The Information Society

An International Journal

ISSN: 0197-2243 (Print) 1087-6537 (Online) Journal homepage: www.tandfonline.com/journals/utis20

Looking Backward and Forward at the Internet

Jon Guice

To cite this article: Jon Guice (1998) Looking Backward and Forward at the Internet, The
Information Society, 14:3, 201-211, DOI: 10.1080/019722498128827
To link to this article: https://doi.org/10.1080/019722498128827

Published online: 29 Jul 2006.

Submit your article to this journal

Article views: 278

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=utis20
Looking Backward and Forward at the Internet

Jon Guice
NASA Ames Research Center, Moffett Field, California, USA

understanding of future social choices. Looking behind us

This article reviews accounts of the origins and growth of the more intelligently will make it easier to see paths ahead.
global Internet with an interest in their implications for the fu- Today’s Internet is often linked in a direct line of descent
ture. Many accounts take overly restricted views of technical de- to plans among engineers and managers working in the
velopment. For instance, the Internet is often seen simply as the 1960s and 1970s with the U.S. Defense Department. The
outgrowth of U.S. Defense Department work and popular enthusi- driving force behind the Internet’s growth is supposed to
asm. Recent interdisciplinary studies of technology present a more be popular enthusiasm.
complex picture of how innovations are developed. They high- Recent interdisciplinary studies of technology present
light technical alternatives,contributions of diverse groups, variety a more complex picture of how innovations are devel-
in meanings of technologies, and the overall surprising charac-
oped and distributed. They highlight the existence of tech-
ter of technology development. They suggest that the Internet has
nical alternatives, including less-than-optimal technical
multiple origins and numerous particular reasons for its spread.
Current discussions of the entry of Internet technologies into con- solutions; the contributions of many different (and often
sumers’ homes and the convergence among computing, media, and con¯ icting) groups; the possibilityof a variety of meanings
telecommunications enterprises represent examples of more com- associated with technologies, and in general the contingent
plex accounts of technology development. Such accounts will likely and often surprising character of technology development.
provide more powerful bases for policy, management, and design. The value of these perspectives is to more accurately por-
tray the complexity of technical development. There may
also be implications for policy debate, technology design,
Keywords communications infrastructure, computer networks, his-
and management strategy.
tory of technology, Internet, social studies of technology,
technology forecasting
A case in point is the convergence of media, comput-
ers, and telecom, of which the Internet is (and I would
argue, always has been) an important aspect. This his-
torical process, which is widely recognized as at once
In ª Internet time,º as in ª dog years,º eons have passed technical, industrial, and political, has been and will con-
since the ® rst messages were sent between computers. tinue to be ® ercely multiparty and multifaceted. There
The Internet now has its own history, though most of the are a variety of technical alternatives being fought over.
events took place just years ago and portentouschanges are Different interests represent different goals for commu-
taking place today. In this article, I want to suggest that nications technologies as well, including ideals of pri-
the way we understand this history is important to our vate pro® t and universal access. Will the future of the
Internet be like its past, like today’s telephone, like in-
teractive television, or like something altogether differ-
Received 13 March 1996; accepted 8 April 1998. ent? It is generally understood that it is dif® cult to
A progenitor of this article was presented in a special session on
predict outcomes for the Internet because there are com-
the history of the Internet at the annual meeting of the Society for
Social Studies of Science, October 1995. I thank all of the participants
plicated relationships among these technologies, markets,
for their contributions and inspiration: Janet Abbate, Geoff Bowker, and industries.
Judy O’ Neill, and Juan Rogers. I also thank Rob Kling, Franois Bar, I have divided the body of this article into two main
and an anonymous reviewer for their comments. sections. I begin with some of the most widely circulating
Address correspondence to Jon Guice, 1365 Bay Laurel Drive, stories and points of view about the Internet. Then I
Menlo Park, CA 94025-5804, USA. E-mail: [email protected]. describe contrasting perspectives taken by recent studies
The Information Society, 14:201±211, 1998
Copyright ° c 1998 Taylor & Francis
0197-2243/98 $12.00 + .00 201
202 J. GUICE

of technology, which I illustrate with a different set of A recent journalistic knock-off, for instance, is described
examples. on the jacket as ª the story of the small group of researchers
and engineers whose invention, daring in its day, became
the foundation for the Internet. With ARPA’ s backing
STANDARD PERSPECTIVES [they] began the quest for a way to connect computers
Many people have heard a little about the origins of the across the countryº (Hafner & Lyon, 1996).
Internet. There are historical tales available on the Inter- The second main feature of the architects’ accounts is
net itself. There have been popular newspaper and maga- that they point out and document key engineering deci-
zine articles that feature retrospective accounts (e.g., sions in the development of the technology. They iden-
Kantrowitz & Rogers, 1994; Kleiner, 1994), practical tify technical accomplishments that appear signi® cant in
handbooks that include sections on the internet’s history hindsight and describe how they were developed. Thus,
(e.g., Evanson, 1994; Krol, 1992), and technical reports for example, the immediate result of ARPA’ s funding, the
that discuss aspects of the topic (e.g., Comer, 1991; FTP ARPANET, has been described as one of the ® rst and
Software, 1991; Nicoll, 1994; Perry et al., 1987; Roberts, largest wide-area networks (networks linking computers
1971; Schultz, 1988; Siewiorek et al., 1982). The general over long distances), the founding location of the TCP/IP
accounts usually run like this: protocol suite, and the ® rst major network to use ª packet
switchingº (van Atta et al., 1990, pp. 20-3, 20-4, 20-13).
The Internet had its origin in a U.S. Department of De-
fense program called ARPANET..., established in 1969 to
A similar approach is taken in a book entitled Architects
provide a secure and survivable communications network of the Web: 1,000 Days That Built the Future of Busi-
for organizations engaged in defense-related research. Re- ness (Reid, 1997). But the greatest amount of technical
searchers and academics in other ® elds began to make use detail is presented on the Internet itself, in various plan-
of the network, and at length the National Science Founda- ning and discussion documents. Accounts such as these
tion (NSF), which had created a similar and parallel network from the perspectives of researchers, engineers, and
called NSFNET, took over much of the TCP/IP technology managers provide extensive detail on the technical aspects
from ARPANET and established a distributed network of of innovations.
networks capable of handling far greater traf® c. (Brittanica
Online, 1994)
This sort of overview assimilates arguments from two dis- Users
tinct bodies of earlier, more detailed historical accounts, Other stories about the history of the Internet take the point
each of which re¯ ects different perspectives on the devel- of view of users, by which is meant individuals who em-
opment of the Internet. ploy and adapt Internet services such as e-mail or ® le trans-
fer. These are accounts by and about enthusiasts. In these
stories, it was the users who drove the development of the
Architects
Internet, not only contributing traf® c for the lines, but ac-
One point of view is that of engineering and managerial tually stimulating new applications and new directions of
architects of the Internet. Their historical re¯ ections are technical development for the engineers.
generally technical or governmental in focus, based on en- The common theme is that the Internet is a popular, lib-
gineering documents, and sometimes include insider anec- eral, democratic, or even anarchistic medium. While hav-
dotes. The architects’ retrospectives can be characterized ing origins in the U.S. Department of DefenseÐ a peren-
by two main features. These features are both insights nial irony for these storiesÐ the Internet has grown to be
and, as I argue later, limitations. an nonhierarchical arena for the free exchange of ideas
First, today’s global Internet is linked to a series of and information. In accounting for its development, ad-
programs within the U.S. government. As Vinton Cerf, a jectives such as ª informal,º ª accidental,º and ª lawlessº
coauthor of the Internet standard of data communications abound (Rogers, 1995). From this perspective, the history
(TCP/IP) and recently known as a ª father of the Internet,º of the Internet is a story of expanding public access to
says, the Internet is ª the direct descendant of strategically Internet services and the ¯ ourishing of a culture of open
motivated fundamental research begun in the 1960s with discussion and technological enthusiasm.
federal sponsorshipº (Cerf, 1995). The original effort was One of the best-known and earliest narratives available
funded by the Defense Department’s Advanced Research on the Internet itself is entirely explicit about this notion.
Projects Agency (ARPA) (Barber Associates, 1975; Bolt With an online collection of historical articles appearing
Beranek and Newman, 1981). since the late 1980s, Michael Hauben and Ronda Hauben
The root of today’s internetworking in a government say they intend to promote civic values that they regard as
program funded by ARPA is one of the best-known ac- central to the formation of the Internet in its currennt form.
counts of the Internet’s origins(Anthes, 1994; Salus, 1995). They describe Usenet news, for example, as a ª poor man’s
 LOOKING BACKWARD AND FORWARD AT THE INTERNET 203

