you
(passive). ...
These new interfaces work with existing media,
such as TV, yet they also work on
hyper-linked text. But most
important, they work on the
emerging universe of networked media
that are spreading across the
telecosm.
Despite the hyperbole, this quotation highlights a useful distinction between
"pull" technology (which requires explicit action by the user) and
"push" technology (which comes to the user without the user's explicit
action). Some activities are easily
categorized by this dichotomy, but others have characteristics of each. Composing and sending an email message and
searching a database require explicit "pull" actions, for
example. Although both the broadcast
mass media and the emerging media services that deliver tailored selections of content
to workstations during idle time can be classified as push technologies
(editors of Wired 1997), the latter form also could be considered
"pull", because the user presumably took action to subscribe to the
service. Similarly, if composing and
sending email is pull technology, then receiving mail can be viewed as a form of
"push". Opening and reading
messages requires explicit actions, but users can decide what to read, delete,
or ignore. They also can sort desirable
and undesirable messages by means of automatic filters. Because subscribing to
desirable content and filtering out undesirable content require parallel
actions, both can be viewed as forms of push technology if one accepts the
Wired definitions of "push" and "pull".
Push and pull combine in other ways as well. People subscribe to distribution lists, which then send messages
at regular or irregular intervals. They
also subscribe to services that alert them when new resources are posted on a
specific network site, but they must take explicit action to view or retrieve
the resources from that site. Truly
interactive forms of communication are difficult to categorize as push or
pull. People engage in conversations in
"chat rooms", play roles in MUDS and MOOS, and hold conferences,
meetings, and classes online in real time.
All require explicit actions, but the characteristics of these two-way
or multi-way conversations are far richer than the solo-action pull of
searching a database or sending a message.
Some of these are the "demassified" communication technologies
that Rogers (1986) predicted more, tailored to individual users and to small
audiences. However, the
"push" technologies of customized desktop news delivery touted by
Wired in 1997, in which messages continually scroll across the subscriber's
screen, have yet to become the commercial success that was predicted. Perhaps they were not sufficiently
customized or "demassified".
Perhaps people found them too disruptive, preferring "pull"
modes in which they could acquire the desired content at their convenience.
The intertwining of communication modes in electronic environments adds new dimensions to information access. Although more study has been devoted to "active" than to "passive" information seeking, even these categories are problematic in this new environment. These are but a few of many communication definitions and concepts being reconsidered in the light of new information technologies.
Task Independence and Task Dependence
The more intertwined tasks and activities become, the more difficult it
becomes to isolate any one task for study.
In the past, most theory and research presumed that the human activities
involved in access to information could be isolated sufficiently to be studied independently. This is particularly true of
information-seeking behavior, a process often viewed as beginning when a person
recognizes the need for information and ending when the person acquires some information
resources that address the need. Such a
narrow view of the process of seeking information simplifies the conduct of
research. For example, information seekers' activities can be studied from the
time they log onto an information retrieval system until they log off with
results in hand. The process can be
continued further by following subsequent activities to determine which
resources discovered online were used, how, and for what purposes. Another approach is to constrain the scope
of study to library-based information seeking.
People can be interviewed when they first enter a library building to
identify their needs as they understood them at that time. Researchers can follow users around the
building (with permission, of course), and can interview the users again before
departure to determine what they learned or accomplished.
Narrowly bounded studies such as these provide insights into detailed activities
and are useful for evaluating specific systems, services, and buildings. However, their value and validity are
declining for the purposes of studying the information environment of today and
assessing the needs of the future. In
the early days of information retrieval, people might reasonably conduct most
or all of their searching on one retrieval system. Only a few systems existed, and each had a limited number of
databases. These were complex systems requiring
lengthy training. Information seekers,
often with the assistance of skilled searchers, would devote considerable
effort to constructing, executing, and iterating a search on a single system (Borgman,
Moghdam, and Corbett 1984). A close
analysis of user-system interaction could provide a rich record of negotiating
a single search query. Even so, such
studies provide little insight into the circumstances from which the
information need arose or into the relationship between a particular system and
the use of other information resources.
In today's environment, most people have access to a vast array of online
resources via the Internet and online resources provided by libraries,
archives, universities, businesses, and other organizations with which they are
affiliated, as well as print and other hard-copy resources. They are much less dependent on any single
system or database. Rather, they are
grazing through a vast array of resources, perhaps "berry picking"
(Bates 1989) from multiple sources and systems. Studying any individual system is far less likely to provide a
comprehensive view of information-seeking activities than it was in the
past. Similarly, people have fewer
reasons to spend time in library buildings, now that they can use many library
resources from the convenience of home, office, dorm, coffee shop, or anywhere
else with network access. And they can
do so at hours of day or night when library buildings normally are closed. Thus, time spent in the library building may
be for narrower and more specific purposes, and may occur only at critical
stages in the search process. The use
of library buildings also reflects patterns that are influenced by age, generation,
culture, discipline of study, and many other factors. Such research should
yield insights into the design of future buildings and services, provided it is
set in a larger context of overall information-use patterns.
Future research on access to information must consider the complex relationships
between information-related activities and the context of work and leisure
practices in which these activities are conducted. Although all scholarship is
constrained by the necessity of studying that which can be studied, particular
caution is necessary when studying tasks that tend to be interdependent.
Technology Adoption and Adaptation
Underlying the design of any information technology are assumptions about
how and why people will use it. The
assumptions are sometimes explicit and sometimes only implicit, whether for
individual communication devices, for information systems, or for the design of
a global information infrastructure. In
identifying design criteria, and making implicit assumptions explicit, many
methods and perspectives can be applied.
We can evaluate which prior technologies were adopted and which were
not, the processes by which they were adopted, how similar technologies are
used, what features and functions are most popular and most effective, and how
their users adapt them to new purposes.
I will highlight three perspectives on assessing how and why people use
information technologies. Though many
other perspectives and methods exist, these three are applicable to our
concerns for access to information.
