Internet

 The Internet is an ever-expanding aggregation of hierarchically interconnected networks that allows easy transfer of information between millions of computers spread across the whole world. On the Internet, "host" computers communicate to one another using the standard TCP/IP protocol over dedicated TELECOMMUNICATIONS lines.

In addition to the million or more such hosts, many millions of computers are directly attached through private local networks (eg, in offices) using various protocols, and untold millions dial into hosts via ordinary telephone lines. If the protocol of the private network or dial-in call is TCP/IP and appropriate software is installed on the user computer (notably a form of SLIP or else PPP for dialed-up connections), then the user is regarded as being directly on the Internet and has access to all facilities. Host computers are owned and operated by universities, corporations and government agencies and for the general public by co-operatives and specialty commercial companies; in this role of providing Internet access, each of these organizations is referred to as an "Internet Service Provider" (ISP).

History

The Internet had its origins in the ARPANet, developed in the 1970s by the US defence department's Advanced Research Projects Agency. Through ARPANet, the department wanted to explore the possibility of establishing a communication network that could survive a nuclear attack. Desirably, communication would continue between networked computers, even if one were shut down or its connection were severed. At first ARPANet connected 4 sites, but very quickly universities, industrial and government laboratories and offices were added, connecting computers involved in pertinent research contracts. The US National Science Foundation established 6 supercomputer centres in the 1980s, initiating what became the Internet for research and education, onto which others have grafted parallel networks of more general purpose.

Structure

Whereas in the US the top level includes several different networks, there is generally just one backbone network for each country; for Canada this is CA*net. Initiated and significantly funded by the NATURAL SCIENCES AND ENGINEERING RESEARCH COUNCIL but developed in conjunction with universities and industry, this network serves as a trunk connecting separate regional and other networks across the country. It has multiple high-capacity cross-connections to the US networks and thereby to other networks around the world.

At the next level is a network for each province, ranging from the giant ONet to the modest PEInet, plus networks for major corporations and institutions. Lesser networks then connect to the relevant provincial network.

Each host computer has a unique numeric identity - its IP address; akin to a world telephone number, it consists of 4 successive numbers separated by periods (dots), descending through the layered networks. As with the telephone network, this numbering scheme is reaching saturation, causing various compromises and demanding its revision.

To ease use, each host also has a unique alphabetic identity consisting of 2 or more names separated by dots, listed in ascending form similar to a mailing address - toronto.cbc.ca being a familiar example. These names do not necessarily equate with networks. They are called "domains," the top-level one being either the 2-letter International Standards Organization (ISO) code for the country or else one of the 3-letter codes assigned to the US (eg, net, com, gov and edu). A host may have a .ca address though be connected to a US network, or have a .com address though be connected to CA*net.

The address is extended down through lesser networks to identify the multitude of local network computers to which personal ones normally connect, that extension adding progressively to the front of the alphabetic address. These extensions usually correspond exactly to the layers involved, except that "aliases" can jump over network names within one organization. The assignment of numbers and of names is strictly controlled, allowing compact yet meaningful unique names to be used. Overall control is vested in the Internet Network Information Center (InterNIC), run by a consortium of US companies, but it is progressively decentralized down the network hierarchy and down the domain hierarchy.

Technology information travels between host computers mostly via dedicated circuits provided to the individual institutions by telephone companies, each circuit typically moving millions of bits per second. Every network is a complex of circuit links controlled by computers, allowing multiple alternative routes between any 2 points. Transmission is in packet mode: the typical message travels as many independent fragments, which are all reassembled at the target host. Each packet travels its overall route in a succession of relay links, with error-checking and formal acknowledgment (or else automatic retransmission) over each link. Thus communication is of multiple store-and-forward style; however, because of the efficiency of the protocol, the speed of the computers and the high-capacity circuits used, worldwide end-to-end travel times for a complete message can be only a few seconds. (This is well illustrated by software that, by transferring digitized sound through the Internet, provides voice conversation close to toll-call quality.) Multiple routing arrangements and a lack of central controlling computers accord maximal reliability to the overall Internet.