ARPANETº created by individuals who had experienced most conservative de® nitions yield results showing what
the bene® ts of ¯ exible, many-to-many communication in any experienced Internet user can attest to: rapid growth.
the military network (Hauben & Hauben, 1995, 1997). There is also no question that access to Internet ser-
The ARPANET itself, in development and use, is inter- vices has expanded to reach new populations. Once the
preted as ª democraticº in nature: province of the U.S. Department of Defense and its con-
The collaboration of the NWG [Networking Working tractors, and still heavily concentrated in the United States,
Group] (mostly graduate students) and ARPA (a component the Internet now reaches of® ces and some homes all
of the military), seems to be contrary to the normal atmo- over the industrialized world. Not only are there more
sphere of the times. Robert Braden of the Internet Activities users of TCP/IP protocols today than in 1970, but this
Board re¯ ects on this collaboration: ª ...One important rea- population represents more walks of life. There has been
son it worked, I believe, is that there were a lot of very bright a cornucopia of applications, including not only some of
people all working more or less in the same direction, led by the original ARPANET services such as e-mail and ® le
some very wise people in the funding agency. The result was transfer, but also bulletin boards, newsgroups, and, of
to create a community of network researchers who believed course, the World Wide Web. As the burgeoning literature
strongly that collaboration is more powerful than competi-
on ª online interactionº and ª computer-mediated commu-
tion among researchers....º (RFC 1336) These ideas point to
a reason why the work of these computer scientists founded
nicationº documents in detail, many users ® nd new plea-
what has come to be one of the most amazing and demo- sures, new ef® ciencies, and new possibilities in experi-
cratic bodies (i.e.: The Net and the culture attached to it) to menting and communicating with Internet technologies.
emerge in a long time. The community that has developedand What is more, the overall growth and greater access
the tools that accompany it form an important democratic of Internet use have had important consequences for the
force. (Hauben & Hauben, 1995, Chapter 6) expansion and diversi® cation of Internet services and in-
These themes have been reiterated in magazine arti- frastructure. This is because users learn by doing. Once
cles, books, and other popular media (Negroponte, 1995; users have ® gured out what the Internet is good forÐ on
Rheingold, 1993). Such accounts sometimes dissolve into their own termsÐ they quickly begin to develop new uses,
an exuberant welter of synonyms for ª autarchic,º such as and the volume and sophistication of traf® c on the In-
the following: ternet are increased. Users are active participants in the
development of a large technological system (Hart et al.,
Still abiding by the Hacker Ethic, these tens of thousands
1992).
of netheads have created myriad computer bulletin boards
and a nonhierarchical linking system called Usenet. At the
same time, they have transformed the Defense Department- Timeline
sponsoredARPANET into what has become the globaldigital
epidemic’s resolutely grass-roots structure.... The Internet... By way of summary of the contributions and limitations
is open (nonproprietary)and rabidly democratic. It is run like of standard accounts, I present a table that combines some
a commune with 4.8 million ® ercely independent members of the most salient themes (Table 1). The table is modeled
(called hosts). It crosses national boundaries and answers to after a type of historical document now widely available
no sovereign. It is literally lawless (Brand in Rogers, 1995). on the Internet: a timeline, which in its most basic form
However, despite the rhetorical excesses, there are im- is a list of selected technical innovations in chronological
portant insights bound up with accounts from the perspec- order. The table illustrates two major lines of progress in
tive of users. The key insights have been articulated by three historical periods.
Hart, Reed, and Bar in terms of two important facts about One line of progress, which re¯ ects architects’ perspec-
the history of the Internet: explosive growth and widening tives, is a series from the ARPANET to today’s global
access by a diversity of users (Hart et al., 1992). infrastructure that supposes particular continuities in tech-
There is no question that TCP/IP-based internetting has nology and government support. Major government initia-
grown. Internet traf® c, de® ned as data ¯ ow on the U.S. tives, usually identi® ed with the titles of particular network
national (NSFNET) backbone, grew explosively through systems, are important landmarks. Some of the most typ-
1988, when it began to roughly double in size each year. ical landmark systems are listed in the period headings of
In 1983 the INTERNET had 200 computer nodes. In 1994 this table.
the global Internet, de® ned as access to e-mail, comprised The other line of progress is a series of ever wider cir-
over 15,000 networks, 2.5 million permanently connected cles of user populations, beginning with the military and
computers, and 25 million people in 125 countries, by one select computer researchers and expanding eventually to
estimate (Cerf, 1995). By the close of the next year, the the computer-owning public at large. Such a series is also
number of networks, computers, and people had roughly re¯ ected in the headings of my table, as each includes
doubled. Statistics such as these are controversial in their a summary of new user populations involved during a
details, but no one disputes the upward curve. Even the period.
204
TABLE 1
Timeline of the Internet
ARPANET: U.S. Department of Defense, NSFNET: Supercomputer centers, NREN: Large and small businesses, private
contractors, computer researchers university faculty, staff, and students citizens, schools, consumers