Adoption
Of the vast number of information technologies that are invented, only a few
make it to the marketplace, and of these, even fewer are successful. The quality of the product is only one
determinant of market success. Many
products that receive critical acclaim fail to garner large market shares. The Beta video recording technology and the
Macintosh computer are the best-known examples. In contrast, many products whose reviews range from skepticism to
scorn achieve great market success.
Business factors such as timing, marketing, and pricing are determinants
of success. Other determinants are
social factors involving how and why people choose to adopt any particular innovation. Rogers (1983, 1986) summarizes the results
of a large number of adoption studies using a five-stage model. The first stage of adoption is knowledge, or
becoming aware of the existence of a new technology that might be useful. This stage is influenced by factors such as
previous practices, felt needs or problems, tendencies toward being innovative,
and norms of the individual's social system. The second stage is persuasion,
which in turn is influenced by the perceived characteristics of the innovation,
how well it might work, how easy it is to try, and how easily the outcome can
be observed. In the third stage, the adopter makes a tentative decision to
accept or to reject the technology.
Acceptance may lead to implementation (fourth stage) and, if the
innovation is deemed sufficiently useful, to a confirmation to continue its use
(fifth stage). If the innovation is
rejected, the individual still may revisit the decision and adopt it later.
Electronic mail (email) provides an instructive example of the adoption
process. A person may first become
aware of its existence through news reports or through discussions with
friends, family, or co-workers. Someone
surrounded by email users will hear about it more quickly and frequently than
someone whose acquaintances are nonusers. Even today, elderly Americans who
have minimal contact with computer users may have at most a vague idea of what
email is, for example. In countries
with minimal telecommunications and computing penetration, only the elite may
be aware of email as a potentially useful technology. In the persuasion stage, a person who has many potential email
correspondents will find the technology more attractive than a person who knows
no one else with an email address.
Similarly, a person who already owns a computer with a modem will find
it far easier to try email than one who must acquire the technology and the skills
to use it. Once they have tried it,
some people will find email sufficiently useful, affordable, and worth the time
and effort to continue using it. Others
will not. Thus, once people become aware
of email, only some will consider trying it, a smaller number will make the
effort to try it; of these, only some will acquire it and continuing using it,
and they may abandon it later.
Conversely, some who rejected email at any of these adoption stages may
consider it again at some later time.
This adoption pattern also operates in the aggregate. The "early adopters" typically are
risk takers who are willing to try unproven techniques, often at great
expense. If they adopt the new
technology, their successes may convince more risk-averse individuals to try it. Conversely, if the early adopters reject it,
others may be more reluctant to try it.
By the time the low-risk late adopters decide to implement a technology,
the early adopters may have moved on to something yet newer and more innovative. Some technologies reach a critical mass of
adoption in a short period of time and are great market successes. Others are unable to find a match with early adopters fast enough, and the
entrepreneurs fail before finding their niche in the market. Others fail because they do not fill a
perceived need. Yet others succeed
because they are good enough, cheap enough, and at the right place at the right
time, although not necessarily an optimal design. Though this explanation is a gross simplification of the adoption
process, it illustrates a few of the many social variables that influence the
success of new information technologies.
Again, email provides a useful case example. Email filled a perceived need early in the development of
computer networks and reached a critical mass of computer users fairly
quickly. Spreadsheets were a similarly
attractive technology that contributed to the adoption of personal
computers. Early adopters of both
technologies were sophisticated computer users who tolerated complex user
interfaces, often unreliable software, and minimal functionality because the technology
was sufficiently valuable for their purposes.
People who are early adopters of one technology tend to be early
adopters of others, willing to tolerate immature technologies in return for
their benefits, and often enjoy the challenge of working at the "bleeding edge"
of technical frontiers.
Conversely, late adopters of one technology tend to be late adopters of
others. These people are far less
likely to appreciate technology for its own sake, preferring mature,
easy-to-use technologies with a high perceived payoff relative to the effort
required in learning to use them. They
are happy to let others "work the bugs out" before spending the time,
effort, and money to adopt them. This
distinction between the personality characteristics and social context of early
and late adopters is an important one to bear in mind when considering technologies
intended for a mass market. If a global
information infrastructure is to achieve wide acceptance, it must be attractive
to late adopters.
Adaptation
Theories of diffusion and adoption are valuable in understanding the social
processes involved in choosing to employ a particular technology. The "diffusion of innovations"
theory originated in rural sociology to explain farmers' choices of
agricultural innovations such as farming equipment, hybrid plants, pesticides, and
techniques for planting, harvesting, and storing crops. The theory was later extended to study the
adoption of a diverse array of innovations including solar energy during a
fossil-fuels shortage and family planning methods in developing countries. One weakness of applying the "diffusion
of innovations" theory to information technologies is the implicit
assumption that the innovation is relatively static. Information technologies
tend to be more dynamic and flexible than farming equipment, for example. Any communication device may be short-lived,
making it difficult to compare the actions of someone who adopted the first
crude implementation to those of someone who adopted a more sophisticated and
less expensive version only months later. Moreover, information technologies
are more malleable and adaptable to individual purposes than are most other
technologies. Thus, we must look not
just at the adoption of information technologies as a binary (adopt / not
adopt) decision, but also at how technologies, once adopted, are adapted over
time.
Books provide an early example of how people adapt information technologies
to their purposes. Manuscripts
(meaning, literally, hand-written) were the first form of written record. Manuscripts on sheepskin or parchment were
easier to create and read than chiseled stone tablets, but still could be read
only by one person in one place at a time.
Manuscripts could be loaned for manual copying, which enabled
duplication, however laborious.
