Before the results of this study can be understood, it is important to clarify what is meant by the various terms used by this project. The most important term to define is what is meant by the "Internet". For people not familiar with it, it can be difficult to grasp what the Internet actually is because it is much more abstract than any other modern communications technology. (This difficulty in understanding what the Internet is is just one of the problems with adopting the Internet as a technology.)⁶

Unlike the telephone network, the Internet is not a specific network. Nor is it a specific place, nor a specific technology or appliance. To mentally conceive of what the Internet is, it helps to break it down into its parts. Envision first just one computer: inside this computer there is memory [a hard drive] and processing power [a chip, like a "Pentium" or a 486]. The user of the computer only has access to this single computer's resources. But if this computer is connected to another somehow, then the user will also have access to the second computer's memory and processing power. (It is not particularly important for this discussion exactly how the two computers are connected, be it by phone wire, by ethernet, or by wireless technology such as cellular links or even satellites.) If a third computer is added, then a network is created which can then be connected, somehow, to another network somewhere else.

However, none of these networks are "the Internet" because, again, the Internet is not a physical entity. What causes these various networks to be part of the collective "Internet" is that there is a standard protocol that all these computers use to talk to each other, called TCP/IP. This protocol is a way of handling data so that any machine, connected to any other via network after network, can receive the data and know what to do with it. What people commonly confuse as the Internet is the various software packages that use this protocol to enable the transmission and processing of data over numerous interconnected wide-area and local-area networks. These applications define the uses of the Internet, but are not actually the Internet itself.

The most common applications for the Internet are email and the World Wide Web. Email is a particular type of transmission of a file, usually a text file, from one address to another. There are various "mail" programs, but all do the same thing which is to take apart the file created on one end and send it off in bits and pieces to the receiving computer where another mail program will put it back together into an "electronic letter."

The World Wide Web [WWW or the Web] is a little more complex, but the essence is still the same. There are various software programs called "browsers" which exist on local computers. Most of these browsers are graphical, Netscape and Mosaic being just two of the more famous examples. These browsers can be told by the user to look at a file on some other computer that is somehow "connected to the Internet" and handles the same protocol. The browser then reads the file and, depending on what kind it is, displays something on the screen for the local computer user. What makes the Web so unique is that the files that the browsers like to read are created in such a way that they become dynamic documents filled with "hypertext." Hypertext is part of a document that can do something when activated, usually by clicking on it. Often this is the address of another file. Clicking on hypertext can send the person using the browser to virtually any other address of any file that is connected to the Internet. This dynamic nature, this ability to hopscotch around the world by a few clicks of the mouse, is one of the most important characteristics of the Internet that give it its value⁷ and also popularity.

Again, the Internet is not restricted to use by Web browsers. It can be put to other functions, some that offer real-time connectivity such as "Talk," "CUSeeMe," or even the Internet "phone". These programs establish a connection between Internet addresses and transmit the data from one computer to another in real time. Typically this data is text, but can also be video or sound. These types of real time connections are not always person-to-person: Internet Relay Chat [IRC] connects several people at once, while programs like MUDs [Multi-User Dungeons] let people interact with each other in a game situation.

Other Internet applications mentioned in the survey include newsgroups (files pertaining to a certain subject are posted in such a way that anyone with newsreading software can read them), gopher (like the WWW except not as dynamic or graphical), finger (a program that looks for basic information at another address such a user's name, whether the user is currently logged on, etc.), ping (ping programs send out data to test whether another machine is responding and how long the transmission takes), telnet or rlogin (these are programs that let people from one machine "log on" and use another even if the user isn't physically at the other machine.), FTP (which stands for "file transfer protocol", commonly used to download files like software), and mailing lists (mass mailings of email which are sometimes automated. Usually they pertain to particular topics of discussion).

Other terminology used deliberately by the survey includes SLIP, PPP, and Ethernet. All of these simplistically describe three possible types of connections a user's computer may have with another on the Internet. These types of connections turn the local computer into a node on the Internet and enable Internet software to be used locally. Computers may offer access to the Internet without using these technologies, but this generally requires using a modem to turn the local computer into a dumb terminal and using a remote server for the actual connection to the Internet. However, at this time, connecting to the Internet via this arrangement usually involves using a remote UNIX server. UNIX servers are quite powerful, but the user interface is generally line-prompt driven and often daunting to learn for new users who would need to learn the precise commands to enter.

The survey also mentions homepages. A homepage is a file written in HyperText Mark-up Language [HTML], the kind of file that Web browsers like to read. These files do not require backgrounds in computer science to write and give the individual a chance to post information to other people using the Internet. A "URL" [Uniform Resource Locator] is the address of a homepage that the web browsing software uses to find to find it. For example, "http://www.berkeley.edu/" is the URL for the homepage of UC Berkeley.

Finally, it is necessary to differentiate the Internet from general Computer Mediated Communication [CMC]. The Internet is CMC, but not all CMC is the Internet. Similarly, not all electronic mail uses the Internet. Electronic mail, an example of CMC, can be on a local, private network. It doesn't even need to use the same protocol as the Internet. Furthermore, not all email using TCP/IP even "gets on" the Internet. In the case of these Berkeley students, if they were emailing each other, they were only using the local campus network. Only when their intended recipient had an account on a computer beyond the campus network did the user actually "use" the Internet.

This survey focused only on the various software applications designed to make use of the Internet's protocol. The analysis set forth therefore generally applies only to the Internet and its applications although some of the analysis is applicable to general CMC.

Back to Contents

Next Section

cathyg@csua.berkeley.edu

http://www.csua.berkeley.edu/~cathyg