1969: ARPANET online, connecting UCLA and a 1986: NSFNET online; management of regional 1991: NSF removes restrictions against
few other university computer centers nets by private ® rms begins; Usenet commercial uses of the Internet; NREN
using NCP (Network Control Protocol). discussion groups reorganized to legislation passed to provide for
1972: With 30 host computers in network, public accommodate greater traf® c. research on ultrafast (gigabit)
demonstration at a computer conference in 1987: Contract to manage NSFNET awarded to networks for commercial and
Washington, DC; InterNetworking Merit Network, Inc., which had supercomputing applications; World
Working Group (INWG) formed to managed a regional educational Wide Web (web) software distributed
establish standard protocols. network with MCI and IBM. by the European Laboratory for
1977: THEORYNET on line, providing e-mail to 1988: Private networks and for-pro® t service Particle Physics (CERN).
over 100 non-defense-contracting providers establish links to the Internet 1992: Faster backbone (T3 leased lines) added
computer researchers at University of for commercial users. to NSFNET.
Wisconsin. 1989: Over 100,000 host computers connected 1993: Contract to manage various functions of
1979: Unix User Network started, to send technical to the Internet. NSFNET awarded to AT&T, General
messages, and later adapted for other 1990: ARPANET decommissioned. Atomics/CERFnet, and Network
discussion groups as Usenet. Solutions, Inc. A graphical browser for
1981: IBM funds BITNET, providing e-mail to the web (Mosaic) distributed by the
some academics. NSF, funded by the National Center for
1982: INWG establishes TCP/IP as standard on Supercomputing Applications.
ARPANET. CSNET on line, funded by 1994: Widespread use of TCP/IP in corporate
National Science Foundation (NSF), for data communications.
full range of network services between 1995: Popular Internet fad catches on,
major U.S. university computer science particularly e-mail and web, extending
departments. to small businesses, homes, and
1983: MILNET on line, devoted to operational schools around the globe.
military nodes formerly on ARPANET. 1996: Number of computer hosts exceeds 12
1984: Over 1000 host computers connected to the million. U.S. President Clinton
Internet. announces ª Next Generation Internetº
1985: Numerous regional networks set up. initiative.
 LOOKING BACKWARD AND FORWARD AT THE INTERNET 205

I have listed three periods as illustrative of a variety of typical focus is often not particular objects but types of
periodizations that the literature offers. The ® rst period is technologies, groups of technical approaches, and large
identi® ed with the network-related research programs co- technological systems. For example, in the economics
ordinated by ARPA. Through the 1970s, the ARPANET of technical change, a key concept is the technological
was elaborated upon and expanded, parallel and simi- paradigm or regime, in which a family of technical ap-
lar networks were set up among defense contractors and proaches and practices, such as microelectronics, estab-
universities, and numerous regional networks, using vari- lish dominance in particular domains (Dosi, 1982; Molina,
ous technical standards, took life. Access to the ARPANET 1993). Typically also, the focus is on new technologyÐ
was limited to the Department of Defense and its contrac- its development, introduction, and adaptation. The goal
tors, particularly in computer science and engineering. they share is to consider an increasingly wide variety of
A second period is de® ned by the establishment of factors and levels of analysis that contribute to technol-
NSFNET, bringing e-mail and other internetworking ser- ogy development. (Basalla, 1988; Bijker & Law, 1992;
vices to supercomputing centers and an increasingly large Bowker et al., 1997; Cowan, 1983; Elliot, 1988b; Foray
number of university departments. The following boom in & Freeman, 1993; Hard, 1994; Hughes, 1983; Kling &
internetworking activity and demand sparked many new Iacono, 1988; MacKenzie, 1990; MacKenzie & Wajcman,
networks and service providers. 1985; Mowery & Rosenberg, 1989; Mukerji, 1994; No-
In the ® nal period, recent events were initially identi- ble, 1986; Rosenberg, 1982; Thomas, 1994; Winner, 1986;
® ed with the 1991 U.S. national announcement of a Na- Yearley, 1988).
tional Research and Engineering Network (NREN), which The new studies arguably have valuable insights for cor-
marked the movement of internetworking to a prominent porate strategy, policy, design, and public debate. While
position in national politics and contributed to further fed- not tightlycoupled to practical outcomes, they are intended
eral investments and the massive expansion in the number to more accurately portray the actual processes of techno-
of users. The last several years have witnessed rapid spread logy development, and may provide general guidelines to
of Internet access among businesses, university students, help think through practical issues. (Rip, 1995) In other
technical professionals, and even primary and secondary words, as in most social science, the new studies of tech-
school students. Corporate advertisingand electronic com- nology sensitize readers to possible analytical approaches
merce grew substantially after the public was given access rather than generate exact or universal guidelines.
to the Internet in 1995. There has also been some political However, the new technology studies have immediate
pressure toward the populist goal of approaching univer- application in counteracting some of the inadequacies of
sal access. Groups in North America, Europe, and parts the existing literature on internetworking. I present this
of Asia have been pushing to bring the Internet to public argument under the headings of four propositions. In each
schools and libraries. case, I ® rst explain the proposition in general terms. Then
This table is not meant to be an authoritative source of I give empirical examples taken from general technology
validated information, but is meant to represent and syn- studies and particular instances in the development of the
thesize other timelines and popular accounts of the history Internet.
of the Internet (e.g., Baker, 1993; Corbin, 1991; Hart et al., There are few published accounts about the Internet that
1992; Zakon, 1996). exemplify the perspectives taken by the new technology
studies. For this reason, I draw on examples described in
popular media in order to suggest some of the depth and
ALTERNATIVE PERSPECTIVES
breadth of research that could be done in this vein.
Having reviewed some of the most in¯ uential perspectives
on the nature and history of the Internet, I want to present
It Could Be Otherwise
an alternative set of perspectives based in recent interdis-
ciplinary studies of technology (cf. Bowker, 1995). With First of all, the new technologystudies come down strongly
roots in economics, the sociology of science, and the his- against any argument that a particular technical solution
tory of technology, among other ® elds, recent technology is the only possible or realistic one. When it is the nature
studies have criticized restrictive models of technologyde- of the technology that is used to justify such a position, it
velopment such as those embedded in what I have called is called ª technical determinism.º But there are other forms
architects’ accounts of the history of the Internet. They of determinism as well, and the new technologystudies are
share many of the perspectives represented by users’ ac- opposed to all of them in principle.
counts of the Internet, but they take users’ points of view While the new studies acknowledge technical con-
even farther, and take other pointsof view seriously as well. straints, possibilities, contingencies, and commitments,
Interdisciplinary technology studies now comprise a they assume there is no internal logic in technological
large body of comparative literature and case studies. The development that merely unfolds in practice. Nor is there
206 J. GUICE

