Computer Networking in Canada: From CA*net to CANARIE

Leslie Regan Shade (McGill University)

Introduction

This paper will discuss the current status of computer networking in Canada, from the establishment of CA*net in the late 1980s, to current plans to upgrade its networking capabilities through the implementation of CANARIE (the Canadian Network for the Advancement of Research, Industry and Education). A preliminary overview of the Internet, the multitude of computer networks that are linked together internationally, will situate Canadian networking in the global context. CANARIE's U.S. equivalent, the National Research and Education Network (NREN) (subsumed within the High-Performance Computing Act of 1991) and the proposed Information Infrastructure and Technology Act will be discussed briefly to give an idea of how contested the debates have been regarding network accessibility. Some of the burgeoning networking communities in Canada that will certainly vie for CANARIE access will be described, including freenets, local area bulletin board services (BBSs) connected via FidoNet, and non-profit organizations such as The Web.

The Internet

The Internet is a vastly interconnected and loosely organized system of computer networks worldwide, composed of local area networks (LANs), city-wide metropolitan networks (MANs), and huge Wide Area Networks (WANs). These various networks are linked together with a variety of systems, from mundane dial-up phone lines to high-speed dedicated leased lines, satellites, microwave links, and fiber-optic links. The Internet is now such a vital communication link connecting not only research and academic communities, but also community colleges, small businesses, public libraries, high schools, and freenets internationally. As well, the growth of commercial Internet providers such as UUNET Communications Services, Sprint's Sprintlink, CERFnet (The California Education and Research Federation Network), and the WELL (Whole Earth 'Lectronic Link), among others, has enabled much broader access to the government-subsidized portions of the Internet.

It is estimated that the number of host machines with direct connection to the Internet's transmission protocol, TCP/IP (Transmission Control Protocol and Internet Protocol) has been doubling every year since 1988. TCP/IP refers to the set of services that allow host computer systems to connect to one another in order to perform the three basic TCP/IP functions: electronic mail, remote log-in, and file transfer. In 1983, TCP/IP became the standard protocol used by both the U.S. Department of Defense's Internet, including the ARPANET (Advanced Research Projects Agency), as well as the technical standard introduced by Sun Microsystems for commercial networking environments (Davidson, 1988, p. 6).

In September 1993, the Internet consisted of over 1.7 million computers and 15,000 nets (6000+ foreign) in over 130 countries. It reaches between 15 and 30 million people, and is estimated to grow by a million users each month (National Science Foundation, 1993; Hart et al., 1992).

Partly this rapid growth can be attributed to the development trajectory of the telecommunications and computer industries towards high-speed, packet-switched networking and their concomitant commitment to invest significant R & D monies in order to develop new switching technologies and network applications. Most importantly, however, the explosive growth of the Internet has been fuelled by the meteoric rise of its user base. Users rely on their daily Internet fix to exchange e-mail; to collaborate on research projects; to participate in multifarious bulletin boards and conferences available through automated fileservers and listservers (including receiving electronic journals); to conduct file transfers via anonymous file transfer protocols (FTPs); and to search hundreds of library catalogues, from the University of California's MELVYL system to those at the University of Pretoria.

CA*net

Currently, the majority of Canadians who can access the Internet do so through the backbone network, CA*net. Most of its user community is typically ensconced in universities, with a secondary user base composed of community colleges, government and industry groups, commercial enterprises, and a growing number of Kindergarten-to-Grade-12 (K-12) schools and freenets, such as the ones sprouting up in Ottawa, Victoria, and Vancouver.

The forerunner to CA*net was NetNorth, established in 1984. NetNorth was the Canadian equivalent to the U.S. BITNET organization (Because It's Time network, consisting of systems connected by point-to-point links running the IBM-based NJE protocol, and designed for the academic community). NetNorth established the underlying co-operative principle that guides Canadian networking today, whereby the regional networks form the basis for the national consortium.

In the mid-1980s, most large universities in Canada began projects to install high-speed networks, with the goal of establishing links between these networks and external networks, such as the NSFnet (National Science Foundation network) in the U.S.

Regional networks utilizing TCP/IP had been created in Ontario, Quebec, and Nova Scotia by 1988, and plans were implemented to link them together. Coincidentally, the National Research Council was developing plans for the creation of a high-speed national network, and its $2 million in start-up funding was one of the catalysts for the creation of CA*net. The University of Toronto, IBM, and INSINC (a value-added telecommunications reseller) became the organization that comprised CA*net Networking Inc. in October 1990.

