Historically, the Internet has evolved from the Advanced Research Projects Agency (ARPA) of the US Department ofDefense (DoD). It was an experimental network intended to help scientists who received research grants from ARPA to communicate with each other. Consequently, the network became the ARPAnet, and started with just four sites in the United States (U.S.): the University of California at Los Angeles (UCLA), the Stanford Research Institute, the University of California at Santa Barbara, and the University of Utah. This network quickly expanded to sites in the United Kingdom and Norway. One and a half years later, ARPAnet was connected across ten widely dispersed sites. Its electronic e-mail and packet switching proved far more flexible than the limited peer-to-peer connections used at the time. By the mid-1980s, when military traffic split off from ARPAnet, the National Science Foundation (NSF) stepped in to coordinate most non-military government traffic. NSF also set up the first high-speed lines connecting many supercomputer centers together as well. ARPAnet visionaries merely wanted to have computer-to- computer communications at a distance. None of them predicted that ARPAnet would become the worldwide network infrastructure known as the Internet today.
In the late 1980s, NSF funded an upgrade for the ARPAnet, called the NSFnet, which initially consisted of five supercomputer centers. They were connected via the 56 kbps telephone lines and provided the resources for a number of scholars. Up to this point, the world's fastest computers had only been available to weapons developers and a few researchers from some federal government research centers and several large, private corporations. Later, NSF allowed a number of regional networks, such as university and corporate, to be linked to the NSFnet. In each area of the country, regional networks were connected to their nearest neighbors, forming a chain. Each chain was connected to a supercomputer center at one point, and these centers are linked together. With this NSFnet configuration, a computer could communicate with other computers on any other network. Computers connected to the NSFnet communicated by forwarding the conversation through its neighbors.
The NSFnet, later called the Internet, has been growing ever since. The use of Internet for supporting the worldwide electronic information society was highly successful. During the late 1980s, some regional networks began to offer accounts to students, ordinary people and businesses, not just those with government research contracts. Sharing supercomputers allowed the connected sites to share a lot of other unrelated research information. The network traffic worldwide was soon overloaded. To solve this problem, NSF awarded a contract to Merit Network Inc. (MNI). MNI ran Michigan's educational network, in partnership with IBM and MCI; they took over management and upgraded the NSFnet backbone. The upgrade allowed the Internet to have modern computers and faster links, such as T-1 (1.544 Mbps) and DS-3 (44.7 Mbps).
Today, the Internet is quite large and complex. Statistically, in the early 1994, it expanded to 128+ countries, over 24,000 domains, over 1.6 million computer hosts, and 14,000+ Internet Protocol connections, and less than 10 millions of users. Table 1 shows the growth of Internet after 1994.
The Internet, originally established as a private channel
for research activities and academics, is now open to a wide range of commercial
services. The Internet provides a number of services and resources. Some of
these are E-Mail, Telnet, file transfer protocol (FTP), Usenet, Finger, Talking,
Internet Relay Chat (IRC), Gopher, WAIS, Web, Mosaic, and Netscape.
Table 1 - The Growth of Internet (1995-2007)
|
|||||||||||||||||||||||||||||||
The Internet services provide a number of client/server facilities, which are multimedia information-retrieval technologies. A client program allows a user's computer to connect to another Internet host to ask for the help of a server program. Popular client/server facilities of the Internet are Usenet, Finger, Talking, IRC, Gopher, WAIS, Web, Mosaic, and Netscape. The Usenet allows an Internet user to post a message to several thousand news groups; the IRC is good for live interactive discussions with users worldwide; the Finger service allows an Internet user to ask for information about a particular user; the Internet's Talking service allows an Internet user's computer to connect to the other users' computers on the Internet, and then type messages back and forth. The Internet also provides Gopher, WAIS, Web, Mosaic, and Netscape as text search and retrieval tools for helpful navigation in the Internet world. The details of Internet services and resources will be discussed later in this paper.
