Chapter One

INTRODUCTION

Robert H. Anderson, Tora K. Bikson, Sally Ann Law, Bridger M. Mitchell

Over the last 15 years, the burgeoning use of personal computers has popularized a number of new information services, including in particular electronic mail or "e- mail." E-mail is a form of information interchange in which messages are sent from one personal computer (or computer terminal) to another via modems and a telecommunications system. The use of e-mail began on the ARPAnet[1] (the precursor of the Internet) in the 1960s and 1970s in the United States, gradually spread along with the use of mainframe- and minicomputer-based local nets in the 1970s, and "exploded" along with the rapid growth of personal computers (PCs) and the Internet in the 1980s. E- mail began as a means of information interchange for small, select groups; its use has spread to encompass millions of people in the United States and all over the world. E-mail has given rise to the formation of many "virtual communities"--groups of individuals, often widely separated geographically, who share common interests. The interpersonal linkages and loyalties associated with these virtual communities can be real and powerful.

E-mail has unique properties that distinguish it from other forms of communication; for example, it supports true interactive communication among many participants. For the first time in human history, we would assert, the means of "broadcasting" or "narrowcasting" are not confined to the few with printing presses, TV stations, money to buy access to those scarce resources, and the like. E-mail is also, unlike telephone calls (with the exception of voice mail and answering machines), asynchronous, so that communication does not depend on the simultaneous availability and attention of sender and recipient. Generalizing greatly, e-mail increases the power of individuals, permitting them to be active participants in a dialog extended in both time and space, rather than passive recipients of "canned" programming and prepackaged information.[2] These characteristics give rise to the question: Can e-mail's novel properties address society's most compelling problems? If so, by what means?

Problem Statement

It is now possible to imagine the arrangement or construction of systems in which nearly universal access to e-mail within the United States could become feasible within a decade-- indeed, that is one aspiration of the U.S. National Information Infrastructure (NII)[3] initiative. Since e-mail use is growing rapidly (e.g., within individual corporations, CompuServe, America Online, Internet, and Bitnet systems and on numerous dial-in electronic bulletin boards), the question may be asked: "Why bother? It's happening anyway." Three important answers to this question are: (1) In spite of the growth of these e- mail systems, the majority of U.S. residents probably will continue to lack access to e-mail well into the next century without societal intervention; (2) there is today a significant lack of active participation by many citizens in the dialog that forms the basis for the U.S. democratic process;[4] and (3) some citizens, such as inner-city minorities and the rural poor, are relatively disenfranchised and constitute groups that will be the last to be reached by commercial e-mail systems that evolve in private markets. Because the properties of e-mail allow individuals to engage in an active civic dialog, with informative and affiliative dimensions, universal e-mail might provide significant benefits in creating interactive communication among U.S. citizens and residents.

The problem, then, is achieving active, responsive citizen participation in our national dialog for all citizens-- participation not only in national politics but in local affairs, job markets, educational systems, health and welfare systems, international discourse, and all other aspects of our society.

There are hints that the distinctive properties of electronic mail systems (including access to, and ability to post and retrieve messages from, various electronic bulletin boards) may well be relevant to this re-enfranchisement of all citizens. The civic networks discussed in Chapter Five of this report exemplify these opportunities.

It is also clear that widespread citizen access to an e-mail system could have profound economic implications that might provide new sources of business and revenue to entrepreneurs providing new services; for example, installation of the French Programme Télétel system resulted in a flourishing of electronic services available to virtually all French citizens (and, for that matter, many visitors--through terminals available in hotel rooms and public sites).

It is important to note, however, that the Minitel terminal used by Télétel was not originally conceived of as access to an e-mail system but rather as an "electronic telephone directory." As is often the case, when some facility for communication becomes possible within a system (e.g., ARPAnet, Télétel, and to a growing extent the Dialog system within the United States), its convenience and empowerment of individuals quickly cause e-mail to become an important form of usage. Lack of true e-mail capabilities may be a major contributor to the failure of other electronic service ventures such as teletext experiments, although too many factors may be involved to confirm this assessment.