CA*net's members are the regional organizations, including those in British Columbia (BCnet), Alberta (ARnet), Saskatchewan (SASK#net), Manitoba (MBnet), Ontario (Onet), Quebec (RISQ), New Brunswick (NB*net), and Newfoundland (NLnet). These regional networks operate as separate entities with their own by-laws, organizational structures, and financing. CA*net's goals and objectives are to support and provide networking connectivity amongst the regional networks; to facilitate and promote the development and use of networking services; and to represent CA*net membership to other related organizations and initiatives (CANARIE Associates, 1992, p. F1).

In order for Canadian networks to connect to the U.S., they must first make a link to the NSFnet, the national backbone network that is used to interconnect regional mid-level networks to one another. The NSFnet currently runs at T3 speeds (44.737 megabits per second) and is managed by Advanced Networks & Services (ANS), a not-for-profit corporation created by Merit, IBM, and MCI. ANS reports that international traffic exchange to the NSFnet is greatest from Canada (Australia, Germany, and Sweden being the other heaviest international users). As of February 1993, Canada had 374 networks linked to the NSFnet (NSF networks by country, 1993).

NREN Initiatives in the U.S.

Recent U.S. initiatives attempting to cope with the growth of data traffic on the Internet have important repercussions for Canada, both in the physical sense (connecting to the NSFnet) and in the social sphere (via articulations concerning an expanded user base, and the legal problematic of computer networking). Plans have been formulated to upgrade one of the major backbone networks constituting the U.S. portion of the Internet, the NSFnet. Often dubbed the "Interim NREN," the NSFnet backbone, which now is capable of transmitting 45 megabits per second (about 1,400 typed pages) will be upgraded to gigabit-per-second network speeds to allow for even more advanced applications and services (such as videoconferencing, multi-media applications, and complex information that can be generated by supercomputers, such as weather prediction models). As an example of what gigabit-per-second speeds entail, it is estimated that an entire encyclopedia could be transferred in less than three seconds (LaQuey & Ryder, 1992), or that its raised capacity of 3 billion bits would be the equivalent of accessing 300 copies of Moby Dick per second (Markoff, 1993).

The U.S. High-Performance Computing Act (HPCA) of 1991 (Public Law 102-194) which was signed into law in December 1991 and had been spearheaded by Al Gore (now U.S. Vice-President), provides legislation to co-ordinate the networking activities of all academic and research institutions and federal agencies (including those at the National Science Foundation, NSF, the National Aeronautics and Space Administration, NASA, the Department of Energy, DOE, and the Department of Defense, DOD) into one high-capacity, high-speed network. Part of the HPCA's five-year, U.S.$2.9 billion budget is the creation of the NREN (National Research and Education Network), which will essentially be the successor to the research and education portion of the Internet in the United States. As currently envisaged, the NREN will link together all academic, government, and industrial sectors to ensure that the United States can remain internationally competitive in high-performance computing. However, Gore vows to push his goals of universal NREN access through his "Son of NREN" bill, the Information Infrastructure and Technology Act of 1992 (IITA). Although the IITA failed to meet Congressional approval in October 1992, Gore has promised to reintroduce the bill, which would authorize a total of U.S.$1.15 billion over the next five years, specifically targeting networking initiatives for K-12 schools, public libraries, health care organizations, and industry, particularly manufacturing.

The IITA's proposed multi-agency Information Infrastructure Development Program, to be co-ordinated by the White House Office of Science and Technology Policy (OSTP), would assign different agencies responsibility for developing applications of networking technology in various areas. For instance, the NSF would be responsible for funding projects to connect primary and secondary schools, as well as co-ordinating the national computer network connecting hundreds of colleges and universities to the NSFnet. The Commerce Department's National Institute of Standards and Technology (NIST) would be given responsibility for developing networking technology for manufacturing, while the National Institute of Health (NIH), in conjunction with the NSF and other agencies, would develop applications of advanced computer and networking technology for health care. This would include connecting hospitals, doctors' offices, and universities so that health-care providers and researchers could share medical data and imagery, like CAT scans and X-rays. The IITA would also provide funding to both NSF and NASA to develop and design prototype technology for digital libraries--huge databases that can store text, images, video, and sound accessible through computer networks (Information Infrastructure Development Act, 1992).

NREN activities have been widely publicized and debated in the media and have spawned a variety of conferences (both real and virtual), actively engaging various interest groups who have a stake in ensuring that their networking interests are met. These include: library organizations such as the American Library Association (ALA) and the Association of Research Libraries (ARL); educational technology groups such as EDUCOM (a Washington, DC-based non-profit consortium of colleges and universities committed to examining the management and ethical uses of information technology in higher education) and the Coalition for Networked Information (CNI); civil liberties groups such as the American Civil Liberties Union (ACLU), Computer Professionals for Social Responsibility (CPSR), and the Electronic Frontier Foundation (EFF). A spirited debate has arisen in the U.S. regarding the conflict between privatization and commercialization of the Internet (Markoff, 1993); broader public access to the Internet through K-12, public-library, and freenet initiatives (Clement, 1991; McClure et al., 1991; Grundner, 1991); and the shifting legal status of networked environments, where notions of copyright, intellectual property, and privacy are being challenged (Branscomb, 1991).