Gutenberg's improvements in movable type in the fifteenth century made
multiple copies economically feasible for the first time. Early printed books retained the shape and
size of manuscripts, following the earlier technology. Although the distribution of multiple copies
meant that more people could own and read a work concurrently, books still were
too bulky for portable use, except by the very rich. Greenberg (1998) recounts the oft-told story of Abdul Kassem
Ismael, who was said to have had a library of 117,000 books in tenth-century
Persia. Not only did he carry his
library with him while he traveled, on the backs of 400 camels, he trained the
camels to walk in alphabetical order. Later innovations led to publishing books
in more portable sizes that fit not only in the saddlebags of yesteryear, but
in the backpacks and briefcases of today.
We find similar adaptations in the use of computer networks. The ARPANET, precursor to the Internet, was
created for remote access to scarce computing resources. Electronic mail was a feature intended to serve
as an ancillary communication function.
Email proved so useful for general communication that it became the
dominant use of the network, much to the surprise of the ARPANET's designers
(Licklider and Vezza 1978; Quarterman 1990).
Email was the "killer application" that attracted most people
to the Internet (Anderson et al. 1995; Quarterman 1990), and it remains the
most important reason for becoming an Internet user (Katz and Aspden 1997).
Email is a far different application today than it was in the early days of
the ARPANET, however. Early email
consisted of very short plain text messages.
Less than a decade ago, messages could take several days to arrive, with
delays caused whenever a server in a store-and-forward network went down. Email was neither fast enough, reliable
enough, nor functional enough to replace most other forms of communication. The technology advanced, as did users'
perceived needs for more capabilities and better services. Today's email supports long messages of
formatted text and is fast, reliable, convenient, and inexpensive (Berghel
1997b). Increasingly, email software allows
people to send and receive file attachments that preserve the integrity of
text, images, graphics, and sound. For
many purposes, email is a suitable substitute for telephone, fax, post, or
express mail.
Email now combines the features of word processors, file transfer (ftp), and
multimedia file management. It also
provides a bridge to the World Wide Web by embedding live links to web
sites. By including a URL (uniform
resource locator) address in an email message, a user can click on an address
to launch a browser application and link to the web site. And the reverse is true. Once at the web site, a user can click on
"email" and send a message to the web site.
Email has evolved from a simple application to one that combines a rich
array of services. As users realized
its value and its constraints, they identified further improvements that could
be made. Yet today's complex email technology has too much functionality to be feasible
for some purposes. Thus, we also find
evidence of complex applications being stripped down to the bare elements that
suit newly identified needs. An example
is the convergence of email with pocket pagers, which themselves were initially
a simple, single-function technology.
Some of today's more elaborate pagers include a full, albeit tiny,
QWERTY keyboard and alphanumeric display, on which people can send and receive
terse messages. Other pagers include
function keys for common responses to email-type messages: yes, no, time, date,
etc. Such devices can convey cryptic
but critical messages, such as "When do you arrive?" (answer:
"AA 75, 8:44pm LAX"), "Did we win the case?", "Running
late, resched Tu at 3?" (answer: "no. Tu 2pm ok?")", "pls get milk", or "get
KT @ school".
These are but a few examples of how people adapt information technologies by
using them. People sometimes adopt only
part of a technology, as illustrated by the example of stripped-down email. Other
times they disable or circumvent features of a technology. Email file
attachments are a case in point. They
are extremely useful for exchanging files quickly between team members, co-authors,
authors and editors, authors or publishers and readers, or teachers and students. But they are useful only when they
work. When exchange partners have
identical hardware and software platforms, fast connections, and (better yet)
the ability to scan for viruses before receipt, file exchange may be seamless.
System designers, along with as those who send file attachments, often are
unaware of the difficulties involved in receiving attachments intact and in a
usable form, however. Despite considerable
progress, the necessary platform independence and software independence
required for reliable exchange of attachments over networks has yet to be achieved. File exchanges between different platforms
(e.g., PC and Macintosh) and different operating systems (Windows 95, Windows
98, Windows NT, Macintosh OS 7.5, Macintosh OS 8.0, Unix, etc.) introduce compatibility
problems. Files created with widely
used word processing software such as Microsoft Word and Corel WordPerfect
often fail to transfer intact. Text may
transfer but formatting may be corrupted, and the likelihood of accurate
transfer decreases with the inclusion of software-specific features such as
tables, graphics, and macros. The more
recent the version of the software used to create a file, the less likely that
earlier versions of the same software or of competing software can open it
intact. Exchanging files of graphics or
sound is yet more problematic. Adding
another layer of concern is the ability of attachments to carry computer
viruses that can contaminate the receiver's computer.
Unsolicited file attachments containing job applications, advertisements,
jokes, cartoons, greeting cards, and myriad other materials clog network and
modem lines and fill disk space. Owing to
problems with technical compatibility, viruses, and bandwidth, many people are
making minimal use of file attachments, and some are setting their email
parameters to reject them entirely.
Local network managers are introducing delays in email delivery to scan
all attachments for viruses, adding another layer of complexity. Sending faxes, or mailing paper and disks,
can be faster, more reliable, and less labor intensive. The email examples offer several lessons in
the adoption and adaptation of information technologies. One lesson is that early adopters are
willing to use an immature technology.
As they use it, they will identify problems, recognize new
possibilities, and demand improvements.
Later adopters will identify yet more problems and more desirable
capabilities as they integrate it into their practices, refining the technology
further. Another lesson is that one
simple technology may spawn so many features that it subdivides into component
parts, as email has done. We also see
that advanced features that are extremely useful in some situations may result
in unintended and undesirable consequences in others, as is the present case
with file attachments. When people have
positive experiences with a technology, they often are more inclined to adopt
another technology. Conversely, when
they have negative experiences, they trust the technology less than before, and
are less inclined to try something new.
All these lessons argue for the importance of studying the use of
information technologies in actual working situations. Though laboratory
experiments are extremely valuable for improving technologies under ideal
conditions, field studies are essential to determine how technologies are
adopted and adapted.
