INTERNET ADDRESSING

Addressing is the practice of assigning a unique identifier (designation) to a particular person, place or machine such that those entities can be mentioned or indicated in commands. Many techniques have been developed throughout history for identifying entities. People are addressed by their names; first, middle and last. However, many institutions identify people by a number such as a social security number or an account number. Machines on the Internet can also be addressed by name or number. Computer names, often called host names, are chosen by the administrator of the network of which they are a part. Machine numbers are called IP numbers or host numbers and are coordinated by a few international agencies to prevent duplication. Network users are identified by a username (or userid). The rules governing userids are defined by each individual network. On many networks, userids may only contain letters or numerals. Some allow a few special characters such as ".", "-", or " " (blank space). Typically, a userid is made up from parts of your name. If your name is "William Johnson", your userid might be chosen as: "JohnWill", "WJohnson", or "William.Johnson". Many networks limit the length of userids, so you may have to abbreviate. Other networks let you use any id you like, such as "Golfer" or "Boss".

Addresses must be unique; otherwise, they will not serve their purpose. No two network users are given the same username on a single network. However two users of different networks might have the same name. In this case, you would have to identify each user by an address that was extended to include the name of their host or network. If two Internet hosts, one named "Comp1" and the other "Comp2", both had users with the username of "WJohnson", then they would be addressed as "WJohnson@Comp1" and "WJohnson@Comp2" respectively to keep them separate. Note the use of the @ symbol as a separator. Users on Comp1 could address WJohnson on their own network without including the Comp1 machine name, but users on Comp2 would have to distinguish the WJohnson identifier on Comp1 from the WJohnson on Comp2 by including the Comp1 machine name in the address.

The Internet is comprised of thousands of individual networks and groups of networks. Identifiers have been created to describe the domain (range or scope) of any network or sub-network. For example, the domain name "fl.us" is used to describe a group of networks in Florida in the United States. The domain name "cc.fl.us" refers to the sub-domain of community colleges in Florida. The group of Internet computers at IRCC (now IRSC) were once identified by the domain name "ircc.cc.fl.us". Presently, two naming conventions exist, the organizational approach and the geographical approach.

Organizational Domain Names

The Organizational Domain Name approach identifies network zones by organizational type, using the following three letter identifiers:

com		Commercial
edu		Educational
gov		Governmental
int		International (mostly NATO)
mil		Military
net		Network Infrastructure
org		Not-for-Profit Organization

Some new high-level domain names (such as biz, name, and web) have recently been added to accommodate the massive demand for machine identifiers and more will appear in the future.

Sub-domains identify such sub-groups as: institutions, divisions, departments, etc. The host name "mail.ircc.cc.fl.us" contains three sub-domains, "ircc", "cc" and "fl".

Geographical Domain Names

The Geographical Domain Name approach identifies network zones by geographical location, using two letter country identifiers. A few examples of Country Codes are:

au		Australia
ca		Canada (note: as a sub-domain, ca stands for California)
de		Germany
fr		French Republic
uk		United Kingdom (or GB for Great Britain)
us		United States

Sub-domains in our country use two letter identifiers for states and three letter identifiers for cities. A few of these identifiers are:

ca		California (note: as a zone, ca stands for Canada)
fl		Florida
ga		Georgia
tlh		Tallahassee

Other locations might take a hybrid approach, sub-dividing their domains into organizational categories such as "ac" for academic or "com" for commercial.

Although it is not typical, a network can have two names, one using each approach. For example, the IRSC network was (at one time) identified as either:

Geographical Domain Name		ircc.cc.fl.us
Organizational Domain Name		ircc.edu

To identify a single unique machine on the Internet, you must use its Fully Qualified Domain Name (FQDN) which includes its host name and its specific domain. The computer at IRSC that receives Internet E-mail has the machine (host) name of "mail". But many other networks also use that name for their mail servers. So to distinguish it from all of the others, you must use the Fully Qualified Domain Name of "mail.irsc.edu", which simply combines its machine name with its precise domain name. Some networks do not publish the names of their individual machines nor their sub-domains. So you may see a domain name that contains only a network name on the left and a zone on the right. The name of the America Online E-mail Host is given simply as aol.com. Their Internet gateway routes all e-mail to the appropriate machine.

Notice that the separate parts of the host names are delimited by periods ("dots") and that they are ordered from left to right going from a narrow to a broad scope. The rightmost part of a host name is called the zone or the high level domain. The middle part(s) describe sub-domains. The leftmost part is the machine name (as defined earlier). This order is fairly standard world-wide, except inside the United Kingdom where they do it in reverse (no joke).