any other grand force that sweeps through history. Hind- Network Architecture. (Drake, 1993; Stix, 1989) A wider
sight can make certain innovations seem like the only real view suggests that the milestones identi® ed in the techni-
contendersÐ especially since history is usually rewritten cal histories are not points on a straight and narrow path,
by the victorsÐ but there were alternatives (Bijker & Law, but islands of agreement, for certain groups with certain
1992, p. 3). purposes at certain times, within an otherwise contentious
Technology development also proceeds in a manner that ® eld.
is strongly dependent on existing practices, resources, and The communicational determinism of users’ accounts
abilities in a particular environment. It thus often proceeds is likewise belied by the details of communication on the
in a way that will, from another point of view, appear less Internet. Usenet, for example, is today a forum in which
than optimal. One of the key concepts in the economic open and free-ranging discussion, including a wide range
literature, for instance, is ª path dependency,º by which of political views, is the norm (Benson, 1996). Was it in-
it is meant that past actions and commitments to partic- evitable that Usenet would evolve to support such norms
ular practices can restrict future choices, even though in and practices? In fact, as a recent study has shown, it
principle other alternatives exist. evolved through the efforts of people and must be contin-
As I have shown already, standard accounts of the his- ually maintained by users as well; ª this consensus stems
tory of the Internet from the point of view of engineering from a long and often painful struggle, as Usenet’s design-
architects tend to start from today’s supposed successes ers, administrators, and users attempted to comprehend,
and work backwards, in a slim line of descent, from earlier de® ne, and govern the communication system they had
ideas and programs to the next major breakthrough. User createdº (Pfaffenberger, 1996, p. 365). In other words,
accounts start from an inevitably different pointÐ namely, the norms and actual practices of Usenet communication
the need to communicateÐ but they take it as a given that are the result of conscious efforts of participants. What is
people will use tools such as the Internet if they are avail- more, Usenet discussions are not completely open. Speech
able. From this point of view, the growth of the Internet is that is thought to restrict the ability of others from speaking
the result of people’s desire for communication, full stop is particularly loathsome to Usenet groups and is censured
(Schuster, 1994). This point of view glosses over all of the by participants.
details of why, how, and which people are communicating.
Studies of the history of other technologies might help
Innovation Requires Many
to provide some distance from the familiarity and apparent
inevitability of today’s Internet. For instance, it is often The new technology studies also argue that technology
forgotten that domestic arti® cial lighting was preceded by development hinges on diverse groups of participants in
large, public displays of electric lights. ª That we no longer two related senses. One is that various individuals, groups,
remember the excitement of electric light spectacles tes- organizations, and institutions are involved in technology
ti® es both to the fact that [electri® cation since the late development. Indeed, the more various groups are in-
19th...century has taken other turns] and to the tendency volved in a technology, the wider its support, and often
of every age to read history backward from the presentº the more deeply rooted and transformative the technol-
(Marvin in Rip, 1995, p. 418). Conversely, studies of elec- ogy. The other is that con¯ ict and competition are at least
trical lighting have shown an enormous variety of mate- as important as cooperation. Con¯ ict is not necessarily
rials and designs intended for use in similar settings. It overt but, rather, is often latent, re¯ ected in the strategies
is necessary to adopt the perspective of a particular time, participants adopt to create advantage in their situations
place, and set of concerns, with accompanying limitations (Bijker & Law, 1992, pp. 8±9; Cambrosio & Limoges,
in technical knowledge, in order to understand technolog- 1991). ª Technology today provides an enormous and di-
ical decisions at any one point. verse range of resources in many social struggles. It is
With regard to the Internet, architects’ retrospective ac- itself constructed and developed out of controversies and
counts have tended to reduce the actual complexity facing contestsº (Elliot, 1988a, p. 4).
researchers and managers at the time that they made par- From the point of view of recent studies, con¯ ict is not
ticular decisions. For instance, while TCP/IP is now taken necessarily bad for technology development. It is simply
as the de® nition of the global Internet, ARPANET ran us- a fact of life that sometimes has useful outcomes. For in-
ing other protocols for about a decade before TCP/IP was stance, today the largest share of wind turbines is sold
established as a standard. Moreover, there have been vari- by Danish companies. However, studies have shown that
ous alternatives to TCP/IP at every time. For instance, OSI Denmark’s prowess in the ® eld is the result of intense
(Open-Systems Interconnection) was created by the Inter- struggle among various groups since the oil crisis of the
national Standardization Organization in the 1970s and by early 1970s, including large engineering ® rms, small craft
around 1990 was still held as the ª European alternativeº shops, utilities, foreign and domestic buyers, and social
to TCP/IP and other protocols such as IBM’ s Systems movements (Grin and van der Graaf, 1996; Karnù e, 1995).
 LOOKING BACKWARD AND FORWARD AT THE INTERNET 207