Organizational Adaptation
Though some technology adoption and adaptation is attributable to individual
choices by individual users, much of it takes place in the context of
organizations. Organizations such as
businesses, governments, universities, and schools make decisions about what hardware,
software, and services to purchase for use by their constituencies. Individuals may have little choice in which
computing platform, Internet provider, or services they use. Organizations usually set policies about how
services such as email and information resources are used. Even in view of these constraints,
individuals often have considerable latitude in how they employ these
technologies in their work practices, however.
Sproull and Kiesler (1991) explain the unpredictable effects of introducing
technology into organizations from a "two-level perspective". They argue that most inventors and early
adopters of technology think primarily about efficiency of the technology. System
designers, as well as early adopters, focus on the instrumental uses to which
the technology is put, whether reducing "telephone tag" through the
use of electronic mail or lowering secretarial costs by replacing typing with
word processing. These are the
"first-level effects" of a technology.
Users rarely implement a new technology in precisely the way that designers
intend, however. Organizations find it
difficult to determine accurate estimates of direct costs, much less to
determine the first-level effects of technology on work practices,
productivity, or profits. Because technologies interact with routine work
practices and policies, implementation leads to "long-term changes in how
people work, treat one another, and structure their organizations"
(Sproull and Kiesler 1991, p. 1). It is
these "second-level effects" on the social system of interdependent
people, events, and behaviors that are most pervasive and most important for
organizations. These effects are also
the most difficult to predict.
Again, email offers illustrations of first- and second-level effects of
introducing an information technology into organizations. The instrumental uses of email are many: it
offers rapid interpersonal communication within the organization and between
the organization and the external world, whether clients, suppliers, members,
customers, citizens, colleagues, friends, or family. Email is convenient and portable. Because it is asynchronous, it can improve time management by
enabling people to send and receive messages at their convenience. It serves as
a broadcast technology, allowing an organization to deliver the same message to
a mass audience of its employees, students, or other groups
simultaneously. Email has radically increased
the speed and volume of communication for most people who use it.
We are finding many second-level effects of email that were not anticipated
at the time of its initial development or adoption. Email is easily abused,
whether by broadcasting messages that are of interest only to a few or by
sending rude and inappropriate messages that are unlikely to be communicated by
other means. Junk email can
proliferate, resulting in inefficient use of staff time to sort through it,
rather than the efficiency of communication intended. Once an organization
adopts email, usually everyone who is provided access is expected to use it
regularly. People are expected to respond
to messages, and to do so quickly. As a
result, memos and other communications that did not require a response in paper
form now result in a flurry of acknowledgments and responses, adding another layer
of communication activity.
Communications that once were oral, or confined to one or a few paper copies
that were controlled by the individuals involved, are now captured in permanent
form on an organization's email servers. As a result, organizations are faced
with a difficult balance between controlling their resources and the rights of
individuals to their privacy (Anderson et al. 1995; Berghel 1997b). Organizations that read employees' email may
defend this practice on the grounds that email is organizational documentation
and that it resides on computers owned by the organization. Individuals, particularly those who have lost
jobs over the content of email messages, may contend that email is the
equivalent of telephone or other oral communications and is subject to
reasonable expectations of privacy.
Conversely, organizations are learning that email can have unexpected and
adverse legal consequences. Conversations
that once were oral and now are recorded can be treated as legal evidence. Among the evidence that convicted Oliver
North in the Iran-Contra affair were email messages that he had deleted; they
were recovered from backup storage as part of the legal discovery process. Similarly, email messages internal to the
Microsoft Corporation are being used by the US government as evidence in an
antitrust case against the corporation. As a result of these and other cases,
many organizations are expanding the scope of their email policies to limit the
content of email messages and to minimize the archival storage of email
transactions (Harmon 1998).
These are only a few of many examples of the positive and negative effects
that email has had on organizational communication. (For more, see Anderson et
al. 1995; Berghel 1997b; Markus 1994.) People's experiences with email and
their perceptions of its role in an organization combine to determine how they
will adapt it to their own practices.
As information technologies are more widely adopted, concern about their
second-level effects is increasing.
These concerns cross many disciplines, levels of analysis, and research
methods. "Social informatics"
is an emerging research area that brings together the concerns of information,
computer, and social scientists with those in the domains of study (Bishop and
Star 1996; Borgman et al. 1996; Bowker et al. 1996). Social informatics scholars are attempting to build upon research
in the design and the use of information systems and upon social studies of
science and technology. This book
brings a social informatics perspective to bear on access to information in
digital libraries and in a global information infrastructure, considering
first-level effects when these are all that can be known and second-level
effects where possible.
Creating a Global Information Infrastructure
The integration, interaction, and interdependence of information- related
tasks and activities leads us to think in terms of an information
infrastructure. Rather than relying on
separate devices for producing text (e.g., typewriters and personal computers),
producing images (e.g., personal computers, photocopy machines, drawing pads),
communicating with individuals (e.g., telephones, telefacsimile (fax) machines,
mailboxes and stamps), and searching for information resources (e.g., personal
computers, local servers, print technologies), all these tasks can be
accomplished via a personal computer connected to the Internet. Conversely, these tasks can be divided up in
many new ways by means of specialized devices such as cell phones, pagers,
palmtops, and other "information appliances" that can share
information. Computer and communication
networks enable the integration of tasks and activities involved in creating,
seeking, and using information, increase the interaction between these
activities, and make them ever more interdependent.
In considering the premise and the promise of a "global information infrastructure",
we must determine what is meant by this phrase. Already it is used in variety
of contexts, with meanings that include a set of technologies, a set of
principles for an international computing and communications network, and a
loose aggregation of people, technology, and content.
What Is Infrastructure?
Terms such as "national information infrastructure" and
"global information infrastructure" are being bandied about with
minimal discussion of what is meant by "infrastructure". Social scientists and historians are
beginning to take a research interest in this concept, particularly as it
relates to organizational communication and work practices. Star and Ruhleder (1996, p. 111-112)
describe infrastructure as follows:
It is both engine and barrier for
change; both customizable
and rigid; both inside and outside
organizational practices.