Enter CANARIE

The Canadian networking community, influenced by U.S. NREN developments, has been formulating its own design and developmental strategy for high-speed networking. The Canadian archetype of the NREN, the Canadian Network for the Advancement of Research, Industry and Education (CANARIE), came into existence at the April 1991 Network Organization Conference sponsored by Industry, Science, and Technology Canada. CANARIE was given the mandate to establish plans for the creation of a high-speed network through the establishment of four working groups: Business, Governance, Marketing, and Network Architecture.

Because of the large number of diverse players in the CANARIE project, including the federal government, provincial governments, territorial governments, the existing CA*net organization, the regional network community, the information technology industry, Canadian common carriers, and the research community, CANARIE is quite ambitious. Essentially, CANARIE will be responsible for the backbone network and its nodes, and for services that it provides to the regional networks. The regional networks, in turn, will be responsible for services to the end-user organizations. As well, CANARIE will provide a test network capability for the use of the Canadian information technology industry and telecommunications research communities.

As stated in the December 1992 Business Plan, the objectives of CANARIE are fivefold. The first strategic priority is to upgrade the capabilities of CA*net, the existing educational and R & D backbone network, from its current 56 kilobits per second to T-1 speeds (1.544 megabits per second) and then finally to T-3 speeds (44.736 megabits per second). The second objective, in recognition of the major support of the backbone financing by regional networks, is to promote CANARIE services in conjunction with the regional networks. The third objective is to establish and operate a high-speed experimental network for use in the development and testing of next-generation networking technologies, applications, software, and services--in particular, facilities and hardware to upgrade the existing R & D and education backbone network to gigabit data-flows, and the development of a gigabit-test network for hardware and software products designed for gigabit networks. The fourth objective is to stimulate the development of new networking applications, software and services that can be sold domestically and internationally. Finally, CANARIE's fifth objective is to support the integration of these new networking technologies into the infrastructure of the operational networks (CANARIE Associates, 1992).

What kinds of networking applications and services will CANARIE concentrate on? Some of the areas for future development that they foresee include: distributed joint authoring (applications that allow multiple authors to work asynchronously, ideal for academic and industrial collaboration); distributed computing (the interaction of multiple computing resources that can process large and complex data sets); the access, monitoring, and operation of remote equipment (for instance, remote imaging instruments for scientific purposes); real-time remote conferencing ("tele-presence" projects which combine computer, audio, video, and audio-graphic material for educational and research/training purposes); multi-media mail and bulletin boards (to stimulate distance education and training); digital libraries (hypertext and hypermedia libraries); and electronic document creation and related library services based on international standards (to aid in interlibrary loans, information retrieval, document creation and delivery) (CANARIE Associates, 1992, p. 10).

Initial plans are for CANARIE to be funded by government, through Industry, Science, and Technology, and through private-sector initiatives, primarily those featuring a co-operative plan engineered between Stentor Canadian Network Management and Unitel Communications Inc., both major Canadian common carrier organizations.

Stentor, previously called Telecom Canada, is the alliance of nine major telephone companies (AGT, B.C. Tel, Bell Canada, Island Tel, Manitoba Telephone System, Maritime Tel & Tel, NBTel, Newfoundland Telephone, and SaskTel) and Telesat Canada. Stentor operates as an unincorporated association that co-ordinates network operations, markets products and services, and distributes revenues from jointly provided services. It has a 41.6% stake in the domestic satellite communications system, Telesat Canada, which leases satellites to broadcasters, cable television operators, and other members and smaller firms. In addition, Telesat's services are fully integrated with the public switched telephone networks (Communications Canada, 1992a).

Unitel Communications Inc., jointly owned by Canadian Pacific Ltd. (48%), Rogers Communications Inc. (32%), and AT&T (20%), operates its own microwave and fiber-optic relay system and switching centres, but generally leases loops from local telephone companies, with interconnection rights to the local exchange facilities of Bell Canada, B.C. Tel, AGT, NBTel, Maritime Tel & Tel, Island Tel, and Newfoundland Telephone. As well, because of the June 1992 CRTC decision deregulating the long-distance telephone market, Unitel may now compete directly with the local networks of Bell Canada in Ontario and Quebec, B.C. Tel, and the four phone utilities in the Atlantic provinces (Surtees, 1992).

The founding members of CANARIE Associates also include a wide range of private stakeholders in the computer and information services industry, such as Digital Equipment of Canada, Inc., Gandalf Technologies Inc., Hewlett-Packard (Canada) Ltd., IBM Canada Ltd., Newbridge Networks Corporation, Northern Telecom Canada Ltd., Positron Industries, Inc., and Unisys Canada Inc.

