In 1963, the United States Postal Service divided the country into small geographical zones and assigned each zone a ZIP code, which is a five-digit number that enables the postal service to very quickly determine how to route mail.
Your local post office sends all its nonlocal outgoing mail to its nearby main office, where the mail is sorted by ZIP code. The sorted mail is then sent to a distribution center, where it is gathered together and shipped out to the appropriate receiving distribution center (see Figure 6.1).
FIGURE 6.1. Internet addresses are used in much the same way as your ZIP code.
If I address an envelope to my pal Lynn at ZIP code 46290, my mail carrier takes the envelope to the local post office (where they check to see whether the envelope is addressed to a local ZIP code); if it doesn't have a local ZIP, they pass it on to a collection center in Portland. There, envelopes are sorted: any with 46 as the first two digits of the ZIP code are destined for Indianapolis in central Indiana. My envelope is dropped into the Indianapolis bag, and off it goes. Once in Indianapolis, the reverse happens, and the envelope lands in Lynn's local post office. Only then does someone look at the street address. A valid five-digit ZIP code removes the necessity of including a city and state on your envelope (although you should include them in case the post office can't read the ZIP code numbers). The final sorting by postal carrier area is done at the local post office, and then Lynn receives the envelope.
More recently, in 1983, the postal service added another four digits to the ZIP code, enabling the post office to determine your street with the nine-digit number.
The ZIP code is decoded left to right, with the first number identifying the largest geographical area. Each next number (moving right) signifies smaller areas, and the last four digits of the nine-digit number represent the actual street.
A similar thing happens over the Internet. Every computer on the Internet has a specific address called the Internet Protocol (IP) address. Each person that uses a computer on the Internet also has a user name that is combined with the IP address to make their full Internet address.
The IP address is made up of four numbers; each number is less than 256. (For example, my Internet provider's address is 192.108.254.10.) The address numbers are separated by periods when written out. As with ZIP codes, IP addresses are decoded from left to right. The first three digits identify the largest divisions of the network. (The numeric IP addresses are not based on geographic location.) The next numbers (moving right) signify smaller virtual locations; the final numbers represent the computer you use to access the Internet.
A computer with an IP address is part of the Internet; but if you use your personal computer to dial in to an Internet access provider, your PC does not need an IP address (unless you are making a SLIP or PPP connection). Your PC serves as an intelligent terminal connected to your Internet provider's computer with a phone line and modem. Your PC is, in effect, just one of several terminals using the common address.
For example, if you want to send me mail, you can do so on some systems by e-mailing to this address:
martinm@192.108.254.10
The martinm is my user name, given to me by the system operator that runs the computer I use to access the Internet. It could just as well have been mmoore, or even just a number, as long as it uniquely identifies me as a user on the system.
The at (@) symbol is used to separate the user name from the IP address. The IP address is the four-part number, described previously. Much more commonly, however, addresses are made up of your user name attached to the front of a domain name: foo@bar.com.
As mentioned in the preceding section, on some systems, you use the IP address to send me mail. More commonly, you use the domain name of the Internet provider's computer. Most people have a hard time remembering a long string of numbers containing periods. For those of us who are challenged by long numbers, the Internet provides domain names.
Domain names represent the IP address and are intended to be simpler to understand and easier to remember. If you want to use a domain name to send me e-mail (instead of the IP address used in the preceding section), you address the e-mail as follows:
martinm@teleport.com
The teleport part of the domain name is the unique name of the computer attached to the Internet. The .com portion of the domain name identifies what kind of operation this computer serves—in this case, a commercial (com) service.
Domain names can have many parts, each separated by a period (usually referred to on the Internet as a dot). However, instead of providing the largest group on the left (as with IP addresses), domain names provide the largest grouping on the right. In other words, in an IP address, the number to the left of the first period is the largest grouping possible within Internet. It defines the network being used. The rightmost number in an IP address identifies one actual computer. Domain names are just the opposite. The largest grouping within the name is the rightmost part of the name, and the specific computer's name is the leftmost part.
Here are some examples:
Domain Name |
What It Means |
|
cs.wisc.edu |
This is a computer in the computer science (cs) department at the University of Wisconsin (wisc), which is an educational (edu) institution. |
|
xcf.berkeley.edu |
The University of California at Berkeley (berkeley) has a computer named xcf somewhere in its hallowed halls. Again, UC Berkeley is an educational (edu) institution. |
|
spacelink.msfc.nasa.gov |
The federal government has its own domain name (gov). This is a computer named spacelink at Marshall Space Flight Center (msfc), which is part of NASA (nasa). |
|
prep.ai.mit.edu |
A computer named prep is probably in the Artificial Intelligence (ai) lab at the Massachusetts Institute of Technology (mit), an educational (edu) institution. |
These domain names are certainly more readily remembered than the IP addresses they represent. Here are several well-recognized, high-level domains in use within the Internet:
Domain Name |
Meaning |
|
com |
Commercial domains used by corporations or companies that have Internet access. The commercial domain also is used by Internet providers such as teleport.com. |
|
edu |
This domain name is used primarily in the United States to identify educational sites. |
|
gov |
The U.S. government. |
|
mil |
U.S. military sites use the mil domain name. |
|
net |
Some networks choose to use this domain name in identifying themselves. For example, some state-run networks use the net domain name because, although educational sites may be tied into the network, the network serves state and local offices as well. |
|
org |
This name is used by organizations. The Internet Society, for example, uses the domain name isoc.org. |
You will discover other high-level domain names in use around the world. For example, Internet traffic from Australia usually uses .au as the final element of the domain name; .nz means the computer is located in New Zealand; and .de is used by Germany.