Universal E-Mail

The initial forays into widespread availability of electronic mail, such as the ARPAnet (and now Internet) experience, Télétel, and growing Prodigy, CompuServe, and America Online usage, lead to an intriguing question: What about "universal e-mail?" What about providing all residents of the United States with access to e-mail service, just as they now all (or almost all) have access to telephone service and postal service? What would be involved in such an undertaking? What are the pros and cons? What are the advantages and disadvantages? Could this have beneficial effects for U.S. society? Greater cohesion? Reduced alienation? Increased participation in the political process? Influence national security? What about beneficial effects for the U.S. economy? Or other productive side effects? And who would pay for the infrastructure and its usage?

More specific questions arise immediately regarding the services and functions to be provided by a universal e-mail system: the required degree of access to such a "universal" service; the provision of privacy; alternative system architectures and implementation schemes; the cost of such a system/service and the method of payment; the likely social and international effect of universal e-mail; and finally, public versus private roles in creating and operating such a service. This report describes our initial study of these and related issues over a two-year period and presents the results of our analyses.

The issue of providing universal e-mail cannot be considered in isolation. As mentioned above, it is part of a larger national debate on a "national information infrastructure" and "global information infra-structure" that is gaining momentum. We hope this report contributes to discussion of the policy and social issues arising from attempts to provide such a service, in addition to technical options regarding implementation of the supporting infrastructure per se, because of these issues' importance in the public policy debate.

Some Definitions

Electronic Mail

For the purposes of this report, we have adopted a definition of electronic mail provided in an earlier RAND report (Anderson et al., 1989):

An electronic mail system:

  1. Permits the asynchronous electronic interchange of information between persons, groups of persons, and functional units of an organization; and

  2. Provides mechanisms supporting the creation, distribution, consumption, processing, and storage of this information . . . .

The words in this definition all have significance. Key among them are the following:

By the above definition, multipart messages containing embedded formatted word-processing documents, video clips, bitmapped pictures, sound clips, and the like are certainly e-mail. Faxes sent from one dedicated fax machine to another, appearing only on output paper are not (because they are not processable in a useful manner), but a "fax" sent from one PC to another meets the definition (because it may be stored for later retransmission, and its contents may be "processed"--e.g., by character recognition or graphics enhancement programs; in fact, some recipients may never get it in paper form). Similarly, using a personal computer to interact with "chat" groups and MUDs[5] usually qualifies as a form of e-mail, because most communication programs through which this interaction is carried out allow the transcript of the interaction to be saved, processed, reused, and so on.

We have tried to use a rather narrow definition of e-mail to focus this report on electronic mail, although it will be clear that most e-mail users will also have facilities at hand to browse the World Wide Web, participate in multiuser simulations and games, and so forth. It is not important to draw too fine a distinction between what is e-mail and what is not; the importance of having some definition becomes clearer in Chapter Three's discussion of access devices and their locations and standards and protocols needed.

Universal Access

The other key concept in this report is "universal access." By this, we simply mean e-mail facilities and services that are

We do not, therefore, envision that every single person will have access, but that e-mail can achieve the same ubiquity that telephones (including the availability of payphones) and TVs have. Table 1.1 shows the penetration of related technologies into U.S. households, for comparison.

Table 1.1

Availability of Related Technologies in U.S. Households

_________________________________
                   Percentage of 
Technology          Households
_________________________________

Television                95
Telephone                 93
Video cassette recorder   85
More than one TV          66
Cable TV                  64
Pay-per-view service      51
Video game system         40
Video camera              28
Fax                        6
_________________________________
  SOURCE:  Times Mirror (1994).
Note that the above percentages are not distributed uniformly across various sectors of our society. For example, a recent report (Mueller and Schement, 1995) describes telephone access in Camden, New Jersey, by family income, ethnicity, age, and other demographic factors. The report indicates that, overall, only 80.6 percent of households in Camden have telephones; notable disparities include families on food stamps, who lag 20.4 percentage points behind households not on food stamps. For many households, "universal access" means traveling to the nearest working payphone (where receiving incoming messages is sometimes precluded either socially or by the technology). Similarly, universal access to e-mail for many may require using public terminals in shared spaces such as libraries and schools (but where barriers to message reception can readily be eliminated).

