The Internet and its Services: A Brief Review

The Basic Structure of the Internet

The Internet is a heterogeneous worldwide network consisting of a large number of host computers and local area networks. The Internet uses the TCP/IP suite of protocols. This allows the integration of a large number of different computers into one single network with highly efficient communication between them. This way, the user can access information on all kinds of host computers from a desktop PC, Macintosh, or whatever he/she has available.

TCP/IP, the communication standard underlying the Internet, originates from work done at the US-Department of Defense in the late 1960s. The first version of the Internet was built by the DoD in 1969 and consisted of just 4 computers (Davidson, 1988, p. 2). In 1982 a set of specifications and protocols have been implemented, which became known as TCP/IP in reference to their two major elements, the "Transmission Control Protocol" (TCP) and the "Internet Protocol" (IP). The development and implementation of TCP/IP stimulated a massive growth process for the Internet. "By late 1987 it was estimated that the growth (of the Internet) had reached 15% per month and remained high for the following two years. By 1990, the connected Internet included over 3,000 active networks and over 3,000 active networks and over 200,000 computers" (Comer, 1991, p. 8). By January 1992 the number of hosts on the Internet was 727,000 (Lottor, 1992), doubling about every 7 months (see figure 2). Various groups of users are connected to the Internet: universities and other educational institutions, government agencies, the military, and at an increasing number private businesses.

Link to Hostcounts

The most fundamental function of the Internet is to pass electronic information from one computer to another. Every computer on the network is identified by a 32 bit Internet Address or IP-Number. This number is commonly represented as four numbers joined by periods. The Internet uses these numbers to guide information through the network ("routing"). For human users, however, such numbers are usually difficult to keep in mind. Therefore, computers are also identified by Domain Names, which are to some extent similar to mailing addresses. Special programs, called "Name Servers", translate Domain Names into IP-Addresses.

Major Internet Services

Internet services can be divided into two groups (Maier, Wildberger, 1993a): communication services and information services. In the first group the Internet mediates in the communication between two or more individuals. In the second group the user turns to the Internet-service in search for some particular information. Communication services can roughly be compared tvices can roughly be compared to a telephone call, information services to a dictionary (of course, the dividing line between the two types of services is fuzzy; for a discussion of the relationship between traditional and electronic forms of communication see Maier, Wildberger, 1993b).

The most important communication services on the Internet are electronic mail (and some derived services) and netnews. Major information services are terminal emulation and file transfer, Gopher, WAIS, and World Wide Web. We will discuss them very briefly below (for a more detailed discussion see e.g., Maier, Wildberger, 1993a; Krol, 1992).

Communication Services

Electronic mail (e-mail) is the most popular and widely used network service. It can be viewed as the electronic equivalent to a regular mail letter. When one user wants to send a message to another Internet user, he/she types the message into a special computer program, adds the e-mail address of the recipient, and sends the message off through the network. Typically the message reaches its destination almost immediately, even when it is on another continent. Practically all gateways between the Internet and other computer networks can handle e-mail messages. Therefore, many more people can be reached by e-mail than are connected to the Internet via TCP/IP.

The major advantage of e-mail over regular mail is that an e- mail message comes in electronic form. Therefore, it cectronic form. Therefore, it can easily be handled and interpreted by a computer program. This feature is use by electronic discussion lists. They operate a database of subscribers and each incoming e-mail message is automatically distributed to each subscriber. LISTSERV, the most popular program for managing discussion lists, also handles subscription via mail messages, archives incoming messages, and allows users to retrieve these and other archived files (for a discussion of LISTSERV see Maier, Wildberger, 1993a,b, Kovacs et al., 1991).

Thousands of discussion lists focusing on all kinds of topics exist on the Internet (LISTSERV is actually a Bitnet-based service). Topics range from various hobbies, political discussions to operational aspects of different computer systems (e.g., GIS) and research questions. For the user, discussion lists are an easy way to identify and contact a large number of people with similar interests. A discussion list can also be considered as a worldwide forum for expressing views and discussing opinions.