Such a preponderance of major-league stakeholders in the CANARIE project illustrates how important advanced networking technology is perceived for both domestic growth and international credibility. Canada's phone companies, which own and maintain the nation's voice communications network at a healthy profit, will stand to gain handsomely if they primarily own and manage CANARIE. Other private concerns that stand to reap profits are AT&T, also involved in the NREN (through their stake in Unitel), and the cable industry, as represented by Rogers.

However, will the quest for the competitive edge lead to more commercialization of CANARIE? In the U.S., commercialization of Internet services is a highly charged issue, and refers to allowing data traffic of a commercially oriented nature (anything not concerned with R & D or educational enterprises) to pass through the federally-funded backbone, i.e., the NSFnet. CA*net's current acceptable usage policy is "deliberately vague on the subject of non-restrictive use" (CANARIE Associates, 1992, p. F4) and CANARIE will be faced with formulating acceptable use statements that please both the R & D and academic community, as well as private investors.

Widening CANARIE Access

So far, there has been little public discussion and debate on CANARIE, aside from those in the academic, industrial, and government sectors. For instance, the media might mention in passing the need to create an "electronic superhighway" but CANARIE has not become a household name.

As currently envisaged, the immediate beneficiaries of CANARIE will be the research and academic communities. There is, however, a growing public that will most certainly demand to have access to CANARIE resources more quickly than the current CANARIE Business Plan foresees (Surtees, 1992, p. 33). These include libraries, freenets and other community networks, local bulletin board systems connected through FidoNet, and non-profit networks, such as The Web.

Libraries

Surprisingly, the library community in Canada has not yet become an organized participant in the CANARIE discussion, as compared to their U.S. counterparts. The possibilities that high-speed networking create for libraries has already been alluded to: the creation of virtual libraries, faster document delivery services, increased access to larger and more diverse databases and research centres, and the maintenance of electronic journals. As well, some commentators believe that CANARIE

may be the way for libraries to respond to the many profound changes in society stemming from the convergence of computer and telecommunications technologies.... CANARIE gives libraries the means to meet the new challenges associated with the changing nature of scholarly communication and the growing importance of electronic publishing. (Silva & Cartwright, 1992, p. 12)

Freenets

Freenets are free public-access computer systems for the exchange of information between members of local and global communities. The first freenet was started in 1986 at Case Western Reserve University Medical School in Cleveland, Ohio, originally a small BBS for family doctors. However, a growing public demand for its resources led to the creation of a community system that now services over 22,000 registered users averaging between 4,500 and 5,000 logins a day. The network offers more than 300 information and communication features, such as e-mail; medical information in the Medical Center; legal information in the Courthouse; the Teleport, where users can skip to computer systems in the U.S. and abroad; Project Hermes, containing the full text of U.S. Supreme Court decisions; the NPTN/U.S.A. Today News Center, which provides the user with a free electronic version of the U.S.A. Today paper; and Academy One, a K-12 electronic school house. In order to establish as many community computer systems throughout the world as possible, the software that operates the freenet is available, on a lease basis, for $1 a year. Each freenet system is made an affiliate of the National Public Telecomputing Network, whose next priority is the development of "cybercasting" services, whereby "a wide variety of quality news and information features will be provided to the affiliates via NPTN feed--a concept very similar to that of any radio or television broadcasting network" (Grundner, 1991).

The National Capital FreeNet in Ottawa (NCF) was established in 1992 as a non-commercial, co-operative, community project with the active participation of volunteers, institutions such as Carleton University, and organizations such as Gandalf Technologies (which donated modems and the communications equipment for connecting the freenet to the Ottawa Public Library), and Sun Microsystems (which donated a SparcStation 10).

The NCF includes a wide range of resources, from a post office where users can send e-mail to other freenet members or the Internet community at large, to access to Usenet (a multifarious collection of diverse newsgroups distributed via the Internet), to a general public discussion group, where the topics run the gamut from pizza to privacy. Specialized discussion groups include ones devoted to social services, health, and the environment (community organizations such as Immigrant and Refugee Services, the Ottawa Chinese Community Service Centre, the Canadian Global Change Programme of the Royal Society of Canada, and committees devoted to AIDS and Alzheimer's Disease are listed). Community associations are given space for their own discussion groups. The Government Centre provides information regarding municipal and regional associations and councils, and federal government departments such as Communications Canada. The Science, Engineering and Technology Centre lists local museums, institutes, outreach programs, university resources and associations such as Women in Engineering and Science. The Daily Newspaper conference includes selections from a variety of sources, such as The Ottawa Citizen, Statistics Canada Daily Newsreleases, Radio Free Europe/Radio Liberty Daily, Eastern Europe Report, China News Digest, South Asia News, Croatia News, Cuba News, New Liberation News Service, Health InfoCom Network News, and travel advisories and daily weather forecasts. Libraries are represented in the freenet by postings of local library resources. Special interest groups are eclectic, including those devoted to computers, media, arts, teaching, sports, dogs, librarians, home beer- and wine-making, astronomy, and mental health.