Advantages of Universal E-Mail

E-mail services can be used both for "telephone-type" messages and for other, usually longer, messages or documents that might otherwise be sent using facsimile or hard-copy postal services, both public and private. Compared with the telephone system, one primary advantage of an e-mail service is that it eliminates "telephone tag." It also provides a content record of the interactions that can be retrieved, printed, studied, selectively forwarded, and in general reused. Other advantages are that it permits (but certainly does not require) more deliberative and reflective, but still interactive, conversational dialogs, as well as one-to-many and many-to-many conversations. These features have led to many new social, commercial, and political groupings of people: the "virtual communities" mentioned above, using e- mail as the linkage.[6] It provides a common context among a set of participants.

Compared with postal services, an e-mail service offers much faster mail delivery--usually minutes between any two locations in the United States (although currently, delays up to a day occur with some Internet access providers), compared with one to several days for postal systems. E-mail systems also afford much more flexibility (both locational and temporal) in that delivery. In the current postal system, a person's mail is delivered to one or two (or at most a few) fixed addresses (e.g., home or office). In most e-mail systems, a person with the proper (portable) terminal equipment can log in to his or her "mailbox" from any location that has electronic access to the system. Today, this means that people can pick up their e-mail from their office, their home, their hotel rooms, another office (perhaps in another city) they are visiting, or any site with a phone jack.[7] In the future, as terminal equipment gets smaller and cellular telephones become more ubiquitous, one will be able to pick up or send e-mail while traveling in a car and flying in an airplane. This results in more geographic independence (where one gets mail) and temporal flexibility (when one gets mail).

These advantages are available in any e-mail system. The additional advantage of a universal e-mail system is that since everyone belongs to the system, a user can send e-mail to anyone, not just a limited group, and receive e-mail from anyone. This makes the special advantages of e-mail available for all of one's correspondence, not just a subset. If the costs of such a service permit attractive pricing, it could take over a significant portion of the business of current postal services[8]--especially when next-generation e-mail systems allow the transmission and viewing of multimedia messages containing high-resolution color pictures, "movie clips" of image sequences, and sound, which could, among other things, support a variety of "electronic commerce."

Disadvantages of Universal E-Mail

The concept of universal e-mail raises serious concerns as well. For example, individual users could get "flooded" with messages, unless some means of "filtering" incoming message traffic is provided. Also, some virtual communities enabled by e-mail could be bad for U.S. society, rather than good; they could conceivably lead to a less-cohesive society, rather than a more-cohesive one.[9] It is also clear that within any e-mail system, some users will be "more equal" than others; they will be able to purchase more powerful equipment, giving them more power over their electronic communication. Some will become more knowledgeable in the features and facilities of the system-- permitting them, for example, to assemble tailored mailing lists for broadcast of their messages--allowing them to take advantage of those features for their own personal benefit or gain. Special-interest groups may in particular be motivated to become further empowered by use of these communication tools. Some (but not all) would also consider it a disadvantage that national borders become more transparent to international commerce and influences (Ronfeldt et al., 1993).

Motivations for Universal E-Mail

The apparent advantages of universal e-mail, despite the possible side effects and disadvantages, lead to a number of possible motivations for establishing such a service in the United States, ranging across the spectrum from the utilitarian to the idealistic. At the utilitarian end is efficiency. Electronic mail uses modern information and telecommunications technology to provide a much faster and more efficient means of conveying information from one point to another than current postal systems, which rely on "technologies"--letters written on paper, put in sealed envelopes, and physically transported from sender to receiver--over two millennia old. The increased speed and efficiency of information delivery by e-mail could have many commercial and economic benefits, contributing to increased U.S. economic competitiveness.

At the idealistic end of the spectrum of motivations, the hypothesis is made that electronic mail makes possible much more egalitarian, deliberative, and reflective dialogs among individuals and groups. (See Sproull and Kiesler, 1991b, for supporting evidence.) It might therefore lead to new social and political linkages within U.S. society, reduce the feelings of alienation that many individuals in the United States feel and give them a new sense of "community," revitalize the involvement of the common citizen in the political process, etc., and in general strengthen the cohesion of U.S. society.