It is product and process. ...
With the rise of decentralized
technologies used across wide
geographical distance, both the
need for common standards and the
need for situated, tailorable
and flexible technologies grow
stronger.
Star and Ruhleder are among the first to describe infrastructure as a social
and technical construct. Their eight
dimensions (ibid., p. 113) can be paraphrased as follows: An infrastructure is
embedded in other structures, social arrangements, and technologies. It is transparent, in that it invisibly
supports tasks. Its reach or scope may
be spatial or temporal, in that it reaches beyond a single event or a single
site of practice. Infrastructure is
learned as part of membership of an organization or group. It is linked with conventions of practice of
day-to-day work. Infrastructure is the
embodiment of standards, so that other tools and infrastructures can
interconnect in a standardized way. It
builds upon an installed base, inheriting both strengths and limitations from
that base. And infrastructure becomes visible
upon breakdown, in that we are most aware of it when it fails to work-when the
server is down, the electrical power grid fails, or the highway bridge
collapses.
As a means to explore the technical and public policy implications of
information infrastructure, the Corporation for National Research Initiatives
has sponsored a series of studies that address historical examples of
large-scale infrastructure. These
include studies of the growth of railroads, telephony and telegraphy,
electricity and light, and banking (Friedlander 1995a,b, 1996a,b). In each case, the technologies involved took
some time to be adopted, to stabilize, and to achieve the critical mass
necessary to form an infrastructure. Railroads, telephones, power companies,
and banks all provided local services for years, or even decades, before
reaching nationwide connectivity. Each
developed with some combination of public and private investment and government
regulation. The means by which an
integrated infrastructure evolved varied, and each involved experimentation
with different forms of technology, regulation, and social arrangements.
Models of infrastructure for railroads, telephones, energy, and banking
could have taken far different forms than they did. Indeed, with the possible
exception of railroads, each of these infrastructures is still evolving actively. Telephony underwent extensive restructuring
in the United States during the 1980s and the 1990s due to changes in
regulatory structure, mergers and acquisitions, and technological
advances. Similar regulatory restructuring
is now underway in Europe and elsewhere.
Meanwhile, technology advances and mergers and acquisitions continue
apace. On the energy front, models for
service provision are changing as energy companies are privatized and global
power relationships shift with variations in supplies and prices of fossil
fuels. On the financial front, models
for banking infrastructure are under scrutiny as markets for stocks,
commodities, currencies, and other financial instruments are becoming much more
tightly coupled.
Each of these infrastructures is deeply embedded in our social fabric, relies
on technical standards, and builds upon an installed base within the scope of
its own and other infrastructures. A
corollary to the notion that infrastructure becomes visible upon breakdown is
that we rarely are aware of it when it is functioning adequately. We often fail to recognize these as
essential infrastructures until telephone service becomes more complex and
expensive, energy services change in cost and character, or the stock market
takes a precipitous fall in value. And,
although Americans make minimal use of railroads, railroads are an essential
form of transportation in much of the world, where people are very much aware
of changes in schedules, routes, prices, and services.
Star and Ruhleder's (1996) set of eight infrastructure dimensions highlights
the complex interaction of technology, social and work practices, and
standards. They also emphasize social
context by noting that infrastructure builds upon an installed base. An
information infrastructure is built upon an installed base of telecommunications
lines, electrical power grids, and computing technology, as well as on
available information resources, organizational arrangements, and people's
practices in using all these aspects.
An installed base establishes a set of capabilities and a set of
constraints that influence future developments. For example, mobile telecommunications must interoperate with
land-based networks, and new computers should be able to read files that were
created on the preceding generation of technology.
The concepts of embeddedness, transparency, and visibility are especially
relevant to a discussion of a global information infrastructure. To be effective, a GII must be embedded in
the technical and social infrastructure of the many nations and cultures it
reaches-so much so that the infrastructure is invisible most of the time. Whether this degree of embeddedness is
possible across countries and cultures is examined throughout this book. When an information infrastructure works
well, people depend on it for critical work, education, and leisure tasks,
taking its reliability for granted.
When it breaks down (for example, when email cannot be sent or received,
when transferred files cannot be read, or when online information stores cannot
be reached), then the information infrastructure becomes very visible. People may resort to alternative means to
complete the task, if those means exist; they may create redundant systems at
considerable effort and expense; and they will trust the infrastructure a bit
less each time it breaks down.
Infrastructure as Public Policy
Infrastructures of many kinds are subject to public policy. For example, the Clinton administration
(1997, 1998) set forth a policy on "critical infrastructure
protection" that is noteworthy for our concerns. The white paper on Presidential Decision Directive 63 (Clinton
Administration 1998) defines "critical infrastructures" as
"those physical and cyber-based systems essential to the minimum operations
of the economy and government. They
include, but are not limited to, telecommunications, energy, banking and
finance, transportation, water systems, and emergency services, both governmental
and private". In the past, these
infrastructures were physically and functionally separate. However, with advances in information
technology these systems are increasingly linked and interdependent. The significance of this interdependence is
that critical systems are ever more vulnerable to "equipment failures, human
error, weather and other natural causes, and physical and cyber attacks". PDD 63 has the goal of protecting critical
infrastructure from intentional attack and minimizing service disruptions due
to any other form of failure.
Information technologies link these critical infrastructures, making them
interdependent, and thus all information technologies could be considered parts
of an information infrastructure.
Information infrastructure usually is more narrowly defined in public
policy documents, however. Typically
the scope includes computing and communications networks, associated
information resources, and perhaps a set of regulations and policies governing
use.