Other freenets are developing: one is up and running in Victoria (the Victoria Free-Net Association) and committees are underway to establish others in Vancouver, Toronto, and other sites. As well, there has been discussion of establishing a national body similar to the NPTN, whose goals will be to liaise with NPTN and other international counterparts.

FidoNet

FidoNet, developed in 1984, is a point-to-point and store-and-forward e-mail WAN using modems on the direct-dial telephone network. Although originally designed for MS-DOS hosts, it has since been ported to environments ranging from UNIX to the Apple II. Gateways exist from FidoNet to the Internet, usually via the uucp (unix-to-unix copy program) network. FidoNet now claims over 13,000 public nodes worldwide (Bush, 1992). In the Ottawa region alone, FidoNet has grown to over 152 nodes hosted by three local networks, representing an average growth of approximately 17 nodes per year (Dodd, 1992).

FidoNet's user base is predominantly located in North America (62%), although Europe, Australia, New Zealand, Asia, Latin America and Africa have their own nodes. The technology is used in both private and public settings. Publicly, FidoNet is available within large corporations, such as AT&T, Georgia Pacific, and Canada Post; through public institutions; and through non-governmental organizations (NGOs), particularly in Africa. Privately, FidoNet users are mostly hobbyists and public BBS users in North America who have a predominant interest in computer hardware and software, gaming, and science fiction.

FidoNet connectivity for local BBSs ensures the most publicly accessible and lowest-cost e-mail and e-news services. Some BBSs are only usable by a single dial-up caller at a time, while others run multi-line systems ranging from 2 to 20 lines. Bush (1992) estimates that the average FidoNet BBS has over 200 active users, 50% of whom use e-news and 5% of whom use private e-mail. "As not all FidoNet nodes have BBS access, we can estimate that on the order of 1,000,000 FidoNet users read or write e-news and on the order of 100,000 of whom use private e-mail" (Bush, 1992).

The Web: "Dial Locally and Act Globally"

The Web, located in Toronto, is a computer network operated by Nirv Community Resource Centre, a non-profit organization that assists other non-profit groups in acquiring and using information technology. The Web is one of the four main nodes for the Association for Progressive Communications (APC, established 1990), the world's largest computer communication network dedicated to promoting human rights and environmental causes. The other APC nodes are located at London (GreenNet), Stockholm (NordNet), and San Francisco (IGC Networks, including PeaceNet, EcoNet, HomeoNet, and ConflictNet). These nodes collect the communication flow from regional nodes (located in the U.S., Nicaragua, Brazil, Russia, Australia, the U.K., Canada, Sweden, and Germany, with affiliated systems in Uruguay, Costa Rica, Czechoslovakia, Bolivia, Kenya, and Cuba), where the messages are then exchanged and distributed internationally. In turn, these messages can be sent to commercial and academic hosts. Then, the entire APC system is conveyed onto the Internet through the IGC Networks, which are a full Internet host.

The APC's membership consists of more than 15,000 subscribers in 90 countries, including many NGOs such as Amnesty International, Friends of the Earth, Oxfam, Greenpeace, and various labour unions. APC carries a variety of alternative news sources, including the InterPress Service; Environmental News Service (Vancouver), the United Nations Information Centre news service; Agencia Latinoamericana de Informacion (Ecuador); Alternet (Washington, DC); New Liberation News Service (Cambridge, MA); Pacific News Service (San Francisco); and World Perspectives Shortwave Monitoring Service (Madison, WI) (Frederick, 1993).

Through the APC, Web users (of which there are approximately 1,400 individuals and organizations) can access over 800 conferences on a range of issues related to the environment, peace, human rights, and social justice. The Web demonstrated its activist stance against the U.S.-Canada Free Trade Agreement (FTA) in the late 1980s when a conference, web.freetrade, was opened with the participation of various groups, such as the Conservation Council of New Brunswick, the Canadian Environmental Law Association, the Canadian Environment Network, Energy Probe, Society of Gaia Principles, and the Ontario Public Interest Research Group.

