| Subsection: |
The Internet, as you remember, is a combination of hardware and software. Another very critical component of the Internet is the protocols, which are a series of agreements about how information will be shared and communicated between machines and networks. The protocols allow computers with different architectures and operating systems to share information.
It is important to note that protocols are NOT the same as programs. A protocol is simply an agreement about how data will be structured and communicated. Internet programs are designed to take advantage of one or more of the protocols.
TCPIP (Transfer Control Protocol / Internet Protocol) is the underlying protocol of the Internet today. It is the protocol that helps to split messages up, helps them find their way to the correct destination, then causes them rejoin at the destination. TCPIP is very important, but you rarely interact with it directly. All the other protocols take advantage of it.
TCPIP has been with us since the earliest days of the ARPANet. It is showing some signs of age, since it is being used for purposes its designers never imagined! Some other approaches to underlying Internet protocols are being studied. One notable possibility is a concept called ATM (Nope, not Automatic Teller Machines, but Asynchronous Transfer Mode.)
One of the earliest protocols to be used regularly is called telnet. The telnet protocol allows a user to access a remote computer. If you have the permission of the owner, you can use your computer as a terminal of another computer. If Terrence has an account on a big unix machine at work, and that computer and his home computer are both attached to the Internet, he can use telnet to have his computer 'call up' the unix machine at work. Once he does this, his computer is acting simply as a terminal. The screen he sees is being sent by the remote computer, and anything he types will be sent there. His own CPU is doing very little work. Any instructions he sends will go to the remote computer.
Why would anyone want to do this?
Telnet allows people to work on computers regardless of geographical limitations. It allows people to set up complex programs on mainframes or other computers, and let people log in to this computer. One good example is the library of Congress Information System (LOCIS). The Library of congress maintains a huge database. Through telnet connections, people can access the LOCIS computer from anywhere in the world.
Telnet is a very inexpensive protocol to implement and utilize. It does have disadvantages, however. The remote computer does all the work. It doesn't matter how nice the graphics capabilities are on your local computer, for example, because you will be forced to use plain text most of the time when working on a telnet connection. Also, not all computers on the Internet allow unlimited telnet access. (In fact, very few do.) It is far more likely that a computer will allow you access only if you already have an account on that machine. In other words, you must have previously arranged permission from the owner of the remote computer in order to use that computer. Telnet is still used extensively by computing professionals, and it is often used by large institutions such as libraries, universities, and government institutions as a very inexpensive way to provide access to information on mainframes. The Indiana University system, for example, uses a system called IUIS (IU Information System) for student scheduling and other purposes. If you have registered for class by computer at an IU campus, you probably used IUIS via telnet (!)
There are special programs designed specifically to help you use Telnet. These programs are generally referred to as terminal programs or telnet clients.
FTP stands for File Transfer Protocol. This protocol is a special protocol which is designed to facilitate copying information between computers through the Internet. Whenever you download a file from the Internet, you are probably using a form of the FTP protocol. Much of the time, FTP happens automatically through Web browsers, but there are special FTP programs as well. These FTP clients look like file management programs (with the directory tree structure and all!) except they also allow you to connect to a remote computer as well, and look at the file structure of that computer. Of course the purpose of this is so you can copy and move files between computers. An FTP program often looks like a file management program with TWO directory structures. You might use such a program if you have access to two Internet computers, and you want to back up some of your files on the other computer. The information you are reading now was created on a unix computer through Telnet, then FTPd to a couple of other computers for review!
If the web is the most glamorous protocol, the mail / news protocols are the work horses of the Internet. These protocols are designed simply to transfer text from computer to computer. There are a number of related protocols, including MNP, SMNP, POP, and on and on... It really isn't that critical to understand the difference between them. The important thing to understand is that when you are reading mail or news, whatever program you use to do this is using a mail/news protocol.
HTTP stands for hypertext transfer protocol. This is the protocol you already know and love. The WWW is based on this protocol. HTTP has some interesting characteristics. It was designed to make the Internet easier to use, and it accomplishes this in two major ways.
First, the HTTP protocol integrates many of the other protocols, so if you have a program that can understand HTTP, it knows how to handle other protocols such as gopher or ftp. Some browsers can also handle mail, news, and telnet.
Secondly, HTTP is designed with HYPERTEXT in mind. The earlier protocols often were implemented on text - only displays, and had command line - style interfaces. This was not a problem when the Internet (and its precursers) was populated by research scientists and the like who actually seemed to ENJOY the confusing unnatural interface.
(Really. They dig that kind of stuff.)
HTTP encourages hypertext multimedia. Hypertext, as you remember, is text with links built in. These links can be activated to allow the user to go to a related topic. Multimedia refers to the ability to incorporate text, graphics, sound, and other media in the same document. Modern users are used to both concepts, and almost expect to find them in programs. The HTTP protocol is thus the most friendly. It is relatively easy through HTTP to share documents that contain all kinds of information.
There are a number of other protocols that are used on the Internet. Many of them are used mainly by computer professionals for maintaining and testing connections (such as ping, which checks to see if a certain site is running). Others (such as gopher) were very popular once, but now have been supplanted by another protocol. The worldwide web has made a number of smaller protocols far less important to the average user, because you don't really have to know which protocol is being used.