Standard accounts of the development of the Internet The history of telephony is full of examples. At var-
tend to recognize a small set of groups and depict them as ious times and places, telephones have been considered
largely cooperative. In a new case study of work among In- worthless toys, a medium for musical concerts, and a
ternet architects that acknowledges con¯ ict among kind of telegraph. ªAn understanding of what the tele-
participants, Rogers (1998) argues that computer network phone `actually is’ does not only affect . . . lexical de® ni-
architectures were designed in interaction with ª the pro- tions, it also has practical implications. The understanding
cess of institutionalizing scienti® c disciplines and satisfy- of the telephone as a telegraph was signi® cant when de-
ing interorganizational constraints in the federal govern- ciding which legal code would regulate the distributionand
ment.º use of the telephone. The understanding of the telephone
Accounts stressing the autarchic character of the In- as a separate technology with special characteristics had
ternet tend to lose sight of numerous forms of control. implicationsfor the research efforts to develop the basic el-
For example, a recent paper argues that the Internet is co- ements of the telephoneº (Bakke, 1994, p. 1). Viewing the
ordinated behind the scenes of many casual users’ expe- history of telephony as a straight path toward the eventual
rience. Its authors liken the Internet to an organization in outcomeÐ hand-held devices for person-to-person voice
which ª 99% of day-to-day functions [are] to be handled communications, delivered by copper wire, to nearly
by empowered employees, leaving the manger free to deal every home and business in industrializedcountriesÐ skips
with the 1% of exceptional issuesº (Gillett & Kapor, 1997). over the long historical process of innovation, research,
Most importantly, standard accounts of the develop- marketing, and diffusion.
ment of the Internet lose sight of stakeholders who are One issue of meaning in the standard history of the In-
neither architects nor users, and in¯ uences that are not ternet is the contrast between its military origins and its
directly technical. Small groups of technicians are gen- autarchic uses. What were the intended contexts of use
erally described as working in solitary splendor, rather and how do they relate to the actual technology designs?
than working, as Rogers’s (1998) study suggests, within One way of resolving this question, for instance, is to argue
conditions set by state agencies, corporations, and other that the ARPANET, while funded by an agency in the Pen-
organizations, and in dialogue with larger organizational tagon, was not designed for directly martial use but to sup-
goals. While user accounts recognize a wider variety of port military-related research (O’ Neill, 1995). Similarly,
individuals, they typically are conceived of either strictly policy debates in Washington, DC, in the early 1990s fea-
as individuals or as a mass. It should be possible to tured two different visions of the meaning and purpose of
identify individuals, groups, organizations, and institu- public spending on information infrastructure: universal
tions that have had galvanizing and organizing effects on public access and grand scienti® c computing (Hart et al.,
both engineers and users. 1992). More recently, and in countries all over the globe,
One of the most immediate in¯ uences might be called another two competing visions have emerged in public
ª media organizersº Ð people who promote resources on debate: ª the net as a pathway of commerce, dominated
the Internet around particular user interests. Individuals by large companies versus a series of places where cit-
who write about, construct, and administer World Wide izens share information and build communityº (Muller,
Web sites, e-mailing lists, and bulletin boards are neither 1996).
architects of the network services nor simply individual From a strictly technical viewpoint,however, there have
users (see, for example, Donnelly& Ross, 1997; Patterson, been twists and turns in the Internet, recon® guring, for
1996; Perry, 1996). example, telephony. As has long been recognized, even
the earliest networking of computers represents signi® cant
points of convergence between computing and telecom-
Technologies Have Meanings
munications, and other points have been developing at an
Another insight of the new technology studies is that there increasingly rapid pace. At the time of writing, one can
is no particular meaning inherent in a particular tech- hold a live audio conversation over the Internet. Telecom
nology. Rather, technologies are associated with various carriers are straining under the weight of new demands
ideas, intentions, purposes, and contexts of use in different for data communications lines. One can send a fax from
times and places, for different people and groups. Since the Internet, send e-mail from a fax, and have one’s e-
it is often the case that people assume that their own pre- mail read by machine over the phone. For reasons such
ferred meaning for a technology is the only meaning, it is as these, it is becoming increasingly anachronistic to re-
enough to say that technologies have more than one mean- fer to computing and telecommunications and other me-
ing. In the development of technologies, this implies that dia by their conventional names. As one observer has ar-
there is generally no direct path from research to full func- gued, ª As a result of the horizontal integration of all me-
tionality because the meaning of technology takes twists dia (voice, audio, video, animation, data) in a common
and turns and unexpected routes. network and terminal infrastructure, telecommunications
208 J. GUICE

and networked-computing applications are no longer dis- `networks.’ The system was built on a preexisting social
tinguishableº (Messerschmitt, 1996). network of computer researchers. Superimposed on this
was a funding network consisting of the individuals and
sites that worked on ARPA contracts. Last in chronologi-
All Technologies Are Network Technologies
cal order was the physical network, the ARPANET itself.
Overall, recent technology studies have built up a con- All three types of networks crossed institutional and geo-
sensus that ª the success of technical innovations depends graphical boundaries that de® ned other environments in
as much on a supportive social environment as on tech- which the ARPANET builders participated.º
nical requirements. Typically, contributions by heteroge- For other examples of the suitability of network meta-
neous actors from different ® elds are needed for the vali- phors to the history of the Internet, one has to look no
dation, production and diffusion of new technologiesº farther than one’s newspaper. Telecom, computing, and
(Elliot, 1988a, p. 3). For this reason, recent technology other media have been coming together not only techno-
studies have tended to reject simple models and embrace logically but also as industries. As one journalist puts it,
complex narrative (Hughes, 1986; Kling, 1987, 1994). ª The convergence of technologies is leading to a conver-
A recurrent metaphor, in the work of an otherwise di- gence of companies. They are merging and acquiring each
verse group of writers, has been the network. Unlike a other at the speed of their high-end productsº (Government
line, or even a tree, a network is taken as potentially limit- Executive, 1997).
less in its topology. Overall, the image of the network is Convergence, connection, and networkinghave all been
used not as a model but as a rhetorical tool to create a more the metaphors of choice in these developments, not only
open-ended conceptual space for investigation of technical for discussing digital technologies, but also for under-
change. It summarizes the notion that innovation results standing relations among major organizations, industries,
from complex, opportunistic interactions among a variety and sectors. The most signi® cant trend has been the
of different entities. emergence of a ª myriad of cross-investments and strate-
In economics, a related concept is what is generally gic alliances among and between all of the information
called ª network externalities,º by which is meant the de- industriesº (Kavounas, 1996). Thus, for example, ª Mi-
pendency that one group has on another in a particular crosoft, Disney, Time Warner, GE/NBC, TCI, and News
technology (Clark, 1985). But the most radical or fun- Corp. have together formed a web of joint partnershipsº
damental network arguments are made by historians and (Auletta, 1997). It need hardly be mentioned, considering
sociologists. In the book punningly titled Networks of its prominence in the press, that the Internet plays many
Power, a historian of electri® cation has argued that the important roles in this competitive ® eld.
establishment of power grids required a complicated set What is more, it is widely held that the evolving net-
of social, organizational, and political arrangements to be work topology is too complicated to be known in advance.
worked out (Hughes, 1983). ªAny long-term assumptions made now about the develop-
Network metaphors contrast with the structure of archi- ment of multimedia, networked multimedia, interactive,
tects’ accounts of the development of the Internet, which video dialtone, new media, and online information will be
tend to describe well-de® ned lines. There is the line from wrong because there will be a real difference in how these
conception to reality, such as the idea in the minds of technologies are appliedº (Communications International,
ARPA managers to the building of a working ARPANET. 1995).
There is the line from one government program to the Although ª convergenceº has just recently emerged as a
next. There is the line from one key idea or invention to buzzword, and will soon be replaced by others, the dynam-
the next. The vagaries, false starts, different perspectives, ics that it speaks toÐ technical, industrial, socialÐ have
and competitive casualties of technical development are been taking place since the Internet began. Although gov-
largely overlooked. ernment agencies took a leading role initially, this should
Users’ perspectives on the development of the Internet not draw attention from the fact that corporations such
illuminate one aspect of the networked character of tech- as MCI, AT&T, and IBM have been involved with in-
nical development: how users have provided feedback in ternetting since the start. There is a long, complicated,
the form of changes in applications and demands to sys- and mostly unwritten history of corporate and other ma-
tem developers. Users’ perspectives on the Internet also jor organizations in a bewildering web of competitive and
implicitly show that the meanings of technologies have cooperative relationships linking computers.
been different in various contexts.
A recent paper informed by technology studies makes
CONCLUSION
a point of elaborating the metaphor of networks. In a
revisionist account of the history of the ARPANET, According to widely circulating accounts, the character
Abbate (1993) argued that it ª comprised several different and development of the Internet have consisted mainly
 LOOKING BACKWARD AND FORWARD AT THE INTERNET 209