Not surprisingly, such activism has continued with the North American Free Trade Agreement (NAFTA) debates. Some of the active participants in the U.S. include the Institute for Agriculture and Trade Policy, a Minneapolis-based NGO, and Mobilization on Development, Trade, Labor and the Environment, a coalition of over 200 environmental, labour, human rights, development, religious, and agricultural organizations. One of the founding members of the U.S.-Mexico-Canada Labor Solidarity Network is Labor Notes, a national monthly labor magazine with articles focusing on current trends in the labor movement and reform efforts. Mexican NGOs are represented by the Red Mexican de Acción Frente al Libre Comercio, also known as the Mexican Free Trade Action Network, composed of labour, women, peasants, and environmentalists from Mexico City and Northern regions, and Servicios Informativos Procesados, a Mexico-City based information centre which promotes communication between Mexico, Canada and the U.S. (Frederick, 1992).

The WEB is also the only network in the world with a direct connection to Cuba. In turn, Cubans can, through their academic and scientific network Cenial, connect to the world at large through the Web connection. The Canadian government does not fund this link, which annually costs $20,000 for mostly telephone hook-ups, and which is derived from private donations (Valdes, 1993).

Legal Issues in Networking

Networking's omnipresent nature has created new legal challenges regarding aspects of copyright, intellectual property, censorship, and privacy. These perplexing issues have been hashed out for several years in the United States by some groups such as the ALA, ARL, EDUCOM, ACLU, EFF, and CPSR. Constitutional expert Laurence Tribe has made suggestions towards updating the U.S. Constitution in light of computer technologies (in particular, the First, Fourth and Fifth Amendments which guarantee freedom of speech, the right to privacy, and protection from unwarranted governmental intrusion, respectively) (Tribe, 1991). Other laws, such as the Computer Fraud and Abuse Act of 1986, the Electronic Communications Privacy Act of 1986, and the Revised Copyright Law of 1976, are being disputed as to their lack of technological currency. As well, specialized conferences, such as the very successful Computers, Freedom, and Privacy (its third was held in San Francisco in March 1993), have created a healthy atmosphere for discussion, debate, and action.

Public debate regarding similar legal issues in Canada, particularly with the CANARIE initiatives, has been, so far, woefully quiet. The exception would be the 1992 CA*net conference, where the question of censorship of Usenet newsgroups, such as the alt.sex hierarchy within academic institutions, was a hot issue. However, given the recent telecommunications privacy initiatives released by the Department of Communications in December 1992, Canada finds itself at an opportune moment for creating an expanded privacy platform for computer networking. There are plenty of laws and proposed legislation, such as the U.S. Electronic Communications Privacy Act of 1986, the European Community's draft Telecom Directive, as well as suggestions for NREN privacy guidelines proposed by CPSR, which could be adapted for Canadian legislation.

Privacy Platforms

In the U.S., the 1986 Electronic Communications Privacy Act (ECPA) protects messages created by digital technologies, data technologies, video communications, and electronic mail. Although similar in protective scope as that afforded to telephone usage under the Communications Act of 1934, which forbids interception and disclosure of information transmitted by that medium, it raises serious privacy issues in the contemporary environment. For instance, under the ECPA, governmental and law enforcement agency requests for information are easily facilitated, and there is no provision made for individuals to challenge or correct the information. As well, this expedites the patchwork creation of personal information on an individual, possibly for targeting for governmental access procedures, as well as allowing the service provider to divulge this same information to non-governmental entities or people.

However, it is in the transborder nature of Internet traffic where the notion of privacy becomes slippery. The legal impact of such international flows has yet to be tested, and the repercussions could be mind-boggling in a decentralized environment that is hard to regulate and manage, and that is further complicated by differing and often conflicting national laws.

Computer Professionals for Social Responsibility (CPSR) in the U.S. has co-ordinated several national efforts to promote privacy protection for network communication, in the belief that the future of network communications depends on the ability to ensure that there is adequate privacy protection for all network users. Because the Federal Communications Commission (FCC), the Computer Science and Technology Board (CSTB) of the National Research Council, and the Office of Technology Assessment (OTA) have been unwilling to address the privacy implications of new network services, CPSR has advanced the idea that there should be a data protection board in the U.S. to assume the task of developing and promulgating privacy principles for network services.

Marc Rotenberg, Washington Director of the CPSR, has noticed that there is a considerable lack of privacy protection within NREN policy. CPSR anticipates three types of privacy problems that the NREN will have to contend with (Rotenberg, 1992). The first problem is that, given the probable commercialization of the NREN, existing privacy problems will be aggravated. For instance, there will now be more incentive to gather personal data through credit-card and telephone-call transactions, which can then be resold by commercial organizations. Address files for the multitude of list servers could be sold, creating unsolicited subscriptions for individuals.

The second problem is that existing legislation, such as the ECPA, is inadequate with respect to NREN policy. Under the ECPA, law enforcement agencies may require that Internet service providers must release records of electronic communications to the government, as well as assisting in the execution of warrants. This is worrisome given that the ECPA has already been criticized for leading to a regression of privacy rights, because "the increased number of crimes which now justify interception [and] the lowered standards of cause required ... constitute an unjustified Orwellian compulsion to prevent crime and possess information" (Burnside, 1987, p. 516). Perhaps the two goals of ensuring secure communication and government oversight are inherently incompatible.