The college has many computers connected to the Internet. All of them have been given an IP number. Only a few of them have been given machine names. Internet machines do not require machine names, but they must each have an IP number. The only machines that are assigned names are the ones that people will be addressing frequently. For example, our public web server has been assigned the machine name of:

www.irsc.edu

Host Numbers

The details of host numbers are beyond the scope of this document. But there are a few basics that you should know. Every machine with a native connection to the Internet has a host number. Machines on the Internet do not address each other by machine names; they use host numbers. Many machines do not have machine names and therefore must be address by number. Most users prefer to use machine names because they are easier to remember. Special Internet hosts called Domain Name Servers (DNS) maintain database listings of the machine names of frequently used machines so that users can address them by name instead of by number. Unfortunately, these databases are not always accurate or available. So you should write down the IP number for any computer that you might have to reach someday, just in case the nearest DNS can't find its machine name that day.

A typical host number looks like: 150.176.240.31

Host numbers are assigned with four parts separated by periods. When saying one of these numbers aloud, you refer to the separating periods as dots. The four-part format used in host numbers is called a dotted quad. The leftmost parts define the domain and the rightmost parts indicate sub-networks or individual machines. Some computers are assigned more than one host number because they are attached to more than one network and as such they have a separate address on each network. Normally it doesn't matter which one you use unless one of the networks is down (off-line) or has an Acceptable Use Policy that conflicts with your type of use.

The only IP number that a user must know, is the IP number of their network's Domain Name Server (DNS). Since the DNS is required to lookup the names of all machines, your computer must know its number to use it. The old network IRCCNet had two DNS's. Their IP numbers were:

Primary DNS:		209.149.16.2
Secondary DNS:		209.149.16.3

Remember the dynamic nature of the Internet. IP numbers can change as network designs are modified. Some Internet Tools (programs) such as Finger, Whois and Netfind can help users to determine host numbers and can be found on many Internet file servers.

Using the Ping Command to Determine a Host Number

One of the simplest tools for determining a host number is the command Ping. This command can be found on most hosts that use the UNIX or Windows operating systems. Ping clients can also be obtained from many online file libraries. The Ping command generates and sends out test packets of data to a remote host that automatically returns the packets to the originating computer. The primary purpose of Ping is to test the link between two machines and to report on the round trip time of the transmissions. If Ping is executed using a host name, the program will interrogate your network's DNS and report the host number on the screen. The only trick in using Ping is in getting it to stop the link testing process. On some machines, this is done by holding down the Control [Ctrl] key and tapping the letter [c]. Some Ping programs can be set to transmit a set quantity of test packets and then stop.

The Windows Ping client is usually stored in the Windows folder on Drive C. It is a DOS-based program, so you must run it from a DOS Command Prompt. To do this, click on [Start], [Programs], [Accessories], [Command (MS-DOS) Prompt] to open a DOS session window and then enter a Ping command such as:

ping www.abc.com

That command should return a reply that looks like:

     Pinging abc.com [199.181.135.201] with 32 bytes of data:
     Reply from 199.181.135.201: bytes=32 time=135ms TTL=251
     Reply from 199.181.135.201: bytes=32 time=133ms TTL=251
     Reply from 199.181.135.201: bytes=32 time=134ms TTL=251
     Reply from 199.181.135.201: bytes=32 time=133ms TTL=251

Internet Security:

Security is an important issue to Internet users.  Two major questions are often asked.

1. "Can other people read my messages sent over the Internet?"  Technically - Yes,  Practically - No.
   In other words, it is possible, but its is also highly unlikely. Just the same way that anyone who is sufficiently determined could also tap your phone or intercept your printed mail, someone could also intercept your e-mail.  But it is not likely.
2. "Can other people steal things from my computer when I am connected to the Internet?" The answer is much the same as above, Yes and No.  It is highly unlikely that anyone would bother to figure out a way to steal from your PC.  Large corporations or major institutions, like the college, protect their networks from invasion by outside (or inside) perpetrators through the use of special software called firewall software.  IRCC's firewall software monitors the packets of data transmitted between the college network and the Internet and will block any packets that seem suspicious.  As you might suspect, no firewall is perfect.  Any one can fail to detect dangerous packets and sometime block innocent packets.  You will probably find that the IRCC firewall is designed to error on the side of caution. In other words, it is purposely conservative as a guardian and may occasionally block legitimate data.  But that is the price we pay for high security.

For more information about Internet addressing, look at the web page:

      http://www.gibsonr.com/classes/internet/toolbox.html#Access

Note that the last word ("Access") in the command above must be capitalized exactly as it appears.

    Return to the Internet Fundamentals Page