While messages are automatically sent to all subscribed users in the case of a discussion list and one has to be subscribed in order to receive the messages, messages in NetNews are distributed between a net of servers. Messages are organized in a hierarchy of newsgroups. Incoming messages are stored for a particular period of time in a publicly accessible area. Each user can connect to this area, ban connect to this area, browse through the stored messages, respond to any one of them. This way NetNews allows for a better overview of ongoing discussions but requires the user to actively connect to the respective area.

Information Services

One of the reasons for the creation of a computer network like the Internet was to give users access to remote computers and to allow them to transfer files to and from this machine. These are typical demands for Telnet and FTP (for "File Transfer Protocol"). In both services the user specifies a specific remote host (through its IP-number or Domain-name) that he/she wants to access. When the user has an account on this remote host, he/she can work there just like on a local machine. With Telnet available, more and more Internet-sites allowed outside users to access some of their information services. Typical examples are electronic library card catalogs, campus information systems and other database applications. Today, the electronic card catalogs of practically all major U.S. libraries and many university libraries in Europe are publicly accessible through the Internet. Many sites run large specialized information systems for the "network community".

A similar type of anonymous access is available through FTP. Many Internet sites allow users to log in as "anonymous" through FTP. This way the sites can make files publicly available. Whoever needs some of these files can deds some of these files can download them through "anonymous FTP". This system has led to a huge supply of freeware- and shareware-software that is distributed through this channel. Practically all network software is available in this way. But also information files, specifications of network standards, research papers, multimedia files and even the complete texts of classical books can be accessed through anonymous FTP.

One of the problems with Telnet and FTP is that the user has to know exactly which information can be found on which computer in order to be able to find it. People began to compile directories of network services and made them available electronically. The most important information in such a directory was usually the IP-number and/or Domain-name of a specific information service. Now, that the Internet "knows" where the requested service resides, why does not it just connect me to itconnect me to it?

In recent years programmers picked up on this ideas and developed more intelligent Internet services. These services try to hide unnecessary technical information from the user and let him/her use the Internet as a whole as an information source. A good example of such a service is Gopher. When a user starts his/her Gopher client it displays a menu for the user to choose from. For each menu item the client receives a descriptive name, a flag that identifies the type of information, and information about where on the network this information can be found (Domain-name, file-name, port number). The client displays only the descriptive name and an identifier for the type of information. This is all the user really needs to know. When he/she selects a menu item, the Gopher client connects to the respective computer, retrieves the requested information, and dispays it on screen.

Because one of the information types can be another Gopher menu, this simple concept yields a very flexible and efficient network service. The user can jump from one Gopher menu to another until he/she has found the required information. The menu items may refer to another Gopher menu, a plain text file, a binary file, a database inquiry module, a picture, sound, or movie, and even a Telnet host (when this item is chosen, Gopher calls the local Telnet client and supplies it with the IP-number or Domain-name of the respective hosame of the respective host; when the Telnet session ends, Gopher resumes control). The items combined in one Gopher menu do not need to point to information on the same computer. A Gopher menu can logically collect related information and display it to the user as alternatives, even when the information is stored in different parts of the world. To the user this dispersed information appears as if it were stored in one subdirectory on the hard disk. We will talk more about the practical side of Gopher in section 5. The user can move from one Gopher server to another and browse the information that is available there. Figure 1 shows the steps one has to take to get from the Gopher server at the University of Minnesota -- the place where Gopher was invented -- to the University of Economics and Business Administration in Vienna, Austria.

An important element in Gopher are queries to WAIS databases (WAIS stands for "wide area information server"; WAIS cannot be used to its full potential in Gopher). WAIS allows for full-text searches in a set of documents. As a result of a database query WAIS returns pointers to the documents or parts of documents that match the request and indicators for the quality of the match. Of course, each document can be downloaded and displayed on screen, stored or printed.

Each WAIS index that is made publicly available has to have a description of its content. These descriptions are again fed into a ons are again fed into a WAIS database. This allows the user to query WAIS for publicly available WAIS databases that match specific criteria.