The third problem is that technical safeguards for privacy deliver merely partial solutions. Technological fixes towards privacy protection, including new techniques in cryptography, such as privacy enhanced mail (PEM) provide ways to protect the content of an electronic message as well as the identity of the message author, but are fraught with controversy (Loen, 1993). Should cryptographic techniques be encouraged and widely disseminated? Should all networks be forced to include encryption of mail as a default standard available to all users? Should law enforcement agencies have access to encryption devices and protected gateways (Denning, 1993)?

CPSR has formulated eight privacy principles for electronic networked communications, which emanate both from existing codes and particularly borrow from the principles developed by the Organization for Economic and Cooperative Development (OECD) in 1981 (Rotenberg, 1992).

Their first recommendation is to ensure that electronic communications remains confidential and protected. Second, they wish to ensure that privacy considerations are recognized explicitly in the provision, use, and regulation of telecommunication services, and that users are informed about the privacy implications of these various services. Third, they recommend that the collection of personal data for telecommunication services should be limited only to the extent necessary to provide the service. Their fourth recommendation is that information should not be disclosed by service providers without the consent of service users. Their fifth recommendation is that routine privacy protection should be free. Their sixth recommendation is that technical means to protect privacy, such as encryption, should be encouraged by service providers. Their seventh recommendation is the development of appropriate security policies. Finally, their eighth recommendation is that a mechanism akin to a privacy protection board should be established to ensure the observance of privacy principles (Rotenberg, 1992).

Suggestions for Canadian Privacy Legislation for Networking

Despite the shortcomings of the ECPA, at least it affords the individual a modicum of privacy protection in the electronic environment. Canada has no law on the books similar to the ECPA, although privacy scholars such as David Flaherty advocate a Canadian counterpart (Flaherty, 1992, p. 104).

One of the crucial points made by the European Community's draft Telecom Directive regarding future network development was that effective privacy protection was developing into an essential prerequisite for social acceptance of new digital networks and services. Flaherty recommends that these directives should be consulted by Canadian regulatory agencies when they plan their equivalent:

A most important trait of the Telecom Directive ... is the importance it attributes to privacy and data protection in the new digital telecommunications environment. North Americans are consistently surprised that Europeans take privacy protection so seriously; in their judgment, one cannot open up telecommunications markets without action to protect the personal data involved too and to harmonize standards for privacy protection within the member countries. Thus the EC Telecom Directive has important implications for what should be done in a small country like Canada in particular. (Flaherty, 1992, p. 75)

Although the 1992 Telecommunications Privacy Principles are a step in the right direction, they remain only principles, and as of yet, are not established in any legal framework (Communications Canada, 1992b). Here, the emphasis on telecommunications privacy is towards the various new telephone technologies (in particular cellular telephones). In light of the proposed CANARIE initiatives, it would seem advantageous to extend these privacy principles to incorporate networked communications by adapting aspects of the ECPA, the EEC Telecom Directive, or the CPSR privacy principles.

Conclusion

It is indisputable that CANARIE's goals of creating a high-speed computer network contain extraordinary possibilities, with applications as exciting as distributed computing and tele-presence projects, to the maintenance of large digital libraries. As well, the possibility exists that CANARIE will stimulate the domestic economy and confer a high level of credibility within the international arena.

It would appear that to date, however, CANARIE has borrowed only the technical spirit, and not the social or legal tone, of its counterpart, the NREN. The lively debate in the U.S. now regarding NREN access and policy issues should galvanize Canadians to consider how they will address such important and fundamental issues. What are the implications of a predominantly privately owned network? Will this increase the commercialization of networking resources? Will Canada's heterogeneous networking community, including K-12 schools, non-profit organizations, freenets, local BBSs, and public libraries have access to CANARIE resources, or will networking still remain within the prevailing provenance of academia and industry?

The legal terrain of networked communications remains elusive, and the problems of censorship, copyright, and privacy need to be addressed before CANARIE's inception. Canada is now well primed to consider expanding its recent telecommunications privacy principles towards computer networking. As Rotenberg reminds us, regardless of the future outcome of technological guarantees towards privacy on the networks, "the protection of privacy is ultimately a policy decision that must be resolved through our political institutions" (Rotenberg, 1992).

References

Branscomb, A. W. (1991, September). Common law for the electronic frontier. Scientific American, 265(3), 154-158.

Burnside, R. S. (1987). The Electronic Communications Privacy Act of 1986: The challenge of applying ambiguous statutory language to intricate telecommunications technologies. Rutgers Computer and Technology Law Journal, 12, 451-517.