Metaphors for Information Infrastructure
Clever metaphors for information infrastructure have helped to capture public
attention. The concept of information
infrastructure is best known in common parlance as the "information
superhighway" (Gore 1994b), or sometimes as the "I-way" or the
"Infobahn". These metaphors for
information infrastructure emphasize the roads or pipes over which data flow,
whether telecommunications, broadcast, cable, or other channels. The highway metaphor captures only a narrow
sense of infrastructure, as it does not encompass information content, communication
processes, or the larger social, political, and economic context. The superhighway metaphor is misleading both
because it skews public understanding toward a low-level infrastructure and because
it suggests that the government would pay the direct costs of the highway's
construction. The Internet was
constructed with a combination of government and private funds. Current public policy, especially in the
United States, is oriented toward private funding for further expansion
(Branscomb and Kahin 1995; Kahin and Abbate 1995; Kahin and Keller 1995).
Though metaphors such as the information superhighway have been extremely
effective in marshalling support for information infrastructure development,
far more is involved than laying roads over which information will travel.
National and International Policies
Individual countries began plans for national information infrastructures in
the early 1990s (see, e.g., Information Infrastructure Program 1992; Karnitas
1996). In the United States, there was
the National Information Infrastructure Act of 1993. In Europe, there was the
European Union's proposal for a European Information Infrastructure (Bangemann
Report 1994). The installed base of
technology on which these plans are predicated includes the Internet, which
began in the late 1960s with the ARPANET (National Research Council 1994;
Quarterman 1990), the "intelligent network" of telecommunications
that followed the deregulation of telephony (Mansell 1993), and related
technologies such as cable and satellite television networks.
In the mid 1990s, national information infrastructure plans began to
converge. In 1994 the United States
proposed formal principles for a global information infrastructure. The following principles were incorporated
into the International Telecommunication Union's "Buenos Aires Declaration
on Global Telecommunication Development for the 21st Century" (1994) and
the United States' "Global Information Infrastructure: Agenda for
Cooperation" (Brown et al. 1995):
o encouraging private sector
investmento promoting open competition
o providing open access to the
network for all information providers and userso creating a flexible regulatory environment that can
keep pace with rapid technological and
market changeso ensuring universal service.
A few months later, the Group of Seven (seven leading industrialized nations,
known as "G-7") met to discuss these principles and agreed to collaborate
"to realize their common vision of the Global Information Society"
and to work cooperatively to construct a global information infrastructure (G-7
Ministerial Conference on the Information Society 1995a, pp. 1-2). These principles emerged from the 1995 G-7
meeting:
o promoting dynamic competitiono
encouraging private investment
o defining an adaptable regulatory frameworko providing open access to networks whileo promoting equality of opportunity to the citizen
o promoting diversity of content, including cultural and linguistic
diversity
o recognizing the necessity of
worldwide cooperation with particular attention to less developed countries.
The G-7 document also included the following.
These principles will apply to the
Global Information Infrastructure by means of:
o promotion of interconnectivity and interoperability
o developing global markets for
networks, services, and applicationso ensuring privacy and data security
o protecting intellectual property
rights
o cooperating in R&D and in the
development of new applications
o monitoring the social and
societal implications of the information society.
The Buenos Aires and G-7 statements have much in common: they are concerned
with technical capabilities ("interconnectivity", "interoperability",
"open access"), promises of rights to provide network services
("open competition", "dynamic competition"), guarantees of
network services ("universal service", "equality of opportunity"),
a means of funding network development ("encouraging private
investment"), and a means of regulating various aspects of its development
and use ("flexible regulatory environment", "adaptable regulatory
framework"). However, they vary on
their treatment of content: the G-7 principles promote diversity of content and
offer some general protections ("privacy", "data security",
"intellectual property"), while the telecommunications principles do
not mention content, addressing only the development and regulation of communication
channels.
Implementing Global Policy
Statements by the G-7 and other multinational bodies such as the United
Nations promote policy agendas of the countries involved, but they lack the
force of law and they provide little if any funding for implementation. Some of the language offers more platitudes
than policy, such as the claim in the European Information Infrastructure plan
that, "as a strategic creation for the whole Union", it will lead to
"a more caring European society with a significantly higher quality of
life" (Bangemann Report 1994).
The G-7 policy statements that frame a global information infrastructure
have raised considerable concern about human rights and social protections from
adverse consequences of its use. Though the G-7 principles include a general
statement about privacy and comment on the need to monitor the social
implications of the information society, they do not ensure legal protection of
rights such as privacy, free expression, and access to information. Despite requests by human rights groups, the
G-7 principles omit references to assurances in the United Nations Declaration
of Human Rights that were approved in 1948 (see United Nations 1998). Particularly relevant are articles 12 and
19:
Article 12: No one shall be
subjected to arbitrary interference with
his privacy, family, home or
correspondence, nor to attacks upon his
honor and reputation. Everyone has the right to the protection of
the law against such interference
or attacks.
Article 19: Everyone has the right
to freedom of opinion and
expression; this right includes
freedom to hold opinions without
interference and to seek, receive
and impart information and ideas
through any media and regardless of
frontiers.
These principles are receiving renewed attention upon the fiftieth anniversary
of their adoption (United Nations 1998).
Computer networks offer unanticipated capabilities for free speech and
access to information. Because
transactions and interactions are easily trackable, computer networks also can
create unanticipated intrusions into privacy (Kang 1998). Many privacy advocates promote an alternative
design model, known as "privacy-enhancing technologies" (Burkert
1997), in which individuals can acquire access to most information services
without revealing their identity if they so choose. Privacy, freedom of speech, and freedom of access to information
are tenets of democracy (Dervin 1994; Lievrouw 1994a,b). People cannot speak
freely or seek information freely if their movements are being tracked and if
they cannot protect and control data about themselves (Agre and Rotenberg 1997;
Diffie and Landau 1998; Information Freedom and Censorship 1988, 1991).