The most sophisticated and most exciting new Internet service is "World Wide Web" (WWW). "The Web" allows everything a user can do with Gopher and WAIS and more with tremendous flexibility. WWW is based on Hypertext links, a concept well known e.g. from the Help system in Microsoft Windows. The main difference is that in WWW Hypertext links may point to documents anywhere on the Internet. When the user activates such a Hypertext link the respective document is retrieved and displayed. It is important to note that documents can be a number of different text files, pictures, sounds, movies, as well as other Internet services like Gopher, WAIS, Telnet, NetNews, etc. This makes WWW an outstandingly general and flexible Internet tool.

When the user follows a Hypertext link and retrieves another document, this new document may contain Hypertext links to other documents with their own Hypertext links, and so on. In this way the user may jump from one document to another and follow an argument the author of the first document has never thought of. The author of a Hypertext document can tie his/her work into an existing network of arguments and link to documents stored somewhere else on the web. This makes the web particularly attractive for scientific publications, where reference to existing e reference to existing documents is important. In a Hypertext document a list of references can really come alive.



Comer, D.E. 1991. Internetworking with TCP/IP, Vol. I: Principles, Protocols, and Architecture, 2nd edition, Englewood Cliffs: Prentice Hall.

Davidson, J. 1988. An Introduction to TCP/IP. New York: Springer-Verlag.

Harnad, St. 1991. Post-Gutenberg Galaxy: The Fourth Revolution in the Means of Production of Knowledge. The Public-Access Computer Systems Review 2: 39-53. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET harnad prv2n1 f=mail).

Harrison, T., Stephen, T., and Winter, J. 1991. Online Journals: Disciplinary Designs for Electronic Scholarship. The Public-Access Computer Systems Review 2: 25-38. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET harrison prv2n1 f=mail).

Hugo, J., and Newell, L. 1991. New Horizons in Adult Education: The First Five Years (1987-1991). The Public-Access Computer Systems Review 2: 77-90. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET hugo prv2n1 f=mail).

Jennings, E.M. 1991. EJournal: An Account of the First Two Years. The Public-Access Computer Systems Review 2: 91-110. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET jenningSERV@UHUPVM1.BITNET: GET jennings prv2n1 f=mail).

Kehoe, B.P. 1993. Zen and the Art of the Internet - A Beginner's Guide to the Internet. 2nd edition, Englewood Cliffs: Prentice Hall.

Kovacs, D., W. McCarty, M. Kovacs, 1991. How to Start and Manage a BITNET LISTSERV Discussion Group: A Beginner's Guide, The Public-Access Computer Systems Review, Vol. 2, Nr. 1, S. 128-143.

Krol, E., E. Hoffman, 1993. What is the Internet?, Network Working Group, Request for Comments: 1462 (FYI: 20). (available - among others - via anonymous FTP from FTP.UNIVIE.AC.AT as /doc/rfc/rfc1462.txt).

Lottor, M. 1992. Internet Growth(1981-1991). Network Working Group, Request for Comments: 1296. (available - among others - via anonymous FTP from FTP.UNIVIE.AC.AT as /doc/rfc/rfc1296.txt).

MacKie-Mason, J.K., H.R. Varian, 1993. Some Economics of the Internet, paper presented at the Tenth Michigan Public Utility Conference at Western Michigan University, March 25--27, 1993.

Maier, G., A. Wildberger, 1993b. Wide Area Computer Networks and Scholarly Communication in Regional Science, Papers in Regional Science, Vol. 72, No. 4.

Maier, G., A. Wildberger, 1993a. In 8 Sekunden um die Welt: Kommunikation ber das Internet, Bonn: Addison Wesley.

Okerson, A. 1991a. The Electronic Journal: What, Whence, and When?. The Public-Accesand When?. The Public-Access Computer Systems Review 2: 5-24. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET okerson prv2n1 f=mail).

Savage, L. 1991. The Journal of the International Academy of Hospitality Research. The Public-Access Computer Systems Review 2: 54-66. (to receive a copy, send e-mail to LISTSERV@UHUPVM1.BITNET: GET savage prv2n1 f=mail).