All these domain names are part of what is called the Domain Name System. The Domain Name System uses a number of computers scattered throughout the Internet to keep track of which computers are located within its geographical area. When you send e-mail to teleport.com, for example, your host computer sends a request to the nameserver to check its database for the IP number for teleport.com. When the IP number is received back by your host computer, it matches that IP number with your message and sends it out on the Internet.
Since the inception of the Domain Name System in 1983 at the University of Wisconsin, the Internet has been easier to use.
The Internet Network Information Center (NIC) manages the naming of Internet nodes. Anyone who wants a domain name must apply for the name.
When you read a Usenet message, you see a line in the message header named Path. Following Path is usually a long list of domain names separated by exclamation points (referred to on the Internet as bangs). This list shows the route that particular message followed to arrive on your computer. The path for one message, for example, was listed as follows:
teleport.com!news.world.net!news.sprintlink.net!demon!uknet!piex!howland. reston.ans.net!math.ohiostate.edu!cyber2.cyberstor.ca!nntp.cs.ubc.ca!unixg.ubc.ca
This pathname lists the most recent transaction first and the first transaction last. In this example the teleport.com computer got the message from a computer on world.net. The message passed through part of the Sprint network, through the United Kingdom's Internet network (uknet), through Ohio State, and so on. The last domain name on the list is the computer from which the message originated. In this case, the University of British Columbia in Canada (ubc.ca). Whew!
The use of the bang (!) was started in the UNIX community with the advent of UUCP. Recall that UUCP (UNIX-to-UNIX Copy) was one of the original network packages that enabled UNIX-based computers to call one another and transmit files and commands back and forth. When UUCP was created, the developers needed a way to identify other UNIX sites by name. They developed a domain name system very similar to the one used today on the Internet. The primary difference between the Internet domain name system and the UUCP system is that an exclamation mark is used to separate the elements of the UUCP address instead of a dot.
Nearly all UNIX sites use the Internet Domain Name System conventions today. At some sites, however, a local network may still use UUCP addressing for internal routing and the Domain Name System for external communications. If you do see a UUCP address, it may look something like this:
zeus!charlieb@teleport.com
Here, the computer that communicates with Internet is named teleport.com. However, teleport.com is connected to a UUCP local area network. One of the computers on the local area network is named zeus, and the user named charlieb has an account on zeus. Therefore, mail comes from the Internet into teleport.com, which sends the file on to a computer called zeus, and to the user named charlieb.
As time goes by, UUCP addresses may disappear entirely.
One other naming system you may run across uses a percent sign (%) as a secondary naming mechanism. In a UUCP address, a percent sign may be used as follows:
charlieb%zeus@teleport.com
Here, the teleport.com computer is responsible for communicating with zeus to make sure that charlieb gets the message.
Another early, major player in the networking world was Bitnet (the Because It's Time Network), formed by City University of New York. Like every other network in the world, Bitnet needed to create a naming system to help users get around their system. Although it is no longer used much in North America, Bitnet is still used in various countries around the globe.
The Bitnet creators limited both user and address names to eight uppercase characters. A Bitnet address looks like this:
NAMEMIKE@BITNIC
Notice that the at (@) symbol is used to separate the user name from the machine name, just as it does with Internet domain names.
If you need to send mail to someone on Bitnet, and you are an Internet user, append the characters .bitnet to the end of the Bitnet address, as follows:
NAMEMIKE@BITNIC.bitnet
The mail will get through.
As the World Wide Web developed on the Internet, it adopted a standardized way of specifying a particular site, or file, on the Web. Currently, the addressing system uses URLs (Uniform Resource Locators). These URLs combine information about the type of protocol being used, the address of the site where the resource is located, the subdirectory location, and (where used) the name of the file.
Here is a typical World Wide Web URL:
HTTP://www.ncsa.uiuc.edu/SDG/Software/Mosaic/MetaIndex.html
This URL is composed of these components:
Occasionally, a URL contains a "port number." When this happens, the port number is placed directly after the site address in this manner:
HTTP://www.ncsa.uiuc.edu:80/SDG/Software/Mosaic/MetaIndex.html
URLs provide a consistent, easily understood method of listing many different kinds of Internet sites—in addition to World Wide Web sites.
The popular menu-oriented Gopher system usually has addresses such as this:
gopher.std.com/pub
In the URL format, Gopher addresses look like this:
gopher://gopher.std.com/pub
The remote computer login system (Telnet) can use a URL such as the following:
telnet://std.com
The location, directory path, and name of a specific file at an FTP file storage site can be specified in a URL such as the following:
ftp://ftp.std.com/pub/oakridge/unleashed.txt
Usenet newsgroups are listed in a different format from other URLs. The newsgroup alt.business is listed in a URL like the following:
news:alt.business
The value of the URL goes far beyond its use as a standardized way of telling humans where to find things on the Internet. The real power of URLs comes when they are used with Web browsers and other programs that provide interfaces between you and the Internet. URLs provide the full set of information to the program—not only about where the resource is located and what its name is, but also what protocol should be used to retrieve it. The URL, therefore, provides a method to shield the Internet user from much of the complexity of the Internet.