These are contentious issues in the United States. One example is that the federal policy on critical infrastructure
protection, discussed above, is being challenged on the basis of its potential
to erode civil liberties (Electronic Privacy Information Center 1998). Public
policy on social aspects of information infrastructure is subject to the laws,
the norms, and the practices of individual countries and jurisdictions, despite
the global reach of computer networks.
When local activities took place only locally, variances in policy and
regulation were less apparent and jurisdiction was rarely an issue. Now that individual communications and
information resources flow quickly and in vast quantities across borders, variances
in policy and regulation can be highly visible and jurisdiction can be highly
contentious. Privacy rights and regulations
have become an international battlefield where many of these issues are being
played out.
The European Union Data Directive, which took effect in late 1998, highlights
fundamental differences in policy approaches to privacy protection. The United States long has taken a
"sector approach", with specific laws governing credit reports,
library borrowing records, videotape rentals, federal government databases,
etc. In the new arena of computer
networks, US policy has favored self-regulation by the private sector over
government-imposed regulation. In contrast,
European countries have favored generalized policies over the control of
personal data, assigning stronger rights to individuals to control information
about themselves than to organizations that collect and manage personal
data. The EU Data Directive
consolidates the policies of individual countries and regulates privacy
protections throughout the European Union.
In view of the extensive commerce between the United States and the
European Union and the volumes of data about personnel, customers, clients, and
suppliers that are subject to regulation, the policies of these jurisdictions
often are in conflict.
For overviews of the rapidly evolving landscape of electronic privacy, see
Agre and Rotenberg 1997, Diffie and Landau 1998, Kang 1998, Rotenberg 1998, and
Schneier and Banisar 1997. Updates,
including pointers to government documents and other primary sources, can be found
at http://www.privacy.org and at http://www.epic.org. Information Infrastructure as a Technical Framework
"Information infrastructure" can refer to a technical framework
rather than to a public policy. As
defined by the (US) National Research Council (1994, p. 22), an information
infrastructure is "a framework in which communications networks support
higher-level services for human communication and access to information. Such an infrastructure has an architectural
aspect-a structure and design-that is manifested in standard interfaces and in
standard objects (voice, video, files, email, and so on) transmitted over the
interfaces".
One of the key components in defining an information infrastructure as a
technical framework is for it to have an open architecture that will enable all
parties to interconnect electronically and to exchange data. The "Open Data Network" concept
(National Research Council 1994) follows both from the Internet (a successful
open architecture for computing) and from established telecommunications policy
principles (Mansell 1993; National Research Council 1994). Under the G-7 principles, closed networks
can interconnect with the open network; closed service networks such as cable
television are allowed under other communications regulations as well. As we move toward ubiquitous computing, a
wider array of devices must interconnect; this makes open systems and
interoperability much more essential.
The emerging global network that interconnects a wide variety of computing
devices located around the world offers great utility for communication between
individuals and organizations, whether for education, work, leisure, or
commerce. The technical framework for
such an information infrastructure is now expected to support a range of tasks
and activities far wider than that for which it was originally designed,
however. The original ARPANET and the
early generations of the Internet were constructed by and for the research, development,
and education communities (Quarterman 1990).
Benign uses by a collegial community were presumed when its technical
architecture was designed (Oppliger 1997).
Substantial enhancements are being made to the technical architecture of the
Internet to support a vastly larger volume and variety of users, capabilities,
and services than was anticipated in the original design. Two new network services illustrate the
scope of the improvements that are under way (Lawton 1998; Lynch 1998). One is
"quality of service": the ability to reserve a set amount of bandwidth,
at a predetermined level of quality, in advance. Rather than the current model, which is largely "first come,
first served" for bandwidth usage, mostly at flat pricing, the new model
supports differential pricing for differential services. Many organizations are willing to pay a
premium to guarantee adequate bandwidth at a specified time (for a
teleconference or a distance-education course, for example). Conversely, many individuals are willing to
tolerate delays in email delivery or Web access in return for lower costs. In
view of the complexity of Internet architecture and the number of political and
service-provider boundaries crossed by an individual transmission, guaranteeing
quality of service will not be a simple accomplishment. Though quality of service is considered an
essential capability of an information infrastructure, precise assessments of
what can be guaranteed and how it can be measured have yet to be established
(Lynch 1998).
Multicasting is another long-awaited service improvement for the technical
framework of a global information infrastructure. At present, most communications are point-to-point
("unicasting"): copies of a message are sent individually to each
intended recipient. The alternative is
broadcasting, in which one message is sent to all users of the network, whether
they want it or not. An intermediate
model is "multicasting": one message is sent to a selected group of
recipients, reducing the amount of bandwidth required. Technically, under multicasting, the
originating server sends one message to each network router on which intended
recipients are located and that router re-sends to its local subscribers
(Lawton 1998). As with quality of service,
the number of providers involved makes multicasting a complex process, but one
that is necessary for efficient use of bandwidth on a global information
infrastructure (Lynch 1998). A variety
of economic and technical models for network service provision are under consideration
for the next generation of network architecture (Shapiro and Varian 1998).
The Internet is already a "network of networks". A global information infrastructure will be
even more so. Though we speak
metaphorically of a single open network, in actuality the Internet links many
layers of networks within organizations, within local geographic areas, within
countries, and within larger geographical regions. These go by various names, including intranets, extranets,
local-area networks (LANs), metropolitan-area networks (MANs), and even
tiny-area networks (TANs). Suffice it
to say that the information infrastructure topography is becoming increasingly
complex, linking together internal organizational networks, closed networks
such as cable TV, and the international Internet.