Connection to the Internet can be accomplished in several different ways. A user may obtain an account on a host connected to the Internet to access the Internet services and resources. Through a commercial Internet service or a federally supported Internet network, a user's computer may be connected directly to the Internet via a modem. This kind of connection allows this computer to become an Internet host with its own Internet address. But it is appropriate only for single remote users or small remote sites with one or two users. Moreover, a corporation may connect its wide area network (WAN), metropolitan area network (MAN) or local area network (LAN) using communications devices, such as bridge, routers and/or gateways, to the Internet. This corporate network becomes a network on the Internet referred to as an Internet subnet. Figure 1 illustrates an Internet access connection scheme that relies on a router, gateway and digital service unit (DSU) to tie directly into the NSFnet backbone or the ANS backbone. Federal subsidies pay for the NSFnet backbone. The ANS backbone is operated by Advanced Network & Services, Inc. (ANS), a commercial Internet provider and a consortium owned by IBM, MCI, and Merit Inc., which charges fee for Internet connection and use.
Within the Internet, each personal computer (PC) is called a terminal. When logging onto the Internet, each PC needs to use a communications software, such as Crosstalk, MLink, or ProComm, to allow it to pretend to be, or emulate, a computer terminal. When connecting to the local Internet system, a user needs to specify what type of terminal they have. Their choice will be Teletype (TTY), ANSI, VT220, or VT100, in order to define for their computer. Since the Internet is designed to allow users to share information resources rather than pieces of hardware, users can send messages to anyone else on a LAN, MAN, or WAN connected to the Internet.
The physical connections of the Internet take a variety of forms. The most prevalent for Internet links are 56 Kbps leased lines, frame relay (64 Kbps), and T1 (1.544 Mbps). Usually, there is a T3 (44.7 Mbps) link to be used as a backbone between two major Internet nodes. Remote PC can be connected to the Internet through modems, which may run from 1,200 baud rate all the way up to as high as 38.4 Kbps.
Like the Postal Service, the Internet uses its own mail address technique to deliver e-mail or a packet. Every computer connected to the Internet has a unique address. This technique uses the Internet Protocol (IP) 32-bit address scheme, consisting of four 8-bit groups separated by a period, and each has a value of less than 256. For example, the addresses: 129.2.98.114 and 167.142.225.6 are typical Internet addresses. The beginning of an Internet address tells the routers what network a user is part of. The right end of an Internet address tells which computer or host on the Internet should receive e-mail. Figure 3 illustrates a "Postal Service" Internet envelope.
The Internet employs the TCP/IP protocol suite that consists of the Internet Protocol (IP) and the Transmission Control Protocol (TCP). The IP takes care of addresses and enables many different types of computers to communicate on the Internet. The TCP takes the information that a user wants to send and breaks it into pieces, called packets.
The TCP numbers each packet so that receipt can be
verified and data can be put back in the proper order. The data within a packet
is usually between 1 and 1500 characters long. Furthermore, if a user sends two
e-mails to the same place at different times on a day, the Internet may use the
same routes. Also, depending on the availability of routers, these e-mails may
not arrive in sequence.
The Internet uses TCP/IP to carry and manage the transfer of data, and employs the UNIX standards as an operating system.
As discussed above, TCP/IP combines two specific protocols: TCP and IP. TCP an IP incorporate most versions of UNIX; and they are supported by virtually every computer vendor. TCP/IP provides the capability that the Internet needs to deliver its services. TCP/IP allows the Internet to provide a reliable full duplex data and graphics transfer. It also offers various networking functions that the Internet needs. These include addressing, connection establishment, flow control, connection release, routing and management, name control and translation, status translation and communications, fragmentation and reassembly, and delivery.