Coaxial cables and terrestrial microwave systems provide many of the circuits, but the relentless pressure to enhance transmission capacities favours fibre-optic cables; the store-and-forward style also allows use of SATELLITE transmissions, with their inherent half-second delay. Major enhancement of the technology for CA*net was the first priority for CANARIE (Canadian Network for the Advancement of Research, Industry and Education), established between government, industry and educators in 1993 "to support the development of the communications infrastructure for a knowledge-based Canada, and thereby contribute to Canadian competitiveness in all sectors of the economy, to prosperity, to job creation and to our quality of life." CANARIE is a private, not-for-profit organization supported by Industry Canada, whose mission is to accelerate Canadian Internet development and use by assisting the development of Canada's communications infrastructure, and to stimulate the development of products and services and their applications.

File Transfer and E-mail

Originally, the underlying network developments provided for the transfer of data files and programs between computers and the delivery of electronic mail-like messages. The latter, under the rubric e-mail, has been the sole purpose of the Internet for probably a majority of its users.

Explicit file transfer of data and programs is of continuing importance to scientists and other technical users, but the burgeoning usage is of implicit file transfer of binary files representing text and pictures and, increasingly, of voice and other sound. This trend relates mostly to home computers, millions of which are able to store and display pictures of high quality.

With the digitization of television following that of voice, this could now extend to moving pictures. However, the capacity of most local telephone lines is only about one-hundredth that needed for moving pictures. Standard coaxial television cables obviously have such capacity, but they are designed for signals that are both one-directional and shared. While special modems and adjusted networks allow some use of these cables for better Internet access, and endless improvement in the sophistication of data transmission edges upwards the capacity of the telephone wires, the achievement of the full potential of the Internet and the widespread realization of an information superhighway require totally new distribution networks. These networks will likely use fibre optics and purely digital transmission. They will also encompass both television and telephony, and will thereby radically change the long-established national policy of separating those 2 technologies.

Using the Internet for any purpose requires the execution of an appropriate client program; this is typically on the user's host computer, but it can be on the user's personal or local network gateway computer if it has special software (eg, PPP). Eudora and FTP are common examples, being applicable to e-mail and file transfer respectively. Telnet allows execution by the user of a program running at a remote computer within the Internet, including client programs not locally available.

Explicit file transfer and e-mail typically occur between established correspondents (though hundreds of mailing lists allow any user to elect to receive anything entered on a given topic), but the growing character of the Internet is for the storage of information intended to be accessed openly by the public. Weather forecasts and general news, calendars of courses in universities, catalogues of books in libraries, commercial product information and promotional material of every type is appearing in files accessible through the Internet. The source computer and the addressing labels of its various contents may be known through other means, but the vast expansion of this aggregation of knowledge on-line is proving to be its own handicap to access.

Searching the Internet

Various programs have been developed to assist users in finding the specific information they need. Archie, created at McGill University, and Hytelnet, created at the University of Saskatchewan, are examples of widely used search software: the former searches across the Internet for files with names containing the submitted query phrase; the latter, oriented to library searching, contains layers of prepared directories (menus) to libraries and databases.

The menu-driven Explorer and Netscape, searching by subject through the World Wide Web (WWW), are perhaps the most widely used. Characteristically, each source on the WWW has a lead directory called a home page, which can lead the user through many layers and into thousands of pages of information, at the same site or at any other site. The information technology that allows the limitless interconnection of pages around the world is called hypertext; its transmission protocol is indicated by the string http//: that has become a familiar part of life. OpenText of Waterloo, Ont, provides software that continually monitors the WWW, analysing descriptive texts of databases to maintain searchable indexes.

Hypertext added pictures to the plain text that had initiated the Internet, but it still has static 2-dimensional messages and major limitations on detail. Newer information technologies are allowing full publishing quality to be transmitted, bringing specifications of 3-dimensional objects that the host computer can manipulate, and incorporating dynamics into the messaging; eg, automatic updating of changing current information, like stock prices.

Costs

The costs of all the computers and communication lines providing the Internet is prodigious, but the marginal cost of any one communication through it is minute. Since the cost of recording each communication would be a multiple of that marginal cost, any charging must be on a broader basis. Universities and other major organizations have contributed on a bulk basis to the development of CA*net; typically, each pays a collective annual fee that is embedded in its financial overhead, giving the misleading impression that usage is free. Many early network providers were funded in a similar manner and allowed public usage without charge, but these, often with the sub-domain name Freenet, are increasingly rarely free of charge. The proliferating commercial ISPs usually have a base monthly charge plus a charge for extra connect time through their dial-up modems.