The boundaries of individual networks can be controlled to varying degrees. A common technique is to protect
organizational or even national networks with "firewalls" that limit
the abilities of authorized users to exit and of outsiders to enter. Some internal resources can be publicly
accessible while others are restricted to internal use, for example. Similarly, firewalls and filtering techniques
can be used to limit external sites that can be reached. Parents can limit
their children's ability to connect to sites known to contain pornography or
other undesirable material. The
definition of "undesirable" varies by context. Companies can limit access to known sites
containing games. Countries can limit
access to sites known to provide undesirable political views. China, for example, currently attempts to
control access to sites outside the country through a single gateway, so that
specific sites deemed objectionable can be blocked. Chinese Internet users are required to register with the police
to gain access to the network (Tan, Mueller, and Foster 1997). A key phrase here is "known
sites". As the Internet proliferates,
new sites appear daily, and sites change names, location, and content
frequently. Reliable filtering software
that can distinguish between acceptable and unacceptable materials is not yet
feasible, and may never be.
For most businesses and governments, security and risk management are far
greater concerns than is pornography.
After connectivity, the most important enabling technology for
electronic commerce is security (Dam and Lin 1996; Geer 1998; Oppliger
1997). One model being studied and
implemented is "trust management", in which mechanisms such as cryptography
are employed to verify the identities of all parties involved in electronic
transactions. Such transactions include
buying and selling goods or services, transferring secure data (such as
financial transactions between banks and stock markets), and proprietary
communications within organizations or between organizations and their clients,
customers, and suppliers. Both retail
transactions between individuals and companies and wholesale transactions
between companies can be accommodated.
An alternative model is "risk management", which focuses on
the likelihood of losses and the size of potential losses from electronic
commerce. Rather than assume that trust
can be guaranteed in all transactions, parties try to determine the degree of
risk exposure and to insure against it. Cryptography is essential to both
models as a means of assuring the authenticity of transactions to the extent
possible. The frontiers of electronic
commerce are being tested in the financial markets today. In view of the size and volume of
transactions among banks, stock markets, investors, and other parties, many
technical and policy aspects of information infrastructure are likely to be
tested first in this arena (Geer 1998).
Information Infrastructure as Technology, People, and Content
Among the broadest conceptualizations of an information infrastructure is
that presented in National Information Infrastructure: Agenda for Action 1993,
where an NII is defined as encompassing a nation's networks, computers,
software, information resources, developers, and producers. This definition comes closer to capturing
the larger sense of infrastructure as a complex set of interactions between
people and technology than do most other public policy statements, technical definitions,
or metaphors.
The above definition is compelling, if vague, because it recognizes that we
are creating something new and something that is more than a sum of its
parts. The information infrastructure
is not a substitute for telephone, broadcast, or cable networks, for computer
systems, for libraries, archives, or museums, for schools and universities, for
banks, or for governments. Rather, it
is a new entity that incorporates and supplements all these technologies and
institutions but is not likely to replace any of them. However, a GII is likely to change each of
these institutions, and how people use them, in profound ways.
The term "global information infrastructure" is used in this broad
sense throughout the present book. A
GII consists of a technical framework of computing and communications
technologies, information content, services, and people, all of which interact
in complex and often unpredictable ways.
No single entity owns, manages, or controls the technical framework of a
GII, although many governments, vast numbers of public and private
organizations, and millions of people contribute to it and use it. The GII is perhaps best understood by the
metaphor of the elephant being examined by a group of blind people-each one
touches a different part of the beast, and thus senses a different entity. From this perspective, a global information infrastructure
is a means for access to information.
However, it can be viewed from many complementary perspectives that also
are valid.
Summary
These are exciting times.
Information technologies are increasing in speed, power, and
sophistication, and they now can link together a vast array of devices into a
network that spans the globe. They
offer new ways of learning, working, and playing, as well as conducting global
commerce. Some contend that these
changes are revolutionary and will change the world; others argue that the
changes are evolutionary, and that individuals and organizations will
incorporate networked information technologies into their practices just as
they incorporated many earlier media and technologies. In this book I take the view that these
changes are neither revolutionary nor evolutionary but somewhere between: that
they are co-evolutionary. New technologies
are based on perceived needs and available capabilities. People adopt these new
technologies if and when they deem the technologies useful and when they deem
the effort and the costs appropriate.
Sometimes individuals make these decisions; sometimes organizations make
them. The result is that some
technologies are adopted by some of the people some of the time. No matter how voluntary or involuntary the
adoption process, individuals and organizations adapt technologies to their
interests and practices, often in ways not anticipated by the designers of
those technologies. Information technologies are more flexible and malleable to
individual practices than are most other innovations, and this makes them especially
adaptable. They also evolve more
quickly than most other innovations, with new and improved versions appearing
at a dizzying rate.
Adoption and adaptation of technology are difficult to predict, owing to the
complex interactions between characteristics of information technologies,
practices of individuals and organizations, economics, public policy, local
cultures, and a host of other factors. Organizations acquiring new technologies
find that estimates of first-level effects, such as those on productivity and
profits, are unreliable. Reliable
predictions of longer-term, second-level effects, such as those on
organizational communication and structure, are nearly impossible. One reason is that external factors, such as
changes in the legal status of electronic communications, can have profound
effects on how individuals and organizations use information technologies.
We are in the process of creating a global information infrastructure that
will interconnect computer networks and various forms of information technologies
around the world. After a review of
some of the many meanings of "information infrastructure", it was
determined that the concept incorporates people, technology, and content and the
interactions between them. This broad
definition incorporates definitions of information infrastructure as a set of
public policies and as a technical framework.
The broader definition is best suited to studying the co-evolution of
technology and behavior as related to access to information, which is the
primary concern of this book. An information infrastructure is only one of
several infrastructures that are essential to a well-functioning society. Others include energy, transportation,
telecommunications, banking and finance, transportation, water systems, and
emergency services. Because each of
these infrastructures is increasingly reliant on information technologies, they
are more interconnected and interdependent.
Their interdependence means that more and more aspects of daily life
depend on the emerging global information infrastructure.
end
Use your browser's BACK button to return to a previous page
For comments, suggestions, or further information on this site, contact Vance Stevens, webmaster. Regarding content of Papyrus-News, contact Mark Warschauer.