of government programs, inventions, and demand from con¯ ict. Existing accounts, as I have pointed out, have
more and more people. However, the standard accounts taken a rather harmonious and unbalanced view of their
of the Internet’s past are mistakenÐ not because they have subject.
gotten their facts wrong, I have argued, but because they Having summarized the main arguments of the article,
recognize a limited range of facts. A technological sys- let me turn in closing to re¯ ect on possible implications
tem that has not only covered the globe geographically,but of a revised view of the Internet’s development for an un-
has been shaped in many different ways by countless indi- derstanding of its future. Extrapolating from the standard
viduals, groups, organizations, and institutions, requires a observations, one would expect the future of the Internet
wide range of analysis. to include more government programs, more inventions,
It is not that standard accounts are without value. Ar- and more users.
chitects’ accounts of the development of the Internet are Here again I would not take issue with the isolated
far from simple in the detail they provide, particularly facts, but with the overall picture that they build up. I
about technical decisions. Users’ accounts offer impor- do not disagree that the future of the Internet will see new
tant insights into processes of adaptation and adoption, the government programs, new inventions, and new enthusi-
changing meaning of technologies, and the contestation of astic users. However, if recent technology studies are a
a variety of social issues in a technical arena. useful guide, the future of the Internet will consist of these
Nonetheless, I have suggested that future scholarship things only in part.
about the Internet would bene® t from extending the gen- As I suggested earlier, the approach of the new technol-
eral ® ndings of recent interdisciplinary studies of other ogy studies is in line with current discussions of ª digital
technologies. These ® ndings I have summarized in terms convergence.º To put an even ® ner point on it, the four
of four aphorisms: aphorisms with which I have summarized recent technol-
ogy studies have direct counterparts in these discussions:
1. It could be otherwise.
2. Innovation requires many. 1. The future of the Internet is dif® cult to predict.
3. Technologies have meanings. 2. It will involve a wide variety of actors in a variety of
4. All technologies are network technologies. relationships, including con¯ ict.
3. Meanings associated with the technologies will con-
For historians of the Internet to make use of the insights
tinue to be various and in ¯ ux.
of recent technologystudies, it will be necessary to expand
4. Overall, the further development of the Internet is a
the empirical basis of investigations. It is necessary to in-
highly rami® ed, complex historical process.
clude at least the variety of actors, institutions, and social
processes that is already acknowledged by the press. Like One point on which talk about convergence is probably
users’ stories about the Internet, new research should in- misleading, however, is suggested by the key phrase ª dig-
clude contributions from nonengineers, but not only from ital convergence.º The problem is that it puts the focus on
individual users. Before the popular adoption and adapta- technology as such, while the central ® nding of technol-
tion of Internet technologies, how were they designed in ogy studies is that important innovations are never strictly
the ® rst place? How have the interests and activities of var- technical in nature. What is more, if there is anything to
ious stakeholders affected the production of technologies extrapolate from current trends of the Internet, it is that
and technical standards? the technology itself is increasingly less important as a
The most apparent omission in existing accounts is the driving force in its development, as it becomes both more
role of large, powerful organizations. As well as showing commercial and more popular. At the same time, it is not
the contributions of small groups of network engineers generally recognized how much logistical and other main-
and enthusiasts, scholars should analyze contributions of tenance work is required to use digital technologies. There
well-organized and funded groups (Roberts, 1992). In the are practical and cultural barriers to use of internetworking
earlier discussion, I mentioned a number of such organiza- applications that will be increasingly salient as the cutting
tions, any one of which would put the history of the Inter- edge of Internet technologies becomes more sophisticated
net in a different light, including the following: national and groups with less technical support attempt to take part
security and other state agencies; cross-agency and quasi- (Kling, 1991, 1997).
governmental bodies; telecommunications, computer, Altogether, these points suggest a different scenario for
electronics, and media companies; scienti® c organizations the future of the Internet: The Internet continues to grow,
and groups; public policy lobby groups; and university in part fueled by extensions in infrastructure and pop-
administrators and organizations (technical and nontech- ulations that could be extrapolated from standard view,
nical). The task, though, is to explore relations and con- but also connects to new infrastructures and populations.
texts among these and other actors. On this point the The Internet fad has little interest for its main sources of
most apparent omissions are the elements of power and growth: corporationsthat want to get work done, especially
210 J. GUICE