Different motivations across this spectrum will appeal to different elements of U.S. society. To achieve widespread appeal--and political/economic support--a U.S. universal e- mail service should satisfy a broad spectrum of these motivations, whether the system is "designed" to meet these objectives (e.g., with heavy U.S. government involvement) or evolves through private initiative and entrepreneurship subject to constraints, incentives, or standards that encourage universal access.

Organization of this Report

The remainder of this report addresses the key questions raised above. Chapter Two describes our analyses of Current Population Survey (CPS) data collected by the U.S. Census Bureau, not only to obtain a picture of the current state of computer and information service usage in U.S. households but also to determine trends in this usage. It indicates groups and sectors of our society with low rates of electronic information system use, to which particular attention must be paid if "universal" access is to be achieved.

Chapter Three discusses technical considerations in achieving universal access, such as: What are the services that will encourage demand? What are the options for providing physical access, including both the nature of the device (e.g., PCs, TV set-top boxes, game machines, dedicated e-mail devices, display telephones) and the location (home, work, schools, kiosks, and other "common" areas)? What would the demands be for training and support? Are current user interfaces adequate for "everyone"? What about standards for addresses, directories, confidentiality, and privacy? What system architectures are most appropriate?

Chapter Four presents economic and regulatory issues raised by the concept of universal access, treating the provision of e-mail from the perspective of market supply, demand, pricing, and market structure.

Chapter Five explores social issues raised by universal access, using as a basis for this analysis descriptions of site visits to five "wired community" prototypes and experiments planned and underway throughout the country. These visits investigated what works today, what does not, and how these findings might be extrapolated to a nationwide system. They also provide additional insight into training and education requirements and the suite of services that attract users.

Universal access to electronic mail services is not just an idea with national implications. There are significant international implications as well. We have been particularly intrigued by correlations between interconnectivity within countries and their degree of democratization. Chapter Six addresses these topics and argues that the implications could affect the allocation of resources underlying some portions of U.S. foreign policy.

Chapter Seven summarizes our key findings and conclusions.

Appendix A contains further analyses of CPS data for 1984, 1989, and 1993, showing additional trends in computer and information system usage by U.S. households. Appendix B contains descriptions of the site visits made to five "wired- community" projects throughout the United States that form the basis of the discussion in Chapter Five.



[1]A brief history of the ARPAnet and Internet can be found in Lynch and Rose (1993), Chapter 1.

[2]Documentation on this point may be found, for example, in Sproull and Kiesler (1991b).

[3]Introductory materials on plans for the NII may be found at Web page: http://nii.nist.gov/.

[4]Documentation about the decline in U.S. "social capital" and its effect on the performance of representative government may be found in Putnam (1993) and Putnam (1995). Among the data cited in his 1995 article: U.S. voter turnout has declined by nearly a quarter from the early 1960s to 1990; Americans who report they have "attended a public meeting on town or school affairs in the past year" declined from 22 percent in 1973 to 13 percent in 1993; participation in parent-teacher organizations dropped from more than 12 million in 1964 to approximately 7 million today; since 1970, volunteering is off for Boy Scouts by 26 percent and for the Red Cross by 61 percent.

[5]A MUD is variously defined as Multiple User Dimension, Multiple User Dungeon, or Multiple User Dialogue. It is a computer program allowing users to explore and help create an on-line environment. Each user takes control of a computerized persona/avatar/incarnation/character. The user can walk around, chat with other characters, explore dangerous monster-infested areas, solve puzzles, and even create his or her very own rooms, descriptions, and items. For further information see, for example, http://www.math.okstate.edu/~jds/mudfaq-p1.html.

[6]As one of many examples documenting this, see Klein (1995).

[7]E-mail can also be forwarded automatically to an alternative mailbox (e.g., closer to a vacation spot or sabbatic location).

[8]In this regard, it should be noted that the current U.S. postal services deliver two things: information (e.g., letters) and bulk material (e.g., packages). A universal e-mail system should, in principle, be able to take over much of the information delivery functions; it obviously cannot handle the bulk material delivery functions. However, some bulk material consists of catalogs and advertising that may, in fact, increasingly become accessible electronically.

[9]Potentially even worse could be hidden virtual communities--on-line underground communities hiding under encryption or just not advertising themselves.


Next chapter
Table of Contents