Bush, R. (1992). FidoNet(tm): Use, technology, and tools. (Available via GOPHER.)

CANARIE Associates. (1992, December). Canadian Network for the Advancement of Research, Industry and Education Business Plan. (Anonymous FTP to unbmvs1.csd.unb.ca: pub/net /CANARIE.)

Clement, J. (1991, August). Educom K-12 networking project. Matrix News, 1(5). (Available from mids@tic.com.)

Communications Canada. (1992a). Telecommunications in Canada: An overview of the carriage industry. Ottawa: Minister of Supply & Services Canada.

Communications Canada. (1992b, December). Telecommunications Privacy Principles. Ottawa: Minister of Supply and Services Canada.

Davidson, J. (1988). An introduction to TCP/IP. New York: Springer Verlag.

Denning, D. E. (1993, March). To tap or not to tap. Communications of the ACM, 36(3), 24-44.

Dodd, C. A. (1992, March). What is a FidoNet: Discussion of the growth and development of an amateur computer network. Carleton University Working Papers in Public Access Networks, #9. (Anonymous FTP to alfred.carleton.ca /pub/freenet /working.papers.)

Flaherty, D. H. (1992, May). Telecommunications privacy: A report to the Canadian Radio-Television and Telecommunications Commission.

Frederick, H. (1993). Computer networks and the emergence of global civil society: The case of the Association for Progressive Communications (APC). In L. Harasim (Ed.), Global networks: Computers and international communication, pp. 283-295. Cambridge, MA: MIT Press.

Frederick, H. H. (1993, March). North American NGO Computer Networking on Trade and Immigration: Computer Communications in Cross-Border Coalition Building. DRU-234-FF (draft), RAND, Santa Monica, California.

Grundner, T. (1991, September/October). Free-Nets: Networking meets Middle America. Link Letter. (Available through nptn Gopher server.)

Hart, J. A., Reed, R. R., & Bar, F. (1992, November). The building of the Internet: Implications for the future of broadband networks. Telecommunications Policy, 16(8), 666-689.

Information Infrastructure and Technology Act: S. 2937. (1992, July). (Anonymous FTP to nic.merit.edu: /nren/iita.1992/gorebill.1992.txt.)

LaQuey, T., & Ryder, J. (1992). The Internet companion: A beginner's guide to global networking. Rockport, MA: Addison-Wesley.

Loen, L. (1993, January 7). Hiding data in plain sight: Some key questions about cryptography. EFFector Online, 4.05. (Anonymous FTP to eff.org.)

Markoff, J. (1993, January 24). Building the electronic superhighway. New York Times, Sec. 3, p. 1.

McClure, C. R., et al. (1991). The National Research and Education Network (NREN): Research and policy perspectives. Norwood, NJ: Ablex.

National Science Foundation. (1993, September). NSFnet Statistics. (Anonymous FTP to nic.merit.edu: statistics/nsfnet /history.netcount, nets.by.country.)

NSF Networks by country. (1993, February 28). (Anonymous FTP to nic.merit.edu: /nsfnet /statistics/nets.by.country.)

Rotenberg, M. (1992, July 21). Proposed privacy guidelines for the NREN. Open forum on library and information service's roles in the National Research and Education Network (NREN). Washington, DC: National Commission on Libraries and Information Science (NCLIS). (Anonymous FTP to cpsr.org: cpsr/nren/nren-privacy-testimony.txt.) Alternative source: Rotenberg (1993).

Rotenberg, M. (1993, July 7). Proposed privacy guidelines for the NREN. The third CPSR cryptography and privacy conference sourcebook. Washington, DC: CPSR. Alternative source: Rotenberg (1992).

Silva, M., & Cartwright, G. F. (1992). The Canadian Network for the Advancement of Research, Industry, and Education (CANARIE). The Public- Access Computer Systems Review, 3(6), 4-14.

Surtees, L. (1992, June 13). Bell to share long distance feeling. The Globe and Mail, p. 1.

Tribe, L. (1991, April). The Constitution in cyberspace: Law and liberty beyond the electronic frontier. Keynote Address to the First U.S. Conference on Computers, Freedom and Privacy (Anonymous FTP to ftp.eff.org.)

U.S. Congress, Senate. (1991, December 9). Public Law 102-194. High-Performance Computing Act of 1991. 102nd Congress, 1st sess., S. 272.

Valdes, A. (1993, March). E-mail bonding. Voice Literary Supplement, p. 14.



  •  Announcements
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Current Issue
    Atom logo
    RSS2 logo
    RSS1 logo
  •  Thesis Abstracts
    Atom logo
    RSS2 logo
    RSS1 logo

We wish to acknowledge the financial support of the Social Sciences and Humanities Research Council for their financial support through theAid to Scholarly Journals Program.

SSHRC LOGO