the coordination of far-¯ ung and diverse information tech- Cambrosio, A., and Limoges, C. 1991. Controversies as governing
nology; and consumers, who don’t care about being techni- processesin technologyassessment.TechnologyAnalysis and Strate-
cally cool but who do care about the activities for which the gic Management 3(4):377±396.
Internet may be used as a tool. Technological enthusiasts Cerf, V. G. 1995. Computer networking: Global infrastructure for
of the 20th century who expected technically demanding the 21st century (World Wide Web document). [http://www.cs.
washington.edu/homes/lazowska/cra/networks.html]
new applications to take off, such as videoconferencing
Clark, N. 1985. The political economy of science and technology. Ox-
in of® ces, schools, and homes, made overly aggressive ford: Basil Blackwell.
predictions. Such applications develop over a longer time CommunicationsInternational.1995. Technologyis only half the story.
than expected by supporters, and unexpected other appli- Communications International 22(2):2.
cations appear. There are extreme differences in the level Comer, D. 1991. Internetworking with TCP/IP, Volume I: Princi-
of services between sectors (such as business and educa- ples, protocols, and architecture. Englewood Cliffs, NJ: Prentice
tion) and between geographical areas and nations. The Hall.
involvement of content providers and gatekeepers has an Corbin, R. A. 1991. The development of the National Research
increasingly important role. Infrastructural crises, such as and Education Network. Information Technology and Libraries
the year 2000 problem and cyberterrorism, trigger shifts September:212±220.
that were not widely anticipated even months before they Cowan, R. S. 1983. More work for Mother: The ironies of household
technology from the open hearth to the microwave. New York: Basic
happened.
Books.
Donnelly, D. F., and Ross, K. 1997. The Internet: Historical media
research on the virtual archives. Historical Journal of Film, Radio
REFERENCES
and Television March:129(8).
Abbate, J. 1993. An archeology of the ARPANET: Unpublished con- Dosi, G. 1982. Technological paradigms and technological trajecto-
ference paper available at: http://www-rci.rutgers.edu/» jea/papers/ ries: A suggested interpretation of the determinants and directions
Arch.html of technical change. Research Policy 11:147±162.
Anthes, G. H. 1994. The history of the future: ARPANET founders Drake, W. J. 1993. The Internet religious war. Telecommunications
talk about the network that emerged as the Internet. Computerworld Policy 17(9):643±649.
28(40):101(3). Elliot, B. 1988a. Introduction. In Technology and social process, ed.
Auletta, K. 1997. American keiretsu: the six most powerful media B. Elliot, pp. 1±7. Edinburgh: Edinburgh University Press.
companies are borrowing the ancient Japanese custom of co-opting Elliot, B., ed. 1988b. Technology and social process. Edinburgh: Ed-
the competition. New Yorker October 20:225(3). inburgh University Press.
Baker, S. 1993. The evolvingInternet backbone.UNIX Review Septem- Evanson, L. 1994. Road maps to the Internet taking off. San Francisco
ber:15(5). Chronicle 3 June:B1.
Bakke, J. W. 1994. Technologies and interpretations: The case of the Foray, D., and Freeman, C., eds. 1993. Technology and the wealth of
telephone. Unpublished conference paper available from the author. nations: the dynamics of constructed advantage. London: Pinter
Barber Associates. 1975. The Advanced Research Projects Agency, Publishers.
1958±1974. Washington, DC: Richard J. Barber Associates, Inc. FTP Software, Inc. 1991. The PC/TCP interoperability guide. Wake-
Basalla, G. 1988. The evolution of technology. Cambridge: Cambridge ® eld, MA: FTP Software, Inc.
University Press. Gillett, S. E., and Kapor, M. 1997. The self-governing Internet: Coor-
Benson, T. W. 1996. Rhetoric, civility, and community: politicaldebate dination by design. In Coordinating the Internet, ed. B. Kahin and
on computer bulletin boards. Communication Quarterly 44(3):359± J. H. Keller, pp. 3±38. Cambridge, MA: MIT Press.
378. Grin, J., and van der Graaf, H. 1996.Technologyassessmentas learning.
Bijker, W. E., and Law, J., eds. 1992. Shaping technology/building so- Science, Technology and Human Values 21(1):72±99.
ciety: Studiesin sociotechnicalchange.Cambridge, MA: MIT Press. Government Executive. 1997. Change is the name of the telecom game.
Bolt Beranek and Newman, Inc. 1981. A History of the ARPANET: The Government Executive 29(6):60±62.
® rst decade. Arlington, VA: Bolt Beranek and Newman, Inc. Pre- Hafner, K., and Lyon, M. 1996. Where wizards stay up late: The origins
pared for U.S. Defense Advanced Research Projects Agency. Stock of the Internet. New York: Simon & Schuster.
number AD/A155 440 NTIS. Hard, M. 1994. Technology as practice. Social Studies of Science
Bowker, G. C. 1995. History of the Internet. (Comments on ses- 24:549±585.
sion on the history of the Internet at the Society for So- Hart, J. A., Reed, R. R., and Bar, F. 1992. The building of the Internet:
cial Studies of Science and Society for the History of Tech- Implications for the future of broadband networks. Telecommunica-
nology, Annual General Meetings, Charlottesville, VA, 18-22 tions Policy 16(8):666±689.
October, 1995.) [http://commons.utopia.usweb.com/cgi-bin/nph- Hauben, M., and Hauben, R. 1995. Netizens: On the history and im-
mailings.cgi/ rre/history.of.the.Internet.html] pact of Usenet and the Internet (on-line book). Hypertext version
Bowker, G. C., Star, S. L., Turner, W., and Gasser, L. 1997. Social [http://www.cs.columbia.edu/» hauben/netbook/]. Central URL for
science, technical systems, and cooperative work: Beyond the great various versions [http://www.columbia.edu/» hauben/netbook/]
divide. Mahwah, NJ: Lawrence Erlbaum Associates. Hauben, M., and Hauben, R. 1997. Netizens: On the history and im-
Brittanica Online. 1994. Internet. Encyclopedia Brittanica. (Subs- pact of usenet and the Internet. Los Alamitos, CA: IEEE Computer
cription-only URL.) Society Press.
 LOOKING BACKWARD AND FORWARD AT THE INTERNET 211