Aside from the costs of owning the equipment and using the services of an ISP, additional costs may be incurred by accessing the Internet. The Internet's expansion has surpassed the realm of communications to include a cyber marketplace, where almost everything can be purchased on-line. Users can shop for clothing, sporting goods, computer equipment, even homes and automobiles, from their homes. Increasingly, web sites require paid subscriptions before allowing access to them, and the user is bombarded with advertising from the moment of logging on.

SchoolNet

Recognizing the immense potential for enhancing education generally, governments, industry and educators in Canada established SchoolNet in 1993 to promote the effective use of information technology by helping Canadian schools and public libraries connect to the Internet. In March 1999 Canada became the first country to have every public library and school (18 263 in total) plugged into the Internet. The SchoolNet program is intended to put every one of Canada's 250 000 elementary and secondary classrooms on-line by 2001. The project is funded by the federal and provincial governments, school boards and corporations, and is not a network itself but a facilitator.

Supporters of the project argue that the Internet makes information more accessible and helps equip students to function in a knowledge-based society. Critics worry that computers in the classroom will interfere with teaching and learning and that the cost of providing computer equipment will over-stress dwindling school budgets, forcing schools to lay off teachers or to save money by not buying updated textbooks. Critics contend that cash-strapped school boards may have to opt for corporate support.

USENET

USENET comprises a collection of hosts receiving network news groups, which are discussion groups or forums on a variety of topics. Network news is a mechanism that broadcasts messages from one host to many hosts, with the message or article sent to a news group being received by a host on USENET. Users accessing the USENET hosts can read the messages, post new ones or reply to particular messages. The messages are archived and are accessible on-line; the vast collection of articles posted daily makes a significant contribution to the expansion of on-line resource accumulation.

Growth

The Internet is growing at a staggering speed; the number of attached networks more than doubles each year, and the number of attached computers and the volume of traffic appear to double every 3 months. For example, in 1988, 58 organizations were registered with CA*net; by May 1996, over 7500 organizations were registered. The number of Internet users increases so rapidly that it is virtually impossible to determine how many people have access to the Net; Internet statistics become obsolete almost at the moment they are calculated.

Implications

The socioeconomic impact of the Internet is already immeasurable, yet has only begun. Its openness offers cheap global communication on a totally unprecedented scale, with all the benefits that it can bring to understanding across the global village and to the development of the individual. By allowing rapid transfer of voluminous textual, pictorial and acoustic information, it could prompt a second industrial revolution commensurate with the first, but reversing it in the sense that it decentralizes rather than centralizes production, leading towards a home-based economy.

The openness of the Net has serious implications, on an Orwellian scale, for the privacy of the individual. Every piece of information a person submits on-line is traceable; the information is available to be tracked, logged, bought and sold. Corporations can obtain information about an individual's preferences in a variety of areas (eg, food, clothing styles, hobbies), income level, debt and state of health and target advertising directly to that individual via e-mail. To combat the invasion of privacy, a Montréal-based company, Zero-Knowledge, has developed Freedom software, which hides a user's web trail by encrypting every communication the user sends, stripping it of all data that would identify the user. However, such software has implications for crime detection, and even national security. Virus makers, pedophiles, terrorists and other cyber miscreants would be untraceable, making the jobs of computer crime investigators even more difficult.

By enhancing 2-way communication within the political realm the Internet could greatly alter our concepts of democracy, while its irrepressible nature can help suppressed people obtain freedom and democracy. However, it facilitates just as easily the promotion of hate, the proliferation of pornography and the dissemination of information on methods of terrorism. ARPANet and its early successors for research and education had well-defined rules for the content and purpose of any information transmitted through them. The evolved Internet is necessarily without such rules, and its vastness makes it difficult to have any. The challenges of hate and other antisocial propaganda are only beginning to be addressed by governments.

See also COMPUTER COMMUNICATIONS; INFORMATION HIGHWAY; MULTIMEDIA COMMUNICATIONS; COMPUTER SCIENCE.