The Internet uses TCP/IP to transfer data, but its computers run on the UNIX operating system. UNIX is a multitasking, multi-user operating system with emphasis on modular design, and is written in the high-level "C" programming language. UNIX offers the Internet a productive computing environment that consists of a rich set of software development tools. UNIX was designed as a 16-bit midrange uniprocessor timesharing system, and consists of a kernel, a shell, and a collection of utilities. The kernel, a small and tightly written module, can implement file, memory, or process management. The shell, an interface between the kernel and users, functions as both a command interpreter and a programming language. It can be customized to meet specific needs. Each standardized shell (i.e., Bourne, C, and Korn) has different features at the user interface, and is equally valuable for some applications. Utilities and other programs (e.g., editors, filters, and formatters), including communications and networking facilities, comprise the remaining UNIX component inventory. It enables the Internet to become a network of networks. The era of open systems began when UNIX was paired with TCP/IP.
UNIX was developed by computer software experts for their own use. It has traditionally not been a pleasant place for novices to tread since the user interface is complex. Workstations/PCs running UNIX are often not capable of supporting vendor-proprietary application architectures, unless the architecture is based on UNIX. Non-support is also due to vendor reluctance to provide an architectural bridge over which users can exit the vendor's proprietary line completely. Although the Internet uses UNIX for its own operating system, an Internet user does not require knowledge of the vast majority of UNIX commands. The Internet takes some parts of the rich set of UNIX commands for its own use. The Internet is a part of the UNIX culture; however, the Internet very much has its own life.
E-mail is the most common service on the Internet as well as any system or network that widely supports e-mail functions. The Internet e-mail service allows a user to send e-mail to a hundred people, simultaneously, as easily as to one. The Simple Mail Transport Protocol (SMTP), an electronic mail protocol used in a TCP/IP network, can be used to provide a facility for a user to send/receive a message to/from anyone else on the Internet. It is often defined as a manager of messages exchanged between mail Internet postmasters. The SMTP allows an Internet user to read, write, edit, print, exit, delete, undelete, send, receive, save, reply, or forward e-mail. An e-mail may be a text file, a binary file, or a copy of pertinent text from another e-mail. It can be distributed across the world to millions of people on the Internet flew in seconds or minutes.
An Internet e-mail is hinged around the concept of an address format. An Internet address provides necessary information used to transfer data between networks connected to the Internet. Technically, several different kinds of
e-mail addresses can be used for the Internet. Like a regular Post Office address, an Internet address has a name and a domain. A name (login name or ID) and a domain (the group that the computer belongs to) are generally separated by an @. In other words, most Internet addresses have a standard format: loginID@domain. An Internet address may contain an !, called a bang or a % sign. These symbols can mean that the user with this kind of address is not directly connected to the Internet, but does have partial access and can receive an e-mail. Table 2 lists some geographic and organizational domains.
Specifically, an e-mail address domain may consist of
many sub-domains. For example, to reach Jim Smith on the system "ceiusa.com", one
would address the mail as "jim.smith@ceiusa.com". In this case, the userid is
"jim.smith", and the domain has two sub-domains: "ceiusa" and "com", with each is
separated by a period (.). The Internet address "jim.smith@ceiusa.com" is one of the
registered addresses provided by the US Internet Information Center (NIC) for
CEI. Every computer on the Internet that CEI owns has an IP address. The
computers attached on the Internet know each
other through the Domain Name Server (DNS) database. When an e-mail is sent, the mail system looks up the machine's name in the DNS and retrieves the address; the mail post office (i.e., the computer that accepts incoming mail) and the router will process the e-mail to the destined address.Domain Name Meaning gov Government Body edu Educational Institution mil Military Site com Commercial Organization org Non-profit Organization net Network Resource us United States uk United Kingdom au Australia
Currently, most Internet providers furnish the E-mail service. To obtain a list of public dial-up Internet providers, for example, a user can send a request to "info-deli@netcom.com", no subject line is needed, and type "Send PDIAl" in the text body; the netcom.com will automatically forward the requested information after receiving the e-mail.
When sending an e-mail over the Internet, carefully address it to get to the right destination. Also, e-mail sometimes fails to be delivered to its destination because machines or pieces of the network normally connected to the Internet are no longer available. Usually the network tries to send this e-mail for days before giving up.
Telnet is used to access many public services, including library card catalogs and other databases. It is the Internet remote login application. Depending on the Telnet software installed at the remote Internet host, a Telnet user can access many Internet services and resources. Once a login occurs, a user can execute various commands to operate the host system. These include, using line-by-line or character-at-a-time mode. Sending special characters, setting operating parameters, checking/printing status information, toggling operating parameters, printing help information, suspending Telnet, closing current connection, or exiting Telnet.
Like most other Internet services, FTP is actually made up of a collection of programs that work together to transfer files between computers on the Internet. This collection of programs includes both client and server software, where the server provides a particular resource while the client makes use of that resource. On LANs, the term "server" is used to refer to the actual computer that runs the server program. On the Internet, since hardware pieces are normally not visible, the words "server" and "client" are usually referred to the programs that request and provide services. Typically, a client sends a request message to a computer running an FTP server. The server sends the requested document back to the client, which then displays the document on the user's computer screen. In other words, the FTP client program allows a local computer to connect to a remote computer on the Internet and transfer files over the Internet; an FTP server program responds to a client's program file transfer requests and processes them. Thus, in the language of the Internet, the FTP client program allows a user to download files from a remote host on the Internet. It sometimes allows a user to upload files to a remote Internet host.
On the Internet, FTP provides local and remote hosts the same commands, such as for transferring files back and forth, and setting the file transfer mode, etc. However, the way in which these commands are implemented varies from host to host. Figure 4 illustrates the relationship between local/remote hosts and client/server software.
The addresses for Internet Providers Information can be found at rtfm.mit.edu /pub/usenet/news.answers/Pdial ftp.netcom.com /pub/inf-deli/public/Pdial ftp.merit.edu /internet/provider/Pdial
The first Usenet was experimented with by the University of North Carolina (USA) and Duke University (USA) in 1979. The first experiment used Usenet as a computerized version of a bulletin board, which allows the posting and reading of news messages and notices. Later, many other organizations joined and, with the explosion of networking in the l980s, Usenet outgrew its original blueprint. However, the legacy of an electronic news network remains. Today, there are about 80,000 different sites that carry Usenet with more than two and half million participants. They post about 30,000 articles a day, which is equivalent of approximately of 40 megabytes of data.
The exciting thing about Usenet is that there is no central administration. Usenet is run by people who use it. Usenet may not be available everywhere since it depends on system administrators who decide whether or not to carry Usenet discussion groups at their sites.
The Internet's talking service also allows more than two persons on the Internet to talk at the same time by using the "ytalk" program. In such case, this program will divide each user's (computer) screen into as many partitions as necessary by drawing horizontal lines.
To use the IRC service, an Internet user needs to have a client program to act as an interface. This client connects to an IRC server. All IRC servers are connected with one another. This creates a global network of IRC users on the Internet. The IRC service maintains a number of different channels including public, private, and secret (or invisible) channels. A user can choose a channel to join a particular conference on the Internet. Once an IRC conference is made, the user's IRC client acts as a window, showing what other conferees are saying at all times.
The original Gopher was developed in 1991 by the Department of Computer and Information Systems, University of Minnesota, to provide an easy way to distribute information available to the campus. Today, there are several thousand Gopher servers founded in many universities and organizations whose networks are connected to the Internet. Each Gopher server contains whatever information that is of interest to local users. An Internet user who desires to get particular information would need to connect his local Gopher client to a specified Gopher server. A Gopher client allows a user to move, smoothly and easily, from one Gopher server to one another. Gopher is always changing and growing.
Based on the ANSI Z39.50 standard, WAIS has been in development for a long time within the library and computing communities. It is used for requesting bibliographic information. Historically, WAIS was originally grown out of a project started by three companies: Apple, Thinking Machines, and Dow Jones. Apple manufactures personal computers with an easy-to-use graphical interface; Thinking Machines designed large computers suitable for searching large amounts of data quickly; and Dow Jones runs a service that sells news and information. One of the ambitious ideas behind the development of WAIS was to provide a tool that allows a computer to keep track of a vast amount of information. It also presents users with only the information that is relevant to their needs.
Like many other Internet services, WAIS uses a client/server system and allows an Internet user to search any of hundreds of collections of data within minutes. Each such collection is called an information source, which is maintained by a WAIS server program. In operation, a user needs to have a WAIS client program to access a public WAIS server on the Internet. There is a variety of WAIS servers available for free, via Anonymous FTP. The use of WAIS is reasonably simple. When a WAIS user makes a request, his WAIS client is connected to an appropriate WAIS server to perform a search. It asks each server, in turn, to search its index for a set of words or the topic that the user requested. The server then sends to that user a list of articles or citations that may be appropriate. There is a limit to the number of articles WAIS reports - usually, between 15 and 50, depending on which client the user uses. The user can pick which articles to view, and WAIS will display them on his computer's screen.
Like many other Internet services, the Web facility is made up of a collection of servers and clients that can exchange information. Most Web clients and servers have been designed to communicate using TCP/IP. Typically, when a Web server receives a request message from a Web client it sends the requested document back to the client. It then displays the document on the requester's (computer) screen. The client program, also called browser, allows an Internet user to read a document, and follow whatever links are selected.
The protocol that a Web client uses to communicate with a Web server is the Hypertext Transmission Protocol (HTTP). All Web clients and servers must use HTTP to send and receive hypermedia (i.e., a combination of hypertext and multimedia) documents. The standard language that Web uses to create and/or recognize a hypermedia document is the HyperText Makeup Language (HTML). HTML is a simple formatting language that makes it possible to include HyperText links and references to other media within documents. A Web document written in HTML is named with the suffix".html", and has a text format similar to that of a 7- bit ASCII document.
The Web service allows a user's computer on the Internet to display the same information on many platforms. Documents with formatting information and graphics might be displayed in full color on an X-terminal. Only text and basic formatting, using the same document, would be shown on a text-only screen. This efficient way provides an extensive menu that allows an Internet user to search quickly through archives and documents. The Web language also provides multimedia presentations, on line forms and database interfaces. They allow an Internet user to interact with and control information rather just view it.
In order to run Mosaic, a PC (e.g., an IBM, Mac, or Sun workstation) connected to the Internet must have the ability to display multiple-screen windows, HyperText and hypermedia documents. It must also have the ability to support interactive graphics within documents, and make basic hypermedia links that support FTP, Gopher, Telnet, WAIS, and others. Mosaic is one of the most popular interfaces to Web today.
Generally, Netscape displays Web pages faster than other Web browsers. It differs from the Web standards in three main areas. First, Netscape has a "tag" command for downloading, centering, and displaying text after first setting aside space on-screen for images. Second, Netscape allows an Internet user to read a document while the system is downloading the Web page; this reduces the waiting time needed while reading that same downloaded Web page. Third, Netscape provides a means of secure data exchange using encryption technology from RSA Data Security.
Netscape immediately displays the text and/or image without waiting for the whole file to arrive unlike Mosaic. As files are loading, Netscape displays status information, as well as providing the total number of bytes of the document at the bottom of the screen. Moreover, while Mosaic can work with newsgroup in a rudimentary way, Netscape has news reading and posting built in. Netscape also allows a user to send (but not receive) mail over the Internet. Netscape is becoming the most popular interface to Web.
When using the NSFnet backbone services, an Internet user needs to read the NSFnet Backbone Services Acceptable Use Policy. An up-to-date version of this policy can be obtained via Anonymous FTP from nic.merit.edu, in the file
"/nsfnet/acceptable.use.policies/nsfnet.txt" (Click to download). If using commercial Internet providers, users should contact their Internet providers or organization's network administrators. They will provide the rules governing their Internet connections.
In general, the following rules may be used by Internet users. These rules may allow users to save time and avoid coming across as undisciplined users on the Internet.
Rule 1. When shipping anything across a national boundary over the Internet, export laws come into effect. When sending software from one place to another over the Internet, an Internet user must consider the intellectual property and license issues.
Rule 2. When preparing a message, be objective factual, professional, friendly and polite; keep messages short and to the point. Be careful when using humor, as gag may be viewed as criticism. If a message exceeds a couple pages, provide a summary up front.
Rule 3. When sending an e-mail, write subject lines like newspaper headlines. Capitalize words only to highlight an important point or to distinguish a title or heading. Note and respect copyright. Abbreviate when possible (e.g., FYI = for your information).
Rule 4. Respect the privacy of others. Do not read other people's e-mails and files. Ask for permission before forwarding other people's messages.
Rule 5. Check and read E-mail daily, delete unwanted messages, download/extract important mail and file them in folders for future reference. Maintain mail within the disk quota. Also, keep files to a minimum and frequently scan the computer for viruses.
Rule 6. Keep questions and comments relevant to the focus of the discussion group. Treat the people on the mailing list with respect. Be tolerant of the group's activity, and possibly provide constructive criticism as opposed to destructive criticism.
Rule 7. When joining a discussion Internet forum, other people on the mailing list may not be interested in reading a desire of a private request, such as subscription and unsubscription. Any requests regarding administrative help should be made to the appropriate places.
Rule 8. Do not oversubscribe. There are thousands of mailing lists and newsgroups. Avoid becoming an Internet junkie. To keep productive, stay with your own interests and keep focused. Drop subscriptions if they chronically exceed a budgeted time.
The Internet is becoming an avenue for individuals and organizations to engage in electronic commerce. The CommerceNet Consortium has recently created CommerceNet, which is an electronic marketplace using an Internet, based infrastructure for electronic commerce. The CommerceNet Consortium, a coalition of organizations (Intel, Sun Microsystems, Pacific Bell, and Apple Computer), is a non- profit corporation. It operates using matching funds from the Technology Reinvestment Program (TRP). TRP is sponsored by ARPA, NIST, NSF, and NASA.
Most people think of the Internet primarily as a government- subsidized research and education tool for universities, but this concept is obsolete. In fact, the Internet is expected to be wholly sustained by commercial business in 1995, followed by cessation of NSF backbone funding. Internet access for some research institutions and universities will still be subsidized by government research grants. The Internet as a whole may be run on a for-profit basis by ANS. However, this does not imply that NSF is going to get out of the Internet business. In fact, in the U.S., NSF has recently awarded multi-million dollars contracts to Merit Inc., MCI, and the University of Southern California Information Sciences Institute (Los Angles) to manage routing of the Internet backbone and upgrade the Internet backbone speeds.
What will the new Internet look like? It may be like a network-wide surge in commercial applications. One disadvantage is that high access cost will become an obstacle to universities and organizations. The Internet will then lose some of its most valuable resources, most of which are supplied to users on a cost effective basis. Moreover, university students may no longer have free accounts that access the Internet, and thus have severely restricted use of Internet resources.
2. H. Hahn & R. Stout, The Internet: Complete Reference, 1994, Osborne McGraw-Hill
3. Ed Krol, The Whole Internet - User's Guide & Catalog, 1992, O'Reilly & Associates, Inc.
4. Thao M. Le, Guide to Data and Computer Communications.
5. K. Hughes, "Everything You Need To Know To Build Your Own Web Server", Open Systems Today Magazine, 6/6/94.
6. J. Johnson, "The Internet: Corporations Worldwide Make
the Connection", Data Communications Magazine, 4/94.