Hughes, T. P. 1983. Networks of power: Electri® cation in Western Nicoll, L. H. 1994. An introductionto the Internet: 1. History, structure,
society, 1880±1930. Baltimore, MD: Johns Hopkins University and access. Journal of Nursing Administration 24(3):9±11.
Press. Noble, D. F. 1986. Forces of production: A social history of industrial
Hughes, T. P. 1986. The seamless web: Technology, science, etcetera, automation. New York: Oxford University Press.
etcetera. Social Studies of Science 16:281±292. O’ Neill, J. E. 1995. The role of ARPA in the development of the
Kahin, B., and Abbate, J., eds. 1995. Standards policy for information ARPANET, 1961±1972. IEEE Annals of the History of Computing
infrastructure.Cambridge, MA: MIT Press. 17(4):76±81.
Kantrowitz, B., and Rogers, A. 1994. The birth of the Internet. News- Patterson, D. A. 1996. An electronic social work knowledge base: A
week 8 August:56±58. strategy for global information sharing. International Social Work
Karnù e, P. 1995. Institutional interpretations and explanations of dif- 39(2):149±150.
ferences in American and Danish approaches to innovation. In The Perry, D. G., Blumenthal, S. H., and Hinden, R. M. 1987. The
institutionalconstructionof organizations: Internationaland longi- ARPANET and the DARPA Internet. Library Hi Tech. 22:51±62.
tudinal studies, eds. W. R. Scott and S. Christensen, pp. 243±276. Perry, L. S. 1996. Internet resources for American and International
Thousand Oaks, CA: Sage Publications. Studies. American Studies International 34(2):25(17).
Kavounas, E. J. 1996. Understanding convergence. Wharton Journal Pfaffenberger, B. 1996. ª If I want it, it’ s OKº : Usenet and the (outer)
40:(20). [http://student.wharton.upenn.edu/journal/v40n20/media. limits of free speech. The Information Society 12(4):365±386.
html] Reid, R. H. 1997. Architects of the web: 1,000 days that built the future
Kleiner, K. 1994. What a tangled web they wove. New Scientist 30 of business. New York: John Wiley & Sons.
July:35±39. Rheingold, H. 1993. The virtual community: Homesteading on the
Kling, R. 1987. De® ning the boundaries of computing across com- electronic frontier. Reading, MA: Addison-Wesley.
plex organizations. In Critical issues in information systems, eds. Rip, A. 1995. Introduction of new technology: Making use of recent
R. Boland and R. Hirschheim, pp. 307±362. New York: John Wiley. insights from sociology and economics of technology. Technology
Kling, R. 1991. Computerization and social transformations. Science, Analysis and Strategic Management 7(4):417±431.
Technology and Human Values 16(3):342±367. Roberts, L. 1971. ARPA network implications. EDUCOM: Bulletin of
Kling, R. 1994. Reading ª all aboutº computerization: How genre con- the Interuniversity Communications Council Fall:4±8.
ventions shape social analyses. The Information Society 10(3):147± Roberts, M. M. 1992. The university role in the United States national
172. research and education network. Computer Networks and ISDN Sys-
Kling, R. 1997. The NGI as an effective support for professional work tems 25(4±5):512±517.
and social life. White Paper for Presidential Advisory Committee Rogers, J. D. 1995. Characterizationsof INTERNET history: Toward a
on High-Performance Computing and Communications, Informa- multistranded account. Paper read in session on the history of the In-
tion Technology, and the Next Generation Internet, December 4. ternet at the Society for Social Studies of Science and Society for the
[http://php.ucs.indiana.edu/» kling] History of Technology, Annual General Meetings, Charlottesville,
Kling, R., and Iacono, S. 1988. The mobilization of support for VA, 18±22 October.
computerization: The role of computerization movements (with Rogers, J. D. 1998. Internetworking and the politics of science:
S. Iacono). Social Problems 35(3):226±243. NSFNET in Internet history. The Information Society.
Krol, E. 1992. The whole Internet: User’s guide and catalog. Sebas- Rosenberg, N. 1982. Inside the BLACK BOX: Technology and eco-
topol, CA: O’ Reilly and Associates, Inc. nomics. Cambridge: Cambridge University Press.
MacKenzie, D. 1990. Inventing accuracy: A historical sociology of Salus, P. 1995. Casting the Net: from ARPANET to INTERNET and
nuclear missile guidance. Cambridge, MA: MIT Press. beyond. Reading, MA: Addison-Wesley.
MacKenzie, D., and Wajcman, J., eds. 1985. The social shaping of Schultz, B. 1988. The evolution of the ARPANET. Datamation 1
technology: How the refrigeratorgot its hum.: Milton Keynes, Open August:71±74.
University. Schuster, J. 1994. The soul of a social machine. Electronic Learning
Messerschmitt, D. G. 1996. The convergence of telecommunications February:16(2).
and computing: What are the implicationstoday? IEEE Proceedings Siewiorek, D. P., Bell, C. G., and Newell, A. 1982. Computer structures:
84(8):1167±1186. Principles and examples. New York: McGraw-Hill.
Molina, A. H. 1993. In search of insights into the generation of techno- Stix, G. 1989. Data communications. IEEE Spectrum 26:43±46.
economic trends: micro- and macro-constituenciesin the micropro- Thomas, R. J. 1994. What machines can’t do. Politics and technologyin
cessor industry. Research Policy 22(5/6):479±506. the industrial enterprise. Berkeley: University of California Press.
Mowery, D., and Rosenberg, N. 1989. Technology and the pursuit of van Atta, R. H., Reed, S. G., and Deitchman, S. J. 1990. DARPA tech-
economic growth. Cambridge: Cambridge University Press. nical accomplishments: An historical review of selected DARPA
Mukerji, C. 1994. Toward a sociology of material culture: Science projects. Washington, DC: Institute for Defense Analyses.
studies, cultural studies and the meanings of things. In The sociology Winner, L. 1986. The whale and the reactor. Chicago: University of
of culture: Emerging theoreticalperspectives, ed. D. Crane, pp. 143± Chicago.
162. Oxford: Blackwell. Yearley, S. 1988. Science, technology and social change. London: Un-
Muller, M. J. 1996. De® ning and designing the Internet. Social Science win Hyman.
Computer Review 14(1):30±33. Zakon, R. H. 1996. Internet timeline. [http://www.bham.ac.uk/docu-
Negroponte, N. 1995. Being digital. New York: Knopf. ments